- Email: [email protected]
Functional Conservation of the Telomerase Protein Est1p in Humans
Current Biology, Vol. 13, 698–704, April 15, 2003, 2003 Elsevier Science Ltd. All rights reserved. DOI 10.1016/S 09 60 - 98 22 ( 03 )0 0 21 0- 0
Functional Conservation of the Telomerase Protein Est1p in Humans Bryan E. Snow,1 Natalie Erdmann,1 Jennifer Cruickshank,1 Hartt Goldman,1 R. Montgomery Gill,2 Murray O. Robinson,4,5 and Lea Harrington1,3,* 1 Ontario Cancer Institute/Advanced Medical Discovery Institute University of Toronto 620 University Avenue Toronto, Ontario M5G 2C1 2 Ontario Cancer Institute 3 Department of Medical Biophysics University of Toronto 610 University Avenue Toronto, Ontario M5G 2M9 Canada 4 Amgen 1 Amgen Drive Thousand Oaks, California 91320
Summary Eukaryotic telomerase contains a telomerase reverse transcriptase (TERT) and an RNA template component that are essential for telomerase catalytic activity and several other telomerase-associated factors of which only a few appear to be integral enzyme components [1–3]. The first essential telomerase protein identified was S. cerevisiae Est1p, whose deletion leads to evershorter telomeres despite the persistence of telomerase activity [4–6]. Extensive genetic and biochemical data show that Est1p, via its interaction with the telomerase RNA and telomere end DNA binding complex Cdc13p/ Stn1p/Ten1p, promotes the ability of telomerase to elongate telomeres in vivo [7–22]. The characterization of Est1p homologs outside of yeast has not been documented. We report the characterization of two putative human homologs of Est1p, hEST1A and hEST1B. Both proteins specifically associated with telomerase activity in human cell extracts and bound hTERT in rabbit reticulocyte lysates independently of the telomerase RNA. Overproduction of hEST1A cooperated with hTERT to lengthen telomeres, an effect that was specific to cells containing telomerase activity. Like Est1p, hEST1A (but not hEST1B) exhibited a single-stranded telomere DNA binding activity. These results suggest that the telomerase-associated factor Est1p is evolutionarily conserved in humans. Results and Discussion Using a reiterative PSI-Blast search of the NCBI nonredundant database with the S. cerevisiae Est1p polypeptide, we identified a region of homology between Est1p (aa 82–341) and an Ebs1-like protein from K. lactis (Gen*Correspondence: [email protected] 5 Present address: GenPath Pharmaceuticals, 300 Technology Square, 7th Floor, Cambridge, Massachusetts 02139.
Bank accession number AAG49579; Expect ⫽ 9 ⫻ 10⫺15) [23]. The second iterative search identified proteins with similarity from A. thaliana (NP_197441; Expect ⫽ 0.006) and N. tabacum (BAB82502; Expect ⫽ 0.072). The final iterative search revealed human KIAA0732 (AA34452; Expect ⫽ 8 ⫻ 10⫺14), KIAA1089 (BAA83041; Expect ⫽ 6 ⫻ 10⫺10), and KIAA0250 (BAA13381; Expect ⫽ 6 ⫻ 10⫺15), as well as uncharacterized proteins in D. melanogaster (AAF56380; Expect ⫽ 9 ⫻ 10⫺9), C. elegans (NP_497566; Expect ⫽ 2 ⫻ 10⫺9), and A. gambiae (EAA12461; Expect ⫽ 2 ⫻ 10⫺5) (Figure 1). A smaller conserved region between these proteins (aa 190–254 of Est1p) was previously noted to contain two putative tetratricopeptide repeat (TPR) sequences, which play a role in protein-protein interactions [24, 25] (Figure 1). Although no specific function has been ascribed to the TPR repeats in Est1p, deletion of a portion of the TPR repeats (aa 159–323) abolishes telomere length maintenance in vivo [10]. KIAA0732 and KIAA1089 contain a C-terminal PIN domain (PilT amino-terminal), which is proposed to confer nucleic acid binding and, in some cases, nuclease activity [24]. The PIN domain is not conserved in Est1p or KIAA0250 [24] (Figure 1 and data not shown). The human KIAA0732 protein (hereafter called hEST1A) showed the highest similarity to Est1p (aa 57– 284) with 17.5% identity and 28.9% similarity, and KIAA1089 (hEST1B) exhibited 13.5% identity and 27% similarity (Figure 1). KIAA0250 (hEST1C) exhibited 15.4% identity and 25.4% similarity to Est1p aa 57–284 and has not yet been characterized. Full-length cDNAs for both hEST1A and hEST1B were used to probe mRNA expression in various human tissues and cell lines. hEST1A and hEST1B mRNAs existed as predominantly single species at 7.5 kb and 4.6 kb, respectively, and were expressed in all human tissues and cell lines examined (see Figure S1 in the Supplemental Data available with this article online). In S. cerevisiae cell extracts, Est1p binds the telomerase RNA independently of the other telomerase components; however, its association with Est2p (TERT) depends on the presence of the telomerase RNA [12, 13, 15, 22]. We tested whether endogenous hEST1A or hEST1B associated with telomerase activity in human cell extracts (Figure 2). Separate antibodies raised against hEST1A or hEST1B peptides specifically immunoprecipitated only their cognate recombinant protein (Figure 2A) and did not exhibit cross-reactivity. Immunoprecipitations from Raji, 293, and HeLa cell lysates with either antibody revealed a specific association of hEST1A and hEST1B with telomerase activity (Figures 2A and 2B, data not shown). In cell lysates and antihEST1A immunoprecipitations, a single ⵑ180 kDa species was detected that comigrated with recombinant hEST1A (Figures 2A and 2B). Using the anti-hEST1B peptide antibody, we detected three species in cell lysates and anti-hEST1B immunoprecipitations (Figures 2A and 2B). Since these three species were competed by using the anti-hEST1B-specific peptide, they must
Functional Conservation of Est1p in Humans 699
Figure 1. Schematic Alignment of Human Est1p Homologs The three human Est1p homologs, hEST1A (KIAA0732), hEST1B (KIAA1089), and hEST1C (KIAA0250), are shown with the region of highest similarity to yeast Est1p (Est One Homology ⫽ EOH) in dark gray, the putative tetratricopeptide repeats (TPR) with a black bar, and the PIN domain in white. Below: multiple sequence alignment over the EOH region; D. melanogaster (Dm), A. thaliana (At), C. elegans (Ce), K. lactis (Kl), S. cerevisiae (Sc), H. sapiens (Hs), and A. gambiae (Ag). The two putative tetratricopeptide repeats (TPRs) (aa 190–254 of Est1p) are indicated with a black bar [24]. The alignment was shaded in conservation mode by using GeneDoc (version 2.6) according to the following amino acid property similarity groups: aliphatic (AVLIM), aromatic (FWYH), acids and amides (QEDN), positive (KR), and tiny (SGA). Residues showing 100% conservation are shaded yellow, those showing 80%–90% conservation are shaded blue, and those showing 50%–80% conservation are shaded green.
represent variants of hEST1B or proteins that specifically cross-react to the peptide antiserum. We were unable to deplete telomerase activity from cell lysates by using anti-hEST1A or anti-hEST1B antibodies. It was also not possible to estimate the stoichiometry of the interaction with hTERT due to the low levels of endogenous hTERT and the inability to detect hTERT in hEST1A immunoprecipitations (data not shown). Nonetheless, an interaction between hEST1A or hEST1B and hTERT was also observed in rabbit reticulocyte lysates (RRL). This interaction was specific to hEST1A and hEST1B, since an irrelevant HA-tagged protein, RGS11 [26], did not coimmunoprecipitate FLAG-tagged hTERT (Figure 2C, lanes 3 and 10–12). In contrast to yeast Est1p, human EST1A and EST1B did not require the telomerase RNA to bind hTERT in vitro (Figure 2C, lanes 5, 8, and 11). S. cerevisiae Est1p demonstrates a weak affinity for
telomeric single-stranded DNA and a nonspecific RNA binding activity in vitro [10]. We therefore tested whether hEST1A or hEST1B could bind telomeric DNA in vitro. Using an oligonucleotide capture assay, in which a neutravidin-captured biotinylated oligonucleotide is incubated with RRL-produced protein, we found that fulllength recombinant hEST1A, but not hEST1B, bound to the neutravidin-captured DNA (Figure S2). The S. cerevisiae telomeric-like sequence (GTGTGG)4GTGT was used for the capture assay, since it exhibited the most efficient precipitation of hEST1A (see Figure S2A, and see below). A fragment of hEST1A (spanning aa 630– 1388) that contained the TPR repeats and PIN domain did not bind DNA (Figure S2A). We tested a number of hEST1A polypeptide fragments for telomere binding and mapped a minimal fragment of hEST1A, spanning aa 114– 503, that contained the DNA binding activity (Figure S2A).
