A Physical and Functional Constituent of Telomerase Anchor Site
2005; Elsevier BV; Volume: 280; Issue: 28 Linguagem: Inglês
10.1074/jbc.m503028200
ISSN1083-351X
Autores Tópico(s)Telomeres, Telomerase, and Senescence
ResumoTelomerase is a ribonucleoprotein reverse transcriptase responsible for the maintenance of one strand of the telomere terminal repeats. It consists minimally of a catalytic protein component (TERT) and an RNA subunit that provides the template. Compared with prototypical reverse transcriptases, telomerase is unique in possessing a DNA binding domain (anchor site) that is distinct from the catalytic site. Yeast TERT mutants bearing deletion or point mutations in an N-terminal domain (known as N-GQ) were found to be selectively impaired in extending primers that form short hybrids with telomerase RNA. The mutants also suffered a significant loss of repeat addition processivity but displayed an enhancement in nucleotide addition processivity. Furthermore, the mutants manifested altered primer utilization properties for oligonucleotides containing non-telomeric residues in the 5′-region. Cross-linking studies indicate that the N-GQ domain physically contacts the 5′-region of the DNA substrate in the context of a telomerase-telomere complex. Together, these results implicate the N-GQ domain of TERT as a physical and functional constituent of the telomerase anchor site. Coupled with previous genetic analysis, our data confirm that anchor site interaction is indeed important for telomerase function in vivo. Telomerase is a ribonucleoprotein reverse transcriptase responsible for the maintenance of one strand of the telomere terminal repeats. It consists minimally of a catalytic protein component (TERT) and an RNA subunit that provides the template. Compared with prototypical reverse transcriptases, telomerase is unique in possessing a DNA binding domain (anchor site) that is distinct from the catalytic site. Yeast TERT mutants bearing deletion or point mutations in an N-terminal domain (known as N-GQ) were found to be selectively impaired in extending primers that form short hybrids with telomerase RNA. The mutants also suffered a significant loss of repeat addition processivity but displayed an enhancement in nucleotide addition processivity. Furthermore, the mutants manifested altered primer utilization properties for oligonucleotides containing non-telomeric residues in the 5′-region. Cross-linking studies indicate that the N-GQ domain physically contacts the 5′-region of the DNA substrate in the context of a telomerase-telomere complex. Together, these results implicate the N-GQ domain of TERT as a physical and functional constituent of the telomerase anchor site. Coupled with previous genetic analysis, our data confirm that anchor site interaction is indeed important for telomerase function in vivo. Telomerase is a ribonucleoprotein that is responsible for maintaining the terminal repeats of telomeres in most organisms (1Greider C.W. Blackburn E.H. Cell. 1985; 43: 405-413Abstract Full Text PDF PubMed Scopus (2604) Google Scholar). It acts as an unusual reverse transcriptase, using a small segment of an integral RNA component as template for the synthesis of the dG-rich strand of telomeres (2Greider C.W. Blackburn E.H. Nature. 1989; 337: 331-337Crossref PubMed Scopus (1303) Google Scholar). The enzymatic core of telomerase consists minimally of two components, a reverse transcriptase (RT) 1The abbreviations used are: RT, reverse transcriptase; CTE, C-terminal extension. -like protein that catalyzes nucleotide addition (named TERT), and an RNA in which the template is embedded (for reviews see Refs. 3Nugent C.I. Lundblad V. Genes Dev. 1998; 12: 1073-1085Crossref PubMed Scopus (400) Google Scholar, 4Kelleher C. Teixeira M. Forstemann K. Lingner J. Trends Biochem. Sci. 2002; 27: 572-579Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar, 5Harrington L. Cancer Lett. 2003; 194: 139-154Crossref PubMed Scopus (86) Google Scholar, 6Chan S. Blackburn E. Philos. Trans. R. Soc. Lond. B Biol. Sci. 2004; 359: 109-121Crossref PubMed Scopus (393) Google Scholar). Telomerase RNAs are evolutionarily divergent in sequence, but recent studies suggest that they share a number of key structural elements, including a template segment, a pseudoknot, and a long-range base-pairing element (for reviews, see Refs. 7Lin J. Ly H. Hussain A. Abraham M. Pearl S. Tzfati Y. Parslow T. Blackburn E. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 14713-14718Crossref PubMed Scopus (90) Google Scholar and 8Chen J. Greider C. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 14683-14684Crossref PubMed Scopus (75) Google Scholar). The TERT protein, on the other hand, is well conserved in evolution and possesses a central domain with significant sequence similarity to prototypical RTs (Fig. 1A) (9Lingner J. Hughes T.R. Shevchenko A. Mann M. Lundblad V. Cech T.R. Science. 1997; 276: 561-567Crossref PubMed Scopus (1041) Google Scholar, 10Nakamura T.M. Cech T.R. Cell. 1998; 92: 587-590Abstract Full Text Full Text PDF PubMed Scopus (282) Google Scholar). Flanking the RT domain are a large N-terminal extension and a short C-terminal extension (CTE). By analogy with prototypical RTs, the CTE is likely to constitute the so-called thumb domain of the polymerase (11Hossain S. Singh S. Lue N. J. Biol. Chem. 2002; 277: 36174-36180Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar, 12Huard S. Moriarty T. Autexier C. Nucleic Acids Res. 2003; 31: 4059-4070Crossref PubMed Scopus (80) Google Scholar). The N-terminal extension consists of a non-conserved region (named N-region), four conserved motifs (named GQ, CP, QFP, and T), and a flexible linker (located between the GQ and CP motif) (13Friedman K.L. Cech T.R. Genes Dev. 1999; 13: 2863-2874Crossref PubMed Scopus (121) Google Scholar, 14Xia J. Peng Y. Mian I.S. Lue N.F. Mol. Cell. Biol. 2000; 20: 5196-5207Crossref PubMed Scopus (121) Google Scholar). The CP, QFP, and T motifs have been shown to mediate interaction with telomerase RNA (15Bryan T.M. Goodrich K.J. Cech T.R. Mol. Cell. 2000; 6: 493-499Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar, 16Lai C.K. Mitchell J.R. Collins K. Mol. Cell. Biol. 2001; 21: 990-1000Crossref PubMed Scopus (145) Google Scholar, 17Bachand F. Autexier C. Mol. Cell. Biol. 2001; 21: 1888-1897Crossref PubMed Scopus (106) Google Scholar, 18Bosoy D. Peng Y. Mian I. Lue N. J. Biol. Chem. 2003; 278: 3882-3890Abstract Full Text Full Text PDF PubMed Scopus (70) Google Scholar). The N-region and GQ motif together apparently comprise a stable domain (named N-GQ) that is required for telomerase function both in vitro and in vivo (see below). An unusual property of telomerase is its ability to mediate realignment of the DNA product relative to the RNA template. This was inferred from the capacity of the enzyme to add more repeats onto a starting primer than are present in the RNA template (19Greider C. Mol. Cell. Biol. 1991; 11: 4572-4580Crossref PubMed Scopus (209) Google Scholar). The addition of multiple repeats depends on two types of movements: simultaneous translocation of the RNA-DNA duplex away from the active site after each nucleotide addition and translocation of the RNA template relative to the DNA product after each cycle of copying the template, such that the 3′-end of the DNA realigns with the 3′-region of the RNA template. These movements have been referred to as type I and type II translocation, and the propensity to carry out the movements referred to as nucleotide addition and repeat addition processivity, respectively (20Collins K. Annu. Rev. Biochem. 1999; 68: 187-218Crossref PubMed Scopus (72) Google Scholar, 21Peng Y. Mian I.S. Lue N.F. Mol. Cell. 2001; 7: 1201-1211Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar, 22Lue N. Krupp G. Parwaresch R. Telomeres and Telomerases: Cancer and Biology. Landes Biosciences, Austin, TX2002: 239-258Google Scholar). Although nucleotide addition processivity is a common property of all RTs, repeat addition processivity is unique to telomerase. Various features of the telomerase protein and RNA have been shown to modulate nucleotide and repeat addition processivity (for review see Ref. 23Lue N. BioEssays. 2004; 26: 955-962Crossref PubMed Scopus (63) Google Scholar). Type II translocation is postulated to require an interaction between telomerase and the more 5′-region of the DNA primer, one that is distinct from the RNA-DNA hybrid near the catalytic site. This "anchor site" interaction is presumed to allow telomerase to remain bound to DNA during type II translocation when the DNA product unpairs from the RNA template prior to realignment. In support of this hypothesis, longer oligonucleotides have been shown to support more processive elongation by Tetrahymena telomerase (24Collins K. Greider C.W. Genes Dev. 1993; 7: 1364-1376Crossref PubMed Scopus (148) Google Scholar). Furthermore, the sequence of the DNA primer 5′ to the RNA-DNA hybrid has been shown to influence the binding affinity and catalytic rate of the enzyme (25Melek M. Davis B. Shippen D. Mol. Cell. Biol. 1994; 14: 7827-7838Crossref PubMed Scopus (37) Google Scholar, 26Lee M. Blackburn E.H. Mol. Cell. Biol. 1993; 13: 6586-6599Crossref PubMed Scopus (100) Google Scholar, 27Prescott J. Blackburn E.H. Genes Dev. 1997; 11: 2790-2800Crossref PubMed Scopus (145) Google Scholar, 28Lue N.F. Peng Y. Nucleic Acids Res. 1998; 26: 1487-1494Crossref PubMed Scopus (39) Google Scholar). Although the precise molecular determinants of anchor site interaction are not well understood, the TERT protein from Euplotes aediculatus and Saccharomyces cerevisiae has been shown to cross-link to the 5′-region of the DNA primer, suggesting that a portion of this protein may mediate anchor site interaction (27Prescott J. Blackburn E.H. Genes Dev. 1997; 11: 2790-2800Crossref PubMed Scopus (145) Google Scholar, 29Hammond P.W. Lively T.N. Cech T.R. Mol. Cell. Biol. 1997; 17: 296-308Crossref PubMed Scopus (91) Google Scholar). In addition, there is physical evidence for the participation of telomerase RNA in anchor site interaction (29Hammond P.W. Lively T.N. Cech T.R. Mol. Cell. Biol. 1997; 17: 296-308Crossref PubMed Scopus (91) Google Scholar). The physiologic significance of anchor site interaction has not been determined. We and others have shown previously that the N-GQ domain of yeast TERT (Est2p) is required for telomerase function both in vivo and in vitro (13Friedman K.L. Cech T.R. Genes Dev. 1999; 13: 2863-2874Crossref PubMed Scopus (121) Google Scholar, 14Xia J. Peng Y. Mian I.S. Lue N.F. Mol. Cell. Biol. 2000; 20: 5196-5207Crossref PubMed Scopus (121) Google Scholar). Some truncations and point mutations in the N-GQ domain cause dramatic telomere shortening, cell senescence, and a drastic reduction in enzyme activity. Similar mutations in the Tetrahymena and human enzymes are also known to adversely affect telomerase function in vitro and in vivo (16Lai C.K. Mitchell J.R. Collins K. Mol. Cell. Biol. 2001; 21: 990-1000Crossref PubMed Scopus (145) Google Scholar, 17Bachand F. Autexier C. Mol. Cell. Biol. 2001; 21: 1888-1897Crossref PubMed Scopus (106) Google Scholar, 30Beattie T.L. Zhou W. Robinson M.O. Harrington L. Mol. Biol. Cell. 2000; 11: 3329-3340Crossref PubMed Scopus (78) Google Scholar, 31Armbruster B. Banik S. Guo C. Smith A. Counter C. Mol. Cell. Biol. 2001; 21: 7775-7786Crossref PubMed Scopus (144) Google Scholar). In addition, the N-GQ domain of Est2p has been shown to bind weakly to nucleic acids when recombinantly expressed and purified from Escherichia coli (14Xia J. Peng Y. Mian I.S. Lue N.F. Mol. Cell. Biol. 2000; 20: 5196-5207Crossref PubMed Scopus (121) Google Scholar). In this study, the enzymatic properties of several Est2p mutants bearing mutations in the N-GQ domain were examined in greater detail. Both physical and function assays supported the notion that this domain of Est2p represents at least part of the telomerase anchor site. Coupled with previous genetic analysis, our data for the first time provide a correlation between loss of anchor site interaction in vitro with loss of telomerase function in vivo. Yeast Strains and Plasmids—The construction of an est2-Δ strain harboring the pSE-Est2-C874 plasmid (containing a protein A-tagged EST2 gene) has been described (14Xia J. Peng Y. Mian I.S. Lue N.F. Mol. Cell. Biol. 2000; 20: 5196-5207Crossref PubMed Scopus (121) Google Scholar). This fully functional Est2p is designated wild type telomerase throughout the text. The plasmid harboring the N-30 mutant has also been described (14Xia J. Peng Y. Mian I.S. Lue N.F. Mol. Cell. Biol. 2000; 20: 5196-5207Crossref PubMed Scopus (121) Google Scholar). The N-50 and N-151 deletion mutant, missing the first 49 and 150 amino acids of Est2p, were similarly constructed by substituting truncated PCR fragments into the N-terminal region of the gene. Purification of and Assay for Yeast Telomerase—Whole cell extracts and IgG-Sepharose-purified telomerase were prepared as previously described (14Xia J. Peng Y. Mian I.S. Lue N.F. Mol. Cell. Biol. 2000; 20: 5196-5207Crossref PubMed Scopus (121) Google Scholar, 21Peng Y. Mian I.S. Lue N.F. Mol. Cell. 2001; 7: 1201-1211Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar, 28Lue N.F. Peng Y. Nucleic Acids Res. 1998; 26: 1487-1494Crossref PubMed Scopus (39) Google Scholar, 32Cohn M. Blackburn E.H. Science. 1995; 269: 396-400Crossref PubMed Scopus (258) Google Scholar). Briefly, the cultures were harvested by centrifugation and resuspended in 1/100 volume of TMG-10 buffer (10 mm Tris·HCl, pH 8.0, 0.1 mm EDTA, 0.1 mm EGTA, 1.2 mm MgCl2, 10% glycerol) (28Lue N.F. Peng Y. Nucleic Acids Res. 1998; 26: 1487-1494Crossref PubMed Scopus (39) Google Scholar). The suspension was mixed with an equal volume of glass beads, and the cells were lyzed by vortexing. The lysates were clarified by centrifugation at 100,000 × g. All of the Est2 variants tested in this study were tagged at the C terminus with protein A. Therefore, telomerase in the extracts was purified on IgG-Sepharose. Typically, protein extract (4 mg) was supplemented with sodium acetate (to 0.4 m) and Tween 20 (to 0.05%) and was incubated with 20 μl of IgG-Sepharose with end-to-end rotation at 4 °C for 2 h. The beads were then washed three time each with 1 ml of TMG-10 buffer containing 600 mm sodium acetate and two times each with TMG-10 buffer. Telomerase prepared in this manner is largely free of nuclease contamination; no degradation of end-labeled primers was observed following incubation with the beads (data not shown). Primer extension assay was carried out using the bead-bound telomerase as previously described (21Peng Y. Mian I.S. Lue N.F. Mol. Cell. 2001; 7: 1201-1211Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar). The primers (Sigma-Genosys) have the following sequences: TEL10, GGGTGTGGTG; TEL15, TGTGTGGTGTGTGGG; OXYT1, GTTTTGGGGTTTTGGG; TEL15(m4,5), TGTGTGGTGTCAGGG; TEL15(m12), TGTCTGGTGTGTGGG. For determination of the processivity of telomerase, the signal for each product was determined by PhosphorImager (Amersham Biosciences) and normalized to the amount of transcript by dividing against the number of labeled residues. The TEL15(m4,5) and OXYT1 primers both end in 3 Gs and can align to only one site along the yeast RNA template, thus supporting the addition of a defined 7-nt sequence (TGTGGTG) up to the 5′-boundary of the template. For nucleotides beyond the +7 position, a sequence of TGTG...was assumed as shown in Equation 1 (calculations were made assuming other plausible sequences, and the conclusions were not altered), Processivity Pi=∑j=i+1N(Tj)/∑j=iN(Tj), (Eq. 1) where Ti denotes the amount of transcript calculated for the primer +i position and N is the highest number such that a visible signal can be discerned in the PhosphorImager file for the primer +N product. Protein Cross-linking—Telomerase was isolated by adsorption to IgG-Sepharose and cross-linked to oligonucleotides containing Iodo-dU substitutions by UV irradiation (310 nm, 15 min at 4 °C). Following cross-linking, radioactive dTTP (3000 Ci/mmol) was added to 0.7 μm and the covalent adduct incubated at 22 °C for 30 min to allow for labeling of the DNA by telomerase. The beads were then washed three times with TMG-10 (300) and subjected to SDS-PAGE and PhosphorImager analysis. For trypsin digestion following cross-linking, the beads were incubated with varying concentrations of trypsin at room temperature for 30 min and washed three times with TMG-10 (200). The Sepharose-bound proteins were then separated in a 9% SDS-polyacrylamide gel and transferred to a nitrocellulose filter by electroblotting. Protein A-tagged species were visualized by Western analysis (ProtoBlot System; Promega). The dried filter was subsequently exposed to a PhosphorImager plate to detect radioactively labeled protein-DNA adducts. Mutations in the N-GQ Domain Caused a Primer-specific Impairment of Telomerase Activity in Vitro—We have previously shown that several deletions in the N-region and several point mutations in the GQ motif (N-10, N-20, N-30, W115A, FH118AA and G123A) caused a severe reduction (>50-fold) in the primer extension activity of telomerase on a particular primer substrate (14Xia J. Peng Y. Mian I.S. Lue N.F. Mol. Cell. Biol. 2000; 20: 5196-5207Crossref PubMed Scopus (121) Google Scholar). Subsequently, two additional deletion mutants (N-50, N-151) were generated and subjected along with the N-30 and W115A mutant to analysis using two other primer oligonucleotides. As shown in Fig. 1B, all of the mutant enzymes exhibited comparable levels of activity as the wild type enzyme in adding just one T residue to the TEL15 primer but were about 10-40-fold less active on the TEL10 primer. Additional studies with other primers suggest that the key determinant of primer utilization by the mutant enzymes is the degree of base complementary between the primer and template RNA: primers that form short hybrids were inefficiently utilized (Table I and data not shown). Notably, primers that form 3-4-bp hybrids with telomerase RNA were extended ∼10-fold less efficiently by N-151 than wild type telomerase, whereas those that form 8-9-bp hybrids supported comparable levels of DNA synthesis by both enzymes.Table IPrimer utilization by wild type and N-151 telomerasePrimerPrimer lengthHybrid lengthaThe length of uninterrupted hybrid that can form between the 3′-end of the primer and yeast telomerase RNA.N-151/wild typebComparable amounts of wild type and N-151 telomerase immobilized on IgG-Sepharose (as judged by Western blotting of the bead preparations) were tested in primer extension assays using labeled dTTP as the sole nucleotide. The activities were quantified by PhosphorImager analysis and the ratio of N-151 to wild type activity presented.TGTGTGGTGTGTGGG1590.62TGTGTGGG881.1TGTGTGGAAACTGGG1540.12TGGGGTTAAACTGGG1540.13GGGTGTGGTG103 or 40.05a The length of uninterrupted hybrid that can form between the 3′-end of the primer and yeast telomerase RNA.b Comparable amounts of wild type and N-151 telomerase immobilized on IgG-Sepharose (as judged by Western blotting of the bead preparations) were tested in primer extension assays using labeled dTTP as the sole nucleotide. The activities were quantified by PhosphorImager analysis and the ratio of N-151 to wild type activity presented. Open table in a new tab Disparate Effects of the N-GQ Mutations on Telomerase Processivity—To assess the effect of the N-GQ mutations on the elongation properties of telomerase, I included both dGTP and dTTP in primer extension assays to allow the addition of multiple nucleotides and then analyzed the processivity of the enzyme at multiple positions during elongation. Earlier analysis suggests that for yeast telomerase polymerization beyond a single repeat can be most easily observed by using primers that form short hybrids with telomerase RNA (33Lue N. Lin Y. Mian I. Mol. Cell. Biol. 2003; 23: 8440-8449Crossref PubMed Scopus (59) Google Scholar, 34Bosoy D. Lue N. Nucleic Acids Res. 2004; 32: 93-101Crossref PubMed Scopus (22) Google Scholar). Thus, to assess nucleotide and repeat addition processivity simultaneously, I used one such primer, named OXYT1. As shown in Fig. 2A, all of the deletion mutants and the W115A point mutant manifested significant alterations in elongation property. Quantitative analysis showed that from the p +2 to +5 position the mutant enzymes were more processive than the wild type enzyme (by ∼15-25%) (Fig. 2B). This effect was reversed at the p +7 position such that the mutant enzymes became less processive (by >3-fold for the N-151 mutant). Because the OXYT1 primer contains 3 G residues at its 3′-end, it can align to a unique position within the yeast telomerase RNA template (9Lingner J. Hughes T.R. Shevchenko A. Mann M. Lundblad V. Cech T.R. Science. 1997; 276: 561-567Crossref PubMed Scopus (1041) Google Scholar, 28Lue N.F. Peng Y. Nucleic Acids Res. 1998; 26: 1487-1494Crossref PubMed Scopus (39) Google Scholar, 35Prescott J. Blackburn E.H. Genes Dev. 1997; 11: 528-540Crossref PubMed Scopus (136) Google Scholar) (Fig. 2A). Based on this alignment position, the p +7 product is generated upon one complete round of nucleotide addition, when the active site of telomerase reaches the 5′-boundary of the template. For further elongation to take place, telomerase must perform a type II translocation. Our observations thus indicate that the N-GQ mutations enhanced type I, but impaired type II, processivity. Notably, the magnitude of the effect correlated with the size of the deletion, suggesting that residues throughout the domain participate in the underlying mechanism(s) for processivity alteration. Similar results were obtained from assays using another primer that forms a short hybrid with telomerase RNA (data not shown). The N-region Is Functionally Required for Interaction between Telomerase and the 5′-Region of the DNA Primer—The altered primer utilization property and processivity for the N-GQ mutants suggest that this domain may be a constituent of the long-hypothesized anchor site. This notion is also consistent with the earlier observation that the N-GQ domain is a low affinity nucleic acid binding domain (14Xia J. Peng Y. Mian I.S. Lue N.F. Mol. Cell. Biol. 2000; 20: 5196-5207Crossref PubMed Scopus (121) Google Scholar). To test this idea functionally, I examined the ability of the N-30 telomerase to utilize a primer with a sequence alteration in the presumed anchor site target. Earlier studies indicate that such a sequence alteration (a G to C mutation in TEL15 to give TEL15(m12), Fig. 3A) weakens anchor site interaction, leading to higher levels of DNA synthesis because of increased enzyme turnover (27Prescott J. Blackburn E.H. Genes Dev. 1997; 11: 2790-2800Crossref PubMed Scopus (145) Google Scholar, 28Lue N.F. Peng Y. Nucleic Acids Res. 1998; 26: 1487-1494Crossref PubMed Scopus (39) Google Scholar). I reasoned that if the N-GQ domain is in fact required for anchor site interactions then the mutant might not exhibit enhanced DNA synthesis on mutated primers. As shown in Fig. 3B, the wild type telomerase mediated 2-fold higher DNA synthesis on the TEL15(m12) primer than on the TEL15 primer (compare lanes 1 and 3 in both the left and right panels), whereas the N-30 enzyme was equally active on both primers (compare lanes 2 and 4). The difference in primer utilization between the wild type and N-30 telomerase was observed regardless of the identity of the labeled nucleotide (Fig. 3, B and C, asterisks), which is present at a much lower concentration than the unlabeled nucleotide. Together, these finding suggests that the N-30 enzyme is in fact defective in interacting with the upstream region of the DNA primer. Aside from the N-GQ domain, two other domains of TERT, named IFD and CTE, have been implicated in primer interaction (11Hossain S. Singh S. Lue N. J. Biol. Chem. 2002; 277: 36174-36180Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar, 12Huard S. Moriarty T. Autexier C. Nucleic Acids Res. 2003; 31: 4059-4070Crossref PubMed Scopus (80) Google Scholar, 33Lue N. Lin Y. Mian I. Mol. Cell. Biol. 2003; 23: 8440-8449Crossref PubMed Scopus (59) Google Scholar). IFD designates a telomerase-specific insertion in the putative fingers domain of TERT, and a mutation in IFD has been demonstrated to cause similar processivity defects as the N-GQ mutants. CTE denotes the C-terminal extension of TERT. Analysis of yeast and human TERT suggests that this region may correspond to the thumb domain of conventional RTs and may contact the nucleic acid duplex from the opposite side of the fingers domain. I tested the ability of an IFD mutant (named LYID589) and a CTE mutant (named C745) to respond to the TEL15(m12) primer mutation. Like the wild type enzyme and in contrast to the N-30 enzyme, these two mutants were able to mediate 2-3-fold greater DNA synthesis on the TEL15(m12) primer than on the TEL15 primer (supplemental Fig. S1). Thus, despite their apparent roles in primer binding, IFD and CTE do not appear to be required for anchor site interactions. The N-GQ Domain Physically Interacts with the 5′-Region of Telomeric Oligonucleotides—To detect the presumed physical interaction between TERT and the primer upstream region, I utilized a previously described cross-linking assay (27Prescott J. Blackburn E.H. Genes Dev. 1997; 11: 2790-2800Crossref PubMed Scopus (145) Google Scholar, 29Hammond P.W. Lively T.N. Cech T.R. Mol. Cell. Biol. 1997; 17: 296-308Crossref PubMed Scopus (91) Google Scholar). Telomerase with an Est2p subunit that is C-terminal tagged with protein A was bound to IgG-Sepharose, incubated with telomere oligonucleotides containing Iodo-dU substitutions in the 5′-region, and irradiated with UV light to allow covalent adduct formation. The adduct was then labeled by telomerase-mediated incorporation of [32P]dTTP to the 3′-end of the oligonucleotide and analyzed by SDS-PAGE and exposure to PhosphorImager plates. As shown in Fig. 4, A and B, a labeled species corresponding in size to Est2p can be detected in this assay. Adduct formation was most efficient with an oligonucleotide derivatized at positions 11 and 13, consistent with its being due to anchor site interactions. Also, as expected, the signal was absent when the enzyme preparation was pretreated with RNase A. Altering the size of Est2p by the addition of a Myc3 tag resulted in a decrease in the mobility of the adduct, thus confirming unequivocally the identity of the cross-linked protein as Est2p (Fig. 4C). To map the site on Est2p that cross-linked to the oligonucleotide, I subjected the cross-linked products to partial proteolysis and analyzed the association of individual proteolytic fragments (as detected in Western blotting using protein A antibodies) with radioactivity. Because of the location of the protein A tag at the C terminus, only fragments that retained the C terminus can be detected in the Western analysis. At least three proteolytic fragments of 95, 60, and 40 kDa can be observed in appreciable quantities (designated A, B, and C in Fig. 4D). These fragments are judged to be caused by cleavage at positions ∼190, 470, and 650 of Est2p. This interpretation was confirmed by analyzing the proteolytic pattern of a C-terminal-truncated Est2p (supplemental Fig. S2). Remarkably, none of the proteolytic fragments is associated with appreciable amounts of radioactivity (Fig. 4D, lanes 2-5 and 7-10). Importantly, the sample shown in lanes 2 and 6 contained similar amounts of full-length protein and fragment A. However, radioactivity was almost exclusively associated with the full-length protein. These observations indicate that the main cross-linking site on Est2p is within the first 190 amino acids. Consistent with this finding, the N-151 mutant exhibited greatly reduced cross-linking to the DNA primer in similar assays (data not shown). In addition to Est2p, another protein of ∼55 kDa can be labeled in the cross-linking assay (labeled species II in Fig. 4D). Labeling of this protein is also sensitive to RNase (data not shown), but its identity is presently unknown. The N-GQ Domain of TERT as the Telomerase Anchor Site—The N-GQ domain appears to fulfill the classic criteria for being a determinant of telomerase anchor site: it physically and functionally interacts with a 5′-region of the DNA primer located outside of the RNA-DNA hybrid, and it is required for optimal type II translocation (repeat addition processivity) (24Collins K. Greider C.W. Genes Dev. 1993; 7: 1364-1376Crossref PubMed Scopus (148) Google Scholar, 26Lee M. Blackburn E.H. Mol. Cell. Biol. 1993; 13: 6586-6599Crossref PubMed Scopus (100) Google Scholar). The current study also reveals two additional biochemical activities of the telomerase anchor site: it facilitates initiation on primers that form short hybrids with telomerase RNA, and it can impair type I translocation or nucleotide addition processivity. All of the activities of the anchor site can be rationalized by its DNA binding function, as follows. I presume that the overall affinity of the telomerase enzyme for DNA is a function of both RNA-DNA and protein-DNA interactions and that a threshold level of affinity is required for efficient binding and polymerization. Given a short RNA-DNA hybrid, the protein-DNA interaction mediated by the N-GQ domain may be critical for this threshold level of affinity, thus accounting for the primer-specific impairment of activity. The realignment step following type II translocation also entails the formation of a short hybrid (36Chen J. Greider C. EMBO J. 2003; 22: 304-314Crossref PubMed Scopus (112) Google Scholar) and may critically depend on the same protein-DNA interaction for complex stability. That the repeat addition processivity was not completely abolished by N-GQ mutations suggests that other domains of TERT (e.g. the IFD motif) can help to stabilize short RNA-DNA hybrids (33Lue N. Lin Y. Mian I. Mol. Cell. Biol. 2003; 23: 8440-8449Crossref PubMed Scopus (59) Google Scholar). Indeed, even very short primers (lacking the presumed anchor site target) can mediate multiple repeat addition by Tetrahymena telomerase, albeit rather inefficiently (37Baran N. Haviv Y. Paul B. Manor H. Nucleic Acids Res. 2002; 30: 5570-5578Crossref PubMed Scopus (24) Google Scholar). Thus, anchor site interactions may greatly stimulate, but not be essential for, multiple repeat addition. Less obvious is the mechanism by which anchor site binding impairs type I translocation. Perhaps the energy barrier for conformational transition that is necessary for nucleotide addition becomes too high with a very stable telomerase-DNA complex (one that contains both a long RNA-DNA hybrid and protein-DNA contacts). Indeed, the nucleotide addition processivity of Kluyveromyces lactis telomerase was shown to be impaired by increasing the length of complementarity between the DNA primer and RNA template, arguing that enhancing complex stability can have a detrimental effect on type I processivity (38Fulton T.B. Blackburn E.H. Mol. Cell. Biol. 1998; 18: 4961-4970Crossref PubMed Google Scholar). Within the putative yeast telomerase anchor site, the GQ motif is clearly conserved in evolution (18Bosoy D. Peng Y. Mian I. Lue N. J. Biol. Chem. 2003; 278: 3882-3890Abstract Full Text Full Text PDF PubMed Scopus (70) Google Scholar). Thus, I propose that telomerase in other species is also likely to utilize the same domain as the anchor site. Consistent with this notion, deletions in this region of hTERT have been demonstrated to alter primer utilization (30Beattie T.L. Zhou W. Robinson M.O. Harrington L. Mol. Biol. Cell. 2000; 11: 3329-3340Crossref PubMed Scopus (78) Google Scholar). Furthermore, Moriarty et al. (39Moriarty T. Marie-Egyptienne D. Autexier C. Mol. Cell. Biol. 2004; 24: 3720-3733Crossref PubMed Scopus (99) Google Scholar) have recently demonstrated a convincing role for the GQ motif of hTERT (named RID1 in their report) in supporting repeat addition processivity. That the N-region is not highly conserved raises the additional possibility that different anchor sites may have evolved slightly different recognition properties for species-specific purposes (e.g. de novo telomere formation, which requires recognition of non-telomeric sequences and appears to be more efficient in ciliates (40Karamysheva Z. Wang L. Shrode T. Bednenko J. Hurley L. Shippen D. Cell. 2003; 113: 565-576Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar)). The N-GQ domain may have functions other than serving as the anchor site. For example, part of the GQ motif has been shown to mediate interaction between Est2p and a regulatory subunit known as Est3p (41Friedman K. Heit J. Long D. Cech T. Mol. Biol. Cell. 2003; 14: 1-13Crossref PubMed Scopus (54) Google Scholar). Even though EST3 is absolutely essential for telomere maintenance in vivo, deletion of this gene does not appear to affect the processivity of telomerase (data not shown), implying that it is not necessary for anchor site function. Nevertheless, the interaction of the anchor site domain with a regulatory subunit underscores the possibility that this domain may be a key target of regulation in vivo. Anchor Site Function Is Required for Telomere Maintenance in Vivo—Earlier analysis has established a critical role for the N-GQ domain of TERT in telomerase function in vivo: yeast strains bearing mutations in this domain of EST2 suffered severe telomere shortening and eventually became senescent (13Friedman K.L. Cech T.R. Genes Dev. 1999; 13: 2863-2874Crossref PubMed Scopus (121) Google Scholar, 14Xia J. Peng Y. Mian I.S. Lue N.F. Mol. Cell. Biol. 2000; 20: 5196-5207Crossref PubMed Scopus (121) Google Scholar). Coupled with the current biochemical analysis, these observations imply that the protein-DNA interaction conferred by the telomerase anchor site is indeed essential for telomerase function in vivo, possibly for one of the following reasons. First, the available molecular genetic data indicate that telomerase preferentially extends short telomeres by multiple repeats within a single cell cycle, very likely through processive repeat addition (42Teixeira M. Arneric M. Sperisen P. Lingner J. Cell. 2004; 117: 323-335Abstract Full Text Full Text PDF PubMed Scopus (408) Google Scholar, 43Marcand S. Brevet V. Gilson E. EMBO J. 1999; 18: 3509-3519Crossref PubMed Scopus (167) Google Scholar, 44Forstemann K. Hoss M. Lingner J. Nucleic Acids Res. 2000; 28: 2690-2694Crossref PubMed Scopus (96) Google Scholar). If this were the case, then the anchor site mutants would be unable to maintain telomeres because they are defective in repeat addition processivity. Alternatively, the synthesis of long telomeric DNA in vivo may be due to multiple cycles of binding, polymerization, and dissociation (i.e. without processive repeat addition). In this latter scenario, the N-GQ mutants may be non-functional because they are unable to rebind the telomere ends using the 3′-region of the RNA template after one round of synthesis. (This latter scenario seems unlikely, given that only a few telomeres are elongated in a cell cycle.) Finally, it has been suggested that telomerase may have a protective role at telomeres by virtue of its ability to bind stably to telomeric DNA (27Prescott J. Blackburn E.H. Genes Dev. 1997; 11: 2790-2800Crossref PubMed Scopus (145) Google Scholar, 35Prescott J. Blackburn E.H. Genes Dev. 1997; 11: 528-540Crossref PubMed Scopus (136) Google Scholar). This protective function can be compromised by the loss of anchor site interaction. Regardless of the precise mechanisms, our results demonstrate that anchor site, a unique structural and biochemical attribute of the telomerase enzyme, is indeed critical for telomere maintenance. I thank Prof. Jim Wang for prior support and advice and for setting a lofty standard in nucleic acid enzymology. The Department of Microbiology and Immunology at Weill Cornell Medical College gratefully acknowledges the support of the William Randolph Hearst Foundation. Download .pdf (.21 MB) Help with pdf files
Referência(s)