SEARCH     advanced search my account e-alerts subscribe register
Journal home
Advance online publication
Current issue
Press releases
For authors
For referees
Contact editorial office
About the journal
For librarians
Contact NPG
Customer services
Site features
NPG Subject areas
Access material from all our publications in your subject area:
Biotechnology Biotechnology
Cancer Cancer
Chemistry Chemistry
Dentistry Dentistry
Development Development
Drug Discovery Drug Discovery
Earth Sciences Earth Sciences
Evolution & Ecology Evolution & Ecology
Genetics Genetics
Immunology Immunology
Materials Materials Science
Medical Research Medical Research
Microbiology Microbiology
Molecular Cell Biology Molecular Cell Biology
Neuroscience Neuroscience
Pharmacology Pharmacology
Physics Physics
Browse all publications
21 January 2002, Volume 21, Number 4, Pages 627-630
Table of contents    Previous  Article  Next   [PDF]
Senescence: does it all happen at the ends?
Sheila A Stewart1 and Robert A Weinberg1,2

1Whitehead Institute for Biomedical Research, Cambridge, Massachusetts, MA 02142, USA

2Department of Biology, Massachusetts Institute of Technology, Cambridge, Massachusetts, MA 02139, USA

Correspondence to: R A Weinberg, Department of Biology, Massachusetts Institute of Technology, Cambridge, MA 02139, USA; E-mail: weinberg@wi.mit.edu


Over 60 years ago Barbara McClintock described the telomere and suggested that it protected the chromosome from illegitimate or end-to-end fusion, thus functioning to protect the genome. Since that time we have discovered that the telomere is a complex structure composed of both DNA and a growing list of associated proteins that together serve to regulate the length of the telomere and, as predicted by McClintock, protect genomic integrity. In addition to its protective role, the telomere has also been hypothesized to serve as a molecular clock that tallies the number of cell divisions and limits further divisions at a predetermined point. However, the precise role of telomeres in predicting and limiting cellular lifespan remains a matter of much debate. In this review, we highlight some of the salient points of basic telomere biology and relate them to the current controversies surrounding the role of telomeres and telomerase in cellular senescence.

Oncogene (2002) 21, 627-630 DOI: 10.1038/sj/onc/1205062


telomere; senescence; telomerase; capping; immortality

Cellular lifespan: a lesson from cancer biology

Normal human cells grown in vitro replicate for a limited period of time before entering senescence, which is currently defined as irreversible growth arrest (Hayflick and Moorhead, 1961). Hayflick first described this limited replicative potential and later others proposed that this limit was genetically defined (Harley et al., 1990, 1992). In many recent experiments, the molecular machinery responsible for determining replicative potential has been associated with the telomeres. As is now well documented, during successive cellular divisions the telomeres in normal human cells shorten progressively. This erosion of telomere length has been attributed to the inability of the general DNA replication machinery to completely replicate the very ends of the chromosomes (Greider and Blackburn, 1996). The observed shortening led in turn to the hypothesis that telomeres function as a 'molecular clock' (Harley et al., 1992). Thus, once the telomeres erode down below a threshold length, the senescence phenotype is provoked via a poorly understood mechanism.

Cancer cells, in stark contrast, proliferate indefinitely and are therefore considered to be immortal (Shay, 1997). (More precisely, it is the lineage of a cancer cell rather than the individual cell that is immortal.) Immortal cells, whether malignant or not, have been found to maintain telomeres at a stable length (Greider and Blackburn, 1996). The central role of telomeres in immortality was formally demonstrated when the enzyme that maintains telomeres, telomerase, was inhibited in telomerase-positive cells (Hahn et al., 1999; Zhang et al., 1999). This inhibition led to progressive telomere shortening which resulted subsequently in chromosome-to-chromosome fusions and cell death. In the opposite type of experiment, cells that were destined to enter into senescence or crisis have been immortalized through the ectopic expression of telomerase (Bodnar et al., 1998; Rufer et al., 2001; Vaziri and Benchimol, 1998; Yang et al., 1999). These observations indicated that telomeres are indeed important determinants of the replicative potential of cells.

Senescence: a limit to cellular proliferation

Cells that undergo replicative senescence fail to divide further but remain viable. In the case of human fibroblasts, pre-senescent cells can be induced to bypass senescence by the introduction of viral oncogenes such as the SV40 Large T antigen or the human papillomavirus E6 and E7 oncogenes (Counter et al., 1992; Shay et al., 1991a). Introduction of these oncogenes leads to the functional inactivation of the tumor suppressor proteins p53 and Rb, both of which must be inactivated in order for human cells to circumvent senescence (Shay et al., 1991b). Once cells bypass senescence they continue to divide and experience further shortening of their telomeres. Cells eventually reach a second proliferative block often referred to as crisis, which is characterized by genomic instability and massive cell death. Rare variants, appearing at a frequency of less than one cell in 107, emerge from a population of cells in crisis (Counter et al., 1992; Wright et al., 1989). These variants have acquired the ability to maintain their telomeres at a stable length and spawn a clone of immortalized descendants.

This scenario depicts two distinct barriers - senescence and crisis - which stand in the way of unlimited proliferative potential. These barriers have been proposed to function as anti-neoplastic mechanisms in vivo, doing so through their ability to confine mutant, pre-neoplastic cell clones to a limited number of replicative divisions. In truth, the in vivo counterparts of these replicative barriers remain poorly defined. In the case of crisis, a close molecular connection can be drawn between telomere shortening, cell death and the avoidance of both by telomerase. However, senescence presents a more complex mechanism whose precise connections with telomere shortening remain unclear. This review focuses on the mechanistic connections between senescence and telomere function.

If telomeres were the only determinants of senescence, then one could predict that the introduction of telomerase into any cell should result in the immortalization of that cell. This prediction became testable after hTERT, the gene encoding the catalytic component of the human telomerase holoenzyme, was cloned. Literature reports indicate that to date telomerase has been introduced into retinal pigment epithelial cells, fibroblasts, mesothelial cells, endothelial cells of both large vessel and microvessel origin, CD8-positive T cells, mammary epithelial cells, keratinocytes, and osteoblasts (Bodnar et al., 1998; Dickson et al., 2000; Ramirez et al., 2001; Rufer et al., 2001; Vaziri and Benchimol, 1998; Yang et al., 1999; Yudoh et al., 2001). Following introduction of hTERT most cell types, including retinal pigment epithelial cells, fibroblasts, mesothelial cells, endothelial cells, CD8-positive T cells, and osteoblasts, were immortalized. However, some types of epithelial cells gave quite different results.

Several studies have reported that mammary epithelial cells and keratinocytes can not be immortalized by hTERT alone (Dickson et al., 2000; Kiyono et al., 1998). Instead, in these studies it was reported that the additional inactivation of the p16INK4A tumor suppressor gene was required to immortalize these cells, calling into question the role of telomere length as the sole determinant of replicative senescence. Yet another report indicated that both mammary epithelial cells and keratinocytes could be immortalized with hTERT alone if adequate growth conditions were supplied in vitro (Ramirez et al., 2001). For example, when keratinocytes were maintained in culture dishes above cell feeder layers, these cells would proliferate indefinitely if they had also acquired the hTERT enzyme. Even these observations remain controversial, however, since others have reported that the continuous growth of keratinocytes even on feeder layers required the additional loss of p16INK4A (Dickson et al., 2000).

Still other reports call into question the role of telomeres as the exclusive determinant of cellular senescence. Introduction of oncogenes such as H-ras and raf result in induction of a senescent phenotype that is indistinguishable from that observed in cells that have undergone replicative senescence (Serrano et al., 1997; Zhu et al., 1998). In the case of H-ras, the additional expression of telomerase did not protect these cells from the onset of senescence (Wei and Sedivy, 1999). In some of these cases, analysis of telomere length indicated that the telomeres were still sufficiently long to protect the chromosomal ends and permit continued proliferation (Wei and Sedivy, 1999) (unpublished results). These conflicting observations have led to confusion at two levels. It remains unclear whether there is a single or multiple types of senescence. In addition, the precise role of telomere length in triggering senescence is also poorly resolved.

Senescence: the role of length versus structure

McClintock and later Muller in maize and flies, respectively, originally proposed that the telomere formed a protective cap on the chromosome that functioned to protect it from chromosomal fusions. Recent studies suggest that the telomere may have a higher order structure that serves this function. The contribution of both DNA and the various telomere-binding proteins to this structure is just beginning to be appreciated. Mammalian telomeres consist of a long stretch of TTAGGG repeats and termini consisting of 150 to 300 bases of single-strand DNA (Blackburn, 2001). In addition, a variety of telomere-binding proteins are associated with the telomere. Electron microscopic studies have suggested that in eukaryotic cells including human cells the telomere ends in a complex structure referred to as a T loop (Griffith et al., 1999; Munoz-Jordan et al., 2001). In this structure, the telomere turns back on itself and the single strand forms a displacement loop by inserting into the double helix. The presence of the T loop, at least in in vitro studies, appears to be dependent on the presence of the telomere-binding protein TRF2 (Griffith et al., 1999). Loss of TRF2 resulted in collapse of the T loop in vitro and inhibition of TRF2 binding in vivo resulted in chromosomal fusions and cell death (Karlseder et al., 1999), suggesting that if the T loop were to exist in vivo, its loss would have catastrophic effects on genomic stability and cell survival.

One major surprise in this list of telomere-capping proteins has come from the inclusion of the telomerase enzyme itself. Blackburn and colleagues reported that introduction of the wild-type telomerase gene into telomerase-negative, precrisis fibroblasts resulted in immortalization while control cells experienced telomere dysfunction and cell death (Zhu et al., 1999). The telomeres in the immortalized cells continued to shorten beyond those present in the control cells that had entered crisis, suggesting that telomerase allowed cells to survive with telomeres that were shorter than those normally required to protect chromosomal ends. Whether telomerase participates directly in this process by assembling into a multi-protein complex that directly protects the chromosomal ends or merely promotes the formation of such a complex by other proteins is unclear at this time. Still, it is interesting to speculate that telomerase may have functions that go beyond the simple maintenance of telomere length as has been suggested by a number of studies (Fu et al., 1999; Holt et al., 1999).

Stated differently, the discovery of a complex molecular structure at the tips of telomeres has raised the possibility that there are two distinct molecular elements that may determine the proliferative capacity of cells: the overall length of the telomeres and, independent of this length, the specialized structures at its tip. The experiments describing the capping of telomeres by telomerase make it plausible that this enzyme is able to influence both of these molecular entities. Thus, it may cause overall telomere lengthening and, independent of this, exert some protective effect on the tips of the telomeres. These effects at the ends of the telomeres may be modulated by the various molecular components present in different cell lineages.

Further evidence that calls into question the role of overall telomere length in senescence comes from a variety of in vitro experiments. When cells from individuals suffering from progeria diseases such as Werner's and Bloom's syndromes are cultured in vitro they typically undergo senescence earlier and sometimes with longer telomeres than age-matched controls (Schulz et al., 1996). Likewise, cells obtained from ataxia telangiectasia (ATM) patients also undergo senescence earlier than age-matched controls (Shiloh et al., 1985). Each of these diseases display variable amounts of genomic instability that has been postulated to be responsible for the early onset of senescence. In addition to cells obtained from individuals with these diseases, cells treated with hydrogen peroxide and gamma radiation also display a senescent phenotype sometimes in the absence of telomere shortening (Chen et al., 2001; Suzuki et al., 2001). As is the case with the genetic diseases mentioned above, both hydrogen peroxide and gamma radiation cause damage throughout the genome in the form of single-strand breaks and double-strand breaks, respectively (Brozmanova, 2001; Khanna and Jackson, 2001; Norbury and Hickson, 2001). Interestingly much of this damage is thought to occur in the telomere and indeed, yeast genetically mutant for the Bloom, Werner, and ATM homologues display telomere defects (Greenwell et al., 1995; Johnson et al., 2001a; Watt et al., 1996). Together these observations further support the notion that telomere shortening is not an absolute requirement for the induction of senescence. In fact, studies using hydrogen peroxide and gamma radiation suggest that damage to the DNA and more likely damage that affects certain aspects of telomere structure without concomitantly causing overall telomere shortening may be sufficient to trigger a senescent response (Chen et al., 2001; Gajdusek et al., 2001; Suzuki et al., 2001; Toussaint et al., 2000).

Does senescence function as an anti-neoplastic mechanism in vivo?

Senescence has been postulated to serve as a tumor-suppressing mechanism that is responsible for limiting the replicative potential of pre-neoplastic cells. This notion, attractive in concept, remains to be proven. Senescence in one form or another certainly occurs in vitro, but the evidence that it mirrors a comparable process occurring in vivo remains elusive. Cells from older individuals do display on average shorter telomeres than do cells from younger donors (Friedrich et al., 2000; Harley et al., 1992). However, even cells obtained from centenarians contain telomeres that are sufficiently long to support further replication in vitro. (Mondello et al., 1999). Indeed, a multitude of alternative, non-telomere-based mechanisms can be invoked to explain the progressive loss of replicative potential of cells isolated from older individuals. For example, the DNA from older individuals contains more accumulated genetic damage, and some have suggested that this may be responsible for the reduced growth capacity of cells obtained from older donors (von Zglinicki et al., 2001; Walter et al., 1997). Other types of cumulative biochemical damage sustained by long-lived cell lineages may also contribute to reduced replicative potential in vitro.

The cell senescence observed in vitro following the introduction of oncogenes has been invoked as yet another indication of its possible in vivo relevance. Thus, such senescence has been portrayed as a mechanism established by the body to block the proliferation of cell clones that sustain mutations that can lead to oncogene activation and resulting cell transformation (Weinberg, 1997). While an attractive hypothesis, it should be noted that the levels of H-Ras required to induce senescence far exceed those observed in the cells of naturally arising tumors (Johnson et al., 2001b). As such, while the evidence for oncogene-induced senescence in vitro is robust and reproducible, its relevance to the tumorigenic processes operating in vivo is questionable.

It is clear that the maintenance of the telomere, both its overall length as well as its complex terminal structure, is critical to genomic integrity and plays a pivotal role in both in vitro and in vivo crisis. Senescence in vitro, however, may not have an in vivo counterpart. Instead, it may simply represent the unhappiness of cells growing under the suboptimal conditions found in culture, which result in the induction of some structural destabilization of the ends of telomeres. This destabilization, in turn, may be detected by molecular machines that are charged with surveying damage to the genome and provoking a senescence response. Alternatively, in the presence of telomerase, this destabilization of the telomere ends may be prevented by the presence of the telomerase enzyme. As a consequence, the normally occurring senescence response will not be triggered. Even if this or a similar mechanism were operative in cultured cells, it provides no reassurance that a similar mechanism operates in vivo, since the initial provoking stimulus for this cascade of events is likely to be inappropriate culture conditions which may not have an in vivo counterpart.

It is likely that as our knowledge of the absolute structure of the mammalian telomere grows and the various signal transduction cascades that impede on the telomere and cell cycle are uncovered, the question of what role senescence plays in vivo will be answered. Even when this issue is resolved, the further question of the role of telomere length in provoking senescence will remain a bone of contention and thus an object of much research.


We thank members of the Weinberg lab for helpful discussions. This work was supported in part by Merck/MIT and Co, Inc. (RA Weinberg), the US National Cancer Institute, NIH/NCI 5R01 CA78461 (RA Weinberg), G Harold and Leila Y Mathers Charitable Fund, a Charles E Culpeper Biomedical Pilot Initiative Grant (RA Weinberg), American Association for Cancer Research (SA Stewart). SA Stewart is a Herman and Margaret Sokol postdoctoral fellow. RA Weinberg is an American Cancer Society Research Professor and a Daniel K Ludwig Cancer Research Professor.


Blackburn EH. (2001). Cell, 106: 661-673. MEDLINE

Bodnar AG, Ouellette M, Frolkis M, Holt SE, Chiu CP, Morin GB, Harley CB, Shay JW, Lichtsteiner S, Wright WE. (1998). Science, 279: 349-352. Article MEDLINE

Brozmanova A. (2001). Neoplasma, 48: 85-93. MEDLINE

Chen QM, Prowse KR, Tu VC, Purdom S, Linskens MH. (2001). Exp. Cell. Res., 265: 294-303. Article MEDLINE

Counter CM, Avilion AA, LeFeuvre CE, Stewart NG, Greider CW, Harley CB, Bacchetti S. (1992). EMBO J., 11: 1921-1929. MEDLINE

Dickson MA, Hahn WC, Ino Y, Ronfard V, Wu JY, Weinberg RA, Louis DN, Li FP, Rheinwald JG. (2000). Mol. Cell. Biol., 20: 1436-1447. MEDLINE

Friedrich U, Griese E, Schwab M, Fritz P, Thon K, Klotz U. (2000). Mech. Ageing Dev., 119: 89-99. MEDLINE

Fu W, Begley JG, Killen MW, Mattson MP. (1999). J. Biol. Chem., 274: 7264-7271. Article MEDLINE

Gajdusek C, Onoda K, London S, Johnson M, Morrison R, Mayberg M. (2001). J. Cell. Physiol., 188: 8-23. Article MEDLINE

Greenwell PW, Kronmal SL, Porter SE, Gassenhuber J, Obermaier B, Petes TD. (1995). Cell, 82: 823-829. MEDLINE

Greider CW, Blackburn EH. (1996). Sci. Am., 274: 92-97. MEDLINE

Griffith JD, Comeau L, Rosenfield S, Stansel RM, Bianchi A, Moss H, de Lange T. (1999). Cell, 97: 503-514. MEDLINE

Hahn WC, Stewart SA, Brooks MW, York SG, Eaton E, Kurachi A, Beijersbergen RL, Knoll JH, Meyerson M, Weinberg RA. (1999). Nat. Med., 5: 1164-1170. Article MEDLINE

Harley CB, Futcher AB, Greider CW. (1990). Nature, 345: 458-460. MEDLINE

Harley CB, Vaziri H, Counter CM, Allsopp RC. (1992). Exp. Gerontol., 27: 375-382. MEDLINE

Hayflick L, Moorhead PS. (1961). Exp. Cell. Res., 25: 585-621.

Holt SE, Glinsky VV, Ivanova AB, Glinsky GV. (1999). Mol. Carcinog., 25: 241-248. Article MEDLINE

Johnson FB, Marciniak RA, McVey M, Stewart SA, Hahn WC, Guarente L. (2001a). EMBO J., 20: 905-913. MEDLINE

Johnson L, Mercer K, Greenbaum D, Bronson RT, Crowley D, Tuveson DA, Jacks T. (2001b). Nature, 410: 1111-1116. Article MEDLINE

Karlseder J, Broccoli D, Dai Y, Hardy S, de Lange T. (1999). Science, 283: 1321-1325. Article MEDLINE

Khanna KK, Jackson SP. (2001). Nat. Genet., 27: 247-254. Article MEDLINE

Kiyono T, Foster SA, Koop JI, McDougall JK, Galloway DA, Klingelhutz AJ. (1998). Nature, 396: 84-88. Article MEDLINE

Mondello C, Petropoulou C, Monti D, Gonos ES, Franceschi C, Nuzzo F. (1999). Exp Cell Res., 248: 234-242. Article MEDLINE

Munoz-Jordan JL, Cross GA, de Lange T, Griffith JD. (2001). EMBO J., 20: 579-588. MEDLINE

Norbury CJ, Hickson ID. (2001). Annu. Rev. Pharmacol. Toxicol., 41: 367-401. MEDLINE

Ramirez RD, Morales CP, Herbert BS, Rohde JM, Passons C, Shay JW, Wright WE. (2001). Genes Dev., 15: 398-403. Article MEDLINE

Rufer N, Migliaccio M, Antonchuk J, Humphries RK, Roosnek E, Lansdorp PM. (2001). Blood, 98: 597-603. Article MEDLINE

Schulz VP, Zakian VA, Ogburn CE, McKay J, Jarzebowicz AA, Edland SD, Martin GM. (1996). Hum. Genet., 97: 750-754. Article MEDLINE

Serrano M, Lin AW, McCurrach ME, Beach D, Lowe SW. (1997). Cell, 88: 593-602. Article MEDLINE

Shay JW. (1997). J. Cell. Physiol., 173: 266-270. Article MEDLINE

Shay JW, Pereira-Smith OM, Wright WE. (1991a). Exp. Cell. Res., 196: 33-39. MEDLINE

Shay M, Braester A, Cohen I. (1991b). Ann. Hematol., 63: 116-118.

Shiloh Y, Tabor E, Becker Y. (1985). Kroc. Found. Ser., 19: 111-121. MEDLINE

Suzuki K, Mori I, Nakayama Y, Miyakoda M, Kodama S, Watanabe M. (2001). Radiat. Res., 155: 248-253. MEDLINE

Toussaint O, Medrano EE, von Zglinicki T. (2000). Exp. Gerontol., 35: 927-945. MEDLINE

Vaziri H, Benchimol S. (1998). Curr. Biol., 8: 279-282. MEDLINE

von Zglinicki T, Burkle A, Kirkwood TB. (2001). Exp. Gerontol., 36: 1049-1062. MEDLINE

Walter CA, Grabowski DT, Street KA, Conrad CC, Richardson A. (1997). Mech. Ageing Dev., 98: 203-222. MEDLINE

Watt PM, Hickson ID, Borts RH, Louis EJ. (1996). Genetics, 144: 935-945. MEDLINE

Wei S, Sedivy JM. (1999). Cancer Res., 59: 1539-1543. MEDLINE

Weinberg RA. (1997). Cell, 88: 573-575. Article MEDLINE

Wright WE, Pereira-Smith OM, Shay JW. (1989). Mol. Cell. Biol., 9: 3088-3092. MEDLINE

Yang J, Chang E, Cherry AM, Bangs CD, Oei Y, Bodnar A, Bronstein A, Chiu CP, Herron GS. (1999). J. Biol. Chem., 274: 26141-26148. Article MEDLINE

Yudoh K, Matsuno H, Nakazawa F, Katayama R, Kimura T. (2001). J. Bone Miner. Res., 16: 1453-1464. MEDLINE

Zhang X, Mar V, Zhou W, Harrington L, Robinson MO. (1999). Genes Dev., 13: 2388-2399. Article MEDLINE

Zhu J, Wang H, Bishop JM, Blackburn EH. (1999). Proc. Natl. Acad. Sci. USA, 96: 3723-3728. Article MEDLINE

Zhu J, Woods D, McMahon M, Bishop JM. (1998). Genes Dev., 12: 2997-3007. MEDLINE

21 January 2002, Volume 21, Number 4, Pages 627-630
Table of contents    Previous  Article  Next    [PDF]