A protein-counting mechanism regulates telomere length

Eukaryotic chromosomes are linear. This might be due to the establishment of sexual reproduction at a very early stage of evolution. Indeed during meiosis, exchanges between circular homologues would create genetically unstable dicentric products (1). However, chromosome linearity results in natural breaks in the double helix, creating difficulties in genome stability. The natural chromosomal termini, or telomeres, should specifically escape checkpoint and repair systems that would normally act on an accidental DNA break (2). Furthermore, the replication of DNA ends involves an absolute loss of sequences since primer degradation at the last Okasaki fragment leaves a single strand gap that cannot be filled in by known DNA polymerases (3). Thus, a mechanism to complete the replication of chromosome ends is needed.

Telomeres are specific DNA-protein complexes. In most organisms, telomeric DNA consists of a tandem array of short guanine-rich repeats, the G-rich strand running 5' to 3' toward the distal end of the chromosome and ending as a short single-stranded 3' overhang (4, 5). In the budding yeast Saccharomyces cerevisiae, probably the best model organism for studying telomeres, telomeric DNA consists of a few hundred base pairs of TG_1-3 repeats organized in a nonnucleosomal structure based on an array of the telomere repeat-binding protein Rap1p (6).

The progressive loss of telomere repeats linked to replication is balanced by a unique ribonucleoprotein reverse transcriptase called telomerase, which specifically extends the 3' G-rich telomeric strand (7). Originally identified in ciliates, the RNA component and the catalytic subunit of telomerase have now been found in a variety of organisms, including yeast and mammals (8). In humans, a loss of telomerase naturally takes place in somatic cells, leading to telomere shortening and, after a generational lag period, to the start of senescence. This programmed repression of telomerase activity in normal human cells also seems to act as a tumor suppressor mechanism that limits the growth potential of transformed cells (9).

In the presence of telomerase, telomere length is maintained at a constant mean value, creating a balance between elongation and shortening. It has been proposed that this equilibrium is determined by negative regulation of telomerase activity by the telomere itself (10). Indeed, in yeast, deletion of the Rap1p carboxyl-terminal domain or telomeric DNA alterations decreasing Rap1p-binding affinity both greatly increases telomere length and leads to heterogeneous telomeric tracts at least 10 times longer than wild type (11). While this implies that Rap1p puts the brakes on telomere elongation, how the protein accomplishes that was an unanswered question.

In an initial experiment, I observed that the number of telomeric repeats, regardless of their orientation, was regulated. I was then able to show that the number of repeats at an individual telomere was reduced when hybrid proteins containing the Rap1p carboxyl terminus were targeted there by a heterologous DNA-binding domain. The extent of this telomere tract shortening was proportional to the number of targeted molecules (12). This indicates that a sensing mechanism can discriminate the precise number of Rap1p molecules bound to the chromosome end. A threshold number of Rap1p molecules at a telomere could therefore be the signal that prevents elongation.

In the process of studying the return to equilibrium of an abnormally shortened telomere in a wild type context, I recently showed that telomere elongation is limited to a few base pairs per generation and progressively inhibited with increasing telomere length (13). This suggests that each individual Rap1p molecule has an additive, probably equivalent, inhibitory effect, up to a number of molecules (i.e., up to a length) sufficient to bring down telomerase activity to a level precisely balancing the constant shortening due to replication.

At a molecular level, Rap1p action on telomerase is still mysterious, but we do know its action is unlikely to be direct. Two factors interacting with Rap1p--Rif1p and Rif2p--are required for telomere length regulation (14). Cdc13p, a telomeric single-strand DNA binding protein, and Stn1p, a Cdc13p-interacting protein, can also negatively regulate telomere elongation, suggesting that the inhibition of telomerase by Rap1p could be mediated by these proteins (15). The inhibitory signal could reduce either the number of bases that an individual telomerase can add, or the likelihood that an enzyme acts.

Yeast telomeres are considered to be heterochromatin-like regions because they confer a generalized transcriptional repression that occurs in a variegated or stochastic fashion (16, 17). This position-effect, also called silencing, requires the same Rap1p carboxy-terminal domain involved in telomere length regulation, as well as two histone-binding factors, Sir3p and Sir4p, that interact with this domain. This interaction seems to explain Rap1p's role in silencing because direct recruitment of the Sir proteins by a heterologous DNA-binding domain is sufficient to initiate stable silencing (18). Remarkably, Rap1p molecules interacting with Sir3p and Sir4p seem to be excluded from the length sensing mechanism, suggesting the existence of two distinct regions within a telomere: one involved in silencing and associated with Sir3p and Sir4p, and a Sir-free region that is kept constant in size and requires Rif1p and Rif2p (5, 12).