Current Biology 700
Figure 2. Human EST1A and EST1B Associate with Telomerase Activity and with hTERT In Vitro (A) Characterization of anti-hEST1A and anti-hEST1B peptide antibodies. Input recombinant, HA-tagged hEST1A and hEST1B were separately synthesized in RRL and then mixed (lane 1, “L,” top and bottom panel). The mixture was incubated with anti-hEST1A (top panel, lanes 2–4) or anti-hEST1B (bottom panel, lanes 2–4) antibody in the presence of no competing peptide (“0”; lane 2), 50 g nonspecific peptide (“NS”; lane 3, the hEST1B and hEST1A specific peptides were swapped), or 50 g specific peptide (“SP”; lane 4). Membranes were probed with anti-HA antibody. Lanes 5–7: similar immunoprecipitations with HeLa cell lysate. Protein markers are shown at the left in kDa. At right, the positions of hEST1A and hEST1B cross-reacting species in HeLa cells are indicated with bars. Note that the slightly lower level of hEST1A in lane 3 is particular to this experiment and therefore does not reflect a nonspecific competition of hEST1B peptide for the association of hEST1A with its cognate peptide antibody. (B) Association of endogenous hEST1A and hEST1B with telomerase activity in human cell extracts. Lanes 1 and 6, input 293 cell lysate; lanes 2–5, immunoprecipitations with anti-hEST1A antibody; lanes 7–10, immunoprecipitations with anti-hEST1B antibody; lane 11, immunoprecipitation in the absence of antibody (⫺A); lanes 12 and 13, immunoprecipitations with anti-hTERT antibody. Immunoprecipitations were carried out with either no competing peptide (lanes 2, 7, and 12) or with 5 g hEST1A peptide (lane 3), 25 g hEST1A peptide (lanes 4 and 9), 25 g hEST1B peptide (lanes 5 and 8), or 25 g hTERT peptide (lanes 10 and 13). Telomerase assays are shown in the top panels (the arrow indicates internal PCR standard, which is faint due to the low number of PCR cycles), and the corresponding Western blots probed with anti-hEST1A antibody (lanes 1–5) or anti-hEST1B are shown in the bottom panels (lanes 6–13). (C) Human EST1A and EST1B interact with hTERT in vitro independently of the telomerase RNA. Input recombinant protein (lanes 1–3) was incubated alone (lanes 4, 7, and 10), with hTERT (lanes 5, 8, and 11), or with hTERT and 0.5 g in vitro-synthesized telomerase RNA (lanes 6, 9, and 12), and was then precipitated onto anti-FLAG resin. The exposure time for lanes containing HA-hEST1B was longer than for HAhEST1A. Top panel: the blots were probed with anti-HA antibody. Bottom panel: the blots were probed with an anti-TERT peptide antibody [37]. The immunoprecipitations were also treated with DNaseI and RNase to rule out nonspecific interactions bridged by nucleic acids (data not shown).
To confirm that the DNA binding activity was conferred by hEST1A, we partially purified a HIS-tagged fragment of hEST1A, spanning aa 114–631, from E. coli (Figure S2B). Electrophoretic mobility shift analysis revealed that partially purified hEST1A could bind a radiolabeled yeast telomere DNA sequence (GTGTGG)4GTGT and, to a lesser extent, the radiolabeled human telomeric sequence (TTAGGG)4TTAG (Figure 3, left panel). In contrast, the nematode telomeric sequence (TTAGGC)4TTAG or a random 28 nucleotide sequence was not detectably bound (Figure 3, left panel). Human telomeric singlestranded, double-stranded, and partially single-stranded DNA substrates did not effectively compete for the interaction with (GTGTGG)4GTGT (Figure 3, right panel, and data not shown). Since the telomeric probe was not quantitatively shifted at the highest protein concentrations tested (ⵑ600 pM), we were unable to determine the
observed dissociation constant of hEST1A for telomeric DNA. S. cerevisiae Est1p shows a preference for free 3⬘ overhangs and S. cerevisiae telomeric DNA sequences [10]. Although the apparent preference of hEST1A for S. cerevisiae telomere-like DNA relative to human telomeric DNA is puzzling, further experiments will be required to determine the precise relative affinity of hEST1A for different telomeric or nontelomeric substrates. We could not detect significant similarity between the hEST1A and S. cerevisiae Est1p DNA binding regions [10]. S. cerevisiae Est1p also binds RNA sequences nonspecifically, but with a higher affinity than for telomeric DNA [10]. A bulged stem within the yeast telomerase RNA, TLC1, is essential for Est1p binding in vivo; however, this stem-loop is not conserved in the human telomerase RNA [27, 28]. In addition, the interaction of Est2p with Est1p is telomerase RNA dependent [27, 28]. We
Functional Conservation of Est1p in Humans 701
Figure 3. Human hEST1A Possesses a Telomere DNA Binding Activity Electrophoretic mobility shift analysis of recombinant hEST1A. Input 5⬘ end-labeled probe (50 pM) was incubated in the absence of added extract (P; lanes 1, 6, 9, and 12) or at increasing concentrations of a partially purified ⵑ40 kDa hEST1A fragment (see the Supplemental Data) (lane 2, 60 pM; lane 3, 120 pM; lanes 4, 7, 10, and 13, 240 pM; lanes 5, 8, 11, and 14, 600 pM). The probes used were (GTGTGG)4GTGT (lanes 1–5), (TTAGGG)4TTAG (lanes 6–8), (TTAGGC)4TTAG (lanes 9–11), or TCGACAGATCGTACATGATCATCATCAC (lanes 12–14). Right panel: the 5⬘ end-labeled probe (GTGTGG)4GTGT was incubated in the absence of added protein (P; lane 1) or with approximately 300 pM partially purified protein (lanes 2–11) in the presence of no added competitor, 5-fold excess of unlabeled (GTG TGG)4GTGT competitor, 50-fold excess of unlabeled (GTGTGG)4GTGT competitor (lanes 2, 3, and 4, respectively), or 500-fold excess of unlabeled competitor, as indicated above each lane (lanes 5–11). Oligonucleotide sequences were as indicated in the left panel, except for dsTTAGGG, which was a purified 28-base duplex of TTAGGG and its complement, and 5⬘-TTAGGG and 3⬘-TTAGGG, which were oligonucleotides containing 13 bases of random duplex DNA with a 28-base 5⬘ or 3⬘ TTAGGG extension, respectively.
