Submitted August 12, 1997.
The tandemly repeated DNA sequence of telomeres is typically specified by the ribonucleoprotein enzyme telomerase. Telomerase copies part of its intrinsic RNA moiety to synthesize one strand of the telomeric repeat DNA. Recent work, taken together with many observations over the past years, has led to the concept of a telomere homeostasis system. We have analyzed the interplay between two key physical components of this system: structural components of the telomere itself and of telomerase. Here we review some of these recent studies. The experimental method used in common in these studies was to make mutations in the template sequence of telomerase RNA, which caused various phenotypes. First, mutating specific residues in the ciliate Tetrahymena thermophila and yeast showed that these residues are required for critical aspects of the enzymatic action of telomerase. Second, certain mutated telomeric sequences caused a strong anaphase block in Tetrahymena micronuclei. Third, specific template mutations in the telomerase RNA gene led to varying degrees of telomere elongation in Tetrahymena and the yeast Kluyveromyces lactis. For some of the K. lactis mutations, the loss of length unregulated elongation was directly related to loss of binding to K. lactis Rap1 protein. Using K. lactis carrying alterations in the telomerase RNA template, and in the gene encoding the Rap1 protein, we found that a crucial determinant of telomere length homeostasis is the nature of the duplex DNA--Rap1 protein complex on the very end repeat of the telomere. We propose that this complex plays a key role in regulating access of telomerase to the telomere.
KEY WORDS: telomeres, telomerase, homeostasis of telomere lengths, Rap1 protein.
In most eukaryotes examined, the telomeric DNA consists of tracts of simple, tandemly repeated sequences extending to the chromosomal ends. These telomeric repeats comprise the essential DNA sequences required for chromosomal stability and complete replication, and are generally short, G-rich tandem repeats on the strand running 5´ to 3´ toward the distal end of the chromosome (reviewed in ).
Several years ago it was discovered that telomerase synthesizes one strand of the telomeric repeats [5, 6]. This is accomplished by copying a short template sequence within the RNA moiety of telomerase ([7-9], reviewed in [4, 10]). Hence, telomerase, a ribonucleoprotein (RNP) complex, is a cellular reverse transcriptase. Telomerase activity has been detected in extracts from a very wide range of eukaryotes [11-13]. One known exception to the normal telomerase mode of maintenance of telomeres is the fruit fly Drosophila, which relies on a retrotransposon additions for maintenance of chromosomal termini [14, 15].
Mutations in the telomerase RNA gene template cause synthesis of the predicted altered telomeric repeat sequences in vivo [9, 16]. In some cases, such template mutations cause effects on cells that can be attributed to the effects of the changed telomeric repeats on telomere functioning. Telomerase RNA mutations can also dramatically affect the enzymatic properties of telomerase activity. Here we review the insights into telomerase and telomere functioning gained from experiments we have performed with telomerase in Tetrahymena and budding yeasts. The experimental approach common to these experiments was to alter the telomerase RNA template sequence. The findings from such experiments have provided some unexpected answers to questions about three quite different facets of telomere biology: the enzymatic function of telomerase, the role of telomeres in chromosome separation, and telomere length homeostasis.
Mutating Telomeric DNA Prevents Chromosome Separation in Anaphase
We recently uncovered a novel role of telomeres in chromosome segregation during mitosis . In this work, Tetrahymena thermophila cells were forced to synthesize mutated telomeric DNA. In T. thermophila, the wild-type telomerase RNA contains the template sequence for the synthesis of GGGGTT telomeric repeats. We changed the template to a sequence predicted to synthesize the GGGGTTTT telomeric repeats found in some other ciliates, and introduced this mutated gene (ter 1-43 AA) on a high copy episomal vector into wild-type T. thermophila cells, leading to overexpression of this mutated gene. Within ten to fifteen cell generations after transformation, the cell population doubling rate had slowed and division of the germ-line nucleus in these cells was found to be severely blocked. Most transformant cells rapidly became grossly enlarged and misshapen, and by 18-30 population doublings completely ceased to divide. Unexpectedly, the block to nuclear division occurred specifically during anaphase. High-resolution three-dimensional fluorescence microscopy demonstrated that even late in anaphase, the mutant chromatids failed to separate at the midzone, often becoming stretched to up to twice their normal length. The lack of separation between daughter nuclei suggested that the mutant telomeres prevented complete chromosome segregation.
Anaphase micronuclei in mutant cells undergoing mitosis were frequently similar in length to wild-type anaphase micronuclei but, as described above, the lagging chromosome ends failed to separate. Despite the inability to complete micronuclear division, some aspects of cell cycle progression took place, as evidenced by the extreme stretching of the mutant anaphase micronucleus, indicating continuation of anaphase, visible attempts at cleavage furrow formation and grossly enlarged cytoplasm. Such a progression is predicted from the known occurrence in ciliates of commitment to cell division during micronuclear anaphase .
Tetrahymena cells contain two nuclei: the conventional, mitotically dividing micronucleus which was the nucleus described above, and the polygenomic, amitotically dividing macronucleus. We previously showed that macronuclear division is also strongly impaired by various telomeric DNA mutations, all created by making different base changes in the telomerase RNA template sequence [9, 19-21]. In wild-type Tetrahymena cells, division of the two nuclei occurs at different times in the cell cycle: the micronucleus completes mitosis before the macronucleus undergoes its amitotic division, after which cytokinesis occurs. In agreement with this order of cell cycle events, in the telomere mutant cells the macronucleus was never observed to initiate division while the micronucleus was arrested in anaphase. Hence, the onset of macronuclear division may require proper completion of micronuclear division. As a consequence, although mutant telomeres may also impair macronuclear division independently of the micronuclear telomere phenotype, such impairment of macronuclear division may not become apparent while the micronucleus is mitotically blocked.
What is the nature of the block to nuclear division caused by mutant telomeric repeats? The frequency with which mutant micronuclear telomeres failed to separate in the experiments described above approached 100% for every chromatid. If mutant telomeres were recognized by the cell as "damaged" DNA, a cell cycle block might be triggered, as in the RAD9-dependent arrest which occurs in response to broken chromosome ends or loss of a yeast telomere [2, 3, 18]. However, such cell cycle arrest would be expected to occur in G2, prior to mitosis . Instead the striking feature of the Tetrahymena telomere mutants was that cell division was blocked only late in the progression of anaphase, with many aspects of the cell cycle continuing during this block. Such a situation is consistent with a physical block in telomere separation. We favor a model in which telomeres of sister chromatids normally are associated until metaphase, and this association must be resolved prior to chromosome segregation in anaphase. We therefore suggest that this is a primary role of telomeres, to promote correct mitotic segregation of sister chromatids, and that the mutated telomeric DNA sequence prevents telomere separation in anaphase.
Little is known about how the telomeres of sister chromatids cohere to one another throughout the G2-phase of the cell cycle and metaphase. Telomere--telomere proximity has been observed cytologically in various organisms , and in mitotically dividing S. pombe cells telomere--telomere association occurs specifically in G2 . Normal telomere associations could be mediated through known telomere binding proteins, such as those in ciliates, the Rap1 and SIR proteins in yeasts , or TRF1 in humans . It has also been suggested that the telomeric single-stranded DNA tail extensions generated at S. cerevisiae telomeres during S-phase might facilitate telomere--telomere interactions between sister chromatids .
If telomeres participate in sister chromatid cohesion, they must then also separate in a timely manner during chromosome segregation. Perhaps, the mutant telomeres cannot bind protein factors properly, or changing the DNA primary sequence of telomeres might change their structure so as to alter the topological recognition of the telomere complex by enzymes such as topoisomerase II. Topoisomerase II is required for anaphase separation of chromosomes in S. pombe and S. cerevisiae . Alternatively, telomeres could become inaccessible to such factors through altered spatial organization in the nucleus. These possibilities remain to be tested.