The function of transcriptional silencing at telomeres has remained elusive. I and others proposed that telomeres serve a regulatory role, acting as a molecular sink for the silencing factors that prevents indiscriminate gene silencing throughout the genome and maintains a silencing compartment within the yeast nucleus (18, 19). Recent studies have suggested that aging, as defined by the life span of mother cells, is in part regulated by the sequestration of Sir proteins at telomeres, strengthening the physiological significance of this model (20).

At telomeres, regulatory processes seem to arise from cooperation and/or competition between multiple factors. This is not peculiar to budding yeast. Recent evidence in fission yeast and in human cells suggests that telomere length regulation also involves a negative regulation of telomerase by a double-stranded telomeric DNA binding protein (21). Studies of heterochromatin in higher eukaryotes also reveal the importance of short- and long-range interactions both in cis and in trans for the establishment of silent domains (16, 22). Finally, the relatively tractable mechanisms of yeast telomeres may provide general insights that will help us understand the regulation of other chromosomal elements, such as replication origins, centromeres and complex transcriptional promoters.

References

1. B. McClintock, Genetics 25, 542 (1940); J.E. Haber and P. C. Thorburn, Genetics 106, 207 (1984).
2. L.L. Sandell and V. A. Zakian, Cell 75, 729 (1993); B. van Steensel, A. Smogorzewska, T. de Lange, Cell 92, 401 (1998).
3. J. D. Watson, Nature (New Biol.) 239, 197 (1972); J. Lingner, J.P. Cooper, T. R. Cech, Science 269, 1533 (1995).
4. V. A. Zakian, Science 270, 1601 (1995).
5. C. Brun, S. Marcand, E. Gilson, Trends Cell Biol. 7, 317 (1997).
6. J.H. Wright, D.E. Gottschling, V. A. Zakian, Genes Dev. 6, 197 (1992).
7. C.W. Greider and E. H. Blackburn, Cell 51, 887 (1987); M.S. Singer and D. E. Gottschling, Science 266, 404 (1994).
8. T.M. Nakamura and T. R. Cech, Cell 92, 587 (1998); C.I. Nugent and V. Lundblad, Genes Dev. 12, 1073 (1998).
9. A. G.Bodnar, et al., Science 279, 289 (1998).
10. A.W. Murray, T.E. Claus, J. W. Szostak, Molec. Cell. Biol. 8, 4642 (1988).
11. G. Kyrion, K.A. Boakye, A. J. Lustig, Molec. Cell. Biol. 12, 5159 (1992); M.J. McEachern and E. H. Blackburn, Nature 376, 403 (1995); A. Krauskopf and E. H. Blackburn, Nature 383, 354 (1996).
12. S. Marcand, E. Gilson, D. Shore, Science 275, 986 (1997).
13. S. Marcand, V. Brevet E. Gilson, submitted
14. D. Wotton and D. Shore, Genes Dev. 11, 748 (1997).
15. C.I. Nugent, T.R. Hughes, N.F. Lue, V. Lundblad, Science 274, 249 (1996); N. Grandin, S.I. Reed, M. Charbonneau, Genes Dev. 11, 512 (1997).
16. S. Marcand, M. S. Gasser, E. Gilson, Curr. Biol. 6, 1222 (1996).
17. A. Lustig, Curr. Opin. Genet. Dev. 8, 233 (1998).
18. S. Marcand, S.W. Buck, P. Moretti, E. Gilson, D. Shore, Genes Dev. 10, 1297 (1996); A.J.C. Lustig, C. Liu, C. Zhang, J. P. Hanish, Mol. Cell. Biol. 16, 2483 (1996).
19. L. Maillet et al., Genes Dev. 10, 1796 (1996); S.W. Buck and D. Shore, Genes Dev. 9, 370 (1995).
20. N. R. Austriaco and P. Guarente, Proc. Natl. Acad. Sci. U.S.A. 94, 9768 (1997).
21. J. P. Cooper, E. R. Nimmo, R. C. Allshire, T. R. Cech, Nature 385, 744 (1997); B. van Steensel and T. de Lange, Nature 385, 740 (1997).
22. L. L. Wallrath, Curr. Opin. Genet. Dev. 8, 147 (1998).

To Advertise   |   Find Products