ABSTRACT
A gel mobility shift assay was developed to examine recognition of yeast telomeres by telomerase. An RNase-sensitive G-rich strand-specific binding activity can be detected in partially purified yeast telomerase fractions. The binding activity was attributed to telomerase, because it co-purifies with TLC1 RNA and telomerase activity over three different chromatographic steps and because the complex co-migrates with TLC1 RNA when subjected to electrophoresis through native gels. Analysis of the binding specificity of yeast telomerase indicates that it recognizes the G-rich strand of yeast telomeres with high affinity and specificity. The Kd for the interaction is ~3 nM. Single-stranded G-rich telomeres from other species, such as human and Tetrahymena, though capable of being extended by yeast telomerase in polymerization assays at high concentrations, bind the enzyme with at least 100-fold lower affinities. The ability of a sequence to be bound tightly by yeast telomerase in vitro correlates with its ability to seed telomere formation in vivo. The implications of these findings for regulation of telomerase activity are discussed.
Telomerase is a ribonucleoprotein (RNP) that is responsible for the synthesis of the dG-rich strand of telomere terminal repeats (1-3). The enzyme was initially identified in Tetrahymena thermophila, using an assay that exploited the ability of telomerase to extend telomere-like oligodeoxynucleotides in the presence of dGTP and dTTP (4). Telomerase was subsequently found to be an unusual reverse transcriptase containing an integral RNA component, a small segment of which acts as the template for synthesis of the dGT-rich strand of telomeric repeats (5). Some telomerase can extend primers in a processive fashion, adding many copies of the telomeric repeats before dissociating (6,7). Telomerase can also cleave the input primer under certain conditions (8-10), a property that is shared by a number of DNA and RNA polymerases.
Telomerase activity has been detected in a wide range of organisms, including protozoa (3), yeast (8,11-13), mouse (14), Xenopus (15) and human (7). Genes encoding the RNA component of the enzyme complex have been cloned for many telomerases, such as that of yeast and human (16,17). Recently some of the polypeptide components of telomerase were cloned. In particular, a yeast protein known as Est2p and homologs of Est2p in Euplotes aediculatus, human and Schizosaccharomyces pombe were shown to be the catalytic components of the respective telomerases (18-21); these polypeptides exhibit significant homology to other reverse transcriptases and mutations that alter Est2p residues that are conserved among reverse transcriptases abolish telomerase activity in vitro and telomerase function in vivo. In addition, two polypeptides, p80 and p95, that co-purify with Tetrahymena telomerase have been cloned and been shown to interact with telomerase RNA and the DNA primer respectively (22). Mouse and human homologs of p80 have also been identified and been shown to associate with the respective telomerases (23,24). These recent developments should greatly facilitate structure-function analysis of telomerase.
How telomerase recognizes the ends of chromosomes in vivo is not clearly understood. In vitro at high primer concentrations telomerase appears capable of extending promiscuously all oligonucleotides that have a high percentage of dG residues. As few as 3 nucleotides at the 3'-end of the primer need to form a hybrid with telomerase RNA to allow polymerization to proceed (25). In the case of ciliated protozoa even AT-rich non-telomeric oligonucleotides can support primer extension efficiently (26). These findings have led to the suggestion that telomerase does not bind telomeres with high sequence specificity. However, only in the case of Tetrahymena was binding of telomerase to primers examined in a direct assay and the results appear consistent with a loose sequence preference for telomerase (27).
To test the generality of promiscuous binding by telomerase we developed a gel mobility shift assay for the enzyme from Saccharomyces cerevisiae. Although yeast telomerase was not purified, an RNA-dependent complex that contained the yeast telomerase RNA (TLC1 RNA) could be readily identified using active fractions and was attributed to telomerase-primer interaction. Surprisingly, analysis of the binding specificity of yeast telomerase indicates that it recognizes the dG-rich strand of yeast telomeres with high affinity and specificity. In addition, high affinity binding to a primer does not correlate with the primer's ability to support extension. However, high affinity binding appears to correlate with the primer's ability to seed telomere formation in vivo. The implications of these results for telomerase function and regulation of telomerase activity are discussed.
The haploid S.cerevisiae strain DG338 (a gift of D.Garfinkel, National Cancer Institute) was used for derivation of active telomerase. The oligonucleotides used are listed in Table 1.
Table 1.
