Portal de Periódicos da CAPES

Identification of Functional Domains and Dominant Negative Mutations in Vertebrate Telomerase RNA Using an in VivoReconstitution System

2001; Elsevier BV; Volume: 276; Issue: 8 Linguagem: Inglês

10.1074/jbc.m008419200

1083-351X

Luis Martı́n-Rivera, Marı́a A. Blasco,

RNA Interference and Gene Delivery

The telomerase holoenzyme consists of two essential components, a reverse transcriptase, TERT (telomerase reverse transcriptase), and an RNA molecule, TR (telomerase RNA, also known as TERC), that contains the template for the synthesis of new telomeric repeats. Telomerase RNA has been isolated from 32 different vertebrates, and a common secondary structure has been proposed (Chen, J.-L., Blasco, M. A., and Greider, C. W. (2000)Cell 100, 503–514). We have generated 25 mutants in the four conserved structural domains of the mousetelomerase RNA molecule, mTR, and assayed their ability to reconstitute telomerase activity inmTR −/− cells in vivo. We found that the pseudoknot and the CR4/CR5 domains are required for telomerase activity but are not essential for mTR stability in the cell, whereas mutations in the BoxH/ACA and the CR7 domains affect mTR accumulation in the cell. We have also identified mTRmutants that are able to inhibit wild type telomerase in vivo. The telomerase holoenzyme consists of two essential components, a reverse transcriptase, TERT (telomerase reverse transcriptase), and an RNA molecule, TR (telomerase RNA, also known as TERC), that contains the template for the synthesis of new telomeric repeats. Telomerase RNA has been isolated from 32 different vertebrates, and a common secondary structure has been proposed (Chen, J.-L., Blasco, M. A., and Greider, C. W. (2000)Cell 100, 503–514). We have generated 25 mutants in the four conserved structural domains of the mousetelomerase RNA molecule, mTR, and assayed their ability to reconstitute telomerase activity inmTR −/− cells in vivo. We found that the pseudoknot and the CR4/CR5 domains are required for telomerase activity but are not essential for mTR stability in the cell, whereas mutations in the BoxH/ACA and the CR7 domains affect mTR accumulation in the cell. We have also identified mTRmutants that are able to inhibit wild type telomerase in vivo. telomerase reverse transcriptase telomerase RNA kilobase pairs polymerase chain reaction Telomerase is a reverse transcriptase that synthesizes telomeric repeats onto chromosome ends (2Blackburn E.H. Nature. 1991; 350: 569-573Crossref PubMed Scopus (3079) Google Scholar, 3Greider C.W. Blackburn E.H. Cell. 1985; 43: 405-413Abstract Full Text PDF PubMed Scopus (2628) Google Scholar, 4Morin G.B. Cell. 1989; 59: 521-529Abstract Full Text PDF PubMed Scopus (1377) Google Scholar, 5Yu G.-L. Bradley J.D. Attardi L.D. Blackburn E.H. Nature. 1990; 344: 126-132Crossref PubMed Scopus (551) Google Scholar). The catalytic protein subunit, TERT,1 shares sequence features with reverse transcriptases (6Counter C.M. Meyerson M. Eaton E.N. Weinberg R.A. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 9202-9207Crossref PubMed Scopus (206) Google Scholar, 7Greenberg R.A. Allsopp R.C. Chin L. Morin G. DePinho R. Oncogene. 1998; 16: 1723-1730Crossref PubMed Scopus (295) Google Scholar, 8Harrington L. Zhou W. McPhail T Oulton R. Yeung D. Mar V. Bass Robinson W. Genes Dev. 1997; 11: 3109-3115Crossref PubMed Scopus (410) Google Scholar, 9Lingner J. Hughes T.R. Schevchenko A. Mann M. Lundblad V. Cech T. Science. 1997; 276: 561-567Crossref PubMed Scopus (1044) Google Scholar, 10Martı́n-Rivera L. Herrera E. Albar J.P. Blasco M.A. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 10471-10476Crossref PubMed Scopus (205) Google Scholar, 11Meyerson M. Counter C.M. Eaton E.Ng. Ellisen L.W. Steiner P. Dickinson Caddle S. Ziaugra L. Beijersbergen R.L. Davidoff M.J. Liu Q. Bacchetti S. Haber D.A. Weinberg R.A. Cell. 1997; 90: 785-795Abstract Full Text Full Text PDF PubMed Scopus (1666) Google Scholar, 12Nakamura T.M. Morin G.B. Chapman K.B. Weinrich S.L. Andrews W.H. Lingner J. Harley C.B. Cech T. Science. 1997; 277: 955-959Crossref PubMed Scopus (2075) Google Scholar). TERT is tightly associated with an RNA molecule, TR, that provides the template for the synthesis of new telomeric repeats (5Yu G.-L. Bradley J.D. Attardi L.D. Blackburn E.H. Nature. 1990; 344: 126-132Crossref PubMed Scopus (551) Google Scholar, 13Greider C.W. Blackburn E.H. Nature. 1989; 337: 331-337Crossref PubMed Scopus (1316) Google Scholar). Several associated proteins interact with the TR and TERT components of the telomerase holoenzyme (14Collins K. Kobayashi R. Greider C.W. Cell. 1995; 81: 677-686Abstract Full Text PDF PubMed Scopus (243) Google Scholar, 15Harrington L. McPhail T. Mar V. Zhou W. Oulton R. Bass M.B. Arruda I. Robinson M.O. Science. 1997; 275: 973-977Crossref PubMed Scopus (631) Google Scholar, 16Holt S.E. Aisner D.L. Baur J. Tesmer V.M. Dy M. Ouellette M. Trager J.B. Morin G.B. Toft D.O. Shay J.W. Wright W.E. White M.A. Genes Dev. 1999; 13: 817-826Crossref PubMed Scopus (477) Google Scholar, 17Nakayama J. Saito M. Nakamura H. Matsuura A. Ishikawa F. Cell. 1997; 88: 875-884Abstract Full Text Full Text PDF PubMed Scopus (373) Google Scholar). The telomerase RNA molecule has been isolated from ciliates, yeasts, and mammals and shown to be essential for telomerase activity (13Greider C.W. Blackburn E.H. Nature. 