Sci. Aging Knowl. Environ., 9 January 2002
Vol. 2002, Issue 1, p. pe1
[DOI: 10.1126/sageke.2002.1.pe1]


Cancer and Aging: Yin, Yang, and p53

Judith Campisi

The author is at the Lawrence Berkeley National Laboratory, MS 84-171, 1 Cyclotron Road, Berkeley, CA 94720, USA. E-mail: jcampisi{at};2002/1/pe1

Key Words: p53 • apoptosis • senescence • premature aging • cancer • Werner

p53 has been intensely studied for more than 20 years (1). First identified as a tumor-associated protein in mice, p53 was considered to be an oncogene for nearly a decade. We now know that p53 is a potent suppressor of tumorigenesis in mammals, and that most malignant tumors harbor mutations that either inactivate or confer new properties (for example, the ability to interact with novel proteins) on p53 or one of its regulators. p53 is a multifