Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Telomere-binding and Stn1p-interacting activities are required for the essential function of Saccharomyces cerevisiae Cdc13p

Telomere-binding and Stn1p-interacting activities are required for the essential function of... © 2000 Oxford University Press Nucleic Acids Research, 2000, Vol. 28, No. 23 4733–4741 Telomere-binding and Stn1p-interacting activities are required for the essential function of Saccharomyces cerevisiae Cdc13p Mei-Jung Wang, Yi-Chien Lin, Te-Ling Pang, Jui-Mei Lee, Chia-Ching Chou and Jing-Jer Lin* Institute of Biopharmaceutical Science, National Yang-Ming University, Shih-Pai, 112, Taipei, Taiwan, Republic of China Received July 31, 2000; Revised and Accepted October 20, 2000 ABSTRACT are composed of short tandem repeated sequences with the strand running in a 5′ to 3′ direction toward the guanine (G)-rich ends Yeast Saccharomyces cerevisiae Cdc13p is the (1). The sequence and the number of repeats vary considerably telomere-binding protein that protects telomeres and in different species. For example, the telomere sequences of yeast regulates telomere length. It is documented that Saccharomyces cerevisiae are ∼250–300 bp of TG /C A, 1–3 1–3 Cdc13p binds specifically to single-stranded TG 1–3 whereas those of human are ∼10 kb TTAGGG/CCCTAA repeats telomeric DNA sequences and interacts with Stn1p. (1,2). Besides the double-stranded telomeric DNA repeats, telo- To localize the region for single-stranded TG DNA meres in all of the organisms that have been analyzed, also contain 1–3 binding, Cdc13p mutants were constructed by deletion a G-rich single-stranded tail (5–7). In S.cerevisiae,a >30 base single-stranded TG tail was detected in late S phase of the cell mutagenesis and assayed for their binding activity. 1–3 cycle (8). This single-stranded tail was postulated as an inter- Based on in vitro electrophoretic mobility shift mediate during telomere replication (8,9). assay, a 243-amino-acid fragment of Cdc13p (amino Protein factors that interact with telomeres participate in acids 451–693) was sufficient to bind single-stranded telomere functions (10). Factors that bind to the double-stranded TG with specificity similar to that of the native 1–3 telomeric DNA sequences have been identified from many protein. Consistent with the in vitro observation, in vivo organisms (11–16). These proteins are essential for the mainte- one-hybrid analysis also indicated that this region of nance of telomere functions and cell viability (11,12,16). For Cdc13p was sufficient to localize itself to telomeres. example, in S.cerevisiae, Rap1p has been shown to bind However, the telomere-binding region of Cdc13p (amino double-stranded telomeric DNA (11,17,18). Mutations on acids 451–693) was not capable of complementing the Rap1p affect the length of telomere, TPE, localization of the growth defects of cdc13 mutants. Instead, a region telomere within the nucleus, telomere recombination and cell viability (19–21). Moreover, protein factors bound to the comprising the Stn1p-interacting and telomere-binding single-stranded telomeric DNA were identified in several region of Cdc13p (amino acids 252–924) comple- organisms (22–30). Among these protein factors, Oxytricha mented the growth defects of cdc13 mutants. These telomere-binding protein is well characterized. The protein is results suggest that binding to telomeres by Cdc13p heterodimeric and is composed of an α subunit and a β subunit is not sufficient to account for the cell viability, inter- (31–33). The α subunit is a single-stranded DNA-binding action with Stn1p is also required. Taken together, protein that binds to the G T single-stranded end of the 4 4 we have defined the telomere-binding domain of telomere. Although the β subunit is not directly involved in Cdc13p and showed that both binding to telomeres binding, it is required for terminus-specific binding. In the and Stn1p by Cdc13p are required to maintain cell yeast genome, there is no homologue of the Oxytricha α-and growth. β-like binding proteins. Instead, Cdc13p binds to single- stranded TG sequences in vitro and interacts with telomeres 1–3 in vivo (24,25,34). It is believed that, in yeast, Cdc13p is the INTRODUCTION functional equivalent of Oxytricha α and β binding proteins Telomeres are the specialized structure at the very ends of despite there being no sequence similarity between Oxytricha eukaryotic chromosomes. Telomeres are essential for the telomere-binding proteins and yeast Cdc13p (J.-J.Lin, unpub- maintenance of chromosome integrity (1,2). They prevent end- lished observation). Furthermore, Cdc13p dissociates from to-end fusion of chromosomes, protect chromosome from single-stranded TG DNA at 300 mM NaCl (J.J.Lin and 1–3 degradation by nucleases and facilitate the complete replication of V.A.Zakian, unpublished result) whereas the binding of chromosomes. They also suppress the expression of nearby Oxytricha proteins to G T DNA remains associated even at 4 4 genes, a phenomenon known as telomere position effect (TPE) 2 M NaCl (33). Thus, Cdc13p might represent a class of (3). Telomeres might also position chromosomes within the single-stranded telomere-binding protein that is different from nucleus (4). In most of the cloned telomeric sequences, telomeres the Oxytricha protein. *To whom correspondence should be addressed. Tel: +86 2 2826 7258; Fax: +86 2 2820 0067; Email: jjlin@ym.edu.tw 4734 Nucleic Acids Research, 2000, Vol. 28, No. 23 CDC13 is an essential gene (35). It is involved in cell cycle (GST) gene of pGEX plasmids (Pharmacia). The 4.4-kb SmaI– control, because a temperature-sensitive allele of CDC13, HpaI fragment containing full length CDC13 from pTHA-CDC13 cdc13-1, causes the cell cycle to arrest in G /M phase at the (25) was ligated to the SmaI-digested pGEX-3X to generate non-permissive temperature (35). This cell cycle control is pGST-CDC13. Plasmid pGST-CDC13 was digested with BglII dependent on the gene product of RAD9, which was postulated and self-ligated to generate pGST-CDC13(1–600), which deleted to be involved in surveying the integrity of chromosomes (36). the BglII–BglII fragment within CDC13.Plasmid pGST- Cdc13p might enable Rad9p to differentiate whether the ends CDC13(451–924) was constructed by ligating the 2.9-kb BamHI– of a linear DNA are telomeres or broken ends by binding to HpaI fragment containing half of CDC13 to BamHI- and SmaI- telomeres. It is also demonstrated that cdc13-1 cells at the non- digested pGEX-3X. Plasmid pGST-CDC13(451–924) was permissive temperature accumulate single-stranded G-rich digested with either PvuII or BglII, and self-ligated to generate DNA near telomeres (35). Nevertheless, it is unclear whether pGST-CDC13(451–871) and pGST-CDC13(451–600), respec- this single-stranded G-rich DNA is recognized as a damage to tively. Plasmid pGST-CDC13(451–693) was constructed by trigger the checkpoint mechanism. Since a mutation allele of digesting pGST-CDC13(451–924) with SalIand NruI, treating est CDC13, cdc13 , causes gradual loss of telomere, Cdc13p also with T4 DNA polymerase and self-ligation. A 1.4-kb EcoRI–SalI appears to regulate the length of telomeres (24). At least a fragment of pTHA-CDC13 was ligated to EcoRI- and SalI- portion of the telomere function of Cdc13p is contributed by its digested pGEX-4T-1 to generate pGST-CDC13(510–924). To interaction with Stn1p. STN1 was identified from a high copy construct pGST-CDC13(601–856), a 0.8-kb BglII–BglII fragment suppressor screening to rescue the temperature-sensitive from pTHA-CDC13 was inserted into BamHI-digested pGEX-1. The NruI–SmaI fragment of pGST-CDC13(601–856) was lethality of cdc13-1 (37) and direct interaction of Cdc13p with deleted to generate pGST-CDC13(601–693). Expression of Stn1p was demonstrated by two-hybrid assays (37). Moreover, these fusion proteins was induced by the addition of isopropyl telomere defects caused by STN1 mutation are similar, if not β-D-thiogalactoside (IPTG) and confirmed by western blotting identical, to CDC13 mutation, suggesting that these two genes analysis using antibody against GST (data not shown). act together in maintaining telomere function (37). Plasmid pET6H-CDC13(451–693), which was used to CDC13 encodes a 924-amino-acid protein with molecular purify Cdc13(451–693)p, was constructed by inserting the mass of 104 895 kDa (35). Sequence comparison between NcoI–NruI fragment of pTHA-NLS-CDC(451–693) into NcoI/ Cdc13p and known DNA- or RNA-binding proteins did not SmaI-digested pET6H (a gift from C.-H. Hu., National Marine provide any information on the region within Cdc13p responsible University, Taiwan). The resulting plasmid was used to for interacting with telomeres (J.-J.Lin, unpublished observation). express 6× His-tagged Cdc13(451–693)p under the control of Thus, Cdc13p might contain a novel DNA-interacting motif the T7 promoter. for binding to telomeres. In order to identify the functional To construct a plasmid for one-hybrid analysis, the DNA domains of Cdc13p, we have generated deletion mutants and fragment encoding Cdc13(451–693)p was amplified by PCR examined the single-stranded TG -binding activity and their 1–3 using pTHA-CDC13 as the template and with primers ML11 ability to interact with Stn1p. Here we showed that (5′-CCGCTCGAGATCCTGTGGGACAATGAC-3′) and ML12 Cdc13(451–693)p, a Cdc13p fragment ranging from amino (5′-CCGCTCGAGATGAGAACCGTTTCT-3′). The 0.7-kb acids 451 to 693, retain the single-stranded TG -binding 1–3 PCR product was subcloned into SmaI-digested pUC18 to activity. We also showed that binding to telomeres by Cdc13p generate pUC-CDC13(451–693). The 0.7-kb DNA fragment is not sufficient to maintain cell viability. Moreover, amino from XhoI-digested pUC-CDC13(451–693) was then ligated acids 252–924, which covered the telomere-binding and with the XhoI-digested pJG4-5 (39) or pRF4-6NL (34) to Stn1p-interacting activities of Cdc13p, are sufficient to generate pJG-CDC13(451–693) or pRF-CDC13(451–693), account for the essential function of Cdc13p. respectively. Since Cdc13p is a telomere-binding protein, the cellular MATERIALS AND METHODS localization of Cdc13p should be in the nucleus. To make cer- tain that the truncated Cdc13p would be delivered into the Strains nucleus, two oligonucleotides NLSw (5′-CTAGCCCCAA- Escherichia coli strain DH5α was used as host for plasmid GAAGAAGCGGAAGGTCGC-3′)and NLSc (5′-CATGGC- construction and propagation, and strain BL21(DE3)pLysS for GACCTTCCGCTTCTTCTTTGGGG-3′) encoding the nuclear Cdc13(451–693)p purification. Yeast strain YPH499 [MATa localization sequence (NLS) of SV40 large T antigen (39,40) amber ochre ura3-52 lys2-801 ade2-101 trp1-∆ 63 his3-∆ 200 leu2-∆ 1 were annealed and ligated with SpeI- and NcoI-digested (38)] with URA3 and ADE2 placed near the telomere of pTHA-CDC13. The resulting plasmid, pTHA-NLS-CDC13, chromosomes VII-L and V-R (YPH499UTAT), respectively, encoded a fusion protein with three HA-tag and NLS at the N- was used as the host for analyzing TPE. Yeast strains, HIS-Tel terminal of Cdc13p. Similarly, plasmid vector pTHA-NLS was and HIS-Int-CA, used for one-hybrid analysis were kindly constructed by inserting NLS sequences into pTHA (25). To provided by V. A. Zakian [Princeton University (34)]. Strain construct plasmid pTHA-NLS-CDC13(451–924), a 1.6-kb 2758-8-4b (MATa cdc13-1 his7 leu2-3, 112 ura3-52 trp1-289, DNA fragment encoding amino acids 451–924 of Cdc13p was provided by L. Hartwell, University of Washington) was used PCR amplified with primers ML01 (5′-TGCCATGGGGATC- as the host for complementation tests. CTGTGGGACAAT-3′)and CDC133 (5′-AACTGCAGACT- AGTCGACTCTTGCTTCTTACC-3′) using Vent DNA poly- Plasmid construction merase (New England Biolabs). The PCR product was digested Deletion mutants of Cdc13p (Fig. 1A) were generated by fusion with NcoIand SalI, and was inserted into NcoI- and SalI-digested of various regions of CDC13 with the glutathione S-transferase pTHA-NLS. To generate pTHA-NLS-CDC13(451–693), plasmid Nucleic Acids Research, 2000, Vol. 28, No. 23 4735 pTHA-NLS-CDC13(451–924) was digested with NruIand buffer (100 mM Tris pH 8.0, 2 mM EDTA, 2 mM DTT, 0.4 M SalI, blunted by T4 DNA polymerase and self-ligated. Plasmid L-arginine, 20% glycerol) at 4°C for 12 h. pTHA-NLS-CDC13(1–252) was constructed by ligating the Electrophoretic mobility shift assay (EMSA) NcoI–EcoRI fragment (0.8 kb) encoding the N-terminal 252- amino-acid polypeptide, of pAS-CDC13(1–252) to NcoI- and Oligonucleotide TG22 (5′-GTGGTGGGTGGGTGTGTGTGGG- EcoRI-digested pTHA-NLS. A 2-kb NcoI–SalI fragment of 3′), TG10 (5′-GGGTGTGGTG-3′), TG15 (5′-TGTGTGGGTG- pAS-CDC13(252–924) was ligated with NcoI- and SalI-digested TGGTG-3′), TG20 (5′-TGGTGTGTGTGGGTGTGGTG-3′), pTHA-NLS to generate pTHA-NLS-CDC13(252–924). TG25 (5′-GGGTGTGGTGTGTGTGGGTGTGGTG-3′), TG30 Expression of full-length and truncated CDC13 was confirmed (5′-TGTGTGGGTGTGGTGTGTGTGGGTGTGGTG-3′)or TG35 (5′-TGTGGTGTGTGGGTGTGGTGTGTGTGGGTG- by western blotting analysis on the extracts prepared from TGGTG-3′) was first 5′ end-labeled with [γ- P]ATP yeast cells with anti-HA antibody 12CA5 (data not shown). (3000 mCi/mM, NEN) using T4 polynucleotide kinase (New Two-hybrid plasmids were constructed by subcloning STN1 England Biolabs) and subsequently purified from a 10% into pACT2 and several fragments of CDC13 into pAS2 sequencing gel after electrophoresis. To perform the assays, (37,41). Plasmid pACT-STN1 was constructed by ligating a cell extracts were mixed with 2.0 ng of P-labeled TG22 or NcoI–BamHI 1.6-kb fragment containing full-length STN1 TG15 DNA with a total volume of 15 µ l containing 50 mM from pJG-STN1 (unpublished construct) into pACT2 with the Tris–HCl pH 7.5, 1 mM EDTA, 50 mM NaCl, 1 mM DTT and same sites. To construct full-length CDC13 into pAS2, the 1 µ g of single stranded poly(dI–dC). The mixtures were allowed to NcoI–SalI fragment (3 kb) of pTHA-CDC13 (25) was incubate at room temperature for 10 min. Then, 3 µ l of 80% inserted into the NcoI- and SalI-digested pAS2. Plasmid glycerol was added and the mixtures were loaded on an 8% non- carrying the temperature-sensitive allele of CDC13, cdc13-1,was denaturing polyacrylamide gel, which was pre-run at 125 V for constructed by replacing the 1.4-kb fragment of pAS-CDC13 10 min. Electrophoresis was carried out in TBE (89 mM Tris– with the equivalent of pTHA-CDC13-1 (25). Plasmid pAS- borate/2 mM EDTA) at 125 V for 105 min. The gels were dried and CDC13(252–924) was constructed by first ligating the 1.4-kb autoradiographed. For competition analysis, 2.0 ng P-labeled EcoRI–SalI CDC13 fragment of pTHA-CDC13 into EcoRI- TG15 was mixed with varying amounts of non-radioactive and SalI-digested pAS2-1 followed by inserting a 0.8-kb competitors before addition of the cell extracts. Binding activity EcoRI–EcoRI fragment of CDC13 isolated from pTHA- was quantified with a PhosphorImager (Molecular Dynamics). CDC13 with EcoRI digestion. One-hybrid analysis Preparation of E.coli extracts One-hybrid analysis was performed using the methods described To prepare protein extracts from E.coli, a 10-ml culture of by Bourns et al. (34). Plasmid pJG4-5, pJG-CDC13(451–693) or E.coli containing a Cdc13p-deletion construct was grown in pRF-CDC13(451–693) was first transformed into yeast strains LB medium with 50 µ g/ml ampicillin at 30°CtoOD =0.6. HIS-Tel and HIS-Int-CA, and selected on plates lacking At this point, IPTG was added to a final concentration of 1 mM tryptophan. To test for telomere-binding activity in vivo, cells and the cells were allowed to grow at 30°C for an additional 16 were grown in liquid medium lacking tryptophan for ∼16 h. h before harvesting by centrifugation. The pellets were resus- Then, cells were spotted in 10-fold serial dilutions on yeast pended in 0.5 ml buffer A [50 mM Tris–HCl pH 7.5, 1 mM synthetic complete medium (YC) plates lacking tryptophan, or EDTA, 1× protease inhibitor cocktail (Calbiochem) and 50 lacking histidine, or with 10, 20 or 40 mM of 3-amino-1,2,4- mM glucose] and were sonicated. Cell debris was removed by triazole (3-AT, Sigma, St Louis, MO). Plates were incubated at centrifugation in an Eppendorf microfuge for 10 min at 4°C, 30°C until the colonies could be observed. Because HIS-Tel the supernatants were aliquoted and frozen in a dry ice-ethanol cells carrying the B42 transcription-activation domain fused bath, and stored at –70°C. The concentration of protein in the with the DNA-binding region of Cdc13p did not grow well on E.coli extract was ∼2–6 mg/ml as determined using a Bio-Rad plates with galactose, all the experiments were performed in protein assay kit with bovine serum albumin as a standard. plates containing 1.5% galactose and 1% glucose. The HIS- Int-CA strain grew better in this condition than in plates Purification of 6× His-tagged Cdc13(451–693)p containing 3% galactose. To purify 6× His-tagged Cdc13(451–693)p, a 500-ml culture Complementation of cdc13∆ by cdc13 fragments using of IPTG-induced E.coli harboring pET6H-CDC13(451–693) plasmid loss experiments was collected by centrifugation. Cells were resuspended in 30 ml of GdHCl buffer (6 M guanidine–HCl, 0.1 M NaH PO , Plasmid loss experiments were carried out to test if binding of 2 4 0.01 M Tris pH 8.0), followed by gentle shaking for 1 h at 4°C. single-stranded TG and/or Stn1p-interacting activity of 1–3 The suspension was then centrifuged at 13 000 g for 15 min at Cdc13p is sufficient to complement the lack of viability caused 4°C. The clear supernatant was collected and applied to a 5 ml by the cdc13∆ mutation. Briefly, plasmid YEP24-CDC13 Ni-NTA–agarose column (Qiagen) previously equilibrated [abbreviation of plasmid YEP24-CDC13-161-4 (35)] was with GdHCl buffer. The column was washed stepwise with transformedintoadiploidstrainYJL401-UTAT [CDC13/ 20 ml of GdHCl buffer containing 1 mM imidazole followed cdc13∆ ::HIS3 (25)] carrying one allele of the cdc13 null by 20 ml of GdHCl buffer containing 20 mM imidazole. The mutation. The transformants were sporulated and subjected to bound protein was eluted by 12 ml of GdHCl buffer containing tetrad analysis. Haploid strain YJL501 was selected from the + + 200 mM imidazole. To renature the purified Cdc13(451–693)p, spores that were Ura (YEP24-CDC13), His (cdc13∆ ::HIS3)and protein eluted from the Ni-NTA–agarose column was diluted to Ade (ADE2 near the telomere of chromosome V-R). Such a strain ∼50 µ g/ml using GdHCl buffer and dialyzed against renaturation requires a plasmid carrying CDC13 (YEP24-CDC13) for its 4736 Nucleic Acids Research, 2000, Vol. 28, No. 23 viability. Subsequently, plasmid pTHA-NLS, pTHA-NLS-CDC13, pTHA-NLS-CDC13(1–252), pTHA-NLS-CDC13(252–924), pTHA-NLS-CDC13(451–924) or pTHA-NLS-CDC13(451–693) was separately transformed into YJL501. The resulting transform- ants were spotted onto plates containing 0.5 mg/ml 5-fluoroorotic acid (5-FOA) and incubated at 30°C until colonies formed. Two-hybrid analysis Plasmids pACT2 and pACT-STN1 were transformed separately into yeast strain Y190. The resulting strains were then transformed with plasmids pAS2 or pAS2 containing CDC13 fragments. The HIS3 reporter system was also used to evaluate the interaction between Cdc13p and Stn1p. In the assays, 5–10 fresh trans- formed colonies from each transformation were mixed and spotted in 10-fold serial dilutions onto YC plates lacking histidine without or with 25 mM 3-AT. Plates were kept at 25 or 30°C until colonies formed. RESULTS Cdc13(451–693)p bound to single-stranded TG DNA in vitro 1–3 To identify the telomeric DNA-interacting region in Cdc13p, plasmids suitable to express GST fusions with various fragments of Cdc13p were constructed (Fig. 1A). Using P-labeled 22-base TG oligonucleotides (TG22) as substrate, the DNA-binding 1–3 activity of these truncated Cdc13p was determined by EMSA. By comparing the gel-shift pattern with that of the vector alone, extra single-stranded TG -binding activity was 1–3 observed in E.coli extracts expressing Cdc13p fragments Figure 1. Single-stranded TG -binding domain of Cdc13p. (A) Schematic 1–3 representation of Cdc13p deletion mutants. The wild-type protein is 924 containing amino acids 451–924, 451–871 or 451–693 (Fig. 1B). amino acids in length. The relative locations of the fragments (thick line) and Among these fusion proteins, Cdc13(451–693)p was the the first as well as the last amino acids are indicated. All these mutants were shortest single-stranded TG DNA-binding fragment of fused in-frame to a GST protein on their N terminus. On the right are the 1–3 single-stranded TG22-binding activities of mutants that were analyzed by Cdc13p. Western blotting analysis using anti-GST antibody EMSA. (B) Cdc13(451–693)p contains the single-stranded TG -binding 1–3 confirmed that these truncated Cdc13p polypeptides were domain of Cdc13p. Escherichia coli extracts (15 µ g) were mixed with P- indeed expressed, albeit degradation of the full-length and labeledTG22in15-µ l reaction mixtures. The reactions were incubated at room temperature for 10 min. A 3-µ l volume of 80% glycerol was added to each some of these truncated forms of Cdc13p was also observed reaction before analysis of the reaction products on an 8% polyacrylamide gel. (data not shown). Therefore, while it is not clear if the fragment Extracts carrying the GST fusion of Cdc13p deletion mutants are indicated. The can be shortened further, it was evident that the single-stranded first lane has no extracts. Migration of free DNA (TG22), Cdc13(451–693)p, TG -binding activity of Cdc13p is located within the fragment Cdc13(451–871)p, Cdc13(451–924)p and Cdc13p are indicated. 