found that hEST1A did not significantly retard the mobility of radiolabeled telomerase RNA (data not shown). Although further experiments are necessary to determine if purified hEST1A shows an RNA binding activity, the RNA-independent interaction of hTERT and hEST1A in RRL (Figure 2C) suggests that the association with human telomerase is bridged by protein-protein interactions. In S. cerevisiae, Est1p appears to play a key role in telomere lengthening in vivo by tethering the telomerase complex to Cdc13p and the telomeric 3⬘ end in vivo [9, 14, 22], or in activating the telomerase complex once it has been recruited to the telomere [21]. Although hEST1A and hEST1B interacted with human telomerase, it was important to ascertain whether either protein could promote the ability of telomerase to lengthen telomeres in vivo. We generated two independent series of stable 293 lines that expressed either HA-tagged hEST1A or HA-hEST1B, alone or in combination with untagged hTERT (Figures 4A and 4C, and data not shown). Since telomere lengths can vary between subclones within a cell population, we analyzed the bulk transformants for alterations in telomere length. After establishment of the lines, terminal restriction fragment (TRF) analysis showed that cells containing empty control vectors or hTERT alone had little effect on average telomere lengths (Figure 4A, lanes 1 and 2). However, expression of even low levels of hEST1A, or a C-terminal fragment of hEST1A (spanning aa 630–1388), led to a marked decrease in average telomere length (Figure 4A, lanes 5 and 8). When hEST1A or the hEST1A C-terminal fragment was coexpressed with hTERT, telomere lengths actually exceeded those of cells expressing only hTERT (Figure 4A, lanes 6 and 7). In each cell line, the telomere length alterations appeared stable and remained relatively constant over the course of 3–4 cell passages (data not shown). We did not detect a significant difference in telomere length in cells expressing hEST1B with or without hTERT, compared with cells expressing hTERT
or vectors alone (Figure 4A, compare lanes 3 and 4 with lanes 1 and 2). These results were corroborated in an independently derived series of 293 stable cell lines by using TRF analysis and a more quantitative measure of total telomere DNA content in a cell population, FlowFISH (Figure 4C and data not shown). The marked telomere length alterations elicited by hEST1A were specific to telomerase-positive cells, since analogous cell lines generated in telomerase-negative ALT (for alternative telomere maintenance) cells showed no effect on telomere lengths (Figure 4B). In the WI38 VA13 ALT cell line, the telomerase RNA is not expressed, and, similar to previous studies, we showed that the cells remained telomerase negative even when expressing exogenous hTERT [29, 30] (data not shown). Since telomeres in ALT cells are already long and heterogeneous, we also assessed total telomere DNA content in each cell line by using an independent quantitative measurement, Flow-FISH [31, 32], and demonstrated that the total telomere DNA content in VA13 stable cell lines paralleled the TRF analysis (Figure 4B). At present, the mechanism of telomere shortening in cells overexpressing hEST1A is unknown. It is possible that excess hEST1A may compete for the limiting levels of hTERT available to form hTERT/hEST1A complexes at the telomere. The lack of telomere shortening upon hEST1A overexpression in ALT cells indirectly supports this hypothesis. We argue that the telomere shortening is not due to nonspecific competition of factors essential for telomere length maintenance, since telomeres became longer when hTERT was coexpressed with hEST1A. It is unclear yet whether the PIN domain contributes to hEST1A or hEST1B function in vivo. The PIN domain within human FEN1 confers an endonuclease activity against “flapped” DNA substrates [24]. A flaplike structure is also postulated to occur at the telomere t loop, which is formed when the 3⬘ single-stranded, G-rich overhang at the telomere terminus loops back and invades the double-stranded telomere DNA [33].
Current Biology 702
Figure 4. Human EST1A Cooperates with hTERT to Lengthen Telomeres In Vivo (A) Terminal restriction fragment analysis of 293 stable cell lines expressing empty vectors alone (lane 1), untagged hTERT alone (lane 2), and HA-tagged hEST1B, hEST1A, or a C-terminal hEST1A fragment (aa 630–1388) (“CT”) alone or together with untagged hTERT (lanes 3–8). Following digestion of 10 g genomic DNA with HinfI and RsaI, the DNA was resolved on a 0.5% w/v agarose gel, transferred to Hybond N⫹, and probed with a 5⬘ end-labeled (CCCTAA)3 probe. DNA markers, in kbp, are shown at the left. Below: Western blotting of 293 cell lysates with anti-hTERT and anti-HA antibodies to confirm protein expression. Protein markers are shown at the left in kDa. Note that lanes 5 and 6 contained 2.5-fold more protein extract, in order to visualize the low levels of hEST1A expression in lane 5. For an as yet unknown reason, we noted a slight decrease in recombinant hTERT expression in cell lines that also expressed recombinant hEST1B (Figure 4A, lane 4; Figure 4B, lane 5). (B) Terminal restriction fragment analysis in VA13 stable cell lines, using the same probe as in (A), and protein expression shown below, as in (A). Below: quantitative Flow-FISH analysis (in arbitrary telomere fluorescence units) of the corresponding VA13 cell lines (n ⫽ 3, p ⬎ 0.05 for all pair-wise comparisons between samples). (C) Flow-FISH analysis of another series of independently derived 293 cell lines (different from those shown in [A]) expressing empty vectors (vector), hTERT alone (T), hEST1A alone (1A), hTERT ⫹ hEST1A (1A⫹T), the hEST1A C terminus, spanning aa 630–1388 (CT), and the hTERT ⫹ hEST1A C terminus (T ⫹ CT). P values for the average telomere fluorescence measurements (n ⫽ 6) are shown above each bar, in comparison to the hTERT cell line.
Regardless of the precise mechanism, the telomere length alterations elicited by hEST1A appeared specific to telomerase-positive cells. Although we cannot completely exclude the possibility that subtle alterations in telomere lengths might be difficult to resolve in ALT cells, an independent quantitative measure of total telomere DNA content (Flow-FISH) corroborated the TRF analysis (Figure 4B). Our data are consistent with the hypothesis that
hEST1A and hEST1B are functional homologs of S. cerevisiae Est1p. Independently, Reichenbach et al. also identified hEST1A, hEST1B, and hEST1C as homologs of Est1p and present compelling functional evidence to support the association of hEST1A with the telomerase complex and its role in telomere integrity in vivo [34]. We found that hEST1A and hEST1B interacted with human telomerase in vitro and in cell extracts. However, hEST1A possessed two properties not shared by
Functional Conservation of Est1p in Humans 703
hEST1B: a DNA binding activity and a marked alteration in average telomere length when overexpressed in cells. Evans and Lundblad previously found that overexpression of Est1p, Est2p, and TLC1 leads to substantial telomere lengthening in S. cerevisiae, whereas overexpression of just Est2p and TLC1 has no effect (V. Lundblad, personal communciation). This observation prompted us to carry out our experiments in human cell lines. It will be important to determine whether hEST1A and hEST1B play cooperative, redundant, or competing roles in the telomerase complex. The similarity of Est1plike factors between yeast and man mirrors that of the telomerase catalytic components, telomere binding proteins such as Rap1p/hRAP1, Ku70/80, Taz1⫹/TRF1/ TRF2 [33], and the telomere end binding proteins (Pot1⫹, hPot1) [35, 36]. Thus, many aspects of telomere length regulation appear to be conserved throughout evolution. Supplemental Data Supplemental Data including additional Experimental Procedures and two figures, one describing the mRNA expression pattern of hEST1A and hEST1B, and the other demonstrating that a region spanning aa 114–503 within hEST1A, but not full-length hEST1B, possesses a DNA binding activity in vitro, are available at http:// images.cellpress.com/supmat/supmatin.htm. Acknowledgments We thank Diane De Abreu, Jennifer Bonderoff, Katie Chiu, and Denis Bouchard for experimental assistance and Brenda Andrews, Ben Blencowe, Dan Durocher, Sara Oster, Frank Sicheri, Mike Tyers, and members of the Harrington lab for helpful discussion. We also thank Joachim Lingner, Sara Evans, and Vicki Lundblad for sharing results prior to publication. L.H. acknowledges funding from the Canadian Institutes of Health Research (MT14340) and the National Institutes of Health (AG16629-04). Received: December 17, 2002 Revised: February 24, 2003 Accepted: February 25, 2003 Published: April 15, 2003 References 1. Nugent, C.I., and Lundblad, V. (1998). The telomerase reverse transcriptase: components and regulation. Genes Dev. 12, 1073–1085. 2. Collins, K., and Mitchell, J.R. (2002). Telomerase in the human organism. Oncogene 21, 564–579. 3. Cong, Y.S., Wright, W.E., and Shay, J.W. (2002). Human telomerase and its regulation. Microbiol. Mol. Biol. Rev. 66, 407–425. 4. Cohn, M., and Blackburn, E.H. (1995). Telomerase in yeast. Science 269, 396–400. 5. Lundblad, V., and Szostak, J.W. (1989). A mutant with a defect in telomere elongation leads to senescence in yeast. Cell 57, 633–643. 6. Lingner, J., Cech, T.R., Hughes, T.R., and Lundblad, V. (1997). Three Ever Shorter Telomere (EST) genes are dispensable for in vitro yeast telomerase activity. Proc. Natl. Acad. Sci. USA 94, 11190–11195. 7. Lin, J.J., and Zakian, V.A. (1996). The Saccharomyces CDC13 protein is a single-strand TG1-3 telomeric DNA-binding protein in vitro that affects telomere behavior in vivo. Proc. Natl. Acad. Sci. USA 93, 13760–13765. 8. Evans, S.K., and Lundblad, V. (1999). Est1 and Cdc13 as comediators of telomerase access. Science 286, 117–120. 9. Pennock, E., Buckley, K., and Lundblad, V. (2001). Cdc13 delivers separate complexes to the telomere for end protection and replication. Cell 104, 387–396.