Derivation of whole cell extracts and the active DEAE fractions was as described elsewhere (9,28). The telomerase activity was further purified on a heparin column, a blue column, a Q column and a Phenyl-Sepharose column as follows. Telomerase-containing DEAE fractions were pooled, diluted with TMG-10(0) (28) to a conductivity of 30 mSi, loaded onto an Affi-Gel Heparin Gel (BioRad) column and the activity eluted with 2 column vol TMG-10(700). Telomerase-containing fractions were pooled and loaded directly onto an Affi-Gel Blue Gel (BioRad) column. The column was washed successively with 2 column vol each TMG-10(500) and TMG-10(1000) and the activity eluted with 2 column vol TMG-10(1000), 1% Triton X-100. Telomerase-containing fractions were pooled, diluted with TMG-10(0) to a conductivity of 40 mSi, loaded onto a Bio-Scale Q (BioRad) column and the activity eluted with 10 column vol of a gradient from TMG-10(400) to TMG-10(1500). Telomerase-containing fractions (conductivity ~110 mSi) were pooled and loaded directly onto a Phenyl-Sepharose (Pharmacia) column. The column was washed successively with 2 column vol each of TMG-10(500) and TMG-10(0) and the activity eluted with 2 column vol TMG-10(0), 1% triton X-100.
For the glycerol gradient fractionation active telomerase from the DEAE column was concentrated 10-fold by use of Centricon-30 (Amicon) and applied to a 20-50% glycerol gradient in TMG-10(200). The gradient was spun at 35 000 r.p.m. at 4°C for 20 h and 15 equal volume fractions collected. The positions of Blue Dextran 2000 and ribosomal particles run in a parallel gradient were used as molecular size standards.
Primer extension assays were carried out using 5 µM primer oligodeoxynucleotides as described elsewhere (28). For quantification of activity the signals from all labeled and RNase-sensitive products (including those that are shorter than the input primer) are summed.
DNA primers used for gel mobility shift experiments were radiolabeled with T4 polynucleotide kinase and purified over a Nick Column (Pharmacia). Complex formation between yeast telomerase and DNA primer was monitored using a previously described gel mobility shift protocol (29). Briefly, partially purified telomerase was incubated with 5 nM (0.4 ng) labeled primer in 16 µl buffer containing 10 mM Tris-HCl, pH 8.0, 2 mM MgCl2, 12% glycerol and 50 ng poly(dI·dC). Where indicated, cold competitor oligonucleotides were added before addition of telomerase. The binding was carried out at 4°C for 20 min and complex formation monitored by electrophoresis through a 4% polyacrylamide gel with running buffer containing 25 mM Tris-HCl, pH 8.3, 190 mM glycine, 5 mM MgCl2 and 1 mM EDTA. The gel was cast in running buffer supplemented with 10% (v/v) glycerol and 0.5 mM DTT.
To follow migration of TLC1 RNA in native gels the polyacrylamide gel used in the mobility shift assay was supplemented with 0.5% agarose and used for separation of the complexes. Following electrophoresis the part of the gel to be analyzed for localization of TLC1 RNA was rinsed briefly in 75 mM Tris-glycine and soaked in 50% urea, 25 mM Tris-glycine, 0.5 mM EDTA with gentle shaking for 30 min. The nucleic acids within the gel were transferred to Hybond-N membrane in 6.0 mM trisodium citrate, 8 mM sodium phosphate (dibasic) at 250 mA for 16 h at 4°C (30). The blot was then probed with a labeled TLC1 DNA fragment using standard protocols.
For derivation of active yeast telomerase we fractionated crude extracts over a DEAE-agarose column, as described by Cohn and Blackburn (9). A high salt eluate from this column has a robust RNase-sensitive polymerization activity that is dependent upon the RNA component of yeast telomerase, TLC1 (9,28). Preparation of the fraction from a strain harboring point mutations in the TLC1 template region yielded an activity that has the expected property for nucleotide utilization; substitution of a GG dinucleotide for an AC dinucleotide in the RNA template resulted in a fraction that incorporates dCMP in addition to dGMP and dTMP (28). Because this fraction appears largely devoid of other contaminating activities, we have used it for further purification and for subsequent biochemical assays.
To gain insights into the mechanisms of complex formation we investigated the binding reaction as a function of divalent metal ion and salt concentrations. Stable binding of yeast telomerase to the telomere required moderate concentrations of magnesium in the gel. As shown in Figure 3A, when the gel and running buffer contained 2-5 mM magnesium and 1 mM EDTA an RNase-sensitive complex can be visualized (bottom three panels). However, reducing the magnesium concentration in the gel to 1 mM while retaining 1 mM EDTA resulted in disappearance of the complex. The magnesium requirement can be satisfied by manganese and calcium, but not by zinc (Fig. 3A, top three panels). When EDTA was omitted from the gel and running buffer, a lower concentration of magnesium (1 mM) was sufficient for complex formation (data not shown). In contrast to the divalent cation requirement in the gel, complex formation was unaffected by the concentration of divalent cations in the initial incubation mixture (as indicated at the top of each panel). Interestingly, the complex visualized in the presence of 5 mM magnesium is considerably more heterogeneous than that in 2 mM magnesium, suggesting that the complex may adopt a greater number of conformations at higher magnesium concentrations.
We tested the sequence specificity of binding by yeast telomerase using a variety of dG-rich oligonucleotides (Table 1). Single-stranded oligonucleotides containing telomere repeats from S.pombe (PTEL16), Tetrahymena (TETRA1 and TETRA2), Oxytrica (OXYT1), Arabidopsis (ARAB1) and human (HS1 and HS2) were used as competitors in the gel mobility shift assay and the affinity of these oligonucleotides relative to a yeast oligonucleotide (TEL15) estimated by the amount required for half-maximal competition. As shown in Figure 4 (top panels) and Table 2, none of the telomeres derived from other species competed significantly for binding, even when present at 100-fold molar excess. Since S.cerevisiae telomeres contain a number of degenerate repeats (e.g. TG, TGG or TGGG), we also tested these different repeats in the competition assay. As shown in Figure 4 (bottom panel) and Table 2, the dinucleotide repeat (TG)n competed well for telomerase binding. In contrast, primers made up of the trinucleotide repeat d(TGG) [(TGG)n], the tetranucleotide repeat d(TGGG) [(TGGG)n] and a repeat with dA insertions [(TAG1-3)n] had much lower affinities for telomerase. We conclude that yeast telomerase exhibits stringent sequence specificity for its own irregular telomeric repeats and for the dinucleotide repeat (TG)n. This finding is supported by another study showing that selected point mutations in a canonical yeast telomere oligonucleotide can also drastically reduce its binding affinity for telomerase (28).
To determine if the binding specificity of yeast telomerase correlates with polymerization specificity we subjected the same set of oligonucleotides to the standard primer extension assay. Unexpectedly, non-yeast primers were found to support greater polymerization by yeast telomerase in general. For example, under the standard reaction conditions the human HS1 primer was 11 times better and the Oxytrica OXYT1 primer 26 times better than the yeast primer TEL15 (Fig. 5A and Table 2). Greater polymerization was also observed for oligonucleotides corresponding to two sub-elements of the yeast telomeric repeats [(TGG)n and (TGGG)n; Table 2]. To confirm that the signals observed were entirely due to telomerase we analyzed incorporation of additional nucleotides by the polymerization activity. As expected, little dCMP incorporation was observed with a non-yeast primer (OXYT1) when telomerase, derived from a TLC1 strain was used. Mutating two of the template residues of TLC1 RNA from r(CA) to r(GG), however, led to incorporation of dCMP, as expected (TLC1-HaeIII fraction; Fig. 5B). This result indicates that the DNA synthesis supported by the non-yeast primers is unlikely to be due to a contaminating DNA polymerase. Thus for the yeast enzyme the primer binding specificity does not match primer elongation specificity, at least in vitro.
Because the bulk of binding and elongation studies have been done using partially purified yeast telomerase, we questioned whether the presence of auxiliary factors in the fraction may have altered the properties of yeast telomerase. Perhaps the properties of yeast telomerase can be attributed to some loosely bound or entirely extrinsic accessory factors. To test this hypothesis we further purified yeast telomerase on four columns and subjected the resulting fraction to binding and elongation analysis. As shown in Figure 6, extensively purified yeast telomerase (Phenyl-Sepharose fraction) continued to exhibit high sequence specificity for binding and a loose sequence requirement for polymerization. For example, primers with the telomeric repeat from Oxytrica (OXYT1) continued to be bound poorly by yeast telomerase (Fig. 6A). Non-yeast primers, however, continued to serve as good substrates for elongation by telomerase (Fig. 6B). We conclude that the binding and polymerization properties of yeast telomerase as determined using partially purified fractions are likely due to a tightly associated subunit or subunits.