1–3 comprising residues 451–693. We also have expressed a 6× His-tagged-Cdc13(451–693)p in E.coli and purified this tagged protein using Ni-NTA–agarose (Fig. 2A). Using P-labeled single-stranded TG oligonucleotides Cdc13(451–693)p specifically bound to single-stranded 1–3 TG DNA as substrate, the DNA-binding ability of this recombinant 1–3 polypeptide was determined by EMSA. The result shown in To evaluate the selectivity of the Cdc13(451–693)p binding Figure 2B further demonstrated that this 6× His-tagged 32 activity, purified protein was mixed with P-labeled single- Cdc13(451–693)p is capable of forming complexes with the stranded TG and various amounts of unlabeled nucleic acid 1–3 single-stranded TG . Evidently, the single-stranded TG 1–3 1–3 competitors before being subjected to EMSA analysis. As binding activity of Cdc13p was located within amino acids shown in Figure 3, unlabeled TG15 competed efficiently with 451–693. Results shown in Figure 2B also indicate that the P-labeled TG15. The binding was reduced by ∼50% when the length of DNA substrate affected the electrophoretic mobility competitor was presented at equal concentrations (Fig. 3A, of the Cdc13(451–693)p–DNA complex. Interestingly, a lanes 3–5 and B). On the other hand, vertebrate (T AG ), 2 3 second migration band was apparent on TG30 or TG35 but not Oxytricha (T G )and Tetrahymena (T G ) telomeric DNA did 4 4 2 4 on TG25; the identity of this second migration band is uncertain not compete for the binding activity of Cdc13(451–693)p to (Fig. 2B, lanes 19–24 and 25–30). Under our assay condition, TG15 (Fig. 3A, lanes 6–14). Total yeast RNA, single-stranded Cdc13(451–693)p bound TG15 with an apparent binding C A DNA or duplex TG /C A DNA did not compete for 1–3 1–3 1–3 constant of 120 nM. Cdc13(451–693)p binding either (Fig. 3A, lanes 15–17, and Nucleic Acids Research, 2000, Vol. 28, No. 23 4737 Figure 2. Purified Cdc13(451–693)p binds single-stranded TG telomeric 1–3 DNA. (A) Purification of Cdc13(451–693)p. A 6× His-tagged Cdc13(451–693)p Figure 3. Competition analysis of Cdc13(451–693)p telomeric DNA-binding was purified from E.coli using a Ni-NTA–agarose column (see Materials and activity. (A) P-labeledTG15(2ng) was mixed with several concentrations of Methods). A Coomassie blue-stained 10% SDS–polyacrylamide gel is given. different competitors and then 0.2 µ g of the purified Cdc13(451–693)p was Lane 1 shows the molecular weight marker; lanes 2 and 3 were 50 µ lof E.coli added to the mixtures. Competitors were TG22 (yeast), (T AG ) (vertebrate), 2 3 3 cultures harboring the pET6H-CDC13(451–693) plasmid grown without and (T G ) (Oxytricha), (T G ) (Tetrahymena) and total yeast RNA. Gel shift 4 4 3 2 4 3 with IPTG induction, respectively; lane 4 was 5 µ g of purified Cdc13(451–693)p. assay was then carried out. An autoradiogram is shown. (B) Quantification of (B) Cdc13(451–693)p contains the single-stranded TG -binding domain of 1–3 32 the Cdc13(451–693)p-binding activity. The relative level of the binding activity Cdc13p. Approximately 27 nM each of P-labeled TG10 (lanes 1–6), TG15 was quantified by a PhosphorImager and the binding activity in the absence of (lanes 7–12), TG20 (lanes 13–18), TG25 (lanes 19–24), TG30 (lanes 25–30) competitor was taken as 100 [(A) lane 2]. The data show the average from and TG35 (lanes 31–36) were mixed with several concentrations of the purified three experiments. Symbols used are: TG15 (yeast, closed circles), (T AG ) 2 3 3 Cdc13(451–693)p and then gel shift assay was carried out. Cdc13(451–693)p used (vertebrate, closed squares), (T G ) (Oxytricha, closed triangles), (T G ) 4 4 3 2 4 3 in each set of experiments were 0, 7, 22, 66, 200 and 600 nM. Autoradiograms are (Tetrahymena, open circles) and total yeast RNA (open squares), respectively. shown. data not shown). This result indicated that Cdc13(451–693)p spotted onto plates without histidine to evaluate the expression bound specifically to single-strand TG telomeric DNA. With 1–3 of HIS3. HIS-Tel cells carrying plasmid vector alone (Act) or the exception of vertebrate telomeric DNA, which partially Cdc13(451–693)p without the B42 transcription-activation competed away the binding of Cdc13p to TG22 (25), domain (Cdc13-DB) cannot grow on plates lacking histidine Cdc13(451–693)p bound specifically to single-stranded TG 1–3 (Fig. 4, left panel). However, cells carrying the B42 transcription- telomeric DNA similarly to Cdc13p. activation domain fused with the DNA-binding region of Cdc13p (Act-Cdc13-DB) grew on plates lacking histidine Cdc13(451–693)p bound telomere in vivo (Fig. 4, left panel). The levels of cell growth with 3-AT for the Previously, a one-hybrid system was developed to examine HIS-Int-CA strain carrying plasmid vector alone (Act), Cdc13-DB whether a protein interacts with telomeres in vivo (34). In that or Act-Cdc13-DB were similar, indicating that Cdc13(451–693)p system, a promoter-defective allele of HIS3 is placed near the would not bind internal TG /C A duplex DNA (Fig. 4, right 1–3 1–3 telomere of chromosome VII-L, HIS-Tel. The protein to be panel). Thus, Cdc13(451–693)p is sufficient to position itself tested is fused to the E.coli B42 transcription-activation to telomeres. Taken together, both in vitro and in vivo evidence domain. When this fusion protein interacts with telomeres, it indicate that the DNA-binding domain of Cdc13p was located activates the expression of HIS3. Thus, expression of His3p in amino acids 451–693. can be used as a means to identify telomere-interacting protein. To verify whether Cdc13(451–693)p binds telomere in vivo, Cdc13(451–693)p was not sufficient to complement cdc13 this fragment was fused with the B42 transcription-activation mutations domain (Act-Cdc13-DB) and expressed in yeast HIS-Tel or HIS-Int-CA. The HIS-Int-CA strain carried internal HIS3 with It has been known that Cdc13p is essential for cell viability. C A sequences at the 5′ region (34). Dilutions of cells were The next question that was considered in this study was whether 1–3 4738 Nucleic Acids Research, 2000, Vol. 28, No. 23 Figure 5. The telomere-binding domain of Cdc13p cannot rescue the growth Figure 4. Cdc13(451–693)p binds telomere in vivo. HIS-Tel (left panel) or defect phenotype of cdc13 mutants. (A) Yeast 2758-8-4b (cdc13-1) carrying HIS-Int-CA (right panel) cells harboring pJG4-5 (Act), pJG-CDC13(451–693) plasmid pTHA-NLS (vector), pTHA-NLS-CDC13 (1–924), pTHA-NLS- (Act-CDC13-DB) or pRF-CDC13(451–693) (CDC13-DB) were spotted in 10-fold CDC13(1–252), pTHA-NLS-CDC13(252–924), pTHA-NLS-CDC13(451–924) serial dilutions onto plates lacking tryptophan (YC-Trp), lacking histidine (YC-His), or pTHA-NLS-CDC13(451–693) were spotted in 10-fold serial dilutions on YC-His plates with 10 mM 3-AT and YC-His plates with 40 mM 3-AT. Photo- YC-Leu plates and grown at 25°C (left), 30°C (middle)or37°C (right) until graphs were taken after the plates were incubated at 30°Cfor 3days. colonies formed. (B) Yeast strain YJL501 (cdc13∆ ::HIS3/YEP24-CDC13) carrying plasmids pTHA-NLS (vector), pTHA-NLS-CDC13 (1–924), pTHA-NLS- CDC13(1–252), pTHA-NLS-CDC13(252–924), pTHA-NLS-CDC13(451–924) or pTHA-NLS-CDC13(451–693) were spotted in 10-fold serial dilutions on YC-Leu plates or plates containing 5-FOA and incubated at 30°C until colonies formed. binding of Cdc13p to telomere is sufficient to account for its Photographs of the plates are shown. essentiality. Here, plasmids expressing Cdc13p, Cdc13(1–252)p, Cdc13(252–924)p, Cdc13(451–924)p or Cdc13(451–693)p were constructed and transformed into yeast strain 2758-8-4b, a temperature-sensitive mutant of CDC13 (cdc13-1). Yeast (Fig. 5B). We have also analyzed the meiotic products from strain 2758-8-4b (cdc13-1) grows normally at 25°Cand arrests CDC13/cdc13∆ ::HIS3 diploid cells harboring plasmid pTHA- at G /M phase of the cell cycle at 30°C. Full-length CDC13 NLS-CDC13 or pTHA-NLS-CDC13(252–924). Having analyzed complemented the temperature-sensitive phenotype of cdc13-1 ∼2000 spore products each from CDC13/cdc13∆ ::HIS3 cells at 30 or 37°C (Fig. 5A). Cells expressing Cdc13(252–924)p carrying plasmid pTHA-NLS-CDC13 (Leu2 marker) or grew at all three temperatures tested. However, cells pTHA-NLS-CDC13(252–924) (Leu2 marker), we obtained + + expressing other fragments of Cdc13p did not complement the 465 and 69 His Leu haploid cells, respectively (data not temperature-sensitive phenotype of cdc13-1 at 30 or 37°C shown). Thus, even though it remains to be tested whether this (Fig. 5A). Since Cdc13(451–693)p could not complement the complementation was caused by overexpression of growth arrest caused by the cdc13-1 mutation, telomere- Cdc13(252–924)p, Cdc13(252–924)p was sufficient to binding activity alone was therefore not sufficient to account complement cdc13-1 and cdc13∆ mutations. for the essentiality of Cdc13p. Cdc13(252–924)p was capable of interacting with Stn1p To test whether Cdc13(252–924)p complements the null allele of cdc13, a plasmid loss experiment was conducted. Cdc13p was shown to interact with Stn1p. To test if this Here, the cdc13∆ ::HIS3 strain YJL501 requires the CDC13- interaction is required for the essential function of Cdc13p, the bearing plasmid (YEP24-CDC13, with URA3 marker) to grow. interaction between Cdc13(252–924)p and Stn1p was evaluated. If a second plasmid introduced into the yeast expresses Two-hybrid analysis was used previously to establish the inter- functional CDC13, YJL501 then no longer requires plasmid action between Cdc13p and Stn1p (37). Here, we used the YEP24-CDC13 for viability. Growth on 5-FOA is used to monitor same approach to dissect the region within Cdc13p that inter- the loss of YEP24-CDC13. As shown in Figure 5B, 5-FOA- acts with Stn1p. Plasmids were constructed in which CDC13 resistant cells were observed in YJL501 transformed with or its fragments were fused to the DNA-binding domain of plasmid expressing Cdc13p. Similarly, 5-FOA-resistant cells GAL4. These plasmids were transformed into yeast strain were observed in YJL501 expressing Cdc13(252–924), Y190 carrying a plasmid with STN1 fused to the activation although this rescue was ∼10- to 100-fold less efficient than domain of GAL4 (pACT-STN1) to analyze for their inter- that of Cdc13p. However, transformation of YJL501 with action. The ability to grow on medium lacking histidine was plasmid vector, pTHA-NLS, or plasmids expressing other used as the criterion to evaluate the interaction between Stn1p fragments of CDC13 did not yield any 5-FOA-resistant cells and various truncated forms of Cdc13p. As shown in Figure 6, Nucleic Acids Research, 2000, Vol. 28, No. 23 4739 Figure 7. Western blotting analysis of Cdc13-1p. Strain 2758-8-4b (cdc13-1) was grown at permissive temperature (25°C) and shifted to non-permissive temperature (30°C) for 2 h. Total cell extracts prepared from these cells were separated by 10% SDS–PAGE and subjected to immunoblotting analysis using polyclonal antibodies raised against Cdc13(1–252)p. Bound antibodies were visualized by chemiluminescence using an ECL kit (Amersham-Pharmacia). Molecular markers are indicated on the left. Here, we applied an immunoblotting assay using polyclonal anti- bodies raised against Cdc13(1–252)p (T.-L.Pang and J.-J.Lin, unpublished data) to evaluate the cellular level of Cdc13p. As shown in Figure 7, under the condition that >90% of the cells hadarrestedat G /M phase, the Cdc13-1p level at the non- Figure 6. Cdc13(251–924)p interacts with Stn1p. Yeast cells Y190/pACT2 or Y190/pACT-STN1 carrying plasmid pAS2 (vector), pAS-CDC13 (1–924), permissive temperature (30°C) was similar to the level of pAS-CDC13-1 (P371S) or pAS-CDC13(252–924) (252–924) were grown on Cdc13-1p at the permissive temperature (25°C). Our results YC medium without leucine and tryptophan for 16 h at 30°C. Ten-fold serial suggested that reduced interaction between Cdc13-1p and dilutions of yeast cells were spotted on plates without leucine and tryptophan Stn1p was not due to the reduced stability of Cdc13-1p. (YC-Leu-Trp), or without leucine, tryptophan and histidine with the addition of 25 mM of 3-AT (YC-Leu-Trp-His+3-AT), and incubated at 25°C(top panel) or 30°C (bottom panel) until colonies formed. The photographs of the plates are shown. DISCUSSION Cdc13p binds specifically to the single-stranded TG tail of 1–3 yeast telomere. Here, we have delineated the regions of under our assay conditions, His colonies were apparent at 25 Cdc13p responsible for this interaction. The single-stranded or 30°C in Y190 harboring plasmids pAS-CDC13 and pACT- TG -binding domain of Cdc13p is within amino acids 451–693, 1–3 STN1. This result was consistent with the previous report that and binding of the single-stranded TG is specific. However, 1–3 Cdc13p interacts with Stn1p (37). His colonies were also binding to telomeres by Cdc13p is not sufficient to account for apparent at 25 or 30°C in Y190 harboring plasmids pAS- the essential function of Cdc13p. Judging from the results the CDC13(252–924) and pACT-STN1 indicating that C-terminal 673-amino-acid polypeptide was sufficient to Cdc13(252–924)p was capable of interacting with Stn1p. We complement the growth defect phenotype of several cdc13 also examined if Cdc13-1p, with Pro371 being replaced by Ser mutants and interaction with Stn1p. Our results indicated that (24,25), might interact with Stn1p. Interestingly, the HIS3 the Stn1p-interaction function and the telomere-binding expression level in Y190/pACT-STN1 carrying pAS-CDC13-1 activity of Cdc13p were essential for cell growth. cells was comparable to those expressing either Cdc13p or Upon proteolytic degradation of the Cdc13p–DNA complex, Cdc13(252–924)p at 25°C (Fig. 6, top panel). However, the a fragment of Cdc13p that covers amino acids 557 to ∼690 was HIS3 expression level of cells carrying Cdc13-1p was shown to associate with single-stranded TG DNA (42). In 1–3 relatively low at 30°C (Fig. 6, bottom panel). Similarly, using this report, our strategy to identify the DNA-binding region of β-galactosidase as the reporter system, the color-forming Cdc13p was the subcloning of restriction enzyme-digested ability of Cdc13-1p was reproducibly reduced, as compared CDC13 fragments followed by evaluation of the single- with that of Cdc13p at 30 or 37°C (data not shown). These stranded TG -binding activities of these expressed CDC13 1–3 results indicated that Cdc13(252–924)p was capable of inter- fragments. Using this approach, the smallest fragment that still acting with Stn1p. Moreover, the interaction of Stn1p with contained the single-stranded TG -binding activities of 1–3 Cdc13-1p appeared reduced at higher temperature, suggesting Cdc13p was within amino acids 451–693 (Fig. 1). Smaller that the temperature-sensitive lethality phenotype of cdc13-1 fragments such as Cdc13(510–693), Cdc13(451–600) or might be due to a decrease in interaction with Stn1p. Cdc13(601–693) did not show detectable single-stranded TG 1–3 To address the possibility that reduced interaction between binding activity. It is unclear whether Cdc13(510–693)p, Cdc13-1p and Stn1p was due to the reduced stability of Cdc13-1p, which covers amino acids 557 to ∼690, did not interact with the Cdc13-1p level at non-permissive temperature was evaluated. single-stranded TG DNA. Nevertheless, our identification of 1–3 4740 Nucleic Acids Research, 2000, Vol. 28, No. 23 Cdc13(451–693)p as a stably expressed telomeric-binding A Pro371 to Ser substitution caused the phenotype of Cdc13-1p domain of Cdc13p provides useful information for future (24,25). On the basis of gel mobility-shift assay, the cdc13-1 mutation did not affect the telomere-binding activity of understanding of how Cdc13p binds and modulates telomere Cdc13p [J.-J.Lin and V.A.Zakian, unpublished result (24,42)]. function. In our two-hybrid system, the interaction of Cdc13-1p with The identity of the second migration bands using TG30 or Stn1p was temperature dependent and the interaction was TG35 as DNA substrate is unclear (Fig. 2). The simplest reduced at the non-permissive temperature (Fig. 6). This result explanation for the appearance of this second migration band indicated that interaction between Cdc13p and Stn1p is essential would be that TG30 or TG35 provides enough space for two for cell survival. While we cannot rule out that another unchar- molecules of Cdc13(451–693)p to bind. This result would also acterized change in Cdc13-1p is responsible for the cell cycle imply that the optimal binding site of Cdc13(451–693)p on arrest at the non-permissive temperature, relatively weak inter- DNA was ∼13–15 bases, an estimation that is in reasonable action between Cdc13-1p and Stn1p might indeed cause the agreement with the results of using a 34-kDa Cdc13 DBD by accumulation of single-stranded G-rich DNA near telomeres Huges et al. (42). However, this explanation is complicated by (35) leading to this phenotype. observations from several reports that telomeric DNA-binding proteins can also promote the formation of a G-quartet struc- ture (43,44). It will be interesting to know whether Cdc13p ACKNOWLEDGEMENTS promotes the formation of a G-quartet structure upon binding. We thank members of J.-J.L.’s laboratory for their help. We also Contrary to the ciliate telomere-binding proteins, both thank Drs W. J. Lin and C. Wang for critical reading of the manu- Cdc13p and Cdc13(451–693)p preferred low salt for binding script. This work was supported by NSC grants 87-2314-B-010- (data not shown). Sequence analysis of this 243-amino-acid 004, 88-2314-010-070 and in part by grant 89-B-FA22-2-4 region did not provide information on which residues within (Program for Promoting Academic Excellence of Universities). this region are responsible for interacting with telomeres. Evidence was presented that single-stranded DNA-binding proteins interact with DNA via hydrophobic interactions REFERENCES between aromatic side chains of the protein and the DNA bases 1. Zakian,V.A. (1995) Science, 270, 1601–1607. (45–47). For example, the single-stranded DNA-binding 2. Blackburn,E.H. (1990) J. Biol. Chem., 265, 5919–5921. protein T4 gp32 utilizes residues Tyr84, 99, 106, 115, 137, 186 3. Gottschling,D.E., Aparicio,O.M., Billington,B.L. and Zakian,V.A. (1990) and Phe138 to form a hydrophobic pocket to interact with the Cell, 63, 751–762. bases of single-stranded DNA (46). In addition, the Oxytricha 4. Klein,F., Laroche,T., Cardenas,M.E., Hofmann,J.F.X., Schweizer,D. and Gasser,S.M. (1992) J. Cell Biol., 117, 935–948. α and β complex uses a series of aromatic amino acids 5. Henderson,E.R. and Blackburn,E.H. (1989) Mol. Cell. Biol., 9, 345–348. including Tyr130α, Tyr293α,Tyr134β, His292α,Phe109β 6. Klobutcher,L.A., Swanton,M.T., Donini,P. and Prescott,D.M. (1981) and Tyr239α to interact with the extended bases of single- Proc. Natl Acad. Sci. USA, 78, 3015–3019. stranded T G DNA (47). It is likely that aromatic residues 7. Pluta,A.F., Kaine,B.P. and Spear,B.B. (1982) Nucleic Acids Res., 10, 4 4 8145–8154. within the 243 amino acids of Cdc13p play a role in the inter- 8. Wellinger,R.J., Wolf,A.J. and Zakian,V.A. (1993) Cell, 72, 51–60. action with telomeres. Interestingly, this region indeed has a 9. Wellinger,R.J., Ethier,K., Labrecque,P. and Zakian,V.A. (1996) Cell, 85, high content of aromatic residues. However, it remains to be 423–433. determined which amino acid residues are responsible for this 10. Shore,D. (1997) Trends Biochem. Sci., 22, 233–235. protein–DNA interaction. 11. Conrad,M.N., Wright,J.H., Wolf,A.J. and Zakian,V.A. (1990) Cell, 63, Using the one-hybrid system, Bourns et al. (34) showed 739–750. 12. Cooper,J.P., Nimmo,E.R., Allshire,R.C. and Cech,T.R. (1997) Nature, previously that the N-terminal 251 amino acids of Cdc13p inter- 385, 744–747. acted with telomere, but that amino acids 252–508 or 509–924 of 13. Bilaud,T., Brun,C., Ancelin,K., Koering,C.E., Laroche,T. and Gilson,E. Cdc13p failed to do so. Since the N-terminal 251 amino acids (1997) Nature Genet., 17, 236–239. were sufficient to target Cdc13p to telomere, it was concluded 14. Broccoli,D., Smogorzewska,A., Chong,L. and de Lange,T. (1997) Nature that the telomere-binding domain was within this region of Genet., 17, 231–235. 15. Chong,L., van Steensel,B., Broccoli,D., Erdjument-Bromage,H., Cdc13p. In contrast to their study, we showed that the Hanish,J., Tempst,P. and de Lange,T. (1995) Science, 270, 1663–1667. telomere-binding domain was mapped to within the 451–693 16. van Steensel,B., Smogorzewska,A. and de Lange,T. (1998) Cell, 92, 401–413. amino acids of Cdc13p both in vitro and in vivo (Figs 1 and 4). 17. Wright,J.H. and Zakian,V.A. (1995) Nucleic Acids Res., 23, 1454–1460. Given that the N-terminal 1–251 amino acids of Cdc13p did 18. Wright,J.H., Gottschling,D.E. and Zakian,V.A. (1992) Genes Dev., 6, not bind to single-stranded TG DNA in vitro (Fig. 1), it 197–210. 1–3 19. Kyrion,G., Boakye,K.A. and Lustig,A.J. (1992) Mol. Cell. Biol., 12, would be interesting to know how this region of Cdc13p inter- 5159–5173. acts with telomeres in vivo. Clearly, in vivo one-hybrid system 20. Lustig,A.J., Kurtz,S. and Shore,D. (1990) Science, 250, 549–553. detected not only proteins that interact directly with telomeric 21. Stavenhagen,J.B. and Zakian,V.A. (1998) Genes Dev., 12, 3044–3058. DNA but also those that interact indirectly (34). For example, 22. Lin,J.-J. (1993) Bioessays, 15, 555–557. Sir and Rif proteins interact with telomeres by their ability to 23. Lin,J.-J. and Zakian,V.A. (1994) Nucleic Acids Res., 22, 4906–4913. bind Rap1p (48–50). One possible explanation is that this N- 24. Nugent,C.I., Hughes,T.R., Lue,N.F. and Lundblad,V. (1996) Science, 274, 249–252. terminal 251-amino-acid fragment interacts indirectly with the 25. Lin,J.-J. and Zakian,V.A. (1996) Proc. Natl Acad. Sci. USA, 93, telomere. Conceivably, protein factors that associated with 13760–13765. telomeres including Rap1p, Rif proteins and Sir proteins, are 26. Petracek,M.E., Konkel,L.M.C., Kable,M.L. and Berman,J. (1998) potential candidates for recruiting the N-terminal 251-amino-acid EMBO J., 13, 3648–3658. polypeptide of Cdc13p to telomeres. 27. McKay,S.J. and Cooke,H. (1992) Nucleic Acids Res., 20, 6461–6464. Nucleic Acids Research, 2000, Vol. 28, No. 23 4741 28. Konkel,L.M.C., Enomoto,S., Chamberlain,E.M., McCune-Zierath,P., 40. Chevray,P.M. and Nathans,D. (1992) Proc. Natl Acad. Sci. USA, 89, Iyadurai,S.J.P. and Berman,J. (1995) Proc. Natl Acad. Sci. USA, 92, 5789–5793. 5558–5562. 41. Harper,J.W., Adami,G.R., Wei,N., Keyomarsi,K. and Elledge,S.J. (1993) 29. Ishikawa,F., Matunis,M.J., Dreyfuss,G. and Cech,T.R. (1993) Mol. Cell. Biol., Cell, 75, 805–816. 13, 4301–4310. 42. Hughes,T.R., Weilbaecher,R.G., Walterscheid,M. and Lundblad,V. 30. Virta-Pearlman,V., Morris,D.K. and Lundblad,V. (1996) Genes Dev., 10, (2000) Proc. Natl Acad. Sci. USA, 97, 6457–6462. 3094–3104. 43. Fang,G. and Cech,T.R. (1993) Cell, 74, 875–885. 31. Fang,G., Gray,J.T. and Cech,T.R. (1993) Genes Dev., 7, 870–882. 44. Giraldo,R. and Rhodes,D. (1998) EMBO J., 13, 2411–2420. 32. Fang,G. and Cech,T.R. (1993) Proc. Natl Acad. Sci. USA, 90, 6056–6060. 45. Venkataram Prasad,B.V. and Chiu,W. (1987) J. Mol. Biol., 193, 579–584. 33. Gottschling,D.E. and Zakian,V.A. (1986) Cell, 47, 195–205. 46. Shamoo,Y., Friedman,A.M., Parsons,M.R., Konigsberg,W.H. and 34. Bourns,B.D., Alexander,M.K., Smith,A.M. and Zakian,V.A. (1998) Steitz,T.A. (1995) Nature, 376, 362–366. Mol. Cell. Biol., 18, 5600–5608. 47. Horvath,M.P., Schweiker,V.L., Bevilacqua,J.M., Ruggles,J.A. and 35. Garvik,B., Carson,M. and Hartwell,L. (1995) Mol. Cell. Biol., 15, 6128–6138. Schultz,S.C. (1998) Cell, 95, 963–974. 36. Weinert,T.A. and Hartwell,L.H. (1988) Science, 241, 317–322. 48. Moretti,P., Freeman,K., Coodly,L. and Shore,D. (1994) Genes Dev., 8, 37. Grandin,N., Reed,S.I. and Charbonneau,M. (1997) Genes Dev., 11, 512–527. 2257–2269. 38. Sikorski,R.S. and Hieter,P. (1989) Genetics, 122, 19–27. 49. Hardy,C.F., Sussel,L. and Shore,D. (1992) Genes Dev., 6, 801–814. 39. Gyuris,J., Golemis,E., Chertkov,H. and Brent,R. (1993) Cell, 75, 791–803. 50. Wotton,D. and Shore,D. (1997) Genes Dev., 11, 748–760. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nucleic Acids Research Oxford University Press

Telomere-binding and Stn1p-interacting activities are required for the essential function of Saccharomyces cerevisiae Cdc13p

Download PDF
9 pages

Loading next page...
 
/lp/oxford-university-press/telomere-binding-and-stn1p-interacting-activities-are-required-for-the-fn5zybhWSC

References (50)

Publisher
Oxford University Press
ISSN
0305-1048
eISSN
1362-4962
DOI
10.1093/nar/28.23.4733
Publisher site
See Article on Publisher Site

Abstract

© 2000 Oxford University Press Nucleic Acids Research, 2000, Vol. 28, No. 23 4733–4741 Telomere-binding and Stn1p-interacting activities are required for the essential function of Saccharomyces cerevisiae Cdc13p Mei-Jung Wang, Yi-Chien Lin, Te-Ling Pang, Jui-Mei Lee, Chia-Ching Chou and Jing-Jer Lin* Institute of Biopharmaceutical Science, National Yang-Ming University, Shih-Pai, 112, Taipei, Taiwan, Republic of China Received July 31, 2000; Revised and Accepted October 20, 2000 ABSTRACT are composed of short tandem repeated sequences with the strand running in a 5′ to 3′ direction toward the guanine (G)-rich ends Yeast Saccharomyces cerevisiae Cdc13p is the (1). The sequence and the number of repeats vary considerably telomere-binding protein that protects telomeres and in different species. For example, the telomere sequences of yeast regulates telomere length. It is documented that Saccharomyces cerevisiae are ∼250–300 bp of TG /C A, 1–3 1–3 Cdc13p binds specifically to single-stranded TG 1–3 whereas those of human are ∼10 kb TTAGGG/CCCTAA repeats telomeric DNA sequences and interacts with Stn1p. (1,2). Besides the double-stranded telomeric DNA repeats, telo- To localize the region for single-stranded TG DNA meres in all of the organisms that have been analyzed, also contain 1–3 binding, Cdc13p mutants were constructed by deletion a G-rich single-stranded tail (5–7). In S.cerevisiae,a >30 base single-stranded TG tail was detected in late S phase of the cell mutagenesis and assayed for their binding activity. 1–3 cycle (8). This single-stranded tail was postulated as an inter- Based on in vitro electrophoretic mobility shift mediate during telomere replication (8,9). assay, a 243-amino-acid fragment of Cdc13p (amino Protein factors that interact with telomeres participate in acids 451–693) was sufficient to bind single-stranded telomere functions (10). Factors that bind to the double-stranded TG with specificity similar to that of the native 1–3 telomeric DNA sequences have been identified from many protein. Consistent with the in vitro observation, in vivo organisms (11–16). These proteins are essential for the mainte- one-hybrid analysis also indicated that this region of nance of telomere functions and cell viability (11,12,16). For Cdc13p was sufficient to localize itself to telomeres. example, in S.cerevisiae, Rap1p has been shown to bind However, the telomere-binding region of Cdc13p (amino double-stranded telomeric DNA (11,17,18). Mutations on acids 451–693) was not capable of complementing the Rap1p affect the length of telomere, TPE, localization of the growth defects of cdc13 mutants. Instead, a region telomere within the nucleus, telomere recombination and cell viability (19–21). Moreover, protein factors bound to the comprising the Stn1p-interacting and telomere-binding single-stranded telomeric DNA were identified in several region of Cdc13p (amino acids 252–924) comple- organisms (22–30). Among these protein factors, Oxytricha mented the growth defects of cdc13 mutants. These telomere-binding protein is well characterized. The protein is results suggest that binding to telomeres by Cdc13p heterodimeric and is composed of an α subunit and a β subunit is not sufficient to account for the cell viability, inter- (31–33). The α subunit is a single-stranded DNA-binding action with Stn1p is also required. Taken together, protein that binds to the G T single-stranded end of the 4 4 we have defined the telomere-binding domain of telomere. Although the β subunit is not directly involved in Cdc13p and showed that both binding to telomeres binding, it is required for terminus-specific binding. In the and Stn1p by Cdc13p are required to maintain cell yeast genome, there is no homologue of the Oxytricha α-and growth. β-like binding proteins. Instead, Cdc13p binds to single- stranded TG sequences in vitro and interacts with telomeres 1–3 in vivo (24,25,34). It is believed that, in yeast, Cdc13p is the INTRODUCTION functional equivalent of Oxytricha α and β binding proteins Telomeres are the specialized structure at the very ends of despite there being no sequence similarity between Oxytricha eukaryotic chromosomes. Telomeres are essential for the telomere-binding proteins and yeast Cdc13p (J.-J.Lin, unpub- maintenance of chromosome integrity (1,2). They prevent end- lished observation). Furthermore, Cdc13p dissociates from to-end fusion of chromosomes, protect chromosome from single-stranded TG DNA at 300 mM NaCl (J.J.Lin and 1–3 degradation by nucleases and facilitate the complete replication of V.A.Zakian, unpublished result) whereas the binding of chromosomes. They also suppress the expression of nearby Oxytricha proteins to G T DNA remains associated even at 4 4 genes, a phenomenon known as telomere position effect (TPE) 2 M NaCl (33). Thus, Cdc13p might represent a class of (3). Telomeres might also position chromosomes within the single-stranded telomere-binding protein that is different from nucleus (4). In most of the cloned telomeric sequences, telomeres the Oxytricha protein. *To whom correspondence should be addressed. Tel: +86 2 2826 7258; Fax: +86 2 2820 0067; Email: jjlin@ym.edu.tw 4734 Nucleic Acids Research, 2000, Vol. 28, No. 23 CDC13 is an essential gene (35). It is involved in cell cycle (GST) gene of pGEX plasmids (Pharmacia). The 4.4-kb SmaI– control, because a temperature-sensitive allele of CDC13, HpaI fragment containing full length CDC13 from pTHA-CDC13 cdc13-1, causes the cell cycle to arrest in G /M phase at the (25) was ligated to the SmaI-digested pGEX-3X to generate non-permissive temperature (35). This cell cycle control is pGST-CDC13. Plasmid pGST-CDC13 was digested with BglII dependent on the gene product of RAD9, which was postulated and self-ligated to generate pGST-CDC13(1–600), which deleted to be involved in surveying the integrity of chromosomes (36). the BglII–BglII fragment within CDC13.Plasmid pGST- Cdc13p might enable Rad9p to differentiate whether the ends CDC13(451–924) was constructed by ligating the 2.9-kb BamHI– of a linear DNA are telomeres or broken ends by binding to HpaI fragment containing half of CDC13 to BamHI- and SmaI- telomeres. It is also demonstrated that cdc13-1 cells at the non- digested pGEX-3X. Plasmid pGST-CDC13(451–924) was permissive temperature accumulate single-stranded G-rich digested with either PvuII or BglII, and self-ligated to generate DNA near telomeres (35). Nevertheless, it is unclear whether pGST-CDC13(451–871) and pGST-CDC13(451–600), respec- this single-stranded G-rich DNA is recognized as a damage to tively. Plasmid pGST-CDC13(451–693) was constructed by trigger the checkpoint mechanism. Since a mutation allele of digesting pGST-CDC13(451–924) with SalIand NruI, treating est CDC13, cdc13 , causes gradual loss of telomere, Cdc13p also with T4 DNA polymerase and self-ligation. A 1.4-kb EcoRI–SalI appears to regulate the length of telomeres (24). At least a fragment of pTHA-CDC13 was ligated to EcoRI- and SalI- portion of the telomere function of Cdc13p is contributed by its digested pGEX-4T-1 to generate pGST-CDC13(510–924). To interaction with Stn1p. STN1 was identified from a high copy construct pGST-CDC13(601–856), a 0.8-kb BglII–BglII fragment suppressor screening to rescue the temperature-sensitive from pTHA-CDC13 was inserted into BamHI-digested pGEX-1. The NruI–SmaI fragment of pGST-CDC13(601–856) was lethality of cdc13-1 (37) and direct interaction of Cdc13p with deleted to generate pGST-CDC13(601–693). Expression of Stn1p was demonstrated by two-hybrid assays (37). Moreover, these fusion proteins was induced by the addition of isopropyl telomere defects caused by STN1 mutation are similar, if not β-D-thiogalactoside (IPTG) and confirmed by western blotting identical, to CDC13 mutation, suggesting that these two genes analysis using antibody against GST (data not shown). act together in maintaining telomere function (37). Plasmid pET6H-CDC13(451–693), which was used to CDC13 encodes a 924-amino-acid protein with molecular purify Cdc13(451–693)p, was constructed by inserting the mass of 104 895 kDa (35). Sequence comparison between NcoI–NruI fragment of pTHA-NLS-CDC(451–693) into NcoI/ Cdc13p and known DNA- or RNA-binding proteins did not SmaI-digested pET6H (a gift from C.-H. Hu., National Marine provide any information on the region within Cdc13p responsible University, Taiwan). The resulting plasmid was used to for interacting with telomeres (J.-J.Lin, unpublished observation). express 6× His-tagged Cdc13(451–693)p under the control of Thus, Cdc13p might contain a novel DNA-interacting motif the T7 promoter. for binding to telomeres. In order to identify the functional To construct a plasmid for one-hybrid analysis, the DNA domains of Cdc13p, we have generated deletion mutants and fragment encoding Cdc13(451–693)p was amplified by PCR examined the single-stranded TG -binding activity and their 1–3 using pTHA-CDC13 as the template and with primers ML11 ability to interact with Stn1p. Here we showed that (5′-CCGCTCGAGATCCTGTGGGACAATGAC-3′) and ML12 Cdc13(451–693)p, a Cdc13p fragment ranging from amino (5′-CCGCTCGAGATGAGAACCGTTTCT-3′). The 0.7-kb acids 451 to 693, retain the single-stranded TG -binding 1–3 PCR product was subcloned into SmaI-digested pUC18 to activity. We also showed that binding to telomeres by Cdc13p generate pUC-CDC13(451–693). The 0.7-kb DNA fragment is not sufficient to maintain cell viability. Moreover, amino from XhoI-digested pUC-CDC13(451–693) was then ligated acids 252–924, which covered the telomere-binding and with the XhoI-digested pJG4-5 (39) or pRF4-6NL (34) to Stn1p-interacting activities of Cdc13p, are sufficient to generate pJG-CDC13(451–693) or pRF-CDC13(451–693), account for the essential function of Cdc13p. respectively. Since Cdc13p is a telomere-binding protein, the cellular MATERIALS AND METHODS localization of Cdc13p should be in the nucleus. To make cer- tain that the truncated Cdc13p would be delivered into the Strains nucleus, two oligonucleotides NLSw (5′-CTAGCCCCAA- Escherichia coli strain DH5α was used as host for plasmid GAAGAAGCGGAAGGTCGC-3′)and NLSc (5′-CATGGC- construction and propagation, and strain BL21(DE3)pLysS for GACCTTCCGCTTCTTCTTTGGGG-3′) encoding the nuclear Cdc13(451–693)p purification. Yeast strain YPH499 [MATa localization sequence (NLS) of SV40 large T antigen (39,40) amber ochre ura3-52 lys2-801 ade2-101 trp1-∆ 63 his3-∆ 200 leu2-∆ 1 were annealed and ligated with SpeI- and NcoI-digested (38)] with URA3 and ADE2 placed near the telomere of pTHA-CDC13. The resulting plasmid, pTHA-NLS-CDC13, chromosomes VII-L and V-R (YPH499UTAT), respectively, encoded a fusion protein with three HA-tag and NLS at the N- was used as the host for analyzing TPE. Yeast strains, HIS-Tel terminal of Cdc13p. Similarly, plasmid vector pTHA-NLS was and HIS-Int-CA, used for one-hybrid analysis were kindly constructed by inserting NLS sequences into pTHA (25). To provided by V. A. Zakian [Princeton University (34)]. Strain construct plasmid pTHA-NLS-CDC13(451–924), a 1.6-kb 2758-8-4b (MATa cdc13-1 his7 leu2-3, 112 ura3-52 trp1-289, DNA fragment encoding amino acids 451–924 of Cdc13p was provided by