10. Virta-Pearlman, V., Morris, D.K., and Lundblad, V. (1996). Est1 has the properties of a single-stranded telomere end-binding protein. Genes Dev. 10, 3094–3104. 11. Chandra, A., Hughes, T.R., Nugent, C.I., and Lundblad, V. (2001). Cdc13 both positively and negatively regulates telomere replication. Genes Dev. 15, 404–414. 12. Zhou, J., Hidaka, K., and Futcher, B. (2000). The Est1 subunit of yeast telomerase binds the Tlc1 telomerase RNA. Mol. Cell. Biol. 20, 1947–1955. 13. Steiner, B.R., Hidaka, K., and Futcher, B. (1996). Association of the Est1 protein with telomerase activity in yeast. Proc. Natl. Acad. Sci. USA 93, 2817–2821. 14. Nugent, C.I., Hughes, T.R., Lue, N.F., and Lundblad, V. (1996). Cdc13p: a single-strand telomeric DNA-binding protein with a dual role in yeast telomere maintenance. Science 274, 249–252. 15. Evans, S.K., and Lundblad, V. (2002). The Est1 subunit of Saccharomyces cerevisiae telomerase makes multiple contributions to telomere length maintenance. Genetics 162, 1101–1115. 16. Evans, S.K., and Lundblad, V. (2000). Positive and negative regulation of telomerase access to the telomere. J. Cell Sci. 113, 3357–3364. 17. Grandin, N., Reed, S.I., and Charbonneau, M. (1997). Stn1, a new Saccharomyces cerevisiae protein, is implicated in telomere size regulation in association with Cdc13. Genes Dev. 11, 512–527. 18. Grandin, N., Damon, C., and Charbonneau, M. (2000). Cdc13 cooperates with the yeast Ku proteins and Stn1 to regulate telomerase recruitment. Mol. Cell. Biol. 20, 8397–8408. 19. Grandin, N., Damon, C., and Charbonneau, M. (2001). Ten1 functions in telomere end protection and length regulation in association with Stn1 and Cdc13. EMBO J. 20, 1173–1183. 20. Qi, H., and Zakian, V.A. (2000). The Saccharomyces telomerebinding protein Cdc13p interacts with both the catalytic subunit of DNA polymerase alpha and the telomerase-associated est1 protein. Genes Dev. 14, 1777–1788. 21. Taggart, A.K., Teng, S.C., and Zakian, V.A. (2002). Est1p as a cell cycle-regulated activator of telomere-bound telomerase. Science 297, 1023–1026. 22. Hughes, T.R., Evans, S.K., Weilbaecher, R.G., and Lundblad, V. (2000). The Est3 protein is a subunit of yeast telomerase. Curr. Biol. 10, 809–812. 23. Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z., Miller, W., and Lipman, D.J. (1997). Gapped BLAST and PSIBLAST: a new generation of protein database search programs. Nucleic Acids Res. 25, 3389–3402. 24. Clissold, P.M., and Ponting, C.P. (2000). PIN domains in nonsense-mediated mRNA decay and RNAi. Curr. Biol. 10, R888–890. 25. Singh, S.M., Steinberg-Neifach, O., Mian, I.S., and Lue, N.F. (2002). Analysis of telomerase in Candida albicans: potential role in telomere end protection. Eukaryotic Cell 1, 967–977. 26. Snow, B.E., Krumins, A.M., Brothers, G.M., Lee, S.F., Wall, M.A., Chung, S., Mangion, J., Arya, S., Gilman, A.G., and Siderovski, D.P. (1998). A G protein gamma subunit-like domain shared between RGS11 and other RGS proteins specifies binding to Gbeta5 subunits. Proc. Natl. Acad. Sci. USA 95, 13307–13312. 27. Seto, A.G., Livengood, A.J., Tzfati, Y., Blackburn, E.H., and Cech, T.R. (2002). A bulged stem tethers Est1p to telomerase RNA in budding yeast. Genes Dev. 16, 2800–2812. 28. Livengood, A.J., Zaug, A.J., and Cech, T.R. (2002). Essential regions of Saccharomyces cerevisiae telomerase RNA: separate elements for Est1p and Est2p interaction. Mol. Cell. Biol. 22, 2366–2374. 29. Wen, J., Cong, Y.S., and Bacchetti, S. (1998). Reconstitution of wild-type or mutant telomerase activity in telomerase-negative immortal human cells. Hum. Mol. Genet. 7, 1137–1141. 30. 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). Two inactive fragments of the integral RNA cooperate to assemble active telomerase with the human protein catalytic subunit (hTERT) in vitro. Mol. Cell. Biol. 19, 6207–6216. 31. Baerlocher, G.M., Mak, J., Tien, T., and Lansdorp, P.M. (2002). Telomere length measurement by fluorescence in situ hybridization and flow cytometry: tips and pitfalls. Cytometry 47, 89–99.
Current Biology 704
32. Rufer, N., Dragowska, W., Thornbury, G., Roosnek, E., and Lansdorp, P.M. (1998). Telomere length dynamics in human lymphocyte subpopulations measured by flow cytometry. Nat. Biotechnol. 16, 743–747. 33. de Lange, T. (2002). Protection of mammalian telomeres. Oncogene 21, 532–540. 34. Reichenbach, P., Ho¨ss, M., Azzalin, C.M., Nabholz, M., Bucher, P., and Lingner, J. (2003). A human homolog of yeast Est1 associates with telomerase and uncaps chromosome ends when overexpressed. Curr. Biol. 13, 568–574. 35. Baumann, P., Podell, E., and Cech, T.R. (2002). Human pot1 (protection of telomeres) protein: cytolocalization, gene structure, and alternative splicing. Mol. Cell. Biol. 22, 8079–8087. 36. Baumann, P., and Cech, T.R. (2001). Pot1, the putative telomere end-binding protein in fission yeast and humans. Science 292, 1171–1175. 37. Harrington, L., Zhou, W., McPhail, T., Oulton, R., Yeung, D.S., Mar, V., Bass, M.B., and Robinson, M.O. (1997). Human telomerase contains evolutionarily conserved catalytic and structural subunits. Genes Dev. 11, 3109–3115. Accession Numbers Sequence data reported in this paper have been deposited in the NCBI GenBank database under Accession Numbers AY168921 and AY168922.
