Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M407768200 on September 27, 2004

J. Biol. Chem., Vol. 279, Issue 51, 53770-53781, December 17, 2004
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/51/53770    most recent
M407768200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Rivera, M. A.
Right arrow Articles by Blackburn, E. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Rivera, M. A.
Right arrow Articles by Blackburn, E. H.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Processive Utilization of the Human Telomerase Template

LACK OF A REQUIREMENT FOR TEMPLATE SWITCHING*

Melissa A. Rivera{ddagger} and Elizabeth H. Blackburn§

From the Department of Biochemistry and Biophysics, University of California San Francisco, California 94143-2200

Received for publication, July 9, 2004 , and in revised form, September 27, 2004.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The ribonucleoprotein telomerase is a specialized reverse transcriptase minimally composed of an RNA, TER, and a protein catalytic subunit, TERT. The TER and TERT subunits of telomerase associate to form a dimeric enzyme in several organisms, including human. A small portion of TER, the template domain, is used by telomerase for the synthesis of tandem repeats of telomeric DNA. We studied some of the requirements for processive template usage by human telomerase. A blunt-ended duplex DNA primer was not utilized by telomerase. With a duplex telomeric DNA primer, a single-stranded 3' overhang with a minimum length of ~6 bases was required for efficient priming activity. Large substitutions in the human TER templating domain did not abolish enzymatic activity, although insertion of two residues into this sequence reduced processivity, as did a template mutation that results in a mismatch between the template region used for copying DNA and the region used for alignment of the substrate primer. Finally, by using a complementary pair of catalytically inactive telomerase RNA pseudoknot mutants in combination with a marked template, we demonstrated that processive synthesis by an obligatory dimer of human telomerase does not require template switching. These results indicate that processive template usage by human telomerase, like that of Tetrahymena telomerase, is strongly dependent on the base identities in the template domain and that a dimeric human telomerase can processively utilize a single template.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Telomeres, the nucleoprotein structures that cap the ends of linear chromosomes, function as protective barriers to distinguish natural chromosome ends from double-strand breaks (for review see Refs. 1 and 2). Telomeres are replenished by telomerase in dividing cells, providing a renewable buffer against the loss of chromosomal DNA that occurs with each cell division as a consequence of the limitations of the replication machinery (for review see Ref. 3). In most eukaryotes, the G-rich telomeric DNA strand is synthesized by the ribonucleoprotein telomerase, minimally composed of an RNA, TER, and a protein catalytic subunit, TERT. The TER moiety of telomerase has several essential roles in telomerase activity (for review see Refs. 4–7). Enzymatically active human telomerase can be minimally reconstituted in rabbit reticulocyte lysates with the protein catalytic subunit, hTERT, and the RNA subunit, hTER (also called hTR).

A short sequence within hTER, the templating domain, can be subdivided into two regions: a 3' region consisting of the alignment bases that can pair with the end of a telomeric DNA primer to be extended, and a 5' region comprised of the template bases that are copied into DNA to elongate the 3' end of the primer (Fig. 1A and Table I, top). The templating domain can base pair to a fully single-stranded telomeric DNA primer or to the single-stranded 3' overhang of a DNA duplex. Telomerase from the ciliates Tetrahymena thermophila and Euplotes aediculatus requires a 3' overhang for processive addition of tandem telomeric repeats, although there is no absolute requirement for the primer substrate to be a telomeric sequence (8–14).



View larger version (44K):
[in this window]
[in a new window]
  		FIG. 1.
Human telomerase requires a short 3' overhang. A, predicted alignment of the double-stranded TAG-U substrate containing a 3' overhang of 6 nucleotides on the hTER template. B, stretch-PCR reactions using HeLa cell S100 extract with double-stranded TAG-U containing 3' overhang lengths of 0, 1, 3, 6, 9, 12, and 21 nucleotides. Reactions were performed in duplicate, and equal amounts of each reaction were loaded in two adjacent lanes. The 38 (ssDNA) lanes are reactions with the 38-mer TAG-U single-stranded substrate alone. Controls were RNase A treatment, +R (lanes 33 and 34), or no extract, Mock (lanes 35 and 36), and were loaded in duplicate. C, graph of the product signals for gel shown in B, as determined by PhosphorImager analysis (see "Materials and Methods"). D, graph of the product signals for telomerase reconstituted in vitro using rabbit reticulocyte lysate. Values were assigned as in C.

 


View this table:
[in this window]
[in a new window]
  		TABLE I
Template mutations of hTER and the corresponding products The putative template region (residues 46–51) of the WT-hTER templating domain is underlined, and the maximal alignment region (residues 52–56) is underlined with dashes. Mutated template residues and their predicted DNA complements (italics) are in red.

 
The enzymatic activity of telomerase can be significantly altered by specific base changes in the templating domain of telomerase (15–18). For example, in Tetrahymena, a single base insertion mutant, 49+C, reduces processivity (the synthesis of multiple repeats) and fidelity (the accuracy of template copying) and causes template slippage (inappropriate re-alignment of the template during synthesis) (16, 19). A single C to U mutation generates premature products and causes significant mis-incorporation (15), whereas complete replacement of the template with U residues reduced enzymatic processivity without affecting fidelity to the new template (18). Similarly, 1- or 2-base template substitutions in the Saccharomyces cerevisiae telomerase RNA cause high levels of misincorporation (20), and finally, a 3-base substitution in the S. cerevisiae telomerase RNA template domain, called tlc1–476GUG, causes complete loss of enzymatic activity (21).

Notably, in S. cerevisiae, the function of the enzymatically dead template mutant, tlc1–476GUG, is restored when wild-type RNA is present in the same RNP1 complex (21). Furthermore, the 476GUG-wild-type heterodimer of telomerase can elongate a primer specific for the tlc1–476GUG template despite the presence of the extension product of a wild-type primer, which remains bound to the enzyme following its elongation. These observations and other evidence showed that this telomerase is a dimer with two active sites and two RNAs that can functionally interact (21, 22). Subsequently, it was reported that human and Euplotes crassus telomerase also exist as dimers (23–26). In one study, a human telomerase reconstituted with an equal mixture of wild-type and mutant template hTers had a significant reduction in the expected levels of wild-type activity, indicating that, as with the S. cerevisiae telomerase, the templates were not completely independent (26).

These observations, made with the S. cerevisiae and human telomerases assembled with mutant and wild-type templates, suggested two possible models. A template switching model postulates that a dimeric telomerase switches templates between each round of template copying, perhaps as a requirement for processive synthesis. In contrast, a parallel synthesis model suggests that a dimeric telomerase catalyzes the addition of repeats onto two substrates simultaneously, resulting in the coordinate elongation of two sister chromatids in vivo (21, 22, 26).

In addition to the template domain, other regions of the telomerase RNA are critical for telomerase function, including a pseudoknot conserved in vertebrate, ciliate, and yeast telomerase RNAs (27–33). Preventing pairing of one stem of the Tetrahymena pseudoknot region disrupts the stable association of TER and TERT in vivo and abolishes the activity of telomerase assembled in vivo (34). Similarly, disruption of the human telomerase RNA (hTER) pseudoknot subdomain P3 impairs reconstitution of telomerase activity in vivo (7) and can reduce or abolish activity in vitro (35, 36). Recently, intermolecular pairing of the P3 pseudoknot region of hTER was shown to be important for the activity of the human telomerase dimer (7). A pair of inactive P3 mutants that eliminated intramolecular P3 base pairing potential could be combined in trans to create a functional heterodimeric telomerase. The activity of the trans P3 pairing appeared as efficient as restoring the P3 base pairing potential in the same RNA molecule (i.e. in cis), although activity was not restored to fully wild-type levels in either case, suggesting that the sequence as well as paired state of P3 is important for telomerase activity (7).

To further our understanding of the mechanism of human telomerase catalytic activity, we investigated a set of requirements for human telomerase template utilization. We first determined that human telomerase requires a 3' overhang of at least ~6 nucleotides to elongate telomeric duplex DNA. We show that, as with Tetrahymena telomerase, either a mismatch between the template and alignment bases in the templating domain or an elongated template reduces the ability of the enzyme to synthesize long products. Such synthesis of long products is reflective of enzyme processivity (9, 10, 37). In addition, larger changes in the template affected activity in unpredicted ways, without preventing DNA polymerization. Finally, we exploited the intermolecular P3 interaction to create an obligatory RNA heterodimer of telomerase containing two different templates, and we demonstrate that template switching is not mechanistically required for a dimeric human telomerase to synthesize long products.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Stretch-PCR Assay—In all assays used in this paper, all the primer oligonucleotides used were gel-purified prior to use to obtain a preparation containing oligonucleotide of only one length. The single-stranded 38-mer oligonucleotide TAG-U (5'-GTAAAACGACGGCCAGTTTGGGGTTGGGGTTGGGGTTG-3') (41) was boiled and then chilled on ice immediately before use to minimize secondary structure formation. Double-stranded telomerase substrates were created by mixing equimolar amounts of the TAG-U oligonucleotide with complementary oligonucleotides of different lengths (38-, 37-, 35-, 32-, 29-, 26-, and 17-mers) in 1x TKG (50 mM Tris-HCl, pH 8, 60 mM potassium glutamate), heating to 90 °C for 5 min, and slow-cooling to room temperature (13). The double-stranded substrates were then stored at -20 °C and thawed on ice (no heat treatment) immediately before addition to the assay. They were analyzed by gel electrophoresis on nondenaturing gels, using radiolabeled tracer oligonucleotides to ensure that the single-stranded oligonucleotides had all shifted to the position for annealed double-stranded oligonucleotides and that no detectable amount of oligonucleotide remained un-annealed. HeLa S100 extracts were made as described (41), with 1x hypotonic buffer containing 10 mM HEPES, pH 8, 3 mM KCl, 1 mM MgCl2, 1 mM dithiothreitol, 0.5% MEGA-9, 1x Complete EDTA-free Protease Inhibitor tablet (Roche Applied Science), and 10 units/ml RNasin. For in vivo assembled telomerase, 20 µl of HeLa S100 extract was assayed. For in vitro assembled telomerase, 1 µl of in vitro reconstituted telomerase containing 1 ng of in vitro transcribed hTER (see below) plus 19 µl of water was assayed. Reaction incubation, termination, stretch-PCR conditions with CTA-R reverse primer (5'-CAGGAAACAGCTATGACCCCTAACCCTAACCCTAACCCT-3'), and sample clean up were performed as described without the hot-start PCR step (41). Samples were mixed 1:1 in formamide loading buffer (80% formamide, 10 mM EDTA plus loading dye), denatured 2–5 min at 95 °C, and chilled before loading 10 µl on 8 M urea, 10% polyacrylamide 30 x 40-cm gels. Dried gels were exposed to a PhosphorImager (Storm, Amersham Biosciences) for signal quantitation. The signal amounts and background detected varied across different experiments; therefore, the data within a single representative experiment was quantified for the graphs, and data signals were not standardized. To compare substrate usage for the various 3' overhang lengths, the product signals across four lanes (duplicate samples with duplicate loading) within a single experiment were averaged using values from the PhosphorImager (ImageQuant version 1.2 software, Amersham Biosciences).