L. Hartwell, University of Washington) was used PCR amplified with primers ML01 (5′-TGCCATGGGGATC- as the host for complementation tests. CTGTGGGACAAT-3′)and CDC133 (5′-AACTGCAGACT- AGTCGACTCTTGCTTCTTACC-3′) using Vent DNA poly- Plasmid construction merase (New England Biolabs). The PCR product was digested Deletion mutants of Cdc13p (Fig. 1A) were generated by fusion with NcoIand SalI, and was inserted into NcoI- and SalI-digested of various regions of CDC13 with the glutathione S-transferase pTHA-NLS. To generate pTHA-NLS-CDC13(451–693), plasmid Nucleic Acids Research, 2000, Vol. 28, No. 23 4735 pTHA-NLS-CDC13(451–924) was digested with NruIand buffer (100 mM Tris pH 8.0, 2 mM EDTA, 2 mM DTT, 0.4 M SalI, blunted by T4 DNA polymerase and self-ligated. Plasmid L-arginine, 20% glycerol) at 4°C for 12 h. pTHA-NLS-CDC13(1–252) was constructed by ligating the Electrophoretic mobility shift assay (EMSA) NcoI–EcoRI fragment (0.8 kb) encoding the N-terminal 252- amino-acid polypeptide, of pAS-CDC13(1–252) to NcoI- and Oligonucleotide TG22 (5′-GTGGTGGGTGGGTGTGTGTGGG- EcoRI-digested pTHA-NLS. A 2-kb NcoI–SalI fragment of 3′), TG10 (5′-GGGTGTGGTG-3′), TG15 (5′-TGTGTGGGTG- pAS-CDC13(252–924) was ligated with NcoI- and SalI-digested TGGTG-3′), TG20 (5′-TGGTGTGTGTGGGTGTGGTG-3′), pTHA-NLS to generate pTHA-NLS-CDC13(252–924). TG25 (5′-GGGTGTGGTGTGTGTGGGTGTGGTG-3′), TG30 Expression of full-length and truncated CDC13 was confirmed (5′-TGTGTGGGTGTGGTGTGTGTGGGTGTGGTG-3′)or TG35 (5′-TGTGGTGTGTGGGTGTGGTGTGTGTGGGTG- by western blotting analysis on the extracts prepared from TGGTG-3′) was first 5′ end-labeled with [γ- P]ATP yeast cells with anti-HA antibody 12CA5 (data not shown). (3000 mCi/mM, NEN) using T4 polynucleotide kinase (New Two-hybrid plasmids were constructed by subcloning STN1 England Biolabs) and subsequently purified from a 10% into pACT2 and several fragments of CDC13 into pAS2 sequencing gel after electrophoresis. To perform the assays, (37,41). Plasmid pACT-STN1 was constructed by ligating a cell extracts were mixed with 2.0 ng of P-labeled TG22 or NcoI–BamHI 1.6-kb fragment containing full-length STN1 TG15 DNA with a total volume of 15 µ l containing 50 mM from pJG-STN1 (unpublished construct) into pACT2 with the Tris–HCl pH 7.5, 1 mM EDTA, 50 mM NaCl, 1 mM DTT and same sites. To construct full-length CDC13 into pAS2, the 1 µ g of single stranded poly(dI–dC). The mixtures were allowed to NcoI–SalI fragment (3 kb) of pTHA-CDC13 (25) was incubate at room temperature for 10 min. Then, 3 µ l of 80% inserted into the NcoI- and SalI-digested pAS2. Plasmid glycerol was added and the mixtures were loaded on an 8% non- carrying the temperature-sensitive allele of CDC13, cdc13-1,was denaturing polyacrylamide gel, which was pre-run at 125 V for constructed by replacing the 1.4-kb fragment of pAS-CDC13 10 min. Electrophoresis was carried out in TBE (89 mM Tris– with the equivalent of pTHA-CDC13-1 (25). Plasmid pAS- borate/2 mM EDTA) at 125 V for 105 min. The gels were dried and CDC13(252–924) was constructed by first ligating the 1.4-kb autoradiographed. For competition analysis, 2.0 ng P-labeled EcoRI–SalI CDC13 fragment of pTHA-CDC13 into EcoRI- TG15 was mixed with varying amounts of non-radioactive and SalI-digested pAS2-1 followed by inserting a 0.8-kb competitors before addition of the cell extracts. Binding activity EcoRI–EcoRI fragment of CDC13 isolated from pTHA- was quantified with a PhosphorImager (Molecular Dynamics). CDC13 with EcoRI digestion. One-hybrid analysis Preparation of E.coli extracts One-hybrid analysis was performed using the methods described To prepare protein extracts from E.coli, a 10-ml culture of by Bourns et al. (34). Plasmid pJG4-5, pJG-CDC13(451–693) or E.coli containing a Cdc13p-deletion construct was grown in pRF-CDC13(451–693) was first transformed into yeast strains LB medium with 50 µ g/ml ampicillin at 30°CtoOD =0.6. HIS-Tel and HIS-Int-CA, and selected on plates lacking At this point, IPTG was added to a final concentration of 1 mM tryptophan. To test for telomere-binding activity in vivo, cells and the cells were allowed to grow at 30°C for an additional 16 were grown in liquid medium lacking tryptophan for ∼16 h. h before harvesting by centrifugation. The pellets were resus- Then, cells were spotted in 10-fold serial dilutions on yeast pended in 0.5 ml buffer A [50 mM Tris–HCl pH 7.5, 1 mM synthetic complete medium (YC) plates lacking tryptophan, or EDTA, 1× protease inhibitor cocktail (Calbiochem) and 50 lacking histidine, or with 10, 20 or 40 mM of 3-amino-1,2,4- mM glucose] and were sonicated. Cell debris was removed by triazole (3-AT, Sigma, St Louis, MO). Plates were incubated at centrifugation in an Eppendorf microfuge for 10 min at 4°C, 30°C until the colonies could be observed. Because HIS-Tel the supernatants were aliquoted and frozen in a dry ice-ethanol cells carrying the B42 transcription-activation domain fused bath, and stored at –70°C. The concentration of protein in the with the DNA-binding region of Cdc13p did not grow well on E.coli extract was ∼2–6 mg/ml as determined using a Bio-Rad plates with galactose, all the experiments were performed in protein assay kit with bovine serum albumin as a standard. plates containing 1.5% galactose and 1% glucose. The HIS- Int-CA strain grew better in this condition than in plates Purification of 6× His-tagged Cdc13(451–693)p containing 3% galactose. To purify 6× His-tagged Cdc13(451–693)p, a 500-ml culture Complementation of cdc13∆ by cdc13 fragments using of IPTG-induced E.coli harboring pET6H-CDC13(451–693) plasmid loss experiments was collected by centrifugation. Cells were resuspended in 30 ml of GdHCl buffer (6 M guanidine–HCl, 0.1 M NaH PO , Plasmid loss experiments were carried out to test if binding of 2 4 0.01 M Tris pH 8.0), followed by gentle shaking for 1 h at 4°C. single-stranded TG and/or Stn1p-interacting activity of 1–3 The suspension was then centrifuged at 13 000 g for 15 min at Cdc13p is sufficient to complement the lack of viability caused 4°C. The clear supernatant was collected and applied to a 5 ml by the cdc13∆ mutation. Briefly, plasmid YEP24-CDC13 Ni-NTA–agarose column (Qiagen) previously equilibrated [abbreviation of plasmid YEP24-CDC13-161-4 (35)] was with GdHCl buffer. The column was washed stepwise with transformedintoadiploidstrainYJL401-UTAT [CDC13/ 20 ml of GdHCl buffer containing 1 mM imidazole followed cdc13∆ ::HIS3 (25)] carrying one allele of the cdc13 null by 20 ml of GdHCl buffer containing 20 mM imidazole. The mutation. The transformants were sporulated and subjected to bound protein was eluted by 12 ml of GdHCl buffer containing tetrad analysis. Haploid strain YJL501 was selected from the + + 200 mM imidazole. To renature the purified Cdc13(451–693)p, spores that were Ura (YEP24-CDC13), His (cdc13∆ ::HIS3)and protein eluted from the Ni-NTA–agarose column was diluted to Ade (ADE2 near the telomere of chromosome V-R). Such a strain ∼50 µ g/ml using GdHCl buffer and dialyzed against renaturation requires a plasmid carrying CDC13 (YEP24-CDC13) for its 4736 Nucleic Acids Research, 2000, Vol. 28, No. 23 viability. Subsequently, plasmid pTHA-NLS, pTHA-NLS-CDC13, pTHA-NLS-CDC13(1–252), pTHA-NLS-CDC13(252–924), pTHA-NLS-CDC13(451–924) or pTHA-NLS-CDC13(451–693) was separately transformed into YJL501. The resulting transform- ants were spotted onto plates containing 0.5 mg/ml 5-fluoroorotic acid (5-FOA) and incubated at 30°C until colonies formed. Two-hybrid analysis Plasmids pACT2 and pACT-STN1 were transformed separately into yeast strain Y190. The resulting strains were then transformed with plasmids pAS2 or pAS2 containing CDC13 fragments. The HIS3 reporter system was also used to evaluate the interaction between Cdc13p and Stn1p. In the assays, 5–10 fresh trans- formed colonies from each transformation were mixed and spotted in 10-fold serial dilutions onto YC plates lacking histidine without or with 25 mM 3-AT. Plates were kept at 25 or 30°C until colonies formed. RESULTS Cdc13(451–693)p bound to single-stranded TG DNA in vitro 1–3 To identify the telomeric DNA-interacting region in Cdc13p, plasmids suitable to express GST fusions with various fragments of Cdc13p were constructed (Fig. 1A). Using P-labeled 22-base TG oligonucleotides (TG22) as substrate, the DNA-binding 1–3 activity of these truncated Cdc13p was determined by EMSA. By comparing the gel-shift pattern with that of the vector alone, extra single-stranded TG -binding activity was 1–3 observed in E.coli extracts expressing Cdc13p fragments Figure 1. Single-stranded TG -binding domain of Cdc13p. (A) Schematic 1–3 representation of Cdc13p deletion mutants. The wild-type protein is 924 containing amino acids 451–924, 451–871 or 451–693 (Fig. 1B). amino acids in length. The relative locations of the fragments (thick line) and Among these fusion proteins, Cdc13(451–693)p was the the first as well as the last amino acids are indicated. All these mutants were shortest single-stranded TG DNA-binding fragment of fused in-frame to a GST protein on their N terminus. On the right are the 1–3 single-stranded TG22-binding activities of mutants that were analyzed by Cdc13p. Western blotting analysis using anti-GST antibody EMSA. (B) Cdc13(451–693)p contains the single-stranded TG -binding 1–3 confirmed that these truncated Cdc13p polypeptides were domain of Cdc13p. Escherichia coli extracts (15 µ g) were mixed with P- indeed expressed, albeit degradation of the full-length and labeledTG22in15-µ l reaction mixtures. The reactions were incubated at room temperature for 10 min. A 3-µ l volume of 80% glycerol was added to each some of these truncated forms of Cdc13p was also observed reaction before analysis of the reaction products on an 8% polyacrylamide gel. (data not shown). Therefore, while it is not clear if the fragment Extracts carrying the GST fusion of Cdc13p deletion mutants are indicated. The can be shortened further, it was evident that the single-stranded first lane has no extracts. Migration of free DNA (TG22), Cdc13(451–693)p, TG -binding activity of Cdc13p is located within the fragment Cdc13(451–871)p, Cdc13(451–924)p and Cdc13p are indicated. 