Functional Conservation of the Telomerase Protein Est1p in Humans Bryan E. Snow,1 Natalie Erdmann,1 Jennifer Cruickshank,1 Hartt Goldman,1 R. Montgomery Gill,2 Murray O. Robinson,4,5 and Lea Harrington1,3,* 1 Ontario Cancer Institute/Advanced Medical Discovery Institute University of Toronto 620 University Avenue Toronto, Ontario M5G 2C1 2 Ontario Cancer Institute 3 Department of Medical Biophysics University of Toronto 610 University Avenue Toronto, Ontario M5G 2M9 Canada 4 Amgen 1 Amgen Drive Thousand Oaks, California 91320
Summary Eukaryotic telomerase contains a telomerase reverse transcriptase (TERT) and an RNA template component that are essential for telomerase catalytic activity and several other telomerase-associated factors of which only a few appear to be integral enzyme components [1–3]. The first essential telomerase protein identified was S. cerevisiae Est1p, whose deletion leads to evershorter telomeres despite the persistence of telomerase activity [4–6]. Extensive genetic and biochemical data show that Est1p, via its interaction with the telomerase RNA and telomere end DNA binding complex Cdc13p/ Stn1p/Ten1p, promotes the ability of telomerase to elongate telomeres in vivo [7–22]. The characterization of Est1p homologs outside of yeast has not been documented. We report the characterization of two putative human homologs of Est1p, hEST1A and hEST1B. Both proteins specifically associated with telomerase activity in human cell extracts and bound hTERT in rabbit reticulocyte lysates independently of the telomerase RNA. Overproduction of hEST1A cooperated with hTERT to lengthen telomeres, an effect that was specific to cells containing telomerase activity. Like Est1p, hEST1A (but not hEST1B) exhibited a single-stranded telomere DNA binding activity. These results suggest that the telomerase-associated factor Est1p is evolutionarily conserved in humans. Results and Discussion Using a reiterative PSI-Blast search of the NCBI nonredundant database with the S. cerevisiae Est1p polypeptide, we identified a region of homology between Est1p (aa 82–341) and an Ebs1-like protein from K. lactis (Gen*Correspondence: [email protected] 5 Present address: GenPath Pharmaceuticals, 300 Technology Square, 7th Floor, Cambridge, Massachusetts 02139.
Bank accession number AAG49579; Expect ⫽ 9 ⫻ 10⫺15) [23]. The second iterative search identified proteins with similarity from A. thaliana (NP_197441; Expect ⫽ 0.006) and N. tabacum (BAB82502; Expect ⫽ 0.072). The final iterative search revealed human KIAA0732 (AA34452; Expect ⫽ 8 ⫻ 10⫺14), KIAA1089 (BAA83041; Expect ⫽ 6 ⫻ 10⫺10), and KIAA0250 (BAA13381; Expect ⫽ 6 ⫻ 10⫺15), as well as uncharacterized proteins in D. melanogaster (AAF56380; Expect ⫽ 9 ⫻ 10⫺9), C. elegans (NP_497566; Expect ⫽ 2 ⫻ 10⫺9), and A. gambiae (EAA12461; Expect ⫽ 2 ⫻ 10⫺5) (Figure 1). A smaller conserved region between these proteins (aa 190–254 of Est1p) was previously noted to contain two putative tetratricopeptide repeat (TPR) sequences, which play a role in protein-protein interactions [24, 25] (Figure 1). Although no specific function has been ascribed to the TPR repeats in Est1p, deletion of a portion of the TPR repeats (aa 159–323) abolishes telomere length maintenance in vivo [10]. KIAA0732 and KIAA1089 contain a C-terminal PIN domain (PilT amino-terminal), which is proposed to confer nucleic acid binding and, in some cases, nuclease activity [24]. The PIN domain is not conserved in Est1p or KIAA0250 [24] (Figure 1 and data not shown). The human KIAA0732 protein (hereafter called hEST1A) showed the highest similarity to Est1p (aa 57– 284) with 17.5% identity and 28.9% similarity, and KIAA1089 (hEST1B) exhibited 13.5% identity and 27% similarity (Figure 1). KIAA0250 (hEST1C) exhibited 15.4% identity and 25.4% similarity to Est1p aa 57–284 and has not yet been characterized. Full-length cDNAs for both hEST1A and hEST1B were used to probe mRNA expression in various human tissues and cell lines. hEST1A and hEST1B mRNAs existed as predominantly single species at 7.5 kb and 4.6 kb, respectively, and were expressed in all human tissues and cell lines examined (see Figure S1 in the Supplemental Data available with this article online). In S. cerevisiae cell extracts, Est1p binds the telomerase RNA independently of the other telomerase components; however, its association with Est2p (TERT) depends on the presence of the telomerase RNA [12, 13, 15, 22]. We tested whether endogenous hEST1A or hEST1B associated with telomerase activity in human cell extracts (Figure 2). Separate antibodies raised against hEST1A or hEST1B peptides specifically immunoprecipitated only their cognate recombinant protein (Figure 2A) and did not exhibit cross-reactivity. Immunoprecipitations from Raji, 293, and HeLa cell lysates with either antibody revealed a specific association of hEST1A and hEST1B with telomerase activity (Figures 2A and 2B, data not shown). In cell lysates and antihEST1A immunoprecipitations, a single ⵑ180 kDa species was detected that comigrated with recombinant hEST1A (Figures 2A and 2B). Using the anti-hEST1B peptide antibody, we detected three species in cell lysates and anti-hEST1B immunoprecipitations (Figures 2A and 2B). Since these three species were competed by using the anti-hEST1B-specific peptide, they must
Functional Conservation of Est1p in Humans 699
Figure 1. Schematic Alignment of Human Est1p Homologs The three human Est1p homologs, hEST1A (KIAA0732), hEST1B (KIAA1089), and hEST1C (KIAA0250), are shown with the region of highest similarity to yeast Est1p (Est One Homology ⫽ EOH) in dark gray, the putative tetratricopeptide repeats (TPR) with a black bar, and the PIN domain in white. Below: multiple sequence alignment over the EOH region; D. melanogaster (Dm), A. thaliana (At), C. elegans (Ce), K. lactis (Kl), S. cerevisiae (Sc), H. sapiens (Hs), and A. gambiae (Ag). The two putative tetratricopeptide repeats (TPRs) (aa 190–254 of Est1p) are indicated with a black bar [24]. The alignment was shaded in conservation mode by using GeneDoc (version 2.6) according to the following amino acid property similarity groups: aliphatic (AVLIM), aromatic (FWYH), acids and amides (QEDN), positive (KR), and tiny (SGA). Residues showing 100% conservation are shaded yellow, those showing 80%–90% conservation are shaded blue, and those showing 50%–80% conservation are shaded green.
represent variants of hEST1B or proteins that specifically cross-react to the peptide antiserum. We were unable to deplete telomerase activity from cell lysates by using anti-hEST1A or anti-hEST1B antibodies. It was also not possible to estimate the stoichiometry of the interaction with hTERT due to the low levels of endogenous hTERT and the inability to detect hTERT in hEST1A immunoprecipitations (data not shown). Nonetheless, an interaction between hEST1A or hEST1B and hTERT was also observed in rabbit reticulocyte lysates (RRL). This interaction was specific to hEST1A and hEST1B, since an irrelevant HA-tagged protein, RGS11 [26], did not coimmunoprecipitate FLAG-tagged hTERT (Figure 2C, lanes 3 and 10–12). In contrast to yeast Est1p, human EST1A and EST1B did not require the telomerase RNA to bind hTERT in vitro (Figure 2C, lanes 5, 8, and 11). S. cerevisiae Est1p demonstrates a weak affinity for
telomeric single-stranded DNA and a nonspecific RNA binding activity in vitro [10]. We therefore tested whether hEST1A or hEST1B could bind telomeric DNA in vitro. Using an oligonucleotide capture assay, in which a neutravidin-captured biotinylated oligonucleotide is incubated with RRL-produced protein, we found that fulllength recombinant hEST1A, but not hEST1B, bound to the neutravidin-captured DNA (Figure S2). The S. cerevisiae telomeric-like sequence (GTGTGG)4GTGT was used for the capture assay, since it exhibited the most efficient precipitation of hEST1A (see Figure S2A, and see below). A fragment of hEST1A (spanning aa 630– 1388) that contained the TPR repeats and PIN domain did not bind DNA (Figure S2A). We tested a number of hEST1A polypeptide fragments for telomere binding and mapped a minimal fragment of hEST1A, spanning aa 114– 503, that contained the DNA binding activity (Figure S2A).