We have shown that partially purified yeast telomerase RNP recognizes the dG-rich strand of the yeast telomere terminal repeat with high affinity and specificity. Although the enzyme preparation used for binding assays was not homogeneous, sequence-specific recognition is likely due to a tightly associated subunit or subunits rather than some loosely associated auxiliary factor. First, even extensively purified telomerase exhibits similar sequence specificity. Second, complex formation can be observed in 1.1 M sodium acetate, a salt concentration that is often sufficient to dissociate loosely bound factors. Finally, the mobility of the complex is similar to that of 1 kb double-stranded DNA (Fig. 1A) and only slightly lower than naked TLC1 RNA (data not shown), suggesting that the complex does not contain many polypeptides in addition to the essential ones.
The subunits of yeast telomerase responsible for high affinity and sequence-specific binding are not known. Because formation of the particular complex is sensitive to both RNase and proteinase K pretreatment, high affinity binding probably involves both an RNA and a protein component(s). Earlier crosslinking studies on the Euplotes telomerase suggest that both the protein catalytic subunit (Est2p homolog) and telomerase RNA come into close contact with the upstream region of the DNA primer, consistent with participation of both in primer binding (31). In addition, a putative yeast telomerase subunit known as Est1p has been shown to bind single-stranded yeast telomeres on its own, consistent with its participation in primer recognition by telomerase (32). However, the affinity of binding for Est1p alone is 250 nM, considerably higher than that of telomerase RNP. Furthermore, the sequence requirement for binding by Est1p appears to be less stringent than telomerase RNP. Thus if Est1p participates in recognition of telomeres by telomerase it must do so in concert with other components. Like Est1p, the CDC13 gene product has also been shown to bind single-stranded yeast telomeres with high affinity (33,34). However, current evidence indicates that Cdc13p is likely to be a telomere binding protein rather than a telomerase component in vivo.
The divalent cation requirement for detection of a yeast telomerase-telomere complex suggests that these cations either participate directly in binding or are required to maintain the correct conformation of the RNP. Indeed, a number of large RNAs have been shown to require magnesium for proper folding (35). In contrast to the recognition properties of yeast telomerase, the Tetrahymena enzyme has been shown to bind a number of dG-rich oligonucleotides with distinct sequence repeats; in addition to d(TTGGGG)3, which contains the canonical Tetrahymena telomere repeat, the Tetrahymena enzyme can also bind d(TTAGGG)3, d(TG)9 and d(TGTGTGGG)2TG with high affinity (27). In addition, stable binding of Tetrahymena telomerase to primer oligonucleotides can be observed even in the absence of divalent cations and in the presence of 0.2 mM EDTA (27). Perhaps the ciliate telomerase RNAs, being smaller, do not need magnesium for proper folding. Alternatively, the molecular mechanisms of binding may be significantly different between the yeast and the ciliate telomerases. That ciliate telomerase should have evolved a different primer recognition mechanism is not at all surprising in the light of its role in the de novo synthesis of telomeres during the normal life cycle of the organism. In contrast, there is no need for yeast telomerase to extend non-homologous telomeric ends or non-telomeric ends for normal chromosome maintenance.
The ability of various GT-rich sequence repeats to serve as the seed for telomere formation in yeast has been carefully examined in two earlier studies (36,37). When these in vivo studies are compared with our in vitro assays a remarkable correlation can be discerned between the ability of a sequence to be bound by telomerase and its ability to contribute to telomere healing. For example, the dinucleotide repeat oligo (TG)n is bound with high affinity by the enzyme and a comparable repeat directs high efficiency healing in vivo. In addition, the trinucleotide repeat oligo (TGG)n, the tetranucleotide repeat oligo (TGGG)n and the irregular repeat oligo (TAG 1-3)n are weakly bound by telomerase in vitro and comparable repeats are unable to promote telomere healing in vivo (Fig. 4, bottom panel and Table 2). In contrast to the binding results, the ability of a sequence to prime DNA synthesis in vitro does not correlate with the ability to direct telomere healing. All four primers tested in this series supported significant polymerization by yeast telomerase in vitro (Table 2). Thus the binding assay appears to more accurately reflect the relative activity of different sequences for telomerase extension in vivo than the standard polymerization assay. Perhaps in vivo the binding of telomeric ends by telomerase is a limiting step for extension. Similar phenomena may occur in humans, where seeding of telomeres in vivo has been shown to require homologous telomeric repeats (38). In addition to binding by telomerase, the ability of a sequence to be recognized by telomere binding proteins may also contribute to telomere healing (36-38). However, because the (TG)n oligo lacks a binding site for RAP1, the major yeast telomere binding protein yet can support telomere formation at high efficiency, the presence of a binding site for RAP1 is not necessary for telomere seeding in vivo.
