DOI: 10.1002/cbic.200700272
Telomerase: A Dimeric Mnage Trois Jean-Louis Mergny,*[a, b] Laurent Lacroix,[a, b] and Jean-FranÅois Riou*[c]
Telomeres protect chromosomal ends from fusion events and provide a means for complete replication of the chromosome. Telomere repeats are added by a specialized enzyme, telomerase. Telomerase is expressed in most human tumours, but is almost undetectable in normal somatic cells; thus telomerase is an almost universal marker for human cancer as well as an attractive drug target. Human telomerase contains two essential components, a telomerase ACHTUNGREreverse-transcriptase catalytic subunit (hTERT) that has homology with the active-site motifs of viral reverse transcriptases and an RNA (hTR, also known as TERC or TER) that has an internal template for telomeric repeat synthesis. What are the core components of the active telomerase complex in cells? In addition to hTR and hTERT, more than 30 proteins have been proposed to be associated with the enzymatic complex (see Table S1 in ref. [1]). Given the cumulative molecular weight of all these factors (> 2 MDa) and the measured size of the active telomerase complex,[2, 3] only a small fraction can be involved in the
active enzyme core complex. Telomerase has been purified from the ciliate Euplotes aediculatus and from Tetrahymena thermophila as a complex of TERT, RNA and an associated protein p43 or p95, ACHTUNGRErespectively.[4, 5] Although telomerase activity was first identified in human cells almost two decades ago and reconstitution of hTR and hTERT provides a mini-
mal activity,[6] it is only now that Reddel and colleagues have identified the actual components of active human telomerase.[1] The low cellular abundance of human telomerase necessitated large-scale methods for its purification; Reddel and co-workers used an elegant approach to purify the complex by ~ 108-fold.[1] The
[a] Dr. J.-L. Mergny, Dr. L. Lacroix INSERM, U565, Acides Nucl$iques: Dynamique, Ciblage et Fonctions Biologiques 43 Rue Cuvier, CP26 Paris Cedex 05, 75231 (France) Fax: (+ 33) 1-40-79-37-05 E-mail:
[email protected] [b] Dr. J.-L. Mergny, Dr. L. Lacroix CNRS, UMR5153, Mus$um National d’Histoire Naturelle USM503 D$partement de “R$gulations, D$veloppement et Diversit$ Mol$culaire” Laboratoire des R$gulations et Dynamique des G$nomes 43 Rue Cuvier, CP26 Paris Cedex 5, 75231 (France) [c] Prof. J.-F. Riou Laboratoire d’Onco-Pharmacologie JE 2428, UFR de Pharmacie Universit$ de Reims Champagne–Ardenne 51 Rue Cognacq-Jay, 51096 Reims (France) Fax: (+ 33) 3-26-91-89-26 E-mail:
[email protected]
ChemBioChem 2007, 8, 1483 – 1485
Scheme 1. Purification scheme used by Cohen et al.[1]
> 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1483
J. Mergny, J.-F. Riou and L. Lacroix purification involved three steps (Scheme 1). The first was a classical immunoprecipitation with a sheep polyclonal antibody generated against an hTERT peptide. The authors then developed a substrate-directed affinity purification, taking advantage of the thermodynamic and kinetic properties of the telomerase enzyme.[7] The enzyme formed a stable complex (t1/2 10 h) with the 5’(TTAGGG)3-3’ substrate, and this lifetime allowed its purification. The addition of dTTP and dATP (but not dGTP) partially restored the ability of telomerase to reverse transcribe its RNA and led to the addition of the nucleotides TTA. This transformed a stable enzyme–substrate complex into a relatively unstable complex (t1/2 < 5 min), allowing rapid elution and providing specificity for catalytically active telomerase. As pointed out by the authors, the lower stability of this partially extended complex illustrates the elegant (but unexplained) structure–activity relationship of human telomerase: Transformation to a less stable enzyme–substrate complex as nucleotide addition approaches the end of the template facilitates efficient translocation and a new round of telomere-repeat extension. ACHTUNGREDespite starting from ~ 1010 human cells (100 g of HEK-293 cells), the concentration of protein in the final purified fraction was too low to measure optically, and the authors were unable to visualize any bands on a silver-stained SDS protein gel. The authors choose to follow telomerase activity during the purification procedure through a direct telomerase activity assay rather than a TRAP assay. The former is less sensitive but also far less prone to artefacts than the second. Importantly, the sedimentation profile of purified telomerase was conserved relative to that of the crude lysate; this implies that no large component was lost or degraded during the purification. Purified telomerase was digested with trypsin, and the resulting peptides were purified and analyzed by mass spectrometry (nanoLC-MS/MS). In the telomerase sample only two proteins were consistently observed: hTERT and dyskerin. The proteins tubulin and actin were found in telomerase samples but also in parallel control experiments in which the
1484
www.chembiochem.org
enzyme was not activated (i.e., in the absence of dTTP and dATP). The presence of the RNA component (hTR) was demonstrated by Northern blot. The data indicate that the active human telomerase enzyme core complex is composed solely of three factors: two protein components, hTERT and dyskerin, and the RNA component, hTR. Given the actual masses of hTERT (127 kDa), hTR (153 kDa) and dyskerin (57 kDa), a complex composed of two of each component (2:2:2) would have a total mass of 674 kDa, close to the observed size (650 to 670 kDa). Note that other combinations (for example 3:1:2) would also give a complex in the observed size range. However, previous publications exclude this possibility, as data have established that human telomerase is a dimer containing two hTRs and most likely two hTERTs.[2, 9] With the notable exception of Tetrahymena,[8] telomerases from most organisms act as dimers. The HEK-293 cell line used to purify the complex was chosen for its ability to grow in a bioreactor at high density, thus providing the preparative quantity of cells required during the purification procedures. However, the authors also measured the size of the active human telomerase complex in a variety of other immortal cell lines. Whole-cell lysates from all cell lines tested exhibited a similar sedimentation profile, with 60 % of total activity eluting in fractions corresponding to an enzyme complex of ~ 650 to 670 kDa (less than what was reported in ref. [9]). This observation tells us that the conclusions reached for the HEK 293 cell line apply to a variety of other human cells and provide some generality for the conclusions. This article is interesting for a number of reasons. From a technical point of view, purifying this complex was equivalent to finding a needle in a haystack. The authors estimate than each HEK-293 cell has only 20 to 50 molecules of telomerase. State-of-the-art techniques allowed Reddel and colleagues to solve this daunting challenge, but the real key was to take advantage of the kinetic properties of this enzyme toward its substrate.[7] Similar methodology could have applications in other purification schemes.
One may argue that this purification was specific for catalytically active enzyme core complexes and therefore represents a partial view of the overall biology of telomerase. Other partners are presumably necessary for the biogenesis, export and stability of individual components of this enzyme. These factors might dissociate from the complex before it reaches its substrate or they could dissociate when the enzyme is activated. Furthermore, the elongation of telomeres might not be the sole function of telomerase: This complex could have “extracurricular”[10] or extratelomeric[11] activities, and perhaps a totally different composition. Mizuno et al. found that human telomerase exists as two distinct active complexes in vivo.[12] Thus, due to the technical bias of the purification strategy, it ACHTUNGREremains to be tested 1) whether this model also applies to normal cells that express relatively high levels of telomerase activity, 2) whether the simple oligonucleotide substrate chosen here recruits the same complex as a natural chromosomal end and 3) whether this model can be generalized to other organisms. It also remains to be determined how the partners interact. It is known that hTR contacts all other factors (hTERT, dyskerin and the DNA substrate). However, it remains to be determined, for example, whether dyskerin establishes a direct contact with the DNA substrate or hTERT. The data provided by Cohen et al. reinforce the role of dyskerin in the telomerase pathway. Dyskerin is a putative pseudouridine synthase that is associated with RNAs that contain the H/ACA motif. These RNAs include not only H/ ACA small nucleolar RNAs, but also human telomerase RNA.[13] The telomeric function of dyskerin might explain the symptoms of the human disease dyskeratosis congenita (DC), a rare bonemarrow-failure syndrome: The physiopathology of DC predominantly relates to defective telomere maintenance (for a recent review, see ref. [14]). In fact, all three components of the active complex might be defective in various forms of DC. One could have expected other factors to be present in the active telomerase complex. Transient associations of
> 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2007, 8, 1483 – 1485
Telomerase hTR and hTERT with chaperone activities occur during the assembly. For example, Keppler et al. recently reported that Hsp90 inhibitors inhibit recombinant telomerase even after telomerase is assembled and suggested a role for Hsp90 in loading telomerase onto the telomere.[15] If this is true, the protein must be easily dissociated from the holoenzyme. The Pif1 helicase has been reported as a negative regulator of telomerase activity. Pif1 appears to unwinds the DNA/hTR hybrid and remains associated with a telomerase activity detected in vitro.[16] Purified hTR and hTERT form an RNP with very limited activity; the addition of dyskerin could perhaps enhance telomerase activity. To our knowledge, this remains to be tested in vitro. However, these observations will hopefully lead to im-
ChemBioChem 2007, 8, 1483 – 1485
proved telomerase reconstitution in vitro for enzymatic and structural studies. Keywords: complexes telomerase · telomeres
·
dyskerin
·
[1] S. B. Cohen, M. E. Graham, G. O. Lovrecz, N. Bache, P. J. Robinson, R. R. Reddel, Science 2007, 315, 1850 – 1853. [2] C. Wenz, B. Enenkel, M. Amacker, C. Kelleher, K. Damm, J. Lingner, EMBO J. 2001, 20, 3526 – 3534. [3] G. Schnapp, H. P. Rodi, W. J. Rettig, A. Schnapp, K. Damm, Nucleic Acids Res. 1998, 26, 3311 – 3313. [4] J. Lingner, T. R. Cech, Proc. Natl. Acad. Sci. USA 1996, 93, 10712 – 10717. [5] K. Collins, R. Kobayashi, C. W. Greider, Cell 1995, 81, 677 – 686. [6] G. B. Morin, Cell 1989, 59, 521 – 529. [7] G. Wallweber, S. Gryaznov, K. Pongracz, R. Pruzan, Biochemistry 2003, 42, 589 – 600.
> 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
[8] T. M. Bryan, K. J. Goodrich, T. R. Cech, Mol. Biol. Cell 2003, 14, 4794 – 4804. [9] T. L. Beattie, W. Zhou, M. O. Robinson, L. Harrington, Mol. Cell. Biol. 2001, 21, 6151 – 6160. [10] S. Chang, R. A. de Pinho, Proc. Natl. Acad. Sci. USA 2002, 99, 12520 – 12522. [11] S. R. Lai, A. P. Cunningham, V. Q. Huynh, L. G. Andrews, T. O. Tollefsbol, Exp. Cell Res. 2007, 313, 322 – 330. [12] H. Mizuno, S. Khurts, T. Seki, Y. Hirota, S. Kaneko, S. Murakamki, J. Biochem. 2007, 141, 641 – 652. [13] J. R. Mitchell, E. Wood, K. Collins, Nature 1999, 402, 551 – 555. [14] T. Vulliamy, I. Dokal, Semin. Hematol. 2006, 43, 157 – 166. [15] B. R. Keppler, A. T. Grady, M. B. Jarstfer, J. Biol. Chem. 2006, 281, 19840 – 19848. [16] J. B. Boule, V. A. Zakian, Nucleic Acids Res. 2006, 34, 4147 – 4153.
Received: May 21, 2007 Published online on July 12, 2007
www.chembiochem.org
1485