Current Biology 700
Figure 2. Human EST1A and EST1B Associate with Telomerase Activity and with hTERT In Vitro (A) Characterization of anti-hEST1A and anti-hEST1B peptide antibodies. Input recombinant, HA-tagged hEST1A and hEST1B were separately synthesized in RRL and then mixed (lane 1, “L,” top and bottom panel). The mixture was incubated with anti-hEST1A (top panel, lanes 2–4) or anti-hEST1B (bottom panel, lanes 2–4) antibody in the presence of no competing peptide (“0”; lane 2), 50 g nonspecific peptide (“NS”; lane 3, the hEST1B and hEST1A specific peptides were swapped), or 50 g specific peptide (“SP”; lane 4). Membranes were probed with anti-HA antibody. Lanes 5–7: similar immunoprecipitations with HeLa cell lysate. Protein markers are shown at the left in kDa. At right, the positions of hEST1A and hEST1B cross-reacting species in HeLa cells are indicated with bars. Note that the slightly lower level of hEST1A in lane 3 is particular to this experiment and therefore does not reflect a nonspecific competition of hEST1B peptide for the association of hEST1A with its cognate peptide antibody. (B) Association of endogenous hEST1A and hEST1B with telomerase activity in human cell extracts. Lanes 1 and 6, input 293 cell lysate; lanes 2–5, immunoprecipitations with anti-hEST1A antibody; lanes 7–10, immunoprecipitations with anti-hEST1B antibody; lane 11, immunoprecipitation in the absence of antibody (⫺A); lanes 12 and 13, immunoprecipitations with anti-hTERT antibody. Immunoprecipitations were carried out with either no competing peptide (lanes 2, 7, and 12) or with 5 g hEST1A peptide (lane 3), 25 g hEST1A peptide (lanes 4 and 9), 25 g hEST1B peptide (lanes 5 and 8), or 25 g hTERT peptide (lanes 10 and 13). Telomerase assays are shown in the top panels (the arrow indicates internal PCR standard, which is faint due to the low number of PCR cycles), and the corresponding Western blots probed with anti-hEST1A antibody (lanes 1–5) or anti-hEST1B are shown in the bottom panels (lanes 6–13). (C) Human EST1A and EST1B interact with hTERT in vitro independently of the telomerase RNA. Input recombinant protein (lanes 1–3) was incubated alone (lanes 4, 7, and 10), with hTERT (lanes 5, 8, and 11), or with hTERT and 0.5 g in vitro-synthesized telomerase RNA (lanes 6, 9, and 12), and was then precipitated onto anti-FLAG resin. The exposure time for lanes containing HA-hEST1B was longer than for HAhEST1A. Top panel: the blots were probed with anti-HA antibody. Bottom panel: the blots were probed with an anti-TERT peptide antibody [37]. The immunoprecipitations were also treated with DNaseI and RNase to rule out nonspecific interactions bridged by nucleic acids (data not shown).
To confirm that the DNA binding activity was conferred by hEST1A, we partially purified a HIS-tagged fragment of hEST1A, spanning aa 114–631, from E. coli (Figure S2B). Electrophoretic mobility shift analysis revealed that partially purified hEST1A could bind a radiolabeled yeast telomere DNA sequence (GTGTGG)4GTGT and, to a lesser extent, the radiolabeled human telomeric sequence (TTAGGG)4TTAG (Figure 3, left panel). In contrast, the nematode telomeric sequence (TTAGGC)4TTAG or a random 28 nucleotide sequence was not detectably bound (Figure 3, left panel). Human telomeric singlestranded, double-stranded, and partially single-stranded DNA substrates did not effectively compete for the interaction with (GTGTGG)4GTGT (Figure 3, right panel, and data not shown). Since the telomeric probe was not quantitatively shifted at the highest protein concentrations tested (ⵑ600 pM), we were unable to determine the
observed dissociation constant of hEST1A for telomeric DNA. S. cerevisiae Est1p shows a preference for free 3⬘ overhangs and S. cerevisiae telomeric DNA sequences [10]. Although the apparent preference of hEST1A for S. cerevisiae telomere-like DNA relative to human telomeric DNA is puzzling, further experiments will be required to determine the precise relative affinity of hEST1A for different telomeric or nontelomeric substrates. We could not detect significant similarity between the hEST1A and S. cerevisiae Est1p DNA binding regions [10]. S. cerevisiae Est1p also binds RNA sequences nonspecifically, but with a higher affinity than for telomeric DNA [10]. A bulged stem within the yeast telomerase RNA, TLC1, is essential for Est1p binding in vivo; however, this stem-loop is not conserved in the human telomerase RNA [27, 28]. In addition, the interaction of Est2p with Est1p is telomerase RNA dependent [27, 28]. We
Functional Conservation of Est1p in Humans 701
Figure 3. Human hEST1A Possesses a Telomere DNA Binding Activity Electrophoretic mobility shift analysis of recombinant hEST1A. Input 5⬘ end-labeled probe (50 pM) was incubated in the absence of added extract (P; lanes 1, 6, 9, and 12) or at increasing concentrations of a partially purified ⵑ40 kDa hEST1A fragment (see the Supplemental Data) (lane 2, 60 pM; lane 3, 120 pM; lanes 4, 7, 10, and 13, 240 pM; lanes 5, 8, 11, and 14, 600 pM). The probes used were (GTGTGG)4GTGT (lanes 1–5), (TTAGGG)4TTAG (lanes 6–8), (TTAGGC)4TTAG (lanes 9–11), or TCGACAGATCGTACATGATCATCATCAC (lanes 12–14). Right panel: the 5⬘ end-labeled probe (GTGTGG)4GTGT was incubated in the absence of added protein (P; lane 1) or with approximately 300 pM partially purified protein (lanes 2–11) in the presence of no added competitor, 5-fold excess of unlabeled (GTG TGG)4GTGT competitor, 50-fold excess of unlabeled (GTGTGG)4GTGT competitor (lanes 2, 3, and 4, respectively), or 500-fold excess of unlabeled competitor, as indicated above each lane (lanes 5–11). Oligonucleotide sequences were as indicated in the left panel, except for dsTTAGGG, which was a purified 28-base duplex of TTAGGG and its complement, and 5⬘-TTAGGG and 3⬘-TTAGGG, which were oligonucleotides containing 13 bases of random duplex DNA with a 28-base 5⬘ or 3⬘ TTAGGG extension, respectively.