1–3 comprising residues 451–693. We also have expressed a 6× His-tagged-Cdc13(451–693)p in E.coli and purified this tagged protein using Ni-NTA–agarose (Fig. 2A). Using P-labeled single-stranded TG oligonucleotides Cdc13(451–693)p specifically bound to single-stranded 1–3 TG DNA as substrate, the DNA-binding ability of this recombinant 1–3 polypeptide was determined by EMSA. The result shown in To evaluate the selectivity of the Cdc13(451–693)p binding Figure 2B further demonstrated that this 6× His-tagged 32 activity, purified protein was mixed with P-labeled single- Cdc13(451–693)p is capable of forming complexes with the stranded TG and various amounts of unlabeled nucleic acid 1–3 single-stranded TG . Evidently, the single-stranded TG 1–3 1–3 competitors before being subjected to EMSA analysis. As binding activity of Cdc13p was located within amino acids shown in Figure 3, unlabeled TG15 competed efficiently with 451–693. Results shown in Figure 2B also indicate that the P-labeled TG15. The binding was reduced by ∼50% when the length of DNA substrate affected the electrophoretic mobility competitor was presented at equal concentrations (Fig. 3A, of the Cdc13(451–693)p–DNA complex. Interestingly, a lanes 3–5 and B). On the other hand, vertebrate (T AG ), 2 3 second migration band was apparent on TG30 or TG35 but not Oxytricha (T G )and Tetrahymena (T G ) telomeric DNA did 4 4 2 4 on TG25; the identity of this second migration band is uncertain not compete for the binding activity of Cdc13(451–693)p to (Fig. 2B, lanes 19–24 and 25–30). Under our assay condition, TG15 (Fig. 3A, lanes 6–14). Total yeast RNA, single-stranded Cdc13(451–693)p bound TG15 with an apparent binding C A DNA or duplex TG /C A DNA did not compete for 1–3 1–3 1–3 constant of 120 nM. Cdc13(451–693)p binding either (Fig. 3A, lanes 15–17, and Nucleic Acids Research, 2000, Vol. 28, No. 23 4737 Figure 2. Purified Cdc13(451–693)p binds single-stranded TG telomeric 1–3 DNA. (A) Purification of Cdc13(451–693)p. A 6× His-tagged Cdc13(451–693)p Figure 3. Competition analysis of Cdc13(451–693)p telomeric DNA-binding was purified from E.coli using a Ni-NTA–agarose column (see Materials and activity. (A) P-labeledTG15(2ng) was mixed with several concentrations of Methods). A Coomassie blue-stained 10% SDS–polyacrylamide gel is given. different competitors and then 0.2 µ g of the purified Cdc13(451–693)p was Lane 1 shows the molecular weight marker; lanes 2 and 3 were 50 µ lof E.coli added to the mixtures. Competitors were TG22 (yeast), (T AG ) (vertebrate), 2 3 3 cultures harboring the pET6H-CDC13(451–693) plasmid grown without and (T G ) (Oxytricha), (T G ) (Tetrahymena) and total yeast RNA. Gel shift 4 4 3 2 4 3 with IPTG induction, respectively; lane 4 was 5 µ g of purified Cdc13(451–693)p. assay was then carried out. An autoradiogram is shown. (B) Quantification of (B) Cdc13(451–693)p contains the single-stranded TG -binding domain of 1–3 32 the Cdc13(451–693)p-binding activity. The relative level of the binding activity Cdc13p. Approximately 27 nM each of P-labeled TG10 (lanes 1–6), TG15 was quantified by a PhosphorImager and the binding activity in the absence of (lanes 7–12), TG20 (lanes 13–18), TG25 (lanes 19–24), TG30 (lanes 25–30) competitor was taken as 100 [(A) lane 2]. The data show the average from and TG35 (lanes 31–36) were mixed with several concentrations of the purified three experiments. Symbols used are: TG15 (yeast, closed circles), (T AG ) 2 3 3 Cdc13(451–693)p and then gel shift assay was carried out. Cdc13(451–693)p used (vertebrate, closed squares), (T G ) (Oxytricha, closed triangles), (T G ) 4 4 3 2 4 3 in each set of experiments were 0, 7, 22, 66, 200 and 600 nM. Autoradiograms are (Tetrahymena, open circles) and total yeast RNA (open squares), respectively. shown. data not shown). This result indicated that Cdc13(451–693)p spotted onto plates without histidine to evaluate the expression bound specifically to single-strand TG telomeric DNA. With 1–3 of HIS3. HIS-Tel cells carrying plasmid vector alone (Act) or the exception of vertebrate telomeric DNA, which partially Cdc13(451–693)p without the B42 transcription-activation competed away the binding of Cdc13p to TG22 (25), domain (Cdc13-DB) cannot grow on plates lacking histidine Cdc13(451–693)p bound specifically to single-stranded TG 1–3 (Fig. 4, left panel). However, cells carrying the B42 transcription- telomeric DNA similarly to Cdc13p. activation domain fused with the DNA-binding region of Cdc13p (Act-Cdc13-DB) grew on plates lacking histidine Cdc13(451–693)p bound telomere in vivo (Fig. 4, left panel). The levels of cell growth with 3-AT for the Previously, a one-hybrid system was developed to examine HIS-Int-CA strain carrying plasmid vector alone (Act), Cdc13-DB whether a protein interacts with telomeres in vivo (34). In that or Act-Cdc13-DB were similar, indicating that Cdc13(451–693)p system, a promoter-defective allele of HIS3 is placed near the would not bind internal TG /C A duplex DNA (Fig. 4, right 1–3 1–3 telomere of chromosome VII-L, HIS-Tel. The protein to be panel). Thus, Cdc13(451–693)p is sufficient to position itself tested is fused to the E.coli B42 transcription-activation to telomeres. Taken together, both in vitro and in vivo evidence domain. When this fusion protein interacts with telomeres, it indicate that the DNA-binding domain of Cdc13p was located activates the expression of HIS3. Thus, expression of His3p in amino acids 451–693. can be used as a means to identify telomere-interacting protein. To verify whether Cdc13(451–693)p binds telomere in vivo, Cdc13(451–693)p was not sufficient to complement cdc13 this fragment was fused with the B42 transcription-activation mutations domain (Act-Cdc13-DB) and expressed in yeast HIS-Tel or HIS-Int-CA. The HIS-Int-CA strain carried internal HIS3 with It has been known that Cdc13p is essential for cell viability. C A sequences at the 5′ region (34). Dilutions of cells were The next question that was considered in this study was whether 1–3 4738 Nucleic Acids Research, 2000, Vol. 28, No. 23 Figure 5. The telomere-binding domain of Cdc13p cannot rescue the growth Figure 4. Cdc13(451–693)p binds telomere in vivo. HIS-Tel (left panel) or defect phenotype of cdc13 mutants. (A) Yeast 2758-8-4b (cdc13-1) carrying HIS-Int-CA (right panel) cells harboring pJG4-5 (Act), pJG-CDC13(451–693) plasmid pTHA-NLS (vector), pTHA-NLS-CDC13 (1–924), pTHA-NLS- (Act-CDC13-DB) or pRF-CDC13(451–693) (CDC13-DB) were spotted in 10-fold CDC13(1–252), pTHA-NLS-CDC13(252–924), pTHA-NLS-CDC13(451–924) serial dilutions onto plates lacking tryptophan (YC-Trp), lacking histidine (YC-His), or pTHA-NLS-CDC13(451–693) were spotted in 10-fold serial dilutions on YC-His plates with 10 mM 3-AT and YC-His plates with 40 mM 3-AT. Photo- YC-Leu plates and grown at 25°C (left), 30°C (middle)or37°C (right) until graphs were taken after the plates were incubated at 30°Cfor 3days. colonies formed. (B) Yeast strain YJL501 (cdc13∆ ::HIS3/YEP24-CDC13) carrying plasmids pTHA-NLS (vector), pTHA-NLS-CDC13 (1–924), pTHA-NLS- CDC13(1–252), pTHA-NLS-CDC13(252–924), pTHA-NLS-CDC13(451–924) or pTHA-NLS-CDC13(451–693) were spotted in 10-fold serial dilutions on YC-Leu plates or plates containing 5-FOA and incubated at 30°C until colonies formed. binding of Cdc13p to telomere is sufficient to account for its Photographs of the plates are shown. essentiality. Here, plasmids expressing Cdc13p, Cdc13(1–252)p, Cdc13(252–924)p, Cdc13(451–924)p or Cdc13(451–693)p were constructed and transformed into yeast strain 2758-8-4b, a temperature-sensitive mutant of CDC13 (cdc13-1). Yeast (Fig. 5B). We have also analyzed the meiotic products from strain 2758-8-4b (cdc13-1) grows normally at 25°Cand arrests CDC13/cdc13∆ ::HIS3 diploid cells harboring plasmid pTHA- at G /M phase of the cell cycle at 30°C. Full-length CDC13 NLS-CDC13 or pTHA-NLS-CDC13(252–924). Having analyzed complemented the temperature-sensitive phenotype of cdc13-1 ∼2000 spore products each from CDC13/cdc13∆ ::HIS3 cells at 30 or 37°C (Fig. 5A). Cells expressing Cdc13(252–924)p carrying plasmid pTHA-NLS-CDC13 (Leu2 marker) or grew at all three temperatures tested. However, cells pTHA-NLS-CDC13(252–924) (Leu2 marker), we obtained + + expressing other fragments of Cdc13p did not complement the 465 and 69 His Leu haploid cells, respectively (data not temperature-sensitive phenotype of cdc13-1 at 30 or 37°C shown). Thus, even though it remains to be tested whether this (Fig. 5A). Since Cdc13(451–693)p could not complement the complementation was caused by overexpression of growth arrest caused by the cdc13-1 mutation, telomere- Cdc13(252–924)p, Cdc13(252–924)p was sufficient to binding activity alone was therefore not sufficient to account complement cdc13-1 and cdc13∆ mutations. for the essentiality of Cdc13p. Cdc13(252–924)p was capable of interacting with Stn1p To test whether Cdc13(252–924)p complements the null allele of cdc13, a plasmid loss experiment was conducted. Cdc13p was shown to interact with Stn1p. To test if this Here, the cdc13∆ ::HIS3 strain YJL501 requires the CDC13- interaction is required for the essential function of Cdc13p, the bearing plasmid (YEP24-CDC13, with URA3 marker) to grow. interaction between Cdc13(252–924)p and Stn1p was evaluated. If a second plasmid introduced into the yeast expresses Two-hybrid analysis was used previously to establish the inter- functional CDC13, YJL501 then no longer requires plasmid action between Cdc13p and Stn1p (37). Here, we used the YEP24-CDC13 for viability. Growth on 5-FOA is used to monitor same approach to dissect the region within Cdc13p that inter- the loss of YEP24-CDC13. As shown in Figure 5B, 5-FOA- acts with Stn1p. Plasmids were constructed in which CDC13 resistant cells were observed in YJL501 transformed with or its fragments were fused to the DNA-binding domain of plasmid expressing Cdc13p. Similarly, 5-FOA-resistant cells GAL4. These plasmids were transformed into yeast strain were observed in YJL501 expressing Cdc13(252–924), Y190 carrying a plasmid with STN1 fused to the activation although this rescue was ∼10- to 100-fold less efficient than domain of GAL4 (pACT-STN1) to analyze for their inter- that of Cdc13p. However, transformation of YJL501 with action. The ability to grow on medium lacking histidine was plasmid vector, pTHA-NLS, or plasmids expressing other used as the criterion to evaluate the interaction between Stn1p fragments of CDC13 did not yield any 5-FOA-resistant cells and various truncated forms of Cdc13p. As shown in Figure 6, Nucleic Acids Research, 2000, Vol. 28, No. 23 4739 Figure 7. Western blotting analysis of Cdc13-1p. Strain 2758-8-4b (cdc13-1) was grown at permissive temperature (25°C) and shifted to non-permissive temperature (30°C) for 2 h. Total cell extracts prepared from these cells were separated by 10% SDS–PAGE and subjected to immunoblotting analysis using polyclonal antibodies raised against Cdc13(1–252)p. Bound antibodies were visualized by chemiluminescence using an ECL kit (Amersham-Pharmacia). Molecular markers are indicated on the left. Here, we applied an immunoblotting assay using polyclonal anti- bodies raised against Cdc13(1–252)p (T.-L.Pang and J.