In Vitro Telomerase RNA Production—Mutant template and P3 hTers were constructed by site-directed mutagenesis as described (43, 44). The T7 promoter was PCR-amplified upstream of the wild-type or mutant hTER genes utilizing the primers T7p-hTR (5'-GGATCCTAATACGACTCACTATAGGGTTGCGGAGGGTGGGCCTG-3') and hTR-451a-EcoRI (5'-GCTGAATTCGCATGTGTGAGCCGAGTCCTG-3'). PCR products were cloned into the BamHI and EcoRI sites of pUC18 and sequence-verified or used directly for transcription. Plasmids were linearized by EcoRI digestion, extracted with phenol/chloroform, and precipitated before RNA transcription. The T7 Megascript kit (Ambion) was used for transcription as per the manufacturer's instructions, with the addition of 20 units of RNase-Inhibitor (Roche Applied Science) and a 4–6-h incubation followed by 4% PAGE for purification.

In Vitro Telomerase Reconstitution—Reconstitution of human telomerase activity was as described previously (36). A 50-µl mixture containing 25 µl of TNT lysate (TNT® T7 Coupled Reticulocyte System, Promega), 0.5 µg of pCR3-FLAG-tagged hTERT (provided by L Harrington), 1 µl of amino acid mixture minus Met, 1 µl of amino acid mixture minus Leu, 1 µl of RNA guard (Amersham Biosciences), 2 µl of TNT Buffer, 18.5 µlofH2O, and 1 µl of T7 polymerase was dispensed in 9-µl aliquots to 1 µg and 100, 10, 1, and 0.1 ng and no RNA of the purified hTERs to a total volume of 10 µl. Reactions were incubated 2 h at 30 °C, and stored at -80 °C.

TRAP Assays—TRAP assays were performed based on previous methods using modified TS and reverse Cx primers designed to detect mutant repeats. The mutant specific primers used are as follows: U11-TS (5'-TTTCCGTCGAGCAGCAAA-3'), U11-Cx2 (5'-GTGTTTTTATTTTTATTTTTATTTTTT-3'), 49A-Cx (5'-GTGCCCTAAATCCCTAAATCCCTAAAA-3'), Wt53 (or 53A-Cx1) (5'-GTGGCCTAAGCCTAAGCCTAAGCC-3'), and 53A-Cx2 (5'-GTGGCCATAGCCATAGCCATAGCCAAA-3'). The reverse primers tested for AU5 are as follows: AU5-Cx1 (5'-GTGATATAAATATAAATATAAATATAT-3'), AU5-Cx2 (5'-GTGATATTTATATTTATATTTATATA-3'), and AU5-Cx3 (5'-GTGTATATATATATTTATAAATATATA-3'). The reverse primers tested for 51G and 53G are as follows: 53G-Cx1 (5'-GTGGGGGTAGGGGTAGGGGTAGGGGA-3'), 53G-Cx2 (5'-GTGGGGTAAGGGTAAGGGTAAGGGGA-3'), and 53G-Cx3 (5'-CTCGGGTAGGGGTAGGGGTAGGGGAA-3'). The wild-type repeat primers, TS (5'-AATCCGTCGAGCAGAGTT-3') and Cx-ext (5'-GTGCCCTTACCCTTACCCTTACCCTAA-3'), were as described previously (45, 46). TRAP reactions (50 µl) each contained 20 mM Tris-HCl, pH 8.5,1.5 mM MgCl2, 63 mM KCl, 1 mM EGTA, 0.005% Tween 20, 2 units of Taq polymerase (Roche Applied Science), 5 µg of bovine serum albumin, 50 µM dNTPs, 10 pmol of {gamma}-32P-labeled TS (or U11-TS), 10 pmol of CX-ext (or mutant specific Cx primer), and 1 µl of in vitro reconstituted telomerase lysates (see above), in which each sample differed only in the amount (at 10-fold dilutions) and type of hTER RNA added. Various annealing times and numbers of cycles of PCR were tested to detect wild-type telomerase activity in a manner reflective of the amount of hTER added to the telomerase reconstitution reactions. For linear TRAP assays, the telomerase reaction was incubated at 30 °C for 30 min followed by PCR for 20 cycles unless noted otherwise in the text. When TRAP conditions were changed, the conditions are as follows: the telomerase reactions were incubated at 30 °C for 2 h, followed by PCR for 25 cycles. PCR cycle conditions for all telomerases except for the U11 and AU5 mutants are as follows: denaturation at 94 °C for 10 s, annealing at 50 °C for 30 s, and extension at 72 °C for 30 s. For the U11 and AU5 mutants (in parallel with wild-type controls), the annealing temperature was lowered to 45 °C for U11 and 40 °C for AU5; the extension temperature was lowered to 60 °C, and the other conditions were unchanged. The PCR annealing temperature of the 49A-Cx primer was tested from 50 to 62 °C without any effect on the amplification of wild-type repeats. Negative controls were pre-digested with 5–10 ng of RNase A for 10 min. Samples and gels were prepared and exposed to a PhosphorImager, as described for the Stretch-PCR assay. The radioactivity signal amounts and background were determined for all bands. After correction for background, which was assessed by taking multiple samples in the region of the bands to be quantified, signal strengths (amount of product in the bands analyzed) were quantified, and ratios were calculated as required (see Table II footnote), using ImageQuant version 1.2 software, Amersham Biosciences.


View this table:
[in this window]
[in a new window]
  		TABLE II
Activity and processivity of obligatory-dimer telomerases containing single type or mixed (WT and 53A) templates Total enzyme levels were measured, and processivity was assessed as described under "Materials and Methods." For comparison of enzyme levels, after background subtraction for each band, all the telomerase-mediated products, from +4 repeats upwards, were added together. The total activity levels were normalized to the wild type telomerase RNA reaction levels (relative activity). The processivity measure was the ratio of the +5 repeat product band(s) to the total longer products (+6 products upwards). The activity and processivity values shown are for the synthesis of the repeats directed by the template indicated in the left column.

 

   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Telomerase Requires a 3' Overhang of ~6 Nucleotides for Elongation of a Duplex Primer—The human telomerase RNA, hTER, contains a templating domain comprised of an 11-nucleotide region, bases 46–56 (Table I, top) (35). Based on the patterns of products seen in conventional assays of human telomerase (14, 38–40) and by analogy to in vitro and in vivo studies of template usage by Tetrahymena telomerase, nucleotides 46–51 are copied into DNA (16). Also by analogy with the Tetrahymena enzyme, the 3' nucleotides of the template domain, bases 52–56, are likely used for alignment (Table I and Fig. 1A).

Human telomerase has been tested with various telomeric and non-telomeric single-stranded primer sequences of different lengths (38, 39), but not with double-stranded substrates containing 3' overhangs. Therefore, to determine the minimal overhang requirements of telomere-like primer substrates necessary to support the processive activity of human telomerase, the stretch-PCR assay was performed on telomerase assembled in vivo and in vitro. The stretch-PCR assay was developed to quantitatively determine relative telomerase levels in telomerase-positive human cells (41). The substrate primer, TAG-U, is a 38-mer oligonucleotide ending in 3.5 repeats of the Tetrahymena telomeric sequence, TTGGGG, and is utilized efficiently by human telomerase in conventional assays (38, 41). The 3' region of TAG-U is predicted to align on the template as indicated in Fig. 1A. Substrates with various 3' overhang lengths were generated by annealing complementary oligonucleotides of various lengths to TAG-U. The 3' overhang lengths assayed were 0, 1, 3, 6, 9, 12, and 21 nucleotides. The fully single-stranded DNA substrate primer of 38 nucleotides, TAG-U, was assayed in parallel for comparison.

S100 extracts of HeLa cells were used to assay human telomerase assembled in vivo. Just before use in the assays, the single-stranded TAG-U primer was boiled and then chilled on ice to prevent formation of any secondary structures. The double-stranded DNA substrates were used directly without heat treatment (see "Materials and Methods").

The observed telomerase activity indicated that the minimum 3' overhang length required for processive telomerase activity is ~6 nucleotides, with only minimal activity being detectable with 1- or 3-nucleotide 3' overhangs. Activity levels reached a plateau at 9 nucleotides (Fig. 1, B and C). Small amounts of products were detectable with double-stranded primers bearing no or very short overhangs. First, however, these products were made in low amounts compared with the products from primers with overhangs 6 nucleotides and longer. Second, they had distinct banding profiles, which allowed us to determine that such products were not produced in significant amounts in the reactions using the longer overhang primers. The modest decrease in activity with the partially double-stranded primers bearing 3' overhangs longer than 12 nucleotides may have reflected some secondary structure formation by these longer 3' extensions, as it was not possible to prevent such potential secondary structure formation by heat treatment, or to prevent any such potential structure formation during the course of the enzyme reaction.

A similar activity profile was obtained with telomerase activity reconstituted in vitro. The human telomerase RNA was transcribed in vitro, purified, and then assembled with the hTERT subunit expressed in a rabbit reticulocyte lysate (as described by Ref. 36) (Fig. 1D). Although the in vitro telomerase activity levels were not standardized relative to in vivo telomerase (see "Materials and Methods"), and cannot be directly compared, the in vitro telomerase activity levels of the various overhang lengths also reached a plateau at and above 9 nucleotides. Thus, the overall similarity between the activity profiles indicates that the 3' overhang requirement of telomerase is a property of the core telomerase RNP that is independent of in vitro or in vivo assembly.

Detection of Mutant Template Telomerase Activity—Several mutants of the templating domain of human telomerase have been made to investigate telomere and telomerase function (26, 42, 43), but few studies have addressed the effects of such template changes on the catalytic action of human telomerase. To address the role of various template residues on telomerase processivity, we performed TRAP assays on telomerase assembled and reconstituted in vitro (as described by Ref. 36) with wild-type or a variety of mutant-template hTERs. The mutants investigated (Table I) included substitutions of 2–3 bases ("53A" and "50G"), moderately large substitutions of 4 or 6 bases ("51G" and "53G"), a 2-base insertion into the template ("49+AA"), and almost complete substitutions of the template ("U11" and "AU5"). All the mutant-template telomerase RNAs tested have been shown to assemble as efficiently with hTERT as wild-type hTER (44).

For telomerase assays, new primers were designed in order to specifically detect synthesis of the repeats predicted for each template. TRAP assays were modified to use the standard TS primer (45) or a new mutant-specific telomerase primer, and a standard Cx-ext primer (46) or new reverse primer. The sets of reactions to be compared were always done with single batches of reaction mix and reticulocyte lysate preparations, to allow direct comparisons within the experiment. Various annealing times and numbers of cycles of PCR were tested to ensure that the activity assays were in the linear range (see "Materials and Methods" for final TRAP conditions chosen for the various hTER mutants). For every template analyzed, a range of PCR cycle numbers and RNA concentrations was tested to ensure that the reaction products synthesized were in the linear range. The dependence of reaction product formation upon added telomerase RNA and hTERT and RNase A sensitivity was validated in all cases, to demonstrate that product formation was attributable to telomerase enzymatic activity. Hence, we did not add an "internal PCR" control to the reactions, because the linearity and telomerase and RNA dependence were determined and controlled within any one set of reactions. The initial radiolabeled TRAP products represent the amplification of 1–4 telomeric repeats. Products smaller than 4 repeats are not distinguished by the TRAP assay because of the length (27-mer) and design of the reverse TRAP primer, Cx-ext. In addition, primer-primer interaction products ("primer-dimer products;" Figs. 3 and 4, asterisks) appeared under the conditions required for some of the less active mutant hTER telomerases tested, due to the small 3-bp overlap between the TS and Cx-ext primers. All these initial TRAP or primer-dimer products are included in all the autoradiograms shown in Figs. 3 and 4 (see lower positions in the gel lanes). In those experiments where primer-dimers appeared, they could be distinguished from telomerase-mediated products on the basis of their presence in control lanes (samples with no hTER or RNase A treatment) and the PCR cycle number (see "Materials and Methods" and "Results"). Furthermore, under the 20 cycles of PCR amplification and 30-min reaction times used in these experiments (unless indicated otherwise), the synthesis of authentic telomerase elongation products longer or equivalent to the primer-dimer products was dependent upon the combined presence of hTER RNA and hTERT cDNA in the reconstitution reaction, and synthesis of these products was RNase A-sensitive.



View larger version (49K):
[in this window]
[in a new window]
  		FIG. 3.
Processivity is reduced with a mutant template with mismatches between the alignment and template regions and with an expanded mutant template. All TRAP reactions used TS as a substrate/forward primer and the mutant-specific reverse primer indicated below for amplification. Relative activity (Rel. Act.) and relative processivity (Rel. Proc.) of mutant hTers was normalized to wild type hTER and assessed as described under "Materials and Methods." ND, not determined. A, 53A-hTer template has matched alignment and template regions and is copied processively. Lanes 1–4, products amplified using the Cx-ext reverse primer specific for wild-type repeats with WT-hTER at 10-fold dilutions (lane 1, 1 ng; lane 2, 0.1 ng; lane 3, 0.01 ng; and lane 4, 0.001 ng); lane 5, no hTER; lane 13, 53A-hTer (1 ng); and lane 14, WT-hTER (1 ng) pretreated with RNase A (R). Lanes amplified using the 53A-Cx2 mutant-specific reverse primer: lanes 6–9, 53A-hTer at 10-fold dilutions (lane 6, 1 ng; lane 7, 0.1 ng; lane 8, 0.01 ng; and lane 9, 0.001 ng); lane 10, no hTER; lane 11, WT-hTER (1 ng); and lane 12, 53A-hTer (1 ng) pretreated with RNase A. B, 50G-hTer has mismatched alignment and template regions and reduced processivity. Lanes amplified using the Cx-ext reverse primer specific for wild-type repeats: lanes 1–4, WT-hTER at 10-fold dilutions (lane 1, 1 ng; lane 2, 0.1 ng; lane 3, 0.01 ng, and lane 4, 0.001 ng); lane 5, no hTER; lane 13, 50G-hTer (1 ng); and lane 14, WT-hTER (1 ng) pretreated with RNase A. Lanes amplified using the 53A-Cx2 mutant-specific reverse primer: lanes 6–9, 50G-hTer at 10-fold dilutions (lane 6, 1 ng; lane 7, 0.1 ng; lane 8, 0.01 ng; and lane 9, 0.001 ng); lane 10, no hTER; lane 11, WT-hTER (1 ng); and lane 12, 53A-hTer (1 ng) pretreated with RNase A. Lane 15, 50G-hTer (100 ng) with adjusted TRAP conditions (see text). C, 49+AA-hTer has a longer template and reduced processivity. Lanes amplified using the Cx-ext reverse primer specific for wild-type repeats: lanes 1–5, WT-hTER at 10-fold dilutions (lane 1, 10 ng; lane 2, 1 ng; lane 3, 0.1 ng; lane 4, 0.01 ng; and lane 5, 0.001 ng); lane 11, 49+AA-hTer (10 ng); lane 12, no hTER; and lane 14, WT-hTER (10 ng) pretreated with RNase A. Lanes amplified using the 49A-Cx mutant reverse primer: lane 6, WT-hTER (10 ng); lane 7, 49+AA-hTer (100 ng) with a longer incubation and increased PCR cycles (see text); lanes 8–10, 49+AA-hTer at 10-fold dilutions (lane 8, 10 ng, lane 9, 1 ng; and lane 10, 0.1 ng); lane 13, no hTER; and lane 15, 49+AA-hTer (10 ng) pretreated with RNase A.

 



View larger version (52K):
[in this window]
[in a new window]
  		FIG. 4.
Two extensive template substitutions have base-specific effects on activity. All TRAP reactions except for U11-hTer reactions (see text) used TS as a substrate/forward primer and the reverse primer indicated below for amplification. Relative activity (Rel. Act.) and relative processivity (Rel. Proc.) of mutant hTers was normalized to wild type HTER and assessed as described under "Materials and Methods." ND, not determined. A, U11-hTer telomerase products lack any 6-base periodicity. Lanes amplified using the Cx-ext reverse primer specific for wild-type repeats: lanes 1–5, WT-hTER at 10-fold dilutions (lane 1,10ng; lane 2, 1 ng; lane 3, 0.1 ng; lane 4, 0.01 ng; and lane 5, 0.001 ng); lane 12, U11-hTer (10 ng), lane 13, no hTER; and lane 14, WT-hTER (10 ng) pretreated with RNase A. Lanes amplified using the U11-TS and U11-Cx2 mutant specific reverse primer: lane 6, WT-hTER (10 ng); lanes 7–11, U11-hTer at 10-fold dilutions (lane 7, 10 ng; lane 8, 1 ng; lane 9, 0.1 ng; lane 10, 0.01 ng, and lane 11, 0.001 ng); lane 15, no hTER; and lane 16, U11-hTer (10 ng) pretreated with RNase A. B, AU5-hTer telomerase is weakly active and retains the 6-base periodicity of products. All lanes were amplified with the TS and Cx-ext primers. Lanes 1–3, WT-hTER (lane 1, 1 ng; lane 2, 0.1 ng; and lane 3, 0.01 ng); lanes 4–9 AU5-hTer (lane 4, 100 ng; lane 5, 10 ng; lane 6, 1 ng; lane 7, 0.1 ng; lane 8, 0.01 ng; and lane 9, 0.001 ng); lane 10, no hTER; lane 11, WT-hTER (1 ng) pretreated with RNase A, and lane 12, AU5-hTer (100 ng) pretreated with RNase A. C, 51G-hTer and 53G-hTer telomerases show weak and processive activity. All lanes were amplified with the TS and Cx-ext primers. Lanes 1–4, WT-hTER (lane 1, 1 ng; lane 2, 0.1 ng; lane 3, 0.01 ng; and lane 4, 0.001 ng); lanes 5–7, 51G-hTer (lane 5, 100 ng; lane 6, 10 ng; and lane 7, 1 ng); lanes 8–10, 53G-hTer (lane 8, 100 ng; lane 9, 10 ng; and lane 10, 1 ng); lane 11, no hTER; lane 12, WT-hTER (1 ng) pretreated with RNase A; lane 13, 51G-hTer (100 ng) pretreated with RNase A, and lane 14, 53G-hTer (100 ng) pretreated with RNase A.

 
The telomerase substrate primer TS is a 18-mer non-telomeric sequence except for its last 3 nucleotides, -GTT-3', and it can potentially align on the wild-type template in two positions as illustrated in Fig. 2A (39, 45, 47). However, studies of Tetrahymena telomerase showed that the enzyme does not elongate from a primer aligned very close to the 5' end of the template unless 4 or 5 contiguous base pairs can form between the template and all 4 or 5 bases at the 3' end of the primer (14). This suggests that an alignment of TS close to the 5' end of the template is unlikely (Fig. 2A, left, upper alignment). Rather, work showing that human telomerase can elongate non-telomeric primers containing at least two of the last 4 nucleotides complementary to the template (39) suggests that the alignment of TS shown as the lower alignment in Fig. 2A, left, is more probable. This alignment would result in the initial addition of 8 dNTPs in a single round, through copying all the way to the 5' end of the template. For the majority of the mutant hTers tested here, the last three residues at the 3' end of the template are unchanged and thus the TS primer could align as illustrated for wild type (Table I and Fig. 2, A–D, left). However, the U11 mutant template is not sufficiently complementary to the TS primer. Hence, a U11-specific TS primer, U11-TS, was designed by modifying the TS primer to end in 3'-AAA (Fig. 2E). This modification was based on previous studies with non-telomeric primers that indicated that substrate base pairing to the template does not require a large amount of complementary sequence (39). Finally, TS could align on the AU5 mutant template in two ways (Table I and Fig. 2F, left).



View larger version (37K):
[in this window]
[in a new window]
  		FIG. 2.
Predicted alignment of TS substrate primers on mutant template hTers. The final 6–8 bases of TS are shown. Mutated bases of the primer (U11-TS) and template residues are in red. A, two possible alignments for TS on the WT-hTER template region. The lower alignment is favored (see text) and initially adds 8 dNTPs before the elongated TS (TS+product) realigns for the next round of synthesis to produce the expected 6-base repeat. B–D, predicted alignment of TS on the 53A-hTer (B), 50G-hTer (C), and 49+AA-hTer (D) mutant templates. Following the first round of synthesis, the TS+product made by 50G-hTer is mismatched to the template. E, two possible alignments of the U11 mutant-specific TS (U11-TS) on U11-hTer. F, two possible alignments for TS on AU5-hTer.

 
In Vitro Processivity Is Affected by Putative Alignment Residue 53—We examined the effect of the 53A and 50G template mutations on telomerase processivity (synthesis of long products). Each mutant template is predicted to produce TTTGGC repeats, assuming that the six 5'-most bases of the hTER template (46–51) are primarily used as template residues (Table I and Fig. 2, A–C). 53A has three templating domain changes at positions 47, 50, and 53, whereas 50G lacks the base change at position 53, which is within the putative alignment residues of the template. To assay the 53A or 50G template mutants, we used a reverse primer specific for TTTGGC mutant repeats, 53A-Cx2.

The 53A mutant has matched template and alignment residues, suggesting it could be processive. As anticipated, this mutant telomerase was processive, and its activity level was quite similar to wild type (Fig. 3A, compare lanes 6–9 for 53A with lanes 1–4 for WT). The banding pattern of amplification products of 53A differed from wild type. This is a likely result of the different sequence dictated by the 53A template, as will be discussed below, and possibly differences in PCR amplification with the 53A-Cx2 reverse primer. When wild-type telomerase activity was assayed by TRAP, using the 53A-Cx2 and the TS primer, no products were visible, indicating that amplification of telomerase products was mutant-specific (Fig. 3A, lane 11). Conversely, the wild-type TS and Cx-ext primers did not amplify mutant products in assays with 53A or 50G telomerase (Fig. 3A, lane 13, and data not shown).

In contrast to 53A, the 50G mutant had greatly diminished processivity (Figs. 2C and 3B). The TS primer could align on the 50G mutant template as illustrated (Fig. 2C), with telomerase initially adding eight dNTPs, starting from position 53 and continuing to the 5' end of the template RNA sequence. Following production of the initial mutant repeat, alignment of the mutant elongation product ending in -TG-3' would inhibit pairing of the complementary G nucleotide with the RNA template at position 52, due to the preceding T mismatch (Fig. 2C). Indeed, only two predominant TRAP products were visible for the 50G mutant, representing the addition of one to five repeats (1–4 repeats in the first product and 5 repeats in the second product; Fig. 3B, lanes 6–9). The range of concentrations of the WT and 50G hTERs were the same, yet for 50G, products are only faintly visible at the highest RNA concentration (Fig. 3B, lanes 1–4 for WT and lanes 6–9 for 50G). Thus, in contrast to 53A, the 50G template may not only hinder translocation but, based on the observation that larger amounts of RNA were required to detect any products, this mutant may also reduce overall telomerase activity (Fig. 3). This difference was consistent across multiple preparations of RNA (data not shown). It was not attributable to a difference in PCR amplification conditions or efficiency, because the 53A mutant produces the same repeat sequence that is amplified in the PCR step, but 53A had robust activity in this assay (compare Fig. 3, A, lanes 6–9 for 53A, and B, lanes 6–9 for 50G).

To detect any longer products made by 50G, the TRAP reaction conditions were changed (see "Materials and Methods") to allow the enzyme more time to align on a noncomplementary substrate sequence or to reveal use of a mode of synthesis not dependent on alignment. Under those conditions, although longer product bands from the 50G mutant became faintly visible, they were much weaker than the first two products, suggesting that the dominant products made by the 50G mutant enzyme still are only one to five repeats long (Fig. 3B, lane 15). Thus, the single base change difference between 50G-hTer and 53A-hTer (the residue at position 53) is sufficient to prevent processivity (Fig. 3, A, lane 6 for 53A, and B, lanes 6 and 15 for 50G). The clear similarity in activity levels of 53A and wild type provides further support that alignment within the templating domain of human telomerase is important for enzyme processivity.

Recent studies have also shown that pairing of the 3' end of the primer with certain templating domain residues in human telomerase affects enzyme processivity (40, 48). Our results extend these findings to different base changes in the templating domain. Taken together, all the results establish a role for pairing between the primer and template, for both Tetrahymena and human telomerase (16, 40, 48).

A 2-Base Insertion into the Template Decreases Processivity and Overall Product Size—Next, we examined the effect on telomerase processivity of the 49+AA template mutation, a 2-base (AA) insertion into the templating domain that expands AA (positions 48 and 49) to an AAAA sequence. This mutation was based on a similar mutation in Tetrahymena, 43AA, which synthesized an 8-base DNA repeat, TTTTGGGG in vivo (49). By analogy, the predicted repeat sequence synthesized by 49+AA is TTTTAGGG (Table I and Fig. 2D). A reverse primer, 49A-Cx, was designed to amplify this repeat. The similarity of this mutant sequence to wild type allowed the wild-type telomerase products synthesized by wild-type telomerase to be amplified with 49A-Cx, even at annealing temperatures higher than those typically used for TRAP (see "Materials and Methods," Fig. 3C, lane 6, and data not shown). Despite this crossprimer amplification, the wild-type enzyme retained its normal 6-base periodicity pattern (compare Fig. 3C, lane 6 with lanes 1–3), and the wild-type Cx-ext reverse primer did not amplify mutant repeats synthesized by the 49+AA mutant (Fig. 3C, lane 11). The processivity of the 49+AA mutant was low, and it made only two predominant TRAP products, corresponding to +1-4 and +5 repeats (Fig. 3C, lanes 8–10). This low processivity was not anticipated, because following translocation, the 49+AA mutant could align with one un-matched base, at position 56, the most 3' end of the template (Fig. 2D). Therefore, in an attempt to increase the processivity by allowing the enzyme more time for synthesis, the TRAP reaction conditions were changed (see "Materials and Methods"). Under those conditions larger products became faintly visible, but the initial two products still strongly predominated (Fig. 3C, lane 7).

We conclude that, like the 50G template, the 49+AA mutant has reduced processivity. Most interestingly, the apparent increase in product size over the wild-type repeats was not as expected, because the 49+AA +5 band and the faint predominant longer products were only one base larger than wild type. This suggests that the 49+AA mutant may not copy all the way to the 5' end of the template to produce a full 8-base repeat (Fig. 3C, compare lanes 7–10 for 49+AA with lanes 1–3 for WT). Although the 49A-Cx primer amplified products smaller than the predicted 8-base repeat, this result is unlikely to be a PCR or other artifact, because the same primer was able to accurately amplify the 6-base wild-type repeat (Fig. 3C, lane 6).

Extensive Template Substitutions Cause Mutant-specific Effects on Telomerase Activity—Finally, we examined the effects of larger template substitutions on telomerase activity. The U11 and AU5 mutant hTers substitute the templating domain with either all Us or 5 alternating AU residues (Table I and Fig. 2, E and F). The analogous mutations made within the Tetrahymena template, U9 and AUN, were enzymatically competent, despite producing non-telomeric sequences. Both Tetrahymena mutants synthesized products expected from faithful template copying, although processivity was much lower than wild type (18). Thus, the U11 and AU5 human mutants were anticipated to be functional, with U11 producing a string of A repeats and the AU5 mutant (TA)n repeats. The mutant-specific reverse primer, U11-Cx2, could potentially also form primer-dimer products as it is primarily T-rich, ending in 3'-TTT, and able to align with the 3'-AAA of U11-TS. Therefore, to favor amplification of products from the U11 mutant enzyme, the PCR conditions were changed (see "Materials and Methods"). For comparison, the wild-type telomerase was assayed, and its products were amplified under the same conditions.

The U11 mutant enzyme was highly active at U11 hTER RNA levels similar to wild-type enzyme. The detection of more U11 than wild-type products at lower RNA concentrations (Fig. 4A, WT, lanes 4–5, and U11, lanes 10 and 11) most likely represents a difference in product amplification by the mutant and wild-type primers, respectively. U11 telomerase generated the expected product pattern, a single-base periodicity ladder of bands. The Tetrahymena U9 mutant also showed a similar banding pattern with low dATP concentrations in the conventional assay (18), suggesting that in these U-rich template sequence mutants the enzyme loses its 6-base periodicity, either from incomplete repeat synthesis or various alignments of its A-rich products along the template (Fig. 2E). The mutant enzyme was not as processive as wild-type telomerase overall (Fig. 4A, compare lanes 1–5 for WT with lanes 7–11 for U11). Under the conditions used, there were no primer-dimer products for the wild-type primers (Fig. 4A, lanes 13 and 14), although the primers specific for U11 products produced a faint primer-dimer product band (Fig. 4A, lanes 15 and 16). In addition, primers specific for mutant products did not amplify wild-type template repeats from hTER, nor did wild-type primers amplify products from the U11 mutant template (Fig. 4A, WT, lane 6, and U11, lane 12).

The AU5 mutant activity was not detectable with any certainty. The last 2 bases of the TS substrate primer (-TT-3'), could align to the last two residues of the AU5 template or could align further into the template (Fig. 2F). Various reverse primers designed to detect the predicted mutant repeats and reaction conditions were tested (data not shown, see "Materials and Methods" for primer sequences). Activity in experiments reconstituting telomerase with the AU5 mutant hTer (Fig. 4B, lanes 4–9) was only detectable with the use of the wild-type Cx-ext primer at high RNA concentrations, after adjusting the TRAP conditions (cycles increased from 20 to 25 and time increased from 30 min to 2 h; see "Materials and Methods"). The only reaction products visible were identical in mobility to those of WT telomerase, suggesting that contaminating activity from WT hTER (possibly in the rabbit reticulocyte lysate) accounted for the product bands seen under these extreme and nonlinear PCR conditions shown in Fig 4B, lanes 4 and 5.

The activity of another pair of mutants, 51G and 53G, was also difficult to detect unambiguously, despite testing various reverse primer sequences designed to allow amplification of the predicted TTCCCC repeats (Table I, data not shown, see "Materials and Methods" for primer sequences). However, with TRAP conditions adjusted for higher sensitivity (see "Materials and Methods") and the use of WT primers, activity could be detected, but again the products were indistinguishable in mobility from the WT telomerase products (Fig. 4C, 51G, lanes 5–7, and 53G, lanes 8–10). However, the enhanced TRAP conditions required to detect these mutant activities were nonlinear for wild-type enzyme (compare Fig. 4C, lanes 1–4 with Fig. 3C, lanes 2–5). The notably weak activity of these mutants may reflect a difficulty in amplification because of a high amount of mismatches to Cx-ext and/or the synthesis of abnormal repeats not amplified by primers specific for mutant repeats. Alternatively, their low activity may be a direct result of the particular RNA base substitutions, as was shown for the 476GUG base substitution mutant in yeast (21).

Template Switching Is Not a Requirement for Processive Synthesis by a Dimeric Telomerase—Processive synthesis is defined for telomerase as the production of long stretches of tandem telomeric DNA repeats (37, 50, 51). Based on observations made with the S. cerevisiae and human telomerases, it was proposed that a dimeric telomerase might elongate its substrate by first copying from one template and then switching to the other template for processive synthesis (22, 26). To test whether such template switching occurs in vitro, we designed a heterodimer of telomerase containing two templates with distinguishable products. We chose the 53A mutant template, which is as processive as wild type relative to its activity and directs synthesis of repeat products with gel electrophoresis mobilities distinct from those of wild-type repeat products (see Fig. 3A).

The telomerase heterodimer was assembled using a pair of RNAs that have complementary mutations in the P3 stem of the pseudoknot of hTER. We have shown previously that in order for telomerase to function, these RNAs must dimerize (Fig. 5, A and B) (7). The P3 stem of the pseudoknot is comprised of the P3-Down strand (hTER residues 107–115) which base pairs to the P3-Up strand (hTER residues 174–183) (Fig. 5A, solid blue bars). The P3-Down-A mutation and the P3-Up-A mutation (Fig. 5A) are complete substitutions of the respective P3-Down and P3-Up strands (see solid red bars, Fig. 5A). Each mutation disrupts base pairing of the P3 pseudoknot and abolishes activity of telomerase assembled in vivo or reconstituted in vitro (7). The P3-Down-A and P3-Up-A mutations are complementary to each other, and combining them restores the potential for P3 base pairing (7). Telomerase activity was shown to be restored by making the double mutant P3-Down/Up-A within the same RNA molecule, or by equimolar mixing of the P3-Down-A RNA with P3-Up-A RNA, thus creating an obligatory heterodimer (P3-Down-A+P3-Up-A) (Fig. 5B) (7). By using the same in vitro reconstitution system described above, telomerase was assembled with each of the P3-Down-A and P3-Up-A mutations, with the P3-Down/Up-A double mutant, and the heterodimer of mixed RNAs (P3-Down-A+P3-Up-A). Telomerase activity was then assayed (Fig. 6A, lanes 12–15). We confirmed that, under our assay conditions, only the P3-Down/Up-A double mutant and the (P3-Down-A+P3-Up-A) heterodimer, which respectively restore intra- and intermolecular P3 base pairing potential, were catalytically active (Fig. 6A, lanes 14 and 15), as shown previously (7). We then constructed these P3 mutant RNAs containing the 53A template, assembled them into telomerase as above, and confirmed the copying of the 53A mutant template in these same P3 mutant contexts (Fig. 6A, lanes 16–19). These results showed that for 53A or WT activity, the base pairing in P3 is critical, i.e. either in cis, as demonstrated for P3-Down/Up-A, or in trans, as demonstrated by the (P3-Down-A+P3-Up-A) heterodimer. To achieve comparable levels of product synthesis, reconstitution of the telomerase activity of the P3 heterodimers required more total hTER RNA than used in the wild-type P3 reactions (Fig. 6A compare lanes 1–11 with 14 and 15 and 18 and 19). This was consistent with previous findings that showed that although restoring base pairing through P3 is required, P3 base composition also influences activity. Although overall enzymatic activity in the P3 mutants is reduced, assembly is unaffected, as described previously (44).



View larger version (27K):
[in this window]
[in a new window]
  		FIG. 5.
Obligatory heterodimers of telomerase formed through intermolecular base pairing of the P3 subdomain of hTER. A, schematic of WT-hTER. The P3 pseudoknot and the templating domain are highlighted in yellow. The residues of the WT P3-Up strand are represented by a blue bar with a bulge; the WT P3-Down strand residues are represented by a blue bar, and the templating domain is a black bar. Legend at right, the pseudoknot mutations, P3-Up-A and P3-Down-A, in red bars (with a bulge for the P3-Up-A mutation); 53A mutant template, black and red stripes. B, the templating domain and mutations of the P3 pseudoknot region of hTER (region highlighted yellow in A). P3-Down-A and P3-Up-A mutations of hTER each represent changes to only one strand of the P3 region and are inactive when each is assembled into telomerase. After mixing both hTER mutants together, dimerization through intermolecular pairing of the P3 pseudoknot region (center) creates a heterodimer (P3-Down-A+P3-Up-A) and restores telomerase activity. C and D, mixing of the P3 pseudoknot mutations with wild-type and 53A-mutant templates. C, the 53A_P3-Down-A mutation of hTER (53A template and P3-Down-A mutation) is mixed with the P3-Up-A mutation (with a wild-type template) to create a mixed template heterodimer (53A_P3-Down-A+P3-Up-A) that is active and produces repeats from each template. In this context the wild-type template is more active than 53A (see text and Fig. 6B). D, the converse of C. The P3-Down-A mutation of hTER (wild-type template) is mixed with 53A_P3-Up-A (53A template), and the heterodimer (P3-Down-A+53A_P3-Up-A) is active and produces wild-type and mutant repeats in similar quantities.

 



View larger version (73K):
[in this window]
[in a new window]
  		FIG. 6.
WT-hTER and 53A-hTer mutant template activity of obligatory heterodimers formed through intermolecular pairing through P3. All TRAP reactions used TS as a substrate/forward primer and the reverse primer indicated below for amplification. A, dimerization through the P3 pseudoknot region is required for activity, and the 53A-hTer template is more sensitive than the wild-type template to the P3 pseudoknot mutations. Lanes amplified using the Cx-ext reverse primer specific for wild-type repeats: lanes 1–5, WT-hTER (lane 1, 10 ng; lane 2, 1 ng; lane 3, 0.1 ng; lane 4, 0.01 ng, and lane 5, 0.001 ng); lane 6, 53A-hTer (10 ng); lanes 12–15, wild-type template (50 ng); lane 12, P3-Down-A; lane 13, P3-Up-A; and lane 14, P3-Down/Up-A with both P3 strands mutated within the same molecule (cis); lane 15 (P3-Down-A+P3-Up-A) heterodimer (refer to Fig. 5B). Lanes amplified using the 53A-Cx2 reverse primer specific for 53A repeats: lanes 7–10, 53A-hTer (lane 7,10ng; lane 8, 1 ng; lane 9, 0.1 ng; and lane 10, 0.01 ng); lane 11, WT-hTER (10 ng); lanes 16–19, 53A template (50 ng); lane 16, 53A_P3-Down-A; lane 17, 53A_P3-Up-A; lane 18, 53A_P3-Down/Up-A with both P3 strands mutated within the same molecule (cis); lane 19 (53A_P3-Down-A+53A_P3-Up-A) heterodimer. B, a mixed heterodimer of WT-hTER and 53A-hTer templates is processive, and the activity of the 53A-hTer template is reduced when the P3-Down-A mutation is on the same RNA. Lanes with a wild-type P3 region and amplified with Wt53, an indiscriminate reverse primer: lanes 1–3, WT-hTER (lane 1, 1 ng; lane 2, 0.1 ng; and lane 3, 0.01 ng); lanes 4–6, 53A-hTer (lane 4, 1 ng; lane 5, 0.1 ng; and lane 6, 0.01 ng); lane 17, no hTER; lane 18, WT-hTER (1 ng) pretreated with RNase A; lane 19, 53A-hTer (1 ng) pretreated with RNase A. Lanes 7–16, P3 mutants with WT or 53A templates (50 ng); lane 7 (WT_P3-Down-A+WT_P3-Up-A) heterodimer amplified with Cx-ext; lane 8 (53A_P3-Down-A+53A_P3-Up-A) heterodimer with two 53A templates amplified with 53A-Cx2; lanes 9–12 (53A_P3-Down-A+P3-Up-A) mixed template heterodimer (Fig. 5C); lane 9, amplified with Cx-ext; lane 10, amplified with 53A-Cx2; lane 11, amplified with Wt53; lane 12, telomerase products amplified with Cx-ext were pooled with products amplified with 53A-Cx2; lanes 13–16 (P3-Down-A+53A_P3-Up-A) mixed template heterodimer (Fig. 5D); lane 13, amplified with Cx-ext; lane 14, amplified with 53A-Cx2; lane 15, amplified with Wt53; lane 16, telomerase products amplified with Cx-ext were pooled with products amplified with 53A-Cx2. C, magnified view of longer telomerase products demonstrating the absence of mixed composition 53A/wild-type products in the template heterodimers. Shown are the telomerase products from lanes 7, 11, and 15 of B after the addition of +10 repeats. D, close-up view of the +5 and +6 repeat products in a separate gel fractionation of independent telomerase reactions performed (left to right in D) as in lanes 8, 13, 14, and 12 in B.