1989; 337: 331-337Crossref PubMed Scopus (1316) Google Scholar, 18Blasco M.A. Funk W.D. Villeponteau B. Greider C.W. Science. 1995; 269: 1267-1270Crossref PubMed Scopus (356) Google Scholar, 19Feng J. Funk W.D. Wang 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. Villeponteau B. Science. 1995; 269: 1236-1241Crossref PubMed Scopus (2087) Google Scholar, 20Singer M.S. Gottschling D.E. Science. 1994; 266: 387-388Crossref PubMed Scopus (653) Google Scholar). The telomerase RNAs from ciliates range in size from 148 to 209 nucleotides and share a conserved secondary structure (21Lingner J. Hendrick L.L. Cech T.R. Genes and Dev. 1994; 8: 1984-1998Crossref PubMed Scopus (177) Google Scholar, 22McCormick-Graham M. Romero D.P. Nucleic Acids Res. 1995; 23: 1091-1097Crossref PubMed Scopus (86) Google Scholar, 23Romero D.P. Blackburn E.H. Cell. 1991; 67: 343-353Abstract Full Text PDF PubMed Scopus (204) Google Scholar). In vertebrates, telomerase RNAs are longer, from 382 to 559 nucleotides (1Chen J.-L. Blasco M.A. Greider C.W. Cell. 2000; 100: 503-514Abstract Full Text Full Text PDF PubMed Scopus (502) Google Scholar, 18Blasco M.A. Funk W.D. Villeponteau B. Greider C.W. Science. 1995; 269: 1267-1270Crossref PubMed Scopus (356) Google Scholar, 19Feng J. Funk W.D. Wang 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. Villeponteau B. Science. 1995; 269: 1236-1241Crossref PubMed Scopus (2087) Google Scholar, 24Tsao D.A. Wu C.W. Lin Y.S. Gene (Amst.). 1998; 9: 51-58Crossref Scopus (18) Google Scholar), and in yeast, they are 1200–1300 nucleotides in length (20Singer M.S. Gottschling D.E. Science. 1994; 266: 387-388Crossref PubMed Scopus (653) Google Scholar, 25McEachern M.J. Blackburn E.H. Genes Dev. 1996; 10: 1822-1834Crossref PubMed Scopus (207) Google Scholar). Several elements of secondary structure have been described both in mammals and in yeasts (26Mitchell J.R. Cheng J. Collins K. Mol. Cell. Biol. 1999; 19: 567-576Crossref PubMed Scopus (437) Google Scholar, 27Roy J. Fulton T.B. Blackburn E.H. Genes Dev. 1998; 12: 3286-3300Crossref PubMed Scopus (71) Google Scholar). In particular, a small nucleolar RNA BoxH/ACA structure has been identified at the 3′ end of mTR and hTR and found to be important for TR stability and processing in vivo (26Mitchell J.R. Cheng J. Collins K. Mol. Cell. Biol. 1999; 19: 567-576Crossref PubMed Scopus (437) Google Scholar). Several structure-function analyses of the full-length human telomeraseRNA, hTR, have been carried out using various in vitro reconstitution assays (28Autexier C. Pruzan R. Funk W. Greider C.W. EMBO J. 1996; 15: 5928-5935Crossref PubMed Scopus (122) Google Scholar, 29Beattie T.L. Zhou W. Robinson M.O. Harrington L. Curr. Biol. 1998; 8: 177-180Abstract Full Text Full Text PDF PubMed Scopus (323) Google Scholar, 30Tesmer V.M. Ford L.P. Holt S.E. Frank B.C. Yi X. Aisner D.L. Ouellette M. Shay J.W. Wright W.E. Mol. Cell. Biol. 1999; 19: 6207-6216Crossref PubMed Scopus (112) Google Scholar). Importantly, a secondary structure for vertebrate TRs has been recently determined by phylogenetic comparative analysis (1Chen J.-L. Blasco M.A. Greider C.W. Cell. 2000; 100: 503-514Abstract Full Text Full Text PDF PubMed Scopus (502) Google Scholar). In this study it is proposed that all vertebrate telomerase RNAs share four highly conserved structural domains as follows: a pseudoknot domain, a CR4-CR5 domain, a BoxH/ACA domain, and a CR7 domain (1Chen J.-L. Blasco M.A. Greider C.W. Cell. 2000; 100: 503-514Abstract Full Text Full Text PDF PubMed Scopus (502) Google Scholar). Here we address the role of the four conserved domains in vertebrate telomerase RNA by assaying the ability of mTR mutations to reconstitute telomerase activity inmTR −/− cells in vivo(31Blasco M.A. Lee H.-W. Hande P. Samper E. Lansdorp P. DePinho R. Greider C.W. Cell. 1997; 91: 25-34Abstract Full Text Full Text PDF PubMed Scopus (1807) Google Scholar). In contrast to previous studies in mammals, mutantmTRs were expressed under the endogenous mouse telomerase gene promoter in an in vivo reconstitution system. In ourin vivo reconstitution analysis, we find regions of mTR that are important for telomerase activity, RNA accumulation, and processingin vivo, and we map these regions onto the recently proposed secondary structure of vertebrate telomerase RNA (1Chen J.-L. Blasco M.A. Greider C.W. Cell. 2000; 100: 503-514Abstract Full Text Full Text PDF PubMed Scopus (502) Google Scholar). Furthermore, we identified mTR unpaired regions in the template, the pseudoknot and the CR4-CR5 domains, which are potential targets for telomerase inhibition. All mutants were generated using as template a plasmid with 1.9 kb of mouse genomic DNA containing the wild typemTR gene under its own promoter sequences. This plasmid was constructed by digesting a 5-kb EcoRI genomic fragment containing the mTR locus (18Blasco M.A. Funk W.D. Villeponteau B. Greider C.W. Science. 