Alterations in Telomerase Enzymatic Action Caused by Telomerase RNA Mutations
An unexpected finding that arose from making mutations in telomerase RNA, initially in the template sequence, was the importance of specific template bases for proper telomerase polymerization. Mutant telomerase RNA genes with altered residues within the 3´-AACCCCAAC-5´ sequence, when introduced into Tetrahymena, caused the synthesis of the specifically altered telomeric repeats, demonstrating that specific residues in the RNA act as the template for telomere synthesis . However, in addition to these expected changes, mutating specific RNA positions in the templating domain also led to aberrant behavior of telomerase in vitro, including different kinds of loss of enzymatic fidelity. For example, a single C to U mutation, at a position within the telomerase RNA template sequence of Tetrahymena telomerase, prevented telomere maintenance in vivo: the telomeres shortened, and after about 25 cell fissions following substitution of telomerase with the mutated RNA, the cells ceased to divide [9, 29]. In vitro analysis of this telomerase showed that it had a greatly increased rate of misincorporation (up to 50% under some conditions) when copying the neighboring template residues. In addition to causing misincorporation, the mutant telomerase was impeded in copying the full template: the elongating product prematurely dissociated from the template two nucleotides before the end of the template was reached. A premature dissociation at the same position in vitro was also caused by another point mutation in the template RNA . An even more dramatic effect on enzymatic activity was caused by a template alteration in the S. cerevisiae, telomerase RNA. A specific three-base substitution (tlc 1-476 GUG mutation) within the 17 base templating domain of the telomerase RNA led to loss of any detectable in vitro telomerase activity in extracts made from these haploid mutant cells. These haploid cells also exhibited telomere instability and senescence, consistent with a lack of telomerase activity in vivo . Hence, base-specific interactions in the template region are critical for enzyme function in this species too.
A highly conserved set of secondary structural features has emerged from phylogenetic comparison of over twenty ciliate telomerase RNAs ( and references therein). Mutations of the Tetrahymena telomerase RNA that disrupted these conserved structures, while allowing some degree of assembly of the mutated telomerase RNA into a telomerase RNP, severely compromised telomerase activity in vivo and in vitro (D. Gilley, T. Ware, and E. Blackburn, unpublished results). The importance of the non-template portion of telomerase RNA was highlighted by performing a cross-species telomerase RNA swap, in which telomerase RNA from the ciliate Glaucoma was expressed in Tetrahymena cells. The Glaucoma telomerase RNA is ~50% divergent in sequence from Tetrahymena telomerase RNA outside the template, but shares a 23 base region of sequence identity centered on the template, and its secondary structure is virtually superimposable upon that of the Tetrahymena telomerase RNA. The Glaucoma telomerase RNA assembled into an RNP in Tetrahymena cells, and partially functioned in vivo and in vitro. However, this hybrid enzyme RNP, made up of Glaucoma RNA complexed with Tetrahymena proteins, polymerized repeats less processively than wild type and displayed an aberrant cleavage activity . This miscleavage reaction of the hybrid enzyme is likely to reflect altered interactions between RNA and protein, which misplace the cleavage active site relative to the template/primer binding site. Together, these findings suggest that telomerase activity can be controlled or regulated by changing molecular interactions within the telomerase RNP mediated through the non-template RNA domains.
Together, these results point to an extraordinary degree of dependence of telomerase on its RNA component--far beyond that expected for the RNA copied by the usual type of reverse transcriptases (in retroviruses or retroelements). Although telomerase is functionally a reverse transcriptase , and its essential catalytic protein component closely resembles other reverse transcriptases in primary amino acid structure , the fact that a trinucleotide base substitution in the 1.3 kb RNA component of yeast destroys detectable enzymatic function is a striking demonstration that the RNA also provides an essential component of the catalytic mechanism beyond that of a template.
Telomere Length Homeostasis
How does the cell regulate synthesis of its telomeres, to ensure their continued presence at the ends of chromosomes? The mode of maintenance of telomeric DNA is different from that of regions of chromosomal DNA in internal regions of chromosomes. Much evidence, direct and indirect, shows that as cells divide, their telomeric DNA is in a constant state of flux, continually being both lost from the terminus, and replenished by the action of telomerase [16, 34]. Thus keeping the processes of addition and loss in balance is a central aspect of telomeric maintenance. Too much telomeric DNA at the chromosome ends can be as deleterious as not enough . Conversely, too little or no telomeric DNA at a chromosomal terminus can cause that chromosome end to be perceived as a broken DNA, in other words, as DNA damage [2, 3]. With too little telomeric DNA, cell cycle arrest ensues. Recombinational mechanisms can repair the "break" by bringing more telomeric DNA to a given chromosomal end, by gene conversion of repeats from either another chromosomal terminal region, or from more internal tracts of telomeric sequences [2, 35]. Recombination events that lead to extra telomeric DNA sequences being placed onto the chromosomal terminus rescue the telomere. In the yeast K. lactis, rescue of telomeres by recombination between the terminal telomeric repeat tracts themselves requires the gene RAD52. This process is probably used in many eukaryotes and we have dubbed it telomere CPR, for telomeric cap-prevented recombination . Telomere CPR is characterized by highly heterogeneous, typically long, telomeric DNA tracts which show episodes of telomere shortening and considerable plasticity in length as cells are propagated. Evidence for a process of telomere maintenance that is the same or similar to telomere CPR has also been found in immortalized human cells lacking detectable telomerase [36, 37]. In one study, 15 out of 50 immortalized cell lines obtained by SV40 transformation of human cells had this alternative, non-telomerase-mediated means of telomere maintenance .
Regulation of Telomere Length Involves the Telomeric DNA--Protein Complex Itself
A striking aspect of telomere maintenance, highlighted by recent findings, is the normal resilience of the telomere to over-shortening or over-lengthening. Cells have a system, called telomere length homeostasis, for keeping the telomere size within well defined bounds. Thus, it is in exceptional rather than normal situations that telomere length falls below its optimal length and requires the backup mechanism of recombination described above.
Telomere homeostasis normally involves not only telomerase, which keeps telomeric DNA "topped up", but also mechanisms that keep its length in check. Regulation of telomere length appears to be exerted at least two levels. A large component of the negative control on telomere length in yeasts and human cells in culture appears to be exerted through the actions of the Rap1 protein, which binds sequence-specifically to duplex telomeric DNA repeats [38, 39], or its functional homolog in mammalian cells, TRF1 . The number of Rap1 proteins on a given telomere is one signal sent to the telomere length regulation machinery that determines length of that telomere . However, a quantitatively much greater contribution to telomere length regulation is made by a different aspect of Rap1 telomere binding: the occupancy of the repeat unit at the very end of the telomeric tract by Rap1. This critical component of telomere length homeostasis was revealed in studies in K. lactis in which both the telomeric DNA sequence and Rap1 protein were altered .
Altering the telomeric repeat sequences caused sequence-specific effects on length regulation . With mutations that disrupted Rap1 binding, regulation was impaired even when essentially full length tracts of wild type repeats were present at the inner portions of the telomeres, to which mutant repeats had been added distally. A mutant telomerase RNA gene that caused synthesis of repeats that cannot bind Rap1 normally was introduced into K. lactis cells, or, conversely, such a mutant telomerase RNA gene, after it had time to cause mutant repeats to be added to telomeres, was replaced by the wild type telomerase RNA [16, 38]. By observing the kinetics of loss of telomere length control in these experiments, it became clear that only the distal one to a few repeats at the telomeric terminus was highly critical for length control. Hence, we proposed that mutating the terminal telomeric sequence reduces the in vivo occupancy of telomeric repeats by Rap1 proteins, so that the telomeric repeat at the end of the telomere has a higher probability of being extended by telomerase. In a simple steric hindrance model, Rap1 bound to the terminal telomeric repeat(s) could normally limit access of telomerase to the telomere terminus. Other possible models are that Rap1 directly interacts with telomerase to negatively regulate telomerase elongation; for example, Rap1 bound to telomeric DNA could induce a conformational change in telomerase that inhibits its polymerization activity. Conversely, such interaction could stimulate a telomerase-intrinsic telomere cleavage activity like that found existing in the Tetrahymena and S. cerevisiae telomerases [32, 42, 43].
The kinetics of loss of telomere length regulation in certain Rap1 and telomeric repeat double mutants of K. lactis  strongly suggested that the telomeric Rap1-containing complex is metastable at its distal tip, existing in a telomerase-inaccessible form and an alternative, accessible, form. Conceptually, this resembles the two-state model for establishment and loss of the silencing chromatin at the more internally-located silent mating type loci in S. cerevisiae (for references and review see ). In the K. lactis experiments, the gradual conversion of the telomere population from regulated to unregulated in K. lactis double mutants suggested that once the first few mutant telomeric repeats are added to a telomeric terminus, that telomeric complex acquires a finite probability of losing its telomerase-inaccessible state. Once this loss of the inaccessible state occurs on any telomere, it is not easily reversed, leading to loss of length control of a increasing fraction of the telomeres over time. Hence, we have proposed that in K. lactis, as in S. cerevisiae [39, 45-48], Rap1 bound to duplex telomeric DNA nucleates a structured, higher order chromatin complex at telomeres . Furthermore, we propose that the major biological function of this complex is to cap chromosome ends and regulate telomere length.
In summary, the dependence of chromosome segregation on the correct telomeric sequence is striking given that the telomeric DNA typically constitutes less than a ten thousandth of the total eukaryotic chromosomal DNA. The importance of telomeric DNA for anaphase separation provides one compelling explanation for the necessity for telomeres in actively dividing cells. What makes the effects of telomeric sequence mutations even more striking is that they are manifested even when only a small fraction of the telomeric repeat tract, specifically that at the extreme terminus, is mutant. Furthermore, these effects of altered telomeric DNA are exerted quite rapidly. The practical significance of these results is that they suggest that altering telomeric DNA, rather than simply loss of telomeric DNA, can cause a block in nuclear division. We have also found that synthesis of the correct telomeric DNA sequence is very dependent on the telomerase RNA structure. Hence, it is possible to envision inducing changes in telomerase RNA to cause it to make deleterious telomeres, in order to attack cancerous or other pathogenic proliferating cells.
Telomerase RNA alterations also reveal interesting properties of the enzyme itself. Our data compellingly argue that the active site of Tetrahymena telomerase (and almost certainly of all telomerases) is one in which base- and structure-specific RNA--protein interactions are required for optimal and correctly templated elongation. How this interdependence between protein and nucleic acid function evolved is an exciting question that is likely to be relevant to considering other RNP collaborative efforts in the modern cell, such as the ribosome and the spliceosome.
Finally, it has become clear that telomeric DNA length is primarily controlled by the antagonistic actions of two cellular complexes that interact with the telomeric DNA: the telomeric DNA--protein complex centered on the structural binding protein Rap1, and the enzymatic activity that elongates telomeres, the ribonucleoprotein telomerase. Thus, the telomere itself is a major controller of telomerase action. It has been hypothesized that telomerase activity is required to maintain telomeres in proliferating human cells and hence that increasing the level of telomerase may promote their conversion to, or maintenance in, a cancerous state. It is likely that in mammalian cells telomere length is negatively regulated by the Rap1-like telomere binding factor TRF1 . Therefore, we need to consider other factors besides telomerase levels as determinants of telomere length maintenance in human cells.
1.Blackburn, E. H., and Greider, C. W. (eds.) (1995)
Telomeres, Cold Spring Harbor Laboratory Press, Cold Spring
2.McEachern, M. J., and Blackburn, E. H. (1996) Genes Dev., 10, 1822-1834.
3.Sandell, L. L., and Zakian, V. A. (1993) Cell, 75, 729-739.
4.Blackburn, E. H. (1994) Cell, 77, 621-623.
5.Greider, C. W., and Blackburn, E. H. (1985) Cell, 43, 405-413.
6.Greider, C. W., and Blackburn, E. H. (1987) Cell, 51, 887-898.
7.Greider, C. W., and Blackburn, E. H. (1989) Nature, 337, 331-337.
8.Shippen-Lentz, D., and Blackburn, E. H. (1990) Science, 247, 546-552.
9.Yu, G. L., Bradley, J. D., Attardi, L. D., and Blackburn, E. H. (1990) Nature, 344, 126-132.
10.Greider, C. W. (1995) in Telomeres (Blackburn, E. H., and Greider, C. W., eds.) Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, pp. 35-68.
11.Blackburn, E. H. (1992) Ann. Rev. Biochem., 61, 113-129.
12.Prowse, K. R., Avilion, A. A., and Greider, C. W. (1993) Proc. Natl. Acad. Sci. USA, 90, 1493-1497.
13.Mantell, L. L., and Greider, C. W. (1994) EMBO J., 13, 3211-3217.
14.Biessmann, H., Mason, J. M., Ferry, K., d'Hulst, M., Valgeirsdottir, K., Traverse, K., and Pardue, M. L. (1990) Cell, 61, 663-673.
15.Levis, R. W., Ganesan, R., Houtchens, K., Tolar, L. A., and Sheen, F. M. (1993) Cell, 75, 1083-1093.
16.McEachern, M. J., and Blackburn, E. H. (1995) Nature, 376, 403-409.
17.Kirk, K. E., Harmon, B. P., Reichardt, I. K., Sedat, J. W., and Blackburn, E. H. (1997) Science, 275, 1478-1481.
18.Adl, S. M., and Berger, J. D. (1996) J. Eukaryot. Microbiol., 43, 77-86.
19.Lee, M. S., Gallagher, R. C., Bradley, J., and Blackburn, E. H. (1993) Cold Spring Harb. Symp. Quant. Biol., 58, 707-718.
20.Yu, G. L., and Blackburn, E. H. (1991) Cell, 67, 823-832.
21.Romero, D. P., and Blackburn, E. H. (1995) J. Eukaryot. Microbiol., 42, 32-43.
22.Weinert, T. A., Kiser, G. L., and Hartwell, L. H. (1994) Genes Dev., 8, 652-665.
23.Yanagida, M. (1995) Bioessays, 17, 519-526.
24.Dernberg, A. F., Sedat, J. W., Cande, W. Z., and Bass, H. W. (1995) in Telomeres (Blackburn, E., and Greider, C., eds.) Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, pp. 96-105.
25.Henderson, E. (1995) in Telomeres (Blackburn, E., and Greider, C., eds.) Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, pp. 11-34.
26.Fang, G., and Cech, T. R. (1995) in Telomeres (Blackburn, E., and Greider, C., eds.) Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, pp. 69-95.
27.Funabiki, H., Hagan, I., Uzawa, S., and Yanagida, M. (1993) J. Cell Biol., 121, 961-976.
28.Chong, L., van Steensel, B., Broccoli, D., Erdjument, B. H., Hanish, J., Tempst, P., and de Lange, T. (1995) Science, 270, 1663-1667.
29.Gilley, D., Lee, M. S., and Blackburn, E. H. (1995) Genes Dev., 9, 2214-2226.
30.Prescott, J., and Blackburn, E. H. (1997) Genes Dev., 11, 528-540.
31.McGormick, G. M., and Romero, D. P. (1996) Mol. Cell Biol., 16, 1871-1879.
32.Bhattacharyya, A., and Blackburn, E. H. (1997) Proc. Natl. Acad. Sci. USA, 94, 2823-2827.
33.Lingner, J., Hughes, T. R., Shevchenko, A., Mann, M., Lundblad, V., and Cech, T. (1997) Science, 276, 561-567.
34.Shampay, J., and Blackburn, E. H. (1988) Proc. Natl. Acad. Sci. USA, 85, 534-538.
35.Lundblad, V., and Blackburn, E. H. (1993) Cell, 73, 347-360.
36.Murnane, J. P., Sabatier, L., Marder, B. A., and Morgan, W. F. (1994) EMBO J., 13, 4953-4962.
37.Bryan, T. M., Englezou, A., Gupta, J., Bacchetti, S., and Reddel, R. R. (1995) EMBO J., 14, 4240-4248.
38.Krauskopf, A., and Blackburn, E. H. (1996) Nature, 383, 354-357.
39.Shore, D. (1997) Nature, 385, 676-677.
40.Van Steensel, B., and de Lange, T. (1997) Nature, 385, 740-743.
41.Shore, D. (1996) Nat. Struct. Biol., 3, 491-493.
42.Collins, K., and Greider, C. W. (1993) Genes Dev., 7, 1364-1376.
43.Cohn, M., and Blackburn, E. H. (1995) Science, 269, 396-400.
44.Strahl, B. D., Bolsinger, S., Hecht, A., Luo, K., and Grunstein, M. (1997) Genes Dev., 11, 83-93.
45.Fang, G., and Cech, T. R. (1995) J. Cell Biol., 130, 243-253.
46.Zakian, V. A. (1995) Science, 270, 1601-1607.
47.Gilson, E., Roberge, M., Giraldo, R., Rhodes, D., and Gasser, S. M. (1993) J. Mol. Biol., 231, 293-310.
48.Palladino, F., Laroche, T., Gilson, E., Axelrod, A., Pillus, L., and Gasser, S. M. (1993) Cell, 75, 543-555.
49.De Lange, T. (1995) in Telomeres (Blackburn, E. H., and Greider, C. W., eds.) Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, pp. 265-293.