The results presented in this paper suggest that yeast telomerase has rather high affinity and sequence specificity for the dG-rich tails of yeast telomeres in vivo. Why should this be necessary? At least four possibilities should be considered. First, the effective concentration of telomeric tails in vivo may be extremely low, for example because of masking by other proteins. Thus a telomerase with high affinity for the tail may be required just to capture the tail prior to extension. Second, the ability of telomerase to discriminate against non-telomeric tails may prevent aberrant healing of chromosomes following breakage, which could lead to large scale deletions. Third, high affinity binding may be required for negative regulation of telomerase activity. As shown in this paper and in another study (28), primers that bind tightly to yeast telomerase tend to support less extension in general, suggesting that tight binding inhibits polymerization. This inhibition may be necessary if telomerase were to bind constitutively to the telomeric end. Finally, telomerase may participate in some telomere functions other than polymerization, such as silencing or capping (16). Binding of telomerase to telomeres may thus be necessary for such additional functions.
We thank Beate Schwer for careful reading of any comments on the manuscript. This work was supported in part by a Basil O'Connor Starter Scholar Award.
Nucleic Acids Research
Pages
Introduction
Materials And Methods
Strain and oligonucleotides
Purification of S.cerevisiae telomerase
Primer extension assay
Gel mobility shift assay
RNA analysis
Results
Identification of a yeast telomerase-telomere complex
Requirements for formation and characteristics of the telomerase-telomere complex
Species-specific recognition of telomeres by yeast telomerase
Lack of correlation between binding and extension by yeast telomerase
Sequence-specific recognition is the property of a tightly associated subunit(s) of yeast telomerase
Discussion
Acknowledgements
References
ACKNOWLEDGEMENTS
REFERENCES
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 27 Feb 1998
Copyright© Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  N. F. Lue, D. Bosoy, T. J. Moriarty, C. Autexier, B. Altman, and S. Leng Telomerase can act as a template- and RNA-independent terminal transferase PNAS, July 12, 2005; 102(28): 9778 - 9783. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Bosoy and N. F. Lue Yeast telomerase is capable of limited repeat addition processivity Nucleic Acids Res., January 2, 2004; 32(1): 93 - 101. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. F. Lue, Y.-C. Lin, and I. S. Mian A Conserved Telomerase Motif within the Catalytic Domain of Telomerase Reverse Transcriptase Is Specifically Required for Repeat Addition Processivity Mol. Cell. Biol., December 1, 2003; 23(23): 8440 - 8449. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Bosoy, Y. Peng, I. S. Mian, and N. F. Lue Conserved N-terminal Motifs of Telomerase Reverse Transcriptase Required for Ribonucleoprotein Assembly in Vivo J. Biol. Chem., January 31, 2003; 278(6): 3882 - 3890. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Hossain, S. Singh, and N. F. Lue Functional Analysis of the C-terminal Extension of Telomerase Reverse Transcriptase. A PUTATIVE "THUMB" DOMAIN J. Biol. Chem., September 20, 2002; 277(39): 36174 - 36180. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Mangahas, M. K. Alexander, L. L. Sandell, and V. A. Zakian Repair of Chromosome Ends after Telomere Loss in Saccharomyces Mol. Biol. Cell, December 1, 2001; 12(12): 4078 - 4089. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Bosoy and N. F. Lue Functional Analysis of Conserved Residues in the Putative "Finger" Domain of Telomerase Reverse Transcriptase J. Biol. Chem., November 30, 2001; 276(49): 46305 - 46312. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Ray and K. W. Runge Yeast telomerase appears to frequently copy the entire template in vivo Nucleic Acids Res., June 1, 2001; 29(11): 2382 - 2394. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Xia, Y. Peng, I. S. Mian, and N. F. Lue Identification of Functionally Important Domains in the N-Terminal Region of Telomerase Reverse Transcriptase Mol. Cell. Biol., July 15, 2000; 20(14): 5196 - 5207. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||