1995; 269: 1267-1270Crossref PubMed Scopus (356) Google Scholar) with XbaI andEcoRV and then subcloning the resulting 1.9-kb fragment in a pBluescript SK−. To generate deletion and substitution mutants, we used the Sculptor in vitro mutagenesis system from Amersham Pharmacia Biotech (32Sayers J.R. Krekel C. Eckstein F. BioTechniques. 1992; 13: 592-596PubMed Google Scholar). All mutants were sequenced using an ABI 377 DNA Sequencer and the ABI Prism dRhodamine Terminator Cycle Sequencing Ready Reaction Kit (PerkinElmer Life Sciences). The following oligonucleotides were used for mutagenesis: Δ3-10, 5′GGGTATTTAAGGTCGAGGGCGGCTAGGCCTCGGCACGATTTTCATTAGCTGTGGGTTCTGG3′; Δ23-102, 5′CACCTAACCCTGATTTTCATTAGCGGGCCACCGCGCGTTCCCG3′; Δ106-185, 5′GAAAGTCCAGACCTGCAGCGGCGCCCAGTCCCGTACCCGCCTACAGGCCG3′; Δ199-294, 5′GCCGGGCCGCCCAGTCCCGTCCAGGCCGGGCGAGCGCCGCG3′; Δ328-389, 5′GCGCCGCGAGGACAGGAATGTCACAACCCCCATTCCCGCTG3′; Δ56-73, 5′GGTCTTTTGTTCTCCTCCGCCCGCTGCAGCGGGCCAGGAAAGTCCAG3′; Δ120-143, 5′CCAGACCTGCAGCGGGCCACCGCGCGTTCCGCAGGAGCTCCAGGTTCG3′; Δ154-167, 5′CCTCAAAAACAAACGTCAGCGCAGGAGCTAGCTCCGCGGCGCCGGGCCGCCCAGTCCCG3′; Δ220-241, 5′CCGTACCCGCCTACAGGCCGCGGCCGGCCCTGCCGCCGCGAAGAGCTCGCCTCTGTCAGC3′; Δ254-266, 5′GGGTCTTAGGACTCCGCTGCCGCCGCGAGTCAGCCGCGGGGCGCCGGGGGCTG3′; Δ357-364, 5′GGTCCCCGTGTTCGGTGTCTTACCGGGAAGTGCACCCGGAACTCGGTTC3′; Δ341-345, 5′GGACAGGAATGGAACTGGTCCCCGGGTGTCTTACCTGAGCTGTGG3′; Δ31-3, 5′CCCTGATTTTCATTAGCTGTGGGTCTTTTGTTCTCCGCCCGCTGTTTTTC3′; Δ90-94, 5′GCTGACTTTCCAGCGGGCCAGGAAAGACCTGCAGCGGGCCACCGCGTTCCCGAGCCTC3′; substitution 57-61 U → A, 5′GGTCTTTTGTTCTCCGCCCGCTGAAAAACTCGCTGACTTCCAGCGGGCCAGGAAAGTCC3′; substitution 129-133 A → U, 5′CACCGCGCGTTCCCGAGCCTCTTTTTCAAACGCAGCGCAGGAGCTCC3′; substitution 65-70 GCUGAC → GUCAGC, 5′GGTCTTTTGTTCTCCGCCCCTGTTTTTCTCGTCAGCTTCCAGCGGGCCAGGAAAGTCCAG3′; substitution 139-144 GUCAGC → GCUGAC, 5′-GTTCCCGAGCCTCAAAAACAAACCTGACGCAGGAGCTCCAGGTTC3′; substitution 71-74 UUCC → GGAA, 5′GGTCTTTTGTTCTCCGCCCGCTGTTTTTCTCGCTGACGGAAAGCGGGCCAGGAAAGTCCAG3′; substitution 71-74 + 84-87 UUCC → GGAA + GGAA → UUCC, 5′GTTTTTCTCGCTGACGGAAAGCGGGCCATTCCAGTCCAGACCTG3′; box H mutant A319U/ A321U/A324U, 5′CAGGCCGGGCGAGCGCCGCAAGGTCTGGTATGGAACTGGTCCCCGTG3′; box ACA mutant A392U/A394U, 5′GCACCCGGAACTCGGTTCTCTCTACCCCCATTCCCGCTGGGGAAATGCC3′. To generate the rest of mTR mutants, we used QuikChange, a PCR-based mutagenesis kit provided by Stratagene. The oligonucleotides used to generate each mutant are as follows: Δ172-314, forward, 5′CCAGGTTCGCCGGGAGCTCCGAGGACAGGAATGGAACTGG3′; Δ172-314, reverse, 5′CCAGTTCCATTCCTGTCCTCGGAGCTCCCGGCGAACCTGG3′; substitution 65-72 GCUGACUU → AAGUCAGC, forward, 5′GGTCTTTTGTTCTCCGCCCGCTGTTTTTCTCCGACTGAACCAGCGGGCCAGGAAAGTCCAG3′; substitution 65-72 GCUGACUU → AAGUCAGC, reverse, 5′CTGGACTTTCCTGGCCCGCAGGTTCAGTCGGAGAAAAACAGCGGGCGGAGAACAAAAGACC3′; substitution 136-144 AACGUCAGC → GCUGACGUU, forward, 5′CGTTCCCGAGCCTCAAAAACATTCCAGTCGGCAGGAGCTCCAGGTTCG3′; substitution 136-144 AACGUCAGC → GCUGACGUU, reverse, 5′CGAACCTGGAGCTCCTGCCGACTGGAATGTTTTTGAGGCTCGGGAACG3′. A NotI-XbaI fragment containing a neo cassette was subcloned in the plasmids containing the selected mTR mutants (see Fig. 5) and in the empty vector (pBluescript SK−, see above). Immortal fibroblasts, derived from wild type embryos (31Blasco M.A. Lee H.-W. Hande P. Samper E. Lansdorp P. DePinho R. Greider C.W. Cell. 1997; 91: 25-34Abstract Full Text Full Text PDF PubMed Scopus (1807) Google Scholar), were transfected with 20 μg of each of the constructs by conventional calcium phosphate transfection. After 2 days, 1/100, 1/50, 1/20, and 1/10 dilutions were plated in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and G418 (200 μg/ml). After 2 weeks of selection several stable clones were picked from the 1/100 and 1/50 dilutions and expanded in selective medium. S-100 extracts, TRAP assays, and telomerase activity quantifications were performed as described below. KO3 p23 cells were transfected with 20 μg of each plasmid using a conventional calcium phosphate protocol (31Blasco M.A. Lee H.-W. Hande P. Samper E. Lansdorp P. DePinho R. Greider C.W. Cell. 1997; 91: 25-34Abstract Full Text Full Text PDF PubMed Scopus (1807) Google Scholar). As control for transfection efficiency, cells were co-transfected with 1 μg of a LacZ expression vector (J7Laz). The transfection efficiency of the lacZ gene was measured by absorbance at 420 nm after addition of O-nitrophenyl β-d-galactopyranoside (Sigma). Cells were divided 48 h post-transfection and further cultured overnight. One dish was used for S-100 extracts and the other for RNA preparation, as described below. Forty eight hours after transfection, total RNA was prepared as described (33Ausubel F.M. Brent R. Kingston R. Moore D. Seidman J.G. Smith J.A. Struhl K. Current Protocols in Molecular Biology. John Wiley & Sons, Inc., New York1993Google Scholar). In the case of the stable transfectants total RNA was prepared from 10-cm dishes after clone amplification as described above. Typically, 30 μg of total RNA was used for Northern blots, except for the control (wild type cells, 10 μg). RNA was separated by electrophoresis on a 6% acrylamide/bisacrylamide (19:1), 7 m urea, TBE 0.6× gel for 2 h at 5 watts. After electrophoresis, RNA was transferred to HybondTM-N+ membranes (Amersham Pharmacia Biotech). Blots were probed with radiolabeled probes against mTR and 5 S RNA. RNA levels were quantified using IQMac version 1.2, and the mean of at least two different experiments was calculated. In all cases, mTR levels in the cell were corrected by the efficiency of transfection of a LacZ plasmid. Final values of mTR abundance in the cell were corrected by transcript size. Finally, as loading control, gels were also probed with the 5 S RNA. S-100 extracts from transient or stable transfectants were prepared as described (31Blasco M.A. Lee H.-W. Hande P. Samper E. Lansdorp P. DePinho R. Greider C.W. Cell. 1997; 91: 25-34Abstract Full Text Full Text PDF PubMed Scopus (1807) Google Scholar). Protein concentration of extracts was determined using the Bio-Rad Protein Assay (Bio-Rad). Reconstituted telomerase activity was measured 48 h after transfection in a modified version of the TRAP assay (34Blasco M.A. Rizen M. Greider C.W. Hanahan D. Nat. Genet. 1996; 12: 200-204Crossref PubMed Scopus (289) Google Scholar, 35Kim 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. Shay J.W. Science. 1994; 266: 2011-2014Crossref PubMed Scopus (6585) Google Scholar). In the case of the stable transfectants, telomerase activity was measured from 10-cm dishes after clone amplification. Telomerase activity levels were quantified using IQMac version 1.2 (the whole telomerase ladder was quantified at the two different concentrations and corrected by dilution, background, and internal control signal). For quantification, films with nonsaturated signals were used. The TRAP results shown in Fig. 3 are for illustrative purposes; some of the films shown are overexposed to allow the visualization of TRAP activity by mutants that are severely affected. The transfection efficiency of the mTR mutants was determined by co-transfecting a plasmid containingLacZ (not shown). In all cases an internal control for PCR efficiency was included in the TRAP assay (TRAPeze kit, Oncor). One hundred μg of total protein from S-100 extracts (see above) were loaded per lane, and Western blots with K-370 antibody were carried out exactly as described in Martı́n-Rivera et al. (10Martı́n-Rivera L. Herrera E. Albar J.P. Blasco M.A. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 10471-10476Crossref PubMed Scopus (205) Google Scholar). For full characterization of K370 antibody specificity see Martı́n-Rivera et al. (10Martı́n-Rivera L. Herrera E. Albar J.P. Blasco M.A. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 10471-10476Crossref PubMed Scopus (205) Google Scholar). In all vertebrates, four distinct conserved structural domains have been defined for the telomerase RNA molecule as follows: (i) the pseudoknot domain, (ii) the CR4-CR5 domain, (iii) the BoxH/ACA domain, and (iv) the CR7 domain (1Chen J.-L. Blasco M.A. Greider C.W. Cell. 2000; 100: 503-514Abstract Full Text Full Text PDF PubMed Scopus (502) Google Scholar). Fig. 1 shows the conserved secondary structure for vertebrate telomerase RNAs mapped onto themouse telomerase RNA (mTR). Eight regions of base pairing interactions are proposed for mTR (P2–P8), which define the four conserved structural domains. A region of long range base pairing, P1, which is conserved in many vertebrates, is not found in mouse or other rodents (1Chen J.-L. Blasco M.A. Greider C.W. Cell. 2000; 100: 503-514Abstract Full Text Full Text PDF PubMed Scopus (502) Google Scholar) (Fig. 1). All 25 mTRmutations were obtained in the context of a genomic construct containing the mTR gene under endogenous promoter sequences shown to be sufficient to direct mTR gene transcription in vivo (18Blasco M.A. Funk W.D. Villeponteau B. Greider C.W. Science. 1995; 269: 1267-1270Crossref PubMed Scopus (356) Google Scholar, 31Blasco M.A. Lee H.-W. Hande P. Samper E. Lansdorp P. DePinho R. Greider C.W. Cell. 1997; 91: 25-34Abstract Full Text Full Text PDF PubMed Scopus (1807) Google Scholar). The telomerase RNA null cell line KO3 p23, derived from first generationmTR −/− mouse embryonic fibroblasts, was used for in vivo reconstitution of telomerase activity with the mutant mTRs (31Blasco M.A. Lee H.-W. Hande P. Samper E. Lansdorp P. DePinho R. Greider C.W. Cell. 1997; 91: 25-34Abstract Full Text Full Text PDF PubMed Scopus (1807) Google Scholar). To study the effect of the mutations on mTR stability in the cell, we performed Northern blot analysis of total RNA isolated from KO3 p23 cells 48 h after transfection with wild type or mutant mTR constructs ("Experimental Procedures"). Fig. 2,a—c, shows representative Northern blot images of mTR accumulation in the cell 48 h after transfection. InmTR −/− cells transfected with wild type mTR, a full-length mTR transcript stably accumulated, showing the same mobility as the endogenous mTR (Fig. 2,a–c, compare mTR band in wild type andcontrol cells lanes). As expected, when themTR −/− cells were transfected with an empty vector, no mTR-specific signal was detected (see empty vector lane in Fig. 2, a and c). At least 2 to 3 different transfections were carried out for everymTR mutant, and average mutant mTR abundance in the cell was quantified and corrected by transfection efficiency of aLacZ plasmid (see "Experimental Procedures"). Mutant RNA accumulation values relative to that of the transfected wild typemTR are shown in Table I.Table ISummary of stability and activity of the different mTR versionsDomainmTR*RNA accumulation 1-aAverage of 2–3 experiments.Telomerase activity 1-aAverage of 2–3 experiments.