found that hEST1A did not significantly retard the mobility of radiolabeled telomerase RNA (data not shown). Although further experiments are necessary to determine if purified hEST1A shows an RNA binding activity, the RNA-independent interaction of hTERT and hEST1A in RRL (Figure 2C) suggests that the association with human telomerase is bridged by protein-protein interactions. In S. cerevisiae, Est1p appears to play a key role in telomere lengthening in vivo by tethering the telomerase complex to Cdc13p and the telomeric 3⬘ end in vivo [9, 14, 22], or in activating the telomerase complex once it has been recruited to the telomere [21]. Although hEST1A and hEST1B interacted with human telomerase, it was important to ascertain whether either protein could promote the ability of telomerase to lengthen telomeres in vivo. We generated two independent series of stable 293 lines that expressed either HA-tagged hEST1A or HA-hEST1B, alone or in combination with untagged hTERT (Figures 4A and 4C, and data not shown). Since telomere lengths can vary between subclones within a cell population, we analyzed the bulk transformants for alterations in telomere length. After establishment of the lines, terminal restriction fragment (TRF) analysis showed that cells containing empty control vectors or hTERT alone had little effect on average telomere lengths (Figure 4A, lanes 1 and 2). However, expression of even low levels of hEST1A, or a C-terminal fragment of hEST1A (spanning aa 630–1388), led to a marked decrease in average telomere length (Figure 4A, lanes 5 and 8). When hEST1A or the hEST1A C-terminal fragment was coexpressed with hTERT, telomere lengths actually exceeded those of cells expressing only hTERT (Figure 4A, lanes 6 and 7). In each cell line, the telomere length alterations appeared stable and remained relatively constant over the course of 3–4 cell passages (data not shown). We did not detect a significant difference in telomere length in cells expressing hEST1B with or without hTERT, compared with cells expressing hTERT
or vectors alone (Figure 4A, compare lanes 3 and 4 with lanes 1 and 2). These results were corroborated in an independently derived series of 293 stable cell lines by using TRF analysis and a more quantitative measure of total telomere DNA content in a cell population, FlowFISH (Figure 4C and data not shown). The marked telomere length alterations elicited by hEST1A were specific to telomerase-positive cells, since analogous cell lines generated in telomerase-negative ALT (for alternative telomere maintenance) cells showed no effect on telomere lengths (Figure 4B). In the WI38 VA13 ALT cell line, the telomerase RNA is not expressed, and, similar to previous studies, we showed that the cells remained telomerase negative even when expressing exogenous hTERT [29, 30] (data not shown). Since telomeres in ALT cells are already long and heterogeneous, we also assessed total telomere DNA content in each cell line by using an independent quantitative measurement, Flow-FISH [31, 32], and demonstrated that the total telomere DNA content in VA13 stable cell lines paralleled the TRF analysis (Figure 4B). At present, the mechanism of telomere shortening in cells overexpressing hEST1A is unknown. It is possible that excess hEST1A may compete for the limiting levels of hTERT available to form hTERT/hEST1A complexes at the telomere. The lack of telomere shortening upon hEST1A overexpression in ALT cells indirectly supports this hypothesis. We argue that the telomere shortening is not due to nonspecific competition of factors essential for telomere length maintenance, since telomeres became longer when hTERT was coexpressed with hEST1A. It is unclear yet whether the PIN domain contributes to hEST1A or hEST1B function in vivo. The PIN domain within human FEN1 confers an endonuclease activity against “flapped” DNA substrates [24]. A flaplike structure is also postulated to occur at the telomere t loop, which is formed when the 3⬘ single-stranded, G-rich overhang at the telomere terminus loops back and invades the double-stranded telomere DNA [33].
Current Biology 702
Figure 4. Human EST1A Cooperates with hTERT to Lengthen Telomeres In Vivo (A) Terminal restriction fragment analysis of 293 stable cell lines expressing empty vectors alone (lane 1), untagged hTERT alone (lane 2), and HA-tagged hEST1B, hEST1A, or a C-terminal hEST1A fragment (aa 630–1388) (“CT”) alone or together with untagged hTERT (lanes 3–8). Following digestion of 10 g genomic DNA with HinfI and RsaI, the DNA was resolved on a 0.5% w/v agarose gel, transferred to Hybond N⫹, and probed with a 5⬘ end-labeled (CCCTAA)3 probe. DNA markers, in kbp, are shown at the left. Below: Western blotting of 293 cell lysates with anti-hTERT and anti-HA antibodies to confirm protein expression. Protein markers are shown at the left in kDa. Note that lanes 5 and 6 contained 2.5-fold more protein extract, in order to visualize the low levels of hEST1A expression in lane 5. For an as yet unknown reason, we noted a slight decrease in recombinant hTERT expression in cell lines that also expressed recombinant hEST1B (Figure 4A, lane 4; Figure 4B, lane 5). (B) Terminal restriction fragment analysis in VA13 stable cell lines, using the same probe as in (A), and protein expression shown below, as in (A). Below: quantitative Flow-FISH analysis (in arbitrary telomere fluorescence units) of the corresponding VA13 cell lines (n ⫽ 3, p ⬎ 0.05 for all pair-wise comparisons between samples). (C) Flow-FISH analysis of another series of independently derived 293 cell lines (different from those shown in [A]) expressing empty vectors (vector), hTERT alone (T), hEST1A alone (1A), hTERT ⫹ hEST1A (1A⫹T), the hEST1A C terminus, spanning aa 630–1388 (CT), and the hTERT ⫹ hEST1A C terminus (T ⫹ CT). P values for the average telomere fluorescence measurements (n ⫽ 6) are shown above each bar, in comparison to the hTERT cell line.
Regardless of the precise mechanism, the telomere length alterations elicited by hEST1A appeared specific to telomerase-positive cells. Although we cannot completely exclude the possibility that subtle alterations in telomere lengths might be difficult to resolve in ALT cells, an independent quantitative measure of total telomere DNA content (Flow-FISH) corroborated the TRF analysis (Figure 4B). Our data are consistent with the hypothesis that
hEST1A and hEST1B are functional homologs of S. cerevisiae Est1p. Independently, Reichenbach et al. also identified hEST1A, hEST1B, and hEST1C as homologs of Est1p and present compelling functional evidence to support the association of hEST1A with the telomerase complex and its role in telomere integrity in vivo [34]. We found that hEST1A and hEST1B interacted with human telomerase in vitro and in cell extracts. However, hEST1A possessed two properties not shared by
Functional Conservation of Est1p in Humans 703
hEST1B: a DNA binding activity and a marked alteration in average telomere length when overexpressed in cells. Evans and Lundblad previously found that overexpression of Est1p, Est2p, and TLC1 leads to substantial telomere lengthening in S. cerevisiae, whereas overexpression of just Est2p and TLC1 has no effect (V. Lundblad, personal communciation). This observation prompted us to carry out our experiments in human cell lines. It will be important to determine whether hEST1A and hEST1B play cooperative, redundant, or competing roles in the telomerase complex. The similarity of Est1plike factors between yeast and man mirrors that of the telomerase catalytic components, telomere binding proteins such as Rap1p/hRAP1, Ku70/80, Taz1⫹/TRF1/ TRF2 [33], and the telomere end binding proteins (Pot1⫹, hPot1) [35, 36]. Thus, many aspects of telomere length regulation appear to be conserved throughout evolution. Supplemental Data Supplemental Data including additional Experimental Procedures and two figures, one describing the mRNA expression pattern of hEST1A and hEST1B, and the other demonstrating that a region spanning aa 114–503 within hEST1A, but not full-length hEST1B, possesses a DNA binding activity in vitro, are available at http:// images.cellpress.com/supmat/supmatin.htm. Acknowledgments We thank Diane De Abreu, Jennifer Bonderoff, Katie Chiu, and Denis Bouchard for experimental assistance and Brenda Andrews, Ben Blencowe, Dan Durocher, Sara Oster, Frank Sicheri, Mike Tyers, and members of the Harrington lab for helpful discussion. We also thank Joachim Lingner, Sara Evans, and Vicki Lundblad for sharing results prior to publication. L.H. acknowledges funding from the Canadian Institutes of Health Research (MT14340) and the National Institutes of Health (AG16629-04). Received: December 17, 2002 Revised: February 24, 2003 Accepted: February 25, 2003 Published: April 15, 2003 References 1. Nugent, C.I., and Lundblad, V. (1998). The telomerase reverse transcriptase: components and regulation. Genes Dev. 12, 1073–1085. 2. Collins, K., and Mitchell, J.R. (2002). Telomerase in the human organism. Oncogene 21, 564–579. 3. Cong, Y.S., Wright, W.E., and Shay, J.W. (2002). Human telomerase and its regulation. Microbiol. Mol. Biol. Rev. 66, 407–425. 4. Cohn, M., and Blackburn, E.H. (1995). Telomerase in yeast. Science 269, 396–400. 5. Lundblad, V., and Szostak, J.W. (1989). A mutant with a defect in telomere elongation leads to senescence in yeast. Cell 57, 633–643. 6. Lingner, J., Cech, T.R., Hughes, T.R., and Lundblad, V. (1997). Three Ever Shorter Telomere (EST) genes are dispensable for in vitro yeast telomerase activity. Proc. Natl. Acad. Sci. USA 94, 11190–11195. 7. Lin, J.J., and Zakian, V.A. (1996). The Saccharomyces CDC13 protein is a single-strand TG1-3 telomeric DNA-binding protein in vitro that affects telomere behavior in vivo. Proc. Natl. Acad. Sci. USA 93, 13760–13765. 8. Evans, S.K., and Lundblad, V. (1999). Est1 and Cdc13 as comediators of telomerase access. Science 286, 117–120. 9. Pennock, E., Buckley, K., and Lundblad, V. (2001). Cdc13 delivers separate complexes to the telomere for end protection and replication. Cell 104, 387–396.
10. Virta-Pearlman, V., Morris, D.K., and Lundblad, V. (1996). Est1 has the properties of a single-stranded telomere end-binding protein. Genes Dev. 10, 3094–3104. 11. Chandra, A., Hughes, T.R., Nugent, C.I., and Lundblad, V. (2001). Cdc13 both positively and negatively regulates telomere replication. Genes Dev. 15, 404–414. 12. Zhou, J., Hidaka, K., and Futcher, B. (2000). The Est1 subunit of yeast telomerase binds the Tlc1 telomerase RNA. Mol. Cell. Biol. 20, 1947–1955. 13. Steiner, B.R., Hidaka, K., and Futcher, B. (1996). Association of the Est1 protein with telomerase activity in yeast. Proc. Natl. Acad. Sci. USA 93, 2817–2821. 14. Nugent, C.I., Hughes, T.R., Lue, N.F., and Lundblad, V. (1996). Cdc13p: a single-strand telomeric DNA-binding protein with a dual role in yeast telomere maintenance. Science 274, 249–252. 15. Evans, S.K., and Lundblad, V. (2002). The Est1 subunit of Saccharomyces cerevisiae telomerase makes multiple contributions to telomere length maintenance. Genetics 162, 1101–1115. 16. Evans, S.K., and Lundblad, V. (2000). Positive and negative regulation of telomerase access to the telomere. J. Cell Sci. 113, 3357–3364. 17. Grandin, N., Reed, S.I., and Charbonneau, M. (1997). Stn1, a new Saccharomyces cerevisiae protein, is implicated in telomere size regulation in association with Cdc13. Genes Dev. 11, 512–527. 18. Grandin, N., Damon, C., and Charbonneau, M. (2000). Cdc13 cooperates with the yeast Ku proteins and Stn1 to regulate telomerase recruitment. Mol. Cell. Biol. 20, 8397–8408. 19. Grandin, N., Damon, C., and Charbonneau, M. (2001). Ten1 functions in telomere end protection and length regulation in association with Stn1 and Cdc13. EMBO J. 20, 1173–1183. 20. Qi, H., and Zakian, V.A. (2000). The Saccharomyces telomerebinding protein Cdc13p interacts with both the catalytic subunit of DNA polymerase alpha and the telomerase-associated est1 protein. Genes Dev. 14, 1777–1788. 21. Taggart, A.K., Teng, S.C., and Zakian, V.A. (2002). Est1p as a cell cycle-regulated activator of telomere-bound telomerase. Science 297, 1023–1026. 22. Hughes, T.R., Evans, S.K., Weilbaecher, R.G., and Lundblad, V. (2000). The Est3 protein is a subunit of yeast telomerase. Curr. Biol. 10, 809–812. 23. Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z., Miller, W., and Lipman, D.J. (1997). Gapped BLAST and PSIBLAST: a new generation of protein database search programs. Nucleic Acids Res. 25, 3389–3402. 24. Clissold, P.M., and Ponting, C.P. (2000). PIN domains in nonsense-mediated mRNA decay and RNAi. Curr. Biol. 10, R888–890. 25. Singh, S.M., Steinberg-Neifach, O., Mian, I.S., and Lue, N.F. (2002). Analysis of telomerase in Candida albicans: potential role in telomere end protection. Eukaryotic Cell 1, 967–977. 26. Snow, B.E., Krumins, A.M., Brothers, G.M., Lee, S.F., Wall, M.A., Chung, S., Mangion, J., Arya, S., Gilman, A.G., and Siderovski, D.P. (1998). A G protein gamma subunit-like domain shared between RGS11 and other RGS proteins specifies binding to Gbeta5 subunits. Proc. Natl. Acad. Sci. USA 95, 13307–13312. 27. Seto, A.G., Livengood, A.J., Tzfati, Y., Blackburn, E.H., and Cech, T.R. (2002). A bulged stem tethers Est1p to telomerase RNA in budding yeast. Genes Dev. 16, 2800–2812. 28. Livengood, A.J., Zaug, A.J., and Cech, T.R. (2002). Essential regions of Saccharomyces cerevisiae telomerase RNA: separate elements for Est1p and Est2p interaction. Mol. Cell. Biol. 22, 2366–2374. 29. Wen, J., Cong, Y.S., and Bacchetti, S. (1998). Reconstitution of wild-type or mutant telomerase activity in telomerase-negative immortal human cells. Hum. Mol. Genet. 7, 1137–1141. 30. 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). Two inactive fragments of the integral RNA cooperate to assemble active telomerase with the human protein catalytic subunit (hTERT) in vitro. Mol. Cell. Biol. 19, 6207–6216. 31. Baerlocher, G.M., Mak, J., Tien, T., and Lansdorp, P.M. (2002). Telomere length measurement by fluorescence in situ hybridization and flow cytometry: tips and pitfalls. Cytometry 47, 89–99.
Current Biology 704
32. Rufer, N., Dragowska, W., Thornbury, G., Roosnek, E., and Lansdorp, P.M. (1998). Telomere length dynamics in human lymphocyte subpopulations measured by flow cytometry. Nat. Biotechnol. 16, 743–747. 33. de Lange, T. (2002). Protection of mammalian telomeres. Oncogene 21, 532–540. 34. Reichenbach, P., Ho¨ss, M., Azzalin, C.M., Nabholz, M., Bucher, P., and Lingner, J. (2003). A human homolog of yeast Est1 associates with telomerase and uncaps chromosome ends when overexpressed. Curr. Biol. 13, 568–574. 35. Baumann, P., Podell, E., and Cech, T.R. (2002). Human pot1 (protection of telomeres) protein: cytolocalization, gene structure, and alternative splicing. Mol. Cell. Biol. 22, 8079–8087. 36. Baumann, P., and Cech, T.R. (2001). Pot1, the putative telomere end-binding protein in fission yeast and humans. Science 292, 1171–1175. 37. Harrington, L., Zhou, W., McPhail, T., Oulton, R., Yeung, D.S., Mar, V., Bass, M.B., and Robinson, M.O. (1997). Human telomerase contains evolutionarily conserved catalytic and structural subunits. Genes Dev. 11, 3109–3115. Accession Numbers Sequence data reported in this paper have been deposited in the NCBI GenBank database under Accession Numbers AY168921 and AY168922.