 
To amplify the products synthesized by the heterodimers with mixed wild-type and 53A templates, the TS substrate primer was used in combination with various reverse primers. One was an indiscriminate reverse primer, Wt53, which could efficiently amplify both wild-type and mutant repeats (Fig. 6B, lanes 1–3 for WT and lanes 4–6 for 53A). The other reverse primers used were the wild-type repeat-specific primer, Cx-ext, or the 53A mutant repeat-specific primer, 53A-Cx2. In addition, prior to gel electrophoresis, the products from amplification with Cx-ext and 53A-Cx2 were mixed after amplification for comparison to the PCR products obtained with the indiscriminate primer Wt53.

To determine whether template switching occurred in this setting, two different heterodimeric, mixed template telomerases were assembled in vitro. The (53A_P3-Down-A+WT_P3-Up-A) heterodimer contained the P3-Down-A mutation and the 53A template on the same RNA molecule, and the P3-Up-A RNA contained a wild-type template (Fig. 5C). Conversely, in the (WT_P3-Down-A+53A_P3-Up-A) heterodimer, the wild-type template resided on the P3-Down molecule, whereas the P3-Up molecule contained the 53A template (Fig. 5D). According to the template-switching model, if processive synthesis requires transfer (crossover) of the elongating product between templates, then mismatched templates would be unable to align crossover products and processivity will drop. This prediction is based on extensive analyses using telomerase from various species, which have shown that pairing of the template with the 3' end of the primer is important for addition of DNA to the primer (13, 16, 21, 22, 38, 52). In the 53A-WT heterodimers, after template switching between the 53A and the wild-type template, there would be a mismatch between position 53 of the wild-type template and the penultimate 3' nucleotide of the 53A mutant product, and vice versa as the primer 3' end switched from the wild type to the 53A template (compare templates and products in Fig. 2, A and B, right side). Therefore, as demonstrated with the mismatch that occurs with the 53A product and the template of the 50G mutant (Fig. 2C), products from a template-switching 53A-WT heterodimer would be predominantly short.

Our results, summarized in Table II, did not support this prediction. For both heterodimers, the wild-type template was functional and highly processive, and there were comparable ratios of long to short products visible with the Cx-ext (Fig. 6B, lanes 9 and 13) or Wt53 primers (Fig. 6B, lanes 11 and 15; Table II) as well as in the mixed products from the Cx-ext and 53A-Cx2 primers (Fig. 6B, lanes 12 and 16). Thus, in these heterodimers, there was no evidence that copying from the wild-type template was stalled after one round of synthesis. The ability to demonstrate differences in processivity among the different mutants, as demonstrated above for the 50G and 53A mutants (Fig. 3, A and B), indicates that the TRAP assay conditions would have revealed any loss of processivity resulting from the inability of a 53A product to be extended on a wild-type template and vice versa. Furthermore, careful analysis of the mobility of the primer extension products from all the mixed template heterodimers revealed no evidence for any individual DNA product molecule containing a mixed composition of 53A and wild-type repeats. The DNA gel electrophoresis system used here to fractionate the telomerase reaction products (see "Materials and Methods") distinguishes differences in the mobility of DNA that result from differences in base composition as well as length (13, 53). For DNA products of equal length, products containing solely wild-type or solely 53A repeats migrate differently from each other, and thus mixed composition 53A/wild-type repeat products are expected to migrate with intermediate and distinct mobilities. Thus, a diverse population of mixed composition 53A/wild-type products would appear as intermediate mobility bands relative to wild type or 53A. No such predicted intermediate mobility bands appeared (Fig. 6, C and D and data not shown).

Most interestingly, the 53A mutant template activity was differentially sensitive to the type of P3 mutation residing within the same molecule; this template failed to be processively copied when the P3-Down-A mutation was present in the same molecule (Table II). This is shown in Fig. 6B for the (53A_P3-Down_A+WT_P3-Up-A) heterodimer (Fig. 6B, 53A-Cx1 primer lane 10, Wt53 primer lane 11, and mixed products lane 12). In contrast, the 53A template was processively copied in the (WT_P3-Down+53A_P3-Up-A) heterodimer (Fig. 6B, 53A-Cx1 primer lane 14, Wt53 primer lane 15, and mixed products lane 16) (Table II). This observed asymmetry in the dependence of the 53A template on the P3 strand indicates that template function is sensitive to mutations in the P3-Down region, supporting the notion that the base composition of the P3 region is important for normal template function/catalysis. However, despite the contextual difference in processive copying of the 53A template, the results obtained with the (WT_P3-Down+53A_P3-Up-A) test heterodimer indicate that processive synthesis in vitro does not require template switching.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Here we have characterized several aspects of human telomerase templating function. First, for a partially duplex telomere-like substrate, the minimal requirement for the length of a 3' overhang, which can pair with a portion of the templating domain, was demonstrated to be ~6 nucleotides. Most interestingly, this human telomerase requirement is similar to that shown for Euplotes telomerase (12), despite the differences between the telomerases of these species in the length of their maximal possible templating domains (11 and 15 nucleotides, respectively) and their product repeat sizes (6- and 8-base repeats, respectively) (12, 28, 35). Human telomerase can also base pair its template to G-rich non-telomeric primers and initiate synthesis by copying the adjacent template residue (39). Thus, the TAG-U substrate, containing the Tetrahymena telomeric repeat, TTGGGG, is predicted to base pair with the template (Fig. 1A). In this case, the observed minimum overhang requirement may reflect competing factors: the DNA and RNA base interactions that promote alignment and synthesis versus steric blockage of the catalytic site caused by the 5' duplex DNA, as proposed previously for Tetrahymena telomerase (13). With both single-stranded and partially double-stranded substrates, as well as base pairing with the templating domain, further interactions of the single-stranded 5' G-rich sequences with a telomerase anchor site may supplement the RNA-DNA base pairing, promoting more efficient elongation (9, 13, 54).

These results suggest that chromosome ends in vivo do not require extensive 3' overhangs to be elongated by telomerase. Single-stranded telomeric DNA-binding proteins, such as Cdc13p or Pot1p, protect telomere ends (55, 56) and, as demonstrated for Cdc13p, may also serve to recruit active telomerase (57, 58). Although long 3' overhangs are found on telomeric DNA extracted from cells (59–61), our findings suggest that such long 3' overhangs are not a requirement for telomerase catalysis per se. The ability of telomerase to extend short overhangs may, rather, be relevant to in vivo situations such as the apparent healing by telomerase of chromosome breaks at non-telomeric sites, which occurs with some limited efficiency in humans (62).

Processivity and Roles of Templating Domain Residues— Processivity of telomerase in vitro, as measured by the ability to synthesize multiple tandem, templated DNA repeats, is modulated both by protein factors and the telomerase RNA (63–66). Protein factors promoting processivity include the p43 subunit of Euplotes telomerase (67), and mutations in the reverse transcriptase homology domain of TERT can also increase processivity (66). Conversely, swapping the telomerase RNA backbone from one ciliate species into another, even though retaining the same sequence of the templating domain and its immediately flanking nucleotides, also reduced processivity of telomerase in in vitro enzymatic reactions (68). Template residues can have particularly marked effects on processivity. A previous study of the human telomerase template alignment sequence concluded that the first two residues of the alignment region, residues 55–56, are dispensable for processive synthesis (48). Here we have shown, using the pair of mutants 50G and 53A, that alignment of products on position 53 is important for processive synthesis. The likely alignments of TS on the template also suggests that position 53 would be the first residue copied (Fig. 2A). Thus, residue 53 may be both a templating and an alignment position. A precedent for this is seen for the corresponding Tetrahymena residue (nucleotide 49 in that species), which can function as an alignment residue with telomeric primers or as a template residue with certain other primers including non-telomeric primers (14, 16). The TS primer is an 18-mer oligonucleotide consisting of a random sequence joined to the 5' end of a 12-mer non-telomeric primer, -1/-11, which was utilized by telomerase in vitro nearly as efficiently as (TTAGGG)2 (39). Thus, the human as well as the Tetrahymena telomerase templates may also demonstrate an overlap of function between alignment and templating residues.

In addition, our results can explain the previous observation that the primer -1/-11 was utilized by telomerase in vitro nearly as efficiently as (TTAGGG)2 (39). In that study it was noted that shortening primer -1/-11 by 1 nucleotide at the 3' end (primer -2/-11; Ref. 39) caused a significant drop in elongation efficiency by telomerase. This can be explained if the last 3 nucleotides of -1/-11, -GTT-3', primarily aligned on template residues 54–56, with position 53 serving as the first template residue (Fig. 2A). In that case, primer -2/-11, which ends in -GT-3', would require position 54 to be the first template residue. We propose that the elongation efficiency of primer -2/-11 is low because it cannot align to a position on the template that is favorable for templated elongation by telomerase. It has been shown for Tetrahymena that the residue corresponding to human template position 54 is not copied (16). Our results and previous findings, including the recent study of the human alignment domain (48), suggest that an analogous situation might occur if human residue 54 cannot function as a template residue. If this is true, extension of the non-telomeric primer ending in -GT-3' would require that the enzyme either elongate the primer in a manner that is independent of alignment or align it on a position closer to the 5' end of the templating domain. However, in the latter case, base pairing along the 5' portion of the template would be incomplete, and it was shown previously for Tetrahymena telomerase that full template RNA-primer base pairing is necessary for elongation from more 5' positions (14). Thus, primer -2/-11 cannot align for efficient elongation by telomerase.

Three of the template mutations tested here, 49+AA, U11, and AU5, were based on analogous mutations analyzed in Tetrahymena TER (43AA, U9, and AUN) (18, 49). Two of these human mutants, 49+AA and U11, were also catalytically functional as expected. However, AU5 had little if any activity, and the 49+AA mutant was unable to produce longer TRAP products. This reduced processivity was not an artifact of the in vitro reconstitution system, as comparable results were found for this mutant telomerase assembled in vivo.2 Another insertion mutant in Tetrahymena, 49+C, also shows decreased processivity in in vitro assays, although, as with the human 49+AA mutant, it was not predicted to disrupt alignment (16). Thus an expanded template size may contribute to low processivity. A specific sequence 5' of the templating domain defines the template boundary in Tetrahymena telomerase (69–71), and in yeast and human telomerases, a template-adjacent stem serves this function (40, 72, 73). Thus, an inherent constraint on the templating domain size in each species' telomerase, perhaps imposed by the geometry of the catalytic site in the RNP, may affect processivity.