-J.Lin, unpublished data) to evaluate the cellular level of Cdc13p. As shown in Figure 7, under the condition that >90% of the cells hadarrestedat G /M phase, the Cdc13-1p level at the non- Figure 6. Cdc13(251–924)p interacts with Stn1p. Yeast cells Y190/pACT2 or Y190/pACT-STN1 carrying plasmid pAS2 (vector), pAS-CDC13 (1–924), permissive temperature (30°C) was similar to the level of pAS-CDC13-1 (P371S) or pAS-CDC13(252–924) (252–924) were grown on Cdc13-1p at the permissive temperature (25°C). Our results YC medium without leucine and tryptophan for 16 h at 30°C. Ten-fold serial suggested that reduced interaction between Cdc13-1p and dilutions of yeast cells were spotted on plates without leucine and tryptophan Stn1p was not due to the reduced stability of Cdc13-1p. (YC-Leu-Trp), or without leucine, tryptophan and histidine with the addition of 25 mM of 3-AT (YC-Leu-Trp-His+3-AT), and incubated at 25°C(top panel) or 30°C (bottom panel) until colonies formed. The photographs of the plates are shown. DISCUSSION Cdc13p binds specifically to the single-stranded TG tail of 1–3 yeast telomere. Here, we have delineated the regions of under our assay conditions, His colonies were apparent at 25 Cdc13p responsible for this interaction. The single-stranded or 30°C in Y190 harboring plasmids pAS-CDC13 and pACT- TG -binding domain of Cdc13p is within amino acids 451–693, 1–3 STN1. This result was consistent with the previous report that and binding of the single-stranded TG is specific. However, 1–3 Cdc13p interacts with Stn1p (37). His colonies were also binding to telomeres by Cdc13p is not sufficient to account for apparent at 25 or 30°C in Y190 harboring plasmids pAS- the essential function of Cdc13p. Judging from the results the CDC13(252–924) and pACT-STN1 indicating that C-terminal 673-amino-acid polypeptide was sufficient to Cdc13(252–924)p was capable of interacting with Stn1p. We complement the growth defect phenotype of several cdc13 also examined if Cdc13-1p, with Pro371 being replaced by Ser mutants and interaction with Stn1p. Our results indicated that (24,25), might interact with Stn1p. Interestingly, the HIS3 the Stn1p-interaction function and the telomere-binding expression level in Y190/pACT-STN1 carrying pAS-CDC13-1 activity of Cdc13p were essential for cell growth. cells was comparable to those expressing either Cdc13p or Upon proteolytic degradation of the Cdc13p–DNA complex, Cdc13(252–924)p at 25°C (Fig. 6, top panel). However, the a fragment of Cdc13p that covers amino acids 557 to ∼690 was HIS3 expression level of cells carrying Cdc13-1p was shown to associate with single-stranded TG DNA (42). In 1–3 relatively low at 30°C (Fig. 6, bottom panel). Similarly, using this report, our strategy to identify the DNA-binding region of β-galactosidase as the reporter system, the color-forming Cdc13p was the subcloning of restriction enzyme-digested ability of Cdc13-1p was reproducibly reduced, as compared CDC13 fragments followed by evaluation of the single- with that of Cdc13p at 30 or 37°C (data not shown). These stranded TG -binding activities of these expressed CDC13 1–3 results indicated that Cdc13(252–924)p was capable of inter- fragments. Using this approach, the smallest fragment that still acting with Stn1p. Moreover, the interaction of Stn1p with contained the single-stranded TG -binding activities of 1–3 Cdc13-1p appeared reduced at higher temperature, suggesting Cdc13p was within amino acids 451–693 (Fig. 1). Smaller that the temperature-sensitive lethality phenotype of cdc13-1 fragments such as Cdc13(510–693), Cdc13(451–600) or might be due to a decrease in interaction with Stn1p. Cdc13(601–693) did not show detectable single-stranded TG 1–3 To address the possibility that reduced interaction between binding activity. It is unclear whether Cdc13(510–693)p, Cdc13-1p and Stn1p was due to the reduced stability of Cdc13-1p, which covers amino acids 557 to ∼690, did not interact with the Cdc13-1p level at non-permissive temperature was evaluated. single-stranded TG DNA. Nevertheless, our identification of 1–3 4740 Nucleic Acids Research, 2000, Vol. 28, No. 23 Cdc13(451–693)p as a stably expressed telomeric-binding A Pro371 to Ser substitution caused the phenotype of Cdc13-1p domain of Cdc13p provides useful information for future (24,25). On the basis of gel mobility-shift assay, the cdc13-1 mutation did not affect the telomere-binding activity of understanding of how Cdc13p binds and modulates telomere Cdc13p [J.-J.Lin and V.A.Zakian, unpublished result (24,42)]. function. In our two-hybrid system, the interaction of Cdc13-1p with The identity of the second migration bands using TG30 or Stn1p was temperature dependent and the interaction was TG35 as DNA substrate is unclear (Fig. 2). The simplest reduced at the non-permissive temperature (Fig. 6). This result explanation for the appearance of this second migration band indicated that interaction between Cdc13p and Stn1p is essential would be that TG30 or TG35 provides enough space for two for cell survival. While we cannot rule out that another unchar- molecules of Cdc13(451–693)p to bind. This result would also acterized change in Cdc13-1p is responsible for the cell cycle imply that the optimal binding site of Cdc13(451–693)p on arrest at the non-permissive temperature, relatively weak inter- DNA was ∼13–15 bases, an estimation that is in reasonable action between Cdc13-1p and Stn1p might indeed cause the agreement with the results of using a 34-kDa Cdc13 DBD by accumulation of single-stranded G-rich DNA near telomeres Huges et al. (42). However, this explanation is complicated by (35) leading to this phenotype. observations from several reports that telomeric DNA-binding proteins can also promote the formation of a G-quartet struc- ture (43,44). It will be interesting to know whether Cdc13p ACKNOWLEDGEMENTS promotes the formation of a G-quartet structure upon binding. We thank members of J.-J.L.’s laboratory for their help. We also Contrary to the ciliate telomere-binding proteins, both thank Drs W. J. Lin and C. Wang for critical reading of the manu- Cdc13p and Cdc13(451–693)p preferred low salt for binding script. This work was supported by NSC grants 87-2314-B-010- (data not shown). Sequence analysis of this 243-amino-acid 004, 88-2314-010-070 and in part by grant 89-B-FA22-2-4 region did not provide information on which residues within (Program for Promoting Academic Excellence of Universities). this region are responsible for interacting with telomeres. Evidence was presented that single-stranded DNA-binding proteins interact with DNA via hydrophobic interactions REFERENCES between aromatic side chains of the protein and the DNA bases 1. Zakian,V.A. (1995) Science, 270, 1601–1607. (45–47). For example, the single-stranded DNA-binding 2. Blackburn,E.H. (1990) J. Biol. Chem., 265, 5919–5921. protein T4 gp32 utilizes residues Tyr84, 99, 106, 115, 137, 186 3. Gottschling,D.E., Aparicio,O.M., Billington,B.L. and Zakian,V.A. (1990) and Phe138 to form a hydrophobic pocket to interact with the Cell, 63, 751–762. bases of single-stranded DNA (46). In addition, the Oxytricha 4. Klein,F., Laroche,T., Cardenas,M.E., Hofmann,J.F.X., Schweizer,D. and Gasser,S.M. (1992) J. Cell Biol., 117, 935–948. α and β complex uses a series of aromatic amino acids 5. Henderson,E.R. and Blackburn,E.H. (1989) Mol. Cell. Biol., 9, 345–348. including Tyr130α, Tyr293α,Tyr134β, His292α,Phe109β 6. Klobutcher,L.A., Swanton,M.T., Donini,P. and Prescott,D.M. (1981) and Tyr239α to interact with the extended bases of single- Proc. Natl Acad. Sci. USA, 78, 3015–3019. stranded T G DNA (47). It is likely that aromatic residues 7. Pluta,A.F., Kaine,B.P. and Spear,B.B. (1982) Nucleic Acids Res., 10, 4 4 8145–8154. within the 243 amino acids of Cdc13p play a role in the inter- 8. Wellinger,R.J., Wolf,A.J. and Zakian,V.A. (1993) Cell, 72, 51–60. action with telomeres. Interestingly, this region indeed has a 9. Wellinger,R.J., Ethier,K., Labrecque,P. and Zakian,V.A. (1996) Cell, 85, high content of aromatic residues. However, it remains to be 423–433. determined which amino acid residues are responsible for this 10. Shore,D. (1997) Trends Biochem. Sci., 22, 233–235. protein–DNA interaction. 11. Conrad,M.N., Wright,J.H., Wolf,A.J. and Zakian,V.A. (1990) Cell, 63, Using the one-hybrid system, Bourns et al. (34) showed 739–750. 12. Cooper,J.P., Nimmo,E.R., Allshire,R.C. and Cech,T.R. (1997) Nature, previously that the N-terminal 251 amino acids of Cdc13p inter- 385, 744–747. acted with telomere, but that amino acids 252–508 or 509–924 of 13. Bilaud,T., Brun,C., Ancelin,K., Koering,C.E., Laroche,T. and Gilson,E. Cdc13p failed to do so. Since the N-terminal 251 amino acids (1997) Nature Genet., 17, 236–239. were sufficient to target Cdc13p to telomere, it was concluded 14. Broccoli,D., Smogorzewska,A., Chong,L. and de Lange,T. (1997) Nature that the telomere-binding domain was within this region of Genet., 17, 231–235. 15. Chong,L., van Steensel,B., Broccoli,D., Erdjument-Bromage,H., Cdc13p. In contrast to their study, we showed that the Hanish,J., Tempst,P. and de Lange,T. (1995) Science, 270, 1663–1667. telomere-binding domain was mapped to within the 451–693 16. van Steensel,B., Smogorzewska,A. and de Lange,T. (1998) Cell, 92, 401–413. amino acids of Cdc13p both in vitro and in vivo (Figs 1 and 4). 17. Wright,J.H. and Zakian,V.A. (1995) Nucleic Acids Res., 23, 1454–1460. Given that the N-terminal 1–251 amino acids of Cdc13p did 18. Wright,J.H., Gottschling,D.E. and Zakian,V.A. (1992) Genes Dev., 6, not bind to single-stranded TG DNA in vitro (Fig. 1), it 197–210. 1–3 19. Kyrion,G., Boakye,K.A. and Lustig,A.J. (1992) Mol. Cell. Biol., 12, would be interesting to know how this region of Cdc13p inter- 5159–5173. acts with telomeres in vivo. Clearly, in vivo one-hybrid system 20. Lustig,A.J., Kurtz,S. and Shore,D. (1990) Science, 250, 549–553. detected not only proteins that interact directly with telomeric 21. Stavenhagen,J.B. and Zakian,V.A. (1998) Genes Dev., 12, 3044–3058. DNA but also those that interact indirectly (34). For example, 22. Lin,J.-J. (1993) Bioessays, 15, 555–557. Sir and Rif proteins interact with telomeres by their ability to 23. Lin,J.-J. and Zakian,V.A. (1994) Nucleic Acids Res., 22, 4906–4913. bind Rap1p (48–50). One possible explanation is that this N- 24. Nugent,C.I., Hughes,T.R., Lue,N.F. and Lundblad,V. (1996) Science, 274, 249–252. terminal 251-amino-acid fragment interacts indirectly with the 25. Lin,J.-J. and Zakian,V.A. (1996) Proc. Natl Acad. Sci. USA, 93, telomere. Conceivably, protein factors that associated with 13760–13765. telomeres including Rap1p, Rif proteins and Sir proteins, are 26. Petracek,M.E., Konkel,L.M.C., Kable,M.L. and Berman,J. (1998) potential candidates for recruiting the N-terminal 251-amino-acid EMBO J., 13, 3648–3658. polypeptide of Cdc13p to telomeres. 27. McKay,S.J. and Cooke,H. (1992) Nucleic Acids Res., 20, 6461–6464. Nucleic Acids Research, 2000, Vol. 28, No. 23 4741 28. Konkel,L.M.C., Enomoto,S., Chamberlain,E.M., McCune-Zierath,P., 40. Chevray,P.M. and Nathans,D. (1992) Proc. Natl Acad. Sci. USA, 89, Iyadurai,S.J.P. and Berman,J. (1995) Proc. Natl Acad. Sci. USA, 92, 5789–5793. 5558–5562. 41. Harper,J.W., Adami,G.R., Wei,N., Keyomarsi,K. and Elledge,S.J. (1993) 29. Ishikawa,F., Matunis,M.J., Dreyfuss,G. and Cech,T.R. (1993) Mol. Cell. Biol., Cell, 75, 805–816. 13, 4301–4310. 42. Hughes,T.R., Weilbaecher,R.G., Walterscheid,M. and Lundblad,V. 30. Virta-Pearlman,V., Morris,D.K. and Lundblad,V. (1996) Genes Dev., 10, (2000) Proc. Natl Acad. Sci. USA, 97, 6457–6462. 3094–3104. 43. Fang,G. and Cech,T.R. (1993) Cell, 74, 875–885. 31. Fang,G., Gray,J.T. and Cech,T.R. (1993) Genes Dev., 7, 870–882. 44. Giraldo,R. and Rhodes,D. (1998) EMBO J., 13, 2411–2420. 32. Fang,G. and Cech,T.R. (1993) Proc. Natl Acad. Sci. USA, 90, 6056–6060. 45. Venkataram Prasad,B.V. and Chiu,W. (1987) J. Mol. Biol., 193, 579–584. 33. Gottschling,D.E. and Zakian,V.A. (1986) Cell, 47, 195–205. 46. Shamoo,Y., Friedman,A.M., Parsons,M.R., Konigsberg,W.H. and 34. Bourns,B.D., Alexander,M.K., Smith,A.M. and Zakian,V.A. (1998) Steitz,T.A. (1995) Nature, 376, 362–366. Mol. Cell. Biol., 18, 5600–5608. 47. Horvath,M.P., Schweiker,V.L., Bevilacqua,J.M., Ruggles,J.A. and 35. Garvik,B., Carson,M. and Hartwell,L. (1995) Mol. Cell. Biol., 15, 6128–6138. Schultz,S.C. (1998) Cell, 95, 963–974. 36. Weinert,T.A. and Hartwell,L.H. (1988) Science, 241, 317–322. 48. Moretti,P., Freeman,K., Coodly,L. and Shore,D. (1994) Genes Dev., 8, 37. Grandin,N., Reed,S.I. and Charbonneau,M. (1997) Genes Dev., 11, 512–527. 2257–2269. 38. Sikorski,R.S. and Hieter,P. (1989) Genetics, 122, 19–27. 49. Hardy,C.F., Sussel,L. and Shore,D. (1992) Genes Dev., 6, 801–814. 39. Gyuris,J., Golemis,E., Chertkov,H. and Brent,R. (1993) Cell, 75, 791–803. 50. Wotton,D. and Shore,D. (1997) Genes Dev., 11, 748–760.

Journal

Nucleic Acids Research – Oxford University Press

Published: Dec 1, 2000

There are no references for this article.