%Full lengthWild-type100100PseudoknotΔ23–1029010PseudoknotΔ120–1434510PseudoknotΔ31–3699130PseudoknotΔ56–735115PseudoknotΔ90–9411214Pseudoknot65GCUGAC70–65GUCAGC706020Pseudoknot189GUCAGC144–189GCUGAC1448922Pseudoknot65GCUGAC70–65GUCAGC707781139GUCAGC144–139GCUGAC144Pseudoknot65GCUGACUU72–65AAGUCAGC729830Pseudoknot186AACGUCAGC144–186GCUGACGUU1447515Pseudoknot65GCUGACUU72–65AAGUCAGC7215018186AACGUCAGC144–186GCUGACGUU144Pseudoknot57UUUUU61–57AAAAA619924Pseudoknot129AAAAA188–129UUUUU18811045Pseudoknot57UUUUU61–57AAAAA6111913129AAAAA188–129UUUUU188Pseudoknot CR4-CR5Δ106–1854120Pseudoknot CR4-CR5Δ154–16747100CR4-CR5Δ199–29416UndetectableCR4-CR5Δ220–2416120CR4-CR5Δ254–26685UndetectableCR4-CR5Δ172–31430UndetectableBoxH/AC A-CR7Δ328–389UndetectableUndetectableCR7Δ357–364UndetectableUndetectableBetween BoxH/AC A-CR7Δ341–34570112BoxH/AC ABox HUndetectableUndetectableBoxH/AC ABox ACAUndetectableUndetectable1-a Average of 2–3 experiments. Open table in a new tab mTR deletions Δ23–102, Δ31–36, and Δ90–94 and all 9 mTR substitutions affecting the pseudoknot domain and the pseudoknot interaction accumulated in the cell at the predicted size to similar levels as the transfected wild type mTR (Fig. 2, a—c, and Table I). Mutants Δ120–143 and Δ56–73, which also disrupted the pseudoknot interaction (helix P3), accumulated to 45 and 51% of wild type mTR levels (Fig. 2 a and Table I). A large mTRdeletion (Δ106–185) disrupting conserved helices P3 and P4 in the pseudoknot and BoxA/ACA domains, respectively, reduced mTR accumulation to 41% of wild type levels (Fig. 2 a and Table I). mTR deletion mutant Δ154–167, which reduces the distance between the pseudoknot and the Box/ACA and CR4-CR5 domains, showed 47% of wild type levels (Fig. 2 a and Table I). Interestingly, deletion of several conserved nucleotides of the J6/5 region in mutant Δ254–266 in the CR4-CR5 domain resulted in accumulation of a lower mobility form of mTR (see asteriskin Fig. 2 a), this mutant showed 85% of wild type mTR levels (Fig. 2 a; Table I). The complete sequencing of allmTR constructs rules out a cloning error as responsible for this higher molecular weight mTR (see "Experimental Procedures"). All the other CR4-CR5 mTR mutants accumulated to the predicted size in the cell indicating that the mutated regions do not play a fundamental role in mTR processing. Complete deletion of the CR4-CR5 domain in mutants Δ199–294 and Δ172–314 accumulated to 16 and 30% of wild type levels, respectively (Fig. 2, a-c; Table I). Deletion of nonconserved loop L6 outside the CR4-CR5 domain in mutant Δ220–241 also resulted in stable mTR accumulation (Fig.2 a; Table I). The decreased RNA accumulation of the large deletion mutants in the pseudoknot and CR4-CR5 domains is likely to be due to an altered secondary structure of these mutant mTRs as a consequence of the big size of the deletion; in this regard, none of the point mutations and small deletions in these regions showed decreased accumulation as compared with wild type mTR. Most mutations affecting any of these two domains resulted in undetectable mTR in the cell 48 h after transfection, indicating that they contain essential regions for mTR stability in vivo. In particular, mTR mutants Δ328–389 (BoxH/ACA and CR7) and Δ357–364 (CR7), and both point mutants in the box H and box ACA conserved sequences, resulted in no detectable mTR accumulation in the cell (Fig. 2, a andb; Table I). Interestingly, not all mutations in these domains abolished mTR accumulation; mutants Δ106–185 and Δ172–314 that disrupt helix P4 in the BoxH/ACA domain accumulated at the predicted size and showed 41 and 30% of wild type mTR levels, respectively (Fig. 2, a and c; Table I). Similarly, deletion of four nonconserved J7b/8a unpaired nucleotides between conserved BoxH/ACA and CR7 domains in mutant Δ341–345 resulted in stable accumulation (70% of wild type levels) (Fig.2 a; Table I). These results indicate that the length of the J7b/8a nonconserved hinge region between the two conserved domains as well as helix P4 are not important for mTR stability in vivo. To identify mTR regions important for reconstitution of telomerase activity in vivo, extracts were prepared 48 h after transfection and were tested for telomerase activity using the TRAP assay (see "Experimental Procedures").mTR −/− cells transiently transfected with wild type mTR showed an efficient reconstitution of telomerase activity in vivo (>50% of the parental wild type cell line activity). This activity was considered 100% telomerase activity reconstitution when compared with the different mTR mutants (see Blasco et al. (31Blasco M.A. Lee H.-W. Hande P. Samper E. Lansdorp P. DePinho R. Greider C.W. Cell. 1997; 91: 25-34Abstract Full Text Full Text PDF PubMed Scopus (1807) Google Scholar) andmTR in Fig. 3,a–e; Table I). No reconstitution of telomerase activity was observed when the cells were transiently transfected with an empty vector (Fig. 3, a and e). To rule out that lack of telomerase reconstitution could be due to absence of mTERT accumulation in the cell, we did Western blot analysis of nuclear extracts from cells transfected with the different mTRversions by using the anti-telomerase-specific antibody K370 (10Martı́n-Rivera L. Herrera E. Albar J.P. Blasco M.A. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 10471-10476Crossref PubMed Scopus (205) Google Scholar). In all cases, except in Box H mutant, similar levels of mTERT protein were detected (Fig. 4). All mutants disrupting the pseudoknot interaction (highly conserved helix P3) affected in vivoreconstitution of telomerase activity. In particular, mutants Δ23–102, Δ56–73, and Δ120–143, all of which disrupt the P3 interaction, showed 10, 15, and 10% of wild type telomerase activity, respectively (Fig. 3, a and b; TableI). In contrast, mutant Δ31–36, which eliminates the unpaired mammalian-specific J2a.1/2a sequence but does not affect the pseudoknot interaction, showed 130% of the wild type mTR telomerase activity (Fig. 3 c), consistent with the fact that the J2a.1/2a sequence is not conserved (1Chen J.-L. Blasco M.A. Greider C.W. Cell. 2000; 100: 503-514Abstract Full Text Full Text PDF PubMed Scopus (502) Google Scholar). Deletion of four unpaired nucleotides in mutant Δ90–94 decreased telomerase activity to 14% (Fig. 3 c; Table I). It has been previously shown in ciliates that the pseudoknot topology rather than the sequence is relevant for telomerase activity (36Gilley D. Blackburn E.H. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 6621-6625Crossref PubMed Scopus (86) Google Scholar). To test whether a similar situation is true for vertebrate telomerase RNA pseudoknot, we generated 9 different substitution mutants. Mutant65GCUGAC70 to65GUCAGC70, which partially disrupted the ability of helix P3 to base pair but did not alter the size of the RNA, showed 20% of the wild type telomerase activity (Fig.3 b; Table I). Mutant 139GUCAGC144 to139GCUGAC144, which partially disrupts helix P3, showed 22% of the wild type activity (Fig. 3 b; TableI). A double substitution mutant 65GCUGAC70 to65GUCAGC70/139GUCAGC144to 139GCUGAC144 which fully restores the base pairing ability in helix P3 showed 81% of the wild type activity (Fig.3 b; Table I). We tried to confirm the above results by constructing new substitution mutants that completely disrupt the P3 helix. Mutant 65GCUGACUU72 to65AAGUCAGC72 showed 30% of the wild type telomerase activity (Fig. 3 e; Table I). Mutant136AACGUCAGC144 to136GCUGACGUU144 showed 15% of the wild type activity (Fig. 3 e; Table I). However, a double substitution mutant 65GCUGACUU72 to65AAGUCAGC72/136AACGUCAGC144to 136GCUGACGUU144 showed 18% of wild type activity (Fig. 3 e; Table I). All together, these results indicate that, in contrast to ciliate pseudoknot, in vertebrate telomerase RNA both pseudoknot topology and sequence are important for reconstitution of telomerase activity. We do not know if the vertebrate pseudoknot interacts with TERT, as has been shown for the ciliate pseudoknot (36Gilley D. Blackburn E.H. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 6621-6625Crossref PubMed Scopus (86) Google Scholar); however, the fact that some of the mutants in this domain inhibit wild type telomerase activity suggests that it does (see below). Three more mutants in the pseudoknot domain were tested that alter two sequences that are not base-paired in the mTR structure. Mutant57UUUUU61 to 57AAAAA61in the J2b/3 region showed 24% of wild type reconstitution activity, indicating that the identity of these residues is important for wild type reconstitution of telomerase activity in agreement with the fact that these residues are highly conserved in all vertebrates (Fig.3 b; Table I) (1Chen J.-L. Blasco M.A. Greider C.W. Cell. 2000; 100: 503-514Abstract Full Text Full Text PDF PubMed Scopus (502) Google Scholar). Mutant 129AAAAA133to 129UUUUU133, which alters a nonconserved sequence in the J2a/3 region, showed 45% of wild type telomerase activity (Fig. 3 b; Table I). A double substitution mutant57UUUUU61 to57AAAAA61/129AAAAA133to 129UUUUU133 showed 13% of the wild type reconstitution activity (Fig. 3 b; Table I), ruling out an interaction between 57UUUUU61 and129AAAAA133 sequences, as predicted from the mTR structure (1Chen J.-L. Blasco M.A. Greider C.W. Cell. 2000; 100: 503-514Abstract Full Text Full Text PDF PubMed Scopus (502) Google Scholar). Deletion of mTR unpaired regions in mutant Δ154–167 reduces the distance between the pseudoknot domain and the CR4-CR5 and BoxH/ACA domains. This mutant showed 100% of wild type telomerase activity (Table I; Fig. 3 b), indicating that the distance between these structural domains is not essential for reconstitution of telomerase activity, in agreement with the variable length of this region in vertebrate telomerase RNAs (1Chen J.-L. Blasco M.A. Greider C.W. Cell. 2000; 100: 503-514Abstract Full Text Full Text PDF PubMed Scopus (502) Google Scholar). Mutant Δ106–185, which removes J2a/3 sequences and disrupts the P3 helix in the pseudoknot domain, and the P4 helix in the BoxH/ACA domain showed 20% of the wild type mTR telomerase activity (Fig.3 a; Table I). In contrast to pseudoknot domain deletion, full deletion of the CR4-CR5 domain in mutants Δ199–294 and Δ172–314 resulted in complete absence of telomerase reconstitutionin vivo (Fig. 3, a and c), indicating that this domain contains essential regions for in vivoreconstitution of telomerase activity. In particular, deletion of several highly conserved residues in the J6/5 unpaired region of the CR4-CR5 domain in mutant Δ254–266 completely abolished the ability of mTR to reconstitute active telomerase complexes in vivo(Fig. 3, b and c; Table I). Absence of telomerase activity reconstitution by mutant Δ254–266 could be due to the altered mobility shown by this mutant (see above). Deletion of loop L6 outside the CR4-CR5 domain in mutant Δ220–241 reduced telomerase activity to 20% that of wild type (Fig. 3, b andc), indicating that this loop is important but not essential for telomerase activity, in agreement with the fact that is not conserved in vertebrates (1Chen J.-L. Blasco M.A. Greider C.W. Cell. 2000; 100: 503-514Abstract Full Text Full Text PDF PubMed Scopus (502) Google Scholar). Some of the mutations in BoxH/ACA and CR7 domains resulted in no accumulation of mTR in vivo (see above) and should result in the absence of active telomerase complexes in vivo. Deletion mutant Δ328–389, which removes part of the highly conserved BoxH/ACA domain (but not the highly conserved H and ACA boxes) and the CR7 domain, resulted in lack of reconstitution of active telomerase complexes (Fig. 3 a; Table I), in agreement with undetectable accumulation of this RNA in the cell (see above). Similarly, a triple mutation in the invariable A residues of box H (A319U/A321U/A324U) and a double mutation in the invariable A residues of the ACA box (A392U/A394U) in the Box/ACA domain also abolished telomerase activity reconstitution (Fig.3 d). Finally, elimination of L8 loop in the CR7 domain in mutant Δ357–364, also resulted in lack of telomerase activity reconstitution (Fig. 3, b and c). Interestingly, elimination of four nucleotides of the junction between the BoxH/ACA and CR7 domains (J7b/8a) in mutant Δ341–345 did not affect telomerase activity, this mutant showing 112% of the wild type telomerase activity (Fig. 3 c; Table I), concurring with the fact that the length of this hinge region is not conserved in vertebrate TRs (1Chen J.-L. Blasco M.A. Greider C.W. Cell. 2000; 100: 503-514Abstract Full Text Full Text PDF PubMed Scopus (502) Google Scholar). Similarly, deletion of P4 helix in the Box/ACA domain in mutant Δ106–185 did not result in lack of reconstitution of telomerase activity (see above). Several of the mTR mutants studied affected telomerase reconstitution in vivo without significantly affecting mTR stability in the cell. Furthermore, some of these mTR mutations consist of small deletions and, therefore, are unlikely to alter correct mTR folding. These mutantmTRs are good candidates to act as telomerase inhibitors when overexpressed in wild type cells. The following mTRmutations were tested as potential dominant negatives: mutants Δ120–143 and Δ56–73 in the pseudoknot domain and mutant Δ220–241 in the CR4-CR5 domain, all of which resulted in 80–90% reduction of wild type mTR reconstitution activity. We also constructed and studied an mTR mutant that lacked the template region, mutant Δ3–10 (see "Experimental Procedures"). To check the ability of the different mTR mutations to inhibit wild type telomerase activity, we transfected a wild type 3T3 cell line (parental cells in Fig. 5) (31Blasco M.A. Lee H.-W. Hande P. Samper E. Lansdorp P. DePinho R. Greider C.W. Cell. 1997; 91: 25-34Abstract Full Text Full Text PDF PubMed Scopus (1807) Google Scholar) with either an empty vector or with the different mutantmTR constructs, and in all cases the constructs contained a neomycin resistance gene for selection of stable clones (see "Experimental Procedures"). After selection with G418, several clones were isolated, and telomerase activity was assayed and quantified as described before. We considered that there was inhibition when telomerase activity was 30% or less that of the parental cell line. In the case of all mTR mutations assayed, but in none of the empty vector-transfected cells, we were able to isolate clones that showed inhibition of telomerase activity. Clones showing telomerase inhibition were 0% for the empty vector (0 of 8), 35% for Δ3–10 mutant (5 of 14), 41% for Δ56–73 mutant (5 of 12), 11.1% for Δ120–143 mutant (1 of 9), and 45.5% for Δ220–241 mutant (5 of 11). Fig. 5 shows telomerase activity quantification of mutantmTR-stable clones as compared with cells transfected with the empty vector or as compared with parental wild type cells (Fig. 5; see parental cells and empty vector lanes). These results show that mutations in the template, pseudoknot, and CR4-CR5 domains render mTR molecules that inhibit telomerase activity when expressed in wild type cells, suggesting that they act as dominant negatives. Furthermore, these results suggest that these regions are not essential for interaction with the mTERT subunit of telomerase. The data presented here show that mutations in any of the four conserved mTR domains affect in vivo reconstitution of active telomerase complexes, suggesting that the full-length mTR molecule is important for in vivo formation of active telomerase complexes. In particular, our structure-function analysis supports a model in which vertebrate telomerase RNAs have two main functional domains as follows: (i) a domain that is essential for mTR stability, including the conserved BoxH/ACA and CR7 regions (highlighted in yellow in Fig.6), and (ii) a domain that is important for activity and that includes the conserved pseudoknot and CR4-CR5 regions (highlighted in blue in Fig. 6). A previous report showed that hTR nucleotides from nucleotides 211 to 241, (including CR4-CR5, BoxH/ACA, and CR7 domains) were important for hTR accumulation and processing in the cell (26Mitchell J.R. Cheng J. Collins K. Mol. Cell. Biol. 1999; 19: 567-576Crossref PubMed Scopus (437) Google Scholar). In this regard, we find that regions within the BoxH/ACA and CR7 domains, but not in the CR4-CR5 domain, are essential for mTR accumulation in vivo. The BoxH/ACA domain has been also shown to be important for stability in small nucleolar RNAs (37Smith C.M. Steitz J.A. Cell. 1997; 89: 669-672Abstract Full Text Full Text PDF PubMed Scopus (283) Google Scholar, 38Weinstein L.B. Steitz J.A. Curr. Opin. Cell Biol. 1999; 11: 378-384Crossref PubMed Scopus (253) Google Scholar). Importantly, mutants affecting helix P4 of BoxH/ACA domain and the hinge region between BoxH/ACA and CR7 domains do not abolish mTR accumulation in the cell and are not highlighted in yellow in Fig. 6. Our results also suggest that the CR4-CR5 domain, formerly included in the "BoxH/ACA region" by Mitchell et al. (26Mitchell J.R. Cheng J. Collins K. Mol. Cell. Biol. 1999; 19: 567-576Crossref PubMed Scopus (437) Google Scholar), contains a highly conserved region that seems important for mTR processing (highlighted ingreen in Fig. 6). As mentioned previously, the mTR domains important for reconstitution of telomerase activity include the pseudoknot and CR4-CR5 domains and are highlighted in blue in Fig. 6. Furthermore, the fact that mutants in these regions are able to inhibit wild type telomerase activity when overexpressed in vivo suggests that mutations in these regions do not abolish binding to mTERT. It is important to notice the fact that none of the pseudoknot mutants resulted in complete absence of telomerase activity reconstitution, suggesting that the pseudoknot domain is not essential for in vivoreconstitution of telomerase activity. In contrast, analysis of mutants in the CR4-CR5 domain suggested an essential role of CR4-CR5 domain in reconstitution of telomerase activity. This coincides with a previous report (26Mitchell J.R. Cheng J. Collins K. Mol. Cell. Biol. 1999; 19: 567-576Crossref PubMed Scopus (437) Google Scholar) that suggested that nucleotides 211–241 in hTR (including the CR4-CR5 domain) contained regions that were important for telomerase activity. In contrast, in vitro reconstitution studies yielded different results regarding the minimal hTR region required for telomerase activity. The studies by Autexier et al. (28Autexier C. Pruzan R. Funk W. Greider C.W. EMBO J. 1996; 15: 5928-5935Crossref PubMed Scopus (122) Google Scholar), which mapped the minimal functional region of hTR to nucleotides 44–203 using a reconstitution system with Micrococcus nuclease-treated, partially purified human 293 cells, and that by Beattie et al. (29Beattie T.L. Zhou W. Robinson M.O. Harrington L. Curr. Biol. 1998; 8: 177-180Abstract Full Text Full Text PDF PubMed Scopus (323) Google Scholar), which mapped the minimal hTR region to nucleotides 10–154 using in vitroproduced hTR and hTERT, do not coincide with the in vivoreconstitution results that show that the conserved CR4-CR5 domain is essential for telomerase activity. The results we obtained with thein vivo reconstitution system are coincidental with those by Tesmer et al. (30Tesmer V.M. Ford L.P. Holt S.E. Frank B.C. Yi X. Aisner D.L. Ouellette M. Shay J.W. Wright W.E. Mol. Cell. Biol. 1999; 19: 6207-6216Crossref PubMed Scopus (112) Google Scholar) that described that hTR nucleotides 1–325 (including the CR4-CR5 domain) but not 1–300 were sufficient to reconstitute wild type telomerase activity, again suggesting that the CR4-CR5 domain is important for telomerase activity. It is important to notice that mutants in mTR regions that are not conserved in other vertebrate telomerase RNAs were less critical for both mTR activity and stability. Structure-function studies in mammalian telomerase RNAs are a necessary step for a rational design of telomerase inhibitors. In this regard, we identify here several short sequences in the template, pseudoknot, and CR4-CR5 conserved domains that when mutated act as dominant negatives. These three mTR unpaired regions in the template, pseudoknot, and CR4-CR5 domains are good candidates for antisense inhibition of telomerase. We thank Carol Greider for sharing the vertebrate telomerase RNA secondary structure before publication and for providing Fig. 1, and Manuel Serrano and Cathy Mark for critical reading of the manuscript. The Department of Immunology and Oncology was founded and is supported by the Spanish Research Council (CSIC) and Amersham Pharmacia Biotech and The Upjohn Co.