The U11 mutant template is extensively altered, yet telomerase retained its ability to polymerize DNA. We speculate that the weak activity of AU5 may have resulted from multiple factors. As discussed above, primers that inefficiently base pair with mutant templates are likely to cause activity to be weak. The primers we used to assay AU5 were not specific for the mutant repeats, and the TS substrate may not align well within the mutant template domain. TS is predicted to align on residues 55–56, with residue 54 being copied, but if this residue functions primarily in alignment, as discussed above, AU5 telomerase would have to utilize another, possibly less efficient mechanism of elongation that is independent of alignment (Fig. 2F). Supporting the possibility that, like the ciliate telomerases, human telomerase can use such an alignment-independent mechanism, the U11 mutant exhibited weak activity with the wild-type TS primer when longer extension times and more PCR cycles were used (data not shown), despite the prediction that this TS primer cannot base pair with the template.

Most interestingly, the pair of mutants 51G and 53G both showed unexpectedly weak activity, although they had fewer template substitutions than U11 or AU5. The unpredictability of these effects of template mutations has precedents in other species. For example, the S. cerevisiae template mutant 476GUG is catalytically dead, producing a null phenotype in vivo, yet similar GUG triple base substitutions only 2 or 3 bases away are functional (21), whereas other 2- or 3-base substitutions cause severe misincorporation (20). The 48C to 48U mutation in Tetrahymena decreases enzyme fidelity and causes premature product dissociation from the template (16), yet the U9 template is faithful in copying (18). Therefore, we propose that human telomerase provides another example of the close relationship between template base identities and the functionality of the telomerase active site.

Template Switching Model—One model proposed to account for a dimeric telomerase is that a dimeric enzyme promotes processivity, mediated by switching of the growing 3' end of the primer between the two templates following each round of copying along the template. We tested such a template-switching model. Such a model, which predicts that telomerase alternates between templates during synthesis, was suggested by in vivo findings as follows: the telomeric DNA of Tetrahymena or yeast cells co-expressing any template mutant, and wild-type telomerase RNA contains closely interdigitated mutant and wild-type repeats (21, 74). Furthermore, in S. cerevisiae and Kluyveromyces lactis, the observation that the enzyme remains bound to its single round elongation product suggested that the enzyme could switch between templates in the same dimeric enzyme complex after each round of extension (21, 22). Although human telomerase is processive in vitro, DNA synthesis from two different templates, one wild-type and one a mutant, was found to be interdependent, and it was suggested that template switching might be a requirement for processive synthesis (26). Significantly, our results indicate this is not the case in vitro. We have shown that in an obligatory dimer telomerase can processively elongate a primer from only one template. Although we only tested a dimeric telomerase containing one particular combination of templates, wild-type and 53A, the 53A mutant template was competent for highly processive synthesis, supported a substantial or stable level of telomerase activity, and overall appeared generally normal in its behavior. As described here, the 50G and 49+AA mutants, and even U11, are much less processive than 53A (as judged by the ratio of long to shorter products; Fig. 3), and our other template mutants had greatly reduced activity levels. Hence, these other mutants were unsuited for use in these experiments, which assayed for decreases in processivity and activity.

Our results, performed with in vitro reconstituted telomerase, do not address whether a dimer may concurrently function on two substrates, i.e. both catalytic sites are active simultaneously. Such a parallel synthesis model of telomerase proposes that telomerase could function on two sister chromatids in vivo, perhaps to coordinate their maintenance. The parallel synthesis model remains to be tested, as does the possibility that only one template site in the dimer is catalytic at a time.

Finally, an intriguing observation was that mutations in the P3 domain of the telomerase pseudoknot affected template function differentially for the 53A versus the wild-type template; 53A template function was strongly dependent on which mutated P3 strand was present in the same RNA molecule. Although mutations of the Tetrahymena TER pseudoknot affect its binding to TERT in vivo as well as telomerase activity (34), decreased binding was not detected for the in vitro reconstituted RNP (70). Similarly, the P3 mutants tested here can still bind hTERT in an in vitro reconstitution system (44).3 How P3 mutations act together with template base changes to affect the action of telomerase and its utilization of the template in vivo remain interesting questions for further investigation.


   FOOTNOTES
 
* This work was supported by grants from the National Institutes of Health and the Steven and Michelle Kirsch Foundation (to E. H. B.) and by a National Science Foundation predoctoral fellowship (to M. A. R.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

{ddagger} Present address: Dept. of Molecular Genetics and Microbiology, University of Texas, Austin, TX 78712. Back

§ To whom correspondence should be addressed. Tel.: 415-476-4912; Fax: 415-514-2913; E-mail: telomer{at}itsa.ucsf.edu.

1 The abbreviations used are: RNP, ribonucleoprotein; WT, wild type; h, human; hTER, human telomerase RNA. Back

2 L. Xu, E. Woo, and E. H. Blackburn, unpublished data. Back

3 H. Ly, L. Xu, and M. Rivera, unpublished results. Back



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Blackburn, E. H. (2001) Cell 106, 661-673[CrossRef][Medline] [Order article via Infotrieve]
  2. de Lange, T. (2001) Science 292, 1075-1076[Free Full Text]
  3. Lingner, J., Cooper, J. P., and Cech, T. R. (1995) Science 269, 1533-1534[Free Full Text]
  4. Blackburn, E. H. (1992) Annu. Rev. Biochem. 61, 113-129[CrossRef][Medline] [Order article via Infotrieve]
  5. Nugent, C. I., and Lundblad, V. (1998) Genes Dev. 12, 1073-1085[Free Full Text]
  6. Blackburn, E. H. (2000) Nat. Struct. Biol. 7, 847-850[CrossRef][Medline] [Order article via Infotrieve]
  7. Ly, H., Xu, L., Rivera, M. A., Parslow, T. G., and Blackburn, E. H. (2003) Genes Dev. 17, 1078-1083[Abstract/Free Full Text]
  8. Collins, K. (1999) Annu. Rev. Biochem. 68, 187-218[CrossRef][Medline] [Order article via Infotrieve]
  9. Lee, M. S., Gallagher, R. C., Bradley, J., and Blackburn, E. H. (1993) Cold Spring Harbor Symp. Quant. Biol. 58, 707-718[Abstract/Free Full Text]
  10. Lee, M. S., and Blackburn, E. H. (1993) Mol. Cell. Biol. 13, 6586-6599[Abstract/Free Full Text]
  11. Melek, M., Greene, E. C., and Shippen, D. E. (1996) Mol. Cell. Biol. 16, 3437-3445[Abstract]
  12. Lingner, J., and Cech, T. R. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 10712-10717[Abstract/Free Full Text]
  13. Wang, H., and Blackburn, E. H. (1997) EMBO J. 16, 866-879[CrossRef][Medline] [Order article via Infotrieve]
  14. Wang, H., D., G., and Blackburn, E. H. (1998) EMBO J. 17, 1152-1160[CrossRef][Medline] [Order article via Infotrieve]
  15. Gilley, D., Lee, M. S., and Blackburn, E. H. (1995) Genes Dev. 9, 2214-2226[Abstract/Free Full Text]
  16. Gilley, D., and Blackburn, E. H. (1996) Mol. Cell. Biol. 16, 66-75[Abstract]
  17. Prescott, J. C., and Blackburn, E. H. (2000) Mol. Cell. Biol. 20, 2941-2948[Abstract/Free Full Text]
  18. Ware, T. L., Wang, H., and Blackburn, E. H. (2000) EMBO J. 19, 3119-3131[CrossRef][Medline] [Order article via Infotrieve]
  19. Yu, G. L., Bradley, J. D., Attardi, L. D., and Blackburn, E. H. (1990) Nature 344, 126-132[CrossRef][Medline] [Order article via Infotrieve]
  20. Lin, J., Smith, D. L., and Blackburn, E. H. (2004) Mol. Biol. Cell 15, 1623-1634[Abstract/Free Full Text]
  21. Prescott, J., and Blackburn, E. H. (1997) Genes Dev. 11, 528-540[Abstract/Free Full Text]
  22. Prescott, J., and Blackburn, E. H. (1997) Genes Dev. 11, 2790-2800[Abstract/Free Full Text]
  23. Beattie, T. L., Zhou, W., Robinson, M. O., and Harrington, L. (2001) Mol. Cell. Biol. 21, 6151-6160[Abstract/Free Full Text]
  24. Tesmer, V. M., Ford, L. P., Holt, S. E., Frank, B. C., Yi, X., Aisner, D. L., Ouellette, M., Shay, J. W., and Wright, W. E. (1999) Mol. Cell. Biol. 19, 6207-6216[Abstract/Free Full Text]
  25. Wang, L., Dean, S. R., and Shippen, D. E. (2002) Nucleic Acids Res. 30, 4032-4039[Abstract/Free Full Text]
  26. Wenz, C., Enenkel, B., Amacker, M., Kelleher, C., Damm, K., and Lingner, J. (2001) EMBO J. 20, 3526-3534[CrossRef][Medline] [Order article via Infotrieve]
  27. Chen, J. L., Blasco, M. A., and Greider, C. W. (2000) Cell 100, 503-514[CrossRef][Medline] [Order article via Infotrieve]
  28. Lingner, J., Hendrick, L. L., and Cech, T. R. (1994) Genes Dev. 8, 1984-1998[Abstract/Free Full Text]
  29. Martin-Rivera, L., and Blasco, M. A. (2001) J. Biol. Chem. 276, 5856-5865[Abstract/Free Full Text]
  30. McCormick-Graham, M., and Romero, D. P. (1995) Nucleic Acids Res. 23, 1091-1097[Abstract/Free Full Text]
  31. Romero, D. P., and Blackburn, E. H. (1991) Cell 67, 343-353[CrossRef][Medline] [Order article via Infotrieve]
  32. ten Dam, E., van Belkum, A., and Pleij, K. (1991) Nucleic Acids Res. 19, 6951[Free Full Text]
  33. Lin, J., Ly, H., Hussain, A., Abraham, M., Pearl, S., Parslow, T. G., Tzfati, Y., and Blackburn, E. H. (2004) Proc. Natl. Acad. Sci. U. S. A., 101, 14713-14718[Abstract/Free Full Text]
  34. Gilley, D., and Blackburn, E. H. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 6621-6625[Abstract/Free Full Text]
  35. Feng, J., Funk, W. D., Wang, S. S., Weinrich, S. L., Avilion, A. A., Chiu, C. P., Adams, R. R., Chang, E., Allsopp, R. C., Yu, J., Le, S., West, M. D., Harley, C. B., Andrews, W. H., Greider, C. W., and Villeponteau, B. (1995) Science 269, 1236-1241[Abstract/Free Full Text]
  36. Beattie, T. L., Zhou, W., Robinson, M. O., and Harrington, L. (1998) Curr. Biol. 8, 177-180[CrossRef][Medline] [Order article via Infotrieve]
  37. Greider, C. W. (1991) Mol. Cell. Biol. 11, 4572-4580[Abstract/Free Full Text]
  38. Morin, G. B. (1989) Cell 59, 521-529[CrossRef][Medline] [Order article via Infotrieve]
  39. Morin, G. B. (1991) Nature 353, 454-456[CrossRef][Medline] [Order article via Infotrieve]
  40. Chen, J. L., and Greider, C. W. (2003) EMBO J. 22, 304-314[CrossRef][Medline] [Order article via Infotrieve]
  41. Tatematsu, K., Nakayama, J., Danbara, M., Shionoya, S., Sato, H., Omine, M., and Ishikawa, F. (1996) Oncogene 13, 2265-2274[Medline] [Order article via Infotrieve]
  42. Guiducci, C., Cerone, M. A., and Bacchetti, S. (2001) Oncogene 20, 714-725[CrossRef][Medline] [Order article via Infotrieve]
  43. Kim, M. M., Rivera, M. A., Botchkina, I. L., Shalaby, R., Thor, A. D., and Blackburn, E. H. (2001) Proc. Natl. Acad. Sci. U. S. A. 98, 7982-7987[Abstract/Free Full Text]
  44. Ly, H., Blackburn, E. H., and Parslow, T. G. (2003) Mol. Cell. Biol. 23, 6849-6856[Abstract/Free Full Text]
  45. Kim, N. W., Piatyszek, M. A., Prowse, K. R., Harley, C. B., West, M. D., Ho, P. L., Coviello, G. M., Wright, W. E., Weinrich, S. L., and Shay, J. W. (1994) Science 266, 2011-2015[Abstract/Free Full Text]
  46. Krupp, G., Kuhne, K., Tamm, S., Klapper, W., Heidorn, K., Rott, A., and Parwaresch, R. (1997) Nucleic Acids Res. 25, 919-921[Abstract/Free Full Text]
  47. Weinrich, S. L., Pruzan, R., Ma, L., Ouellette, M., Tesmer, V. M., Holt, S. E., Bodnar, A. G., Lichtsteiner, S., Kim, N. W., Trager, J. B., Taylor, R. D., Carlos, R., Andrews, W. H., Wright, W. E., Shay, J. W., Harley, C. B., and Morin, G. B. (1997) Nat. Genet. 17, 498-502[CrossRef][Medline] [Order article via Infotrieve]
  48. Gavory, G., Farrow, M., and Balasubramanian, S. (2002) Nucleic Acids Res. 30, 4470-4480[Abstract/Free Full Text]
  49. Kirk, K. E., Harmon, B. P., Reichardt, I. K., Sedat, J. W., and Blackburn, E. H. (1997) Science 275, 1478-1481[Abstract/Free Full Text]
  50. Greider, C. W., and Blackburn, E. H. (1985) Cell 43, 405-413[CrossRef][Medline] [Order article via Infotrieve]
  51. Harrington, L. A., and Greider, C. W. (1991) Nature 353, 451-454[CrossRef][Medline] [Order article via Infotrieve]
  52. Shippen-Lentz, D., and Blackburn, E. H. (1990) Science 247, 546-552[Abstract/Free Full Text]
  53. Lerman, L. S., Fischer, S. G., Hurley, I., Silverstein, K., and Lumelsky, N. (1984) Annu. Rev. Biophys. Bioeng. 13, 399-423[CrossRef][Medline] [Order article via Infotrieve]
  54. Collins, K., and Greider, C. W. (1993) Genes Dev. 7, 1364-1376[Abstract/Free Full Text]
  55. Baumann, P., and Cech, T. R. (2001) Science 292, 1171-1175[Abstract/Free Full Text]
  56. Nugent, C. I., Hughes, T. R., Lue, N. F., and Lundblad, V. (1996) Science 274, 249-252[Abstract/Free Full Text]
  57. Grandin, N., Damon, C., and Charbonneau, M. (2001) EMBO J. 20, 1173-1183[CrossRef][Medline] [Order article via Infotrieve]
  58. Pennock, E., Buckley, K., and Lundblad, V. (2001) Cell 104, 387-396[CrossRef][Medline] [Order article via Infotrieve]
  59. Dionne, I., and Wellinger, R. J. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 13902-13907[Abstract/Free Full Text]
  60. Henderson, E. R., and Blackburn, E. H. (1989) Mol. Cell. Biol. 9, 345-348[Abstract/Free Full Text]
  61. Makarov, V. L., Hirose, Y., and Langmore, J. P. (1997) Cell 88, 657-666[CrossRef][Medline] [Order article via Infotrieve]
  62. Cooke, H. (1995) in Telomeres (Blackburn, E. H., and Greider, C. W., eds) pp. 219-245, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY
  63. Lai, C. K., Miller, M. C., and Collins, K. (2003) Mol. Cell 11, 1673-1683[CrossRef][Medline] [Order article via Infotrieve]
  64. Mason, D. X., Goneska, E., and Greider, C. W. (2003) Mol. Cell. Biol. 23, 5606-5613[Abstract/Free Full Text]
  65. Moriarty, T. J., Marie-Egyptienne, D. T., and Autexier, C. (2004) Mol. Cell. Biol. 24, 3720-3733[Abstract/Free Full Text]
  66. Lue, N. F. (2004) BioEssays 26, 955-962[CrossRef][Medline] [Order article via Infotrieve]
  67. Aigner, S., and Cech, T. R. (2004) RNA (N. Y.) 10, 1108-1118
  68. Bhattacharyya, A., and Blackburn, E. H. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 2823-2827[Abstract/Free Full Text]
  69. Autexier, C., and Greider, C. W. (1995) Genes Dev. 9, 2227-2239[Abstract/Free Full Text]
  70. Autexier, C., and Greider, C. W. (1998) Nucleic Acids Res. 26, 787-795[Abstract/Free Full Text]
  71. Lai, C. K., Miller, M. C., and Collins, K. (2002) Genes Dev. 16, 415-420[Abstract/Free Full Text]
  72. Tzfati, Y., Fulton, T. B., Roy, J., and Blackburn, E. H. (2000) Science 288, 863-867[Abstract/Free Full Text]
  73. Seto, A. G., Umansky, K., Tzfati, Y., Zaug, A. J., Blackburn, E. H., and Cech, T. R. (2003) RNA (N. Y.) 9, 1323-1332
  74. Yu, G. L., and Blackburn, E. H. (1991) Cell 67, 823-832[CrossRef][Medline] [Order article via Infotrieve]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 Nucleic Acids ResHome page
J. Tang, Z.-y. Kan, Y. Yao, Q. Wang, Y.-h. Hao, and Z. Tan
G-quadruplex preferentially forms at the very 3' end of vertebrate telomeric DNA
Nucleic Acids Res., March 27, 2008; 36(4): 1200 - 1208.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
C. L. Gipson, Z.-T. Xin, S. C. Danzy, T. G. Parslow, and H. Ly
Functional Characterization of Yeast Telomerase RNA Dimerization
J. Biol. Chem., June 29, 2007; 282(26): 18857 - 18863.
[Abstract] [Full Text] [PDF]

Home page
 BloodHome page
Z.-T. Xin, A. D. Beauchamp, R. T. Calado, J. W. Bradford, J. A. Regal, A. Shenoy, Y. Liang, P. M. Lansdorp, N. S. Young, and H. Ly
Functional characterization of natural telomerase mutations found in patients with hematologic disorders
Blood, January 15, 2007; 109(2): 524 - 532.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Cell. Biol.Home page
E. J. Middleman, J. Choi, A. S. Venteicher, P. Cheung, and S. E. Artandi
Regulation of Cellular Immortalization and Steady-State Levels of the Telomerase Reverse Transcriptase through Its Carboxy-Terminal Domain.
Mol. Cell. Biol., March 1, 2006; 26(6): 2146 - 2159.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
W. C. Drosopoulos, R. DiRenzo, and V. R. Prasad
Human Telomerase RNA Template Sequence Is a Determinant of Telomere Repeat Extension Rate
J. Biol. Chem., September 23, 2005; 280(38): 32801 - 32810.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
A. J. Zaug, E. R. Podell, and T. R. Cech
Human POT1 disrupts telomeric G-quadruplexes allowing telomerase extension in vitro
PNAS, August 2, 2005; 102(31): 10864 - 10869.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Biol. CellHome page
T. J. Moriarty, R. J. Ward, M. A.S. Taboski, and C. Autexier
An Anchor Site-Type Defect in Human Telomerase That Disrupts Telomere Length Maintenance and Cellular Immortalization
Mol. Biol. Cell, July 1, 2005; 16(7): 3152 - 3161.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
M. Lei, A. J. Zaug, E. R. Podell, and T. R. Cech
Switching Human Telomerase On and Off with hPOT1 Protein in Vitro
J. Biol. Chem., May 27, 2005; 280(21): 20449 - 20456.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/51/53770    most recent
M407768200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Rivera, M. A.
Right arrow Articles by Blackburn, E. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Rivera, M. A.
Right arrow Articles by Blackburn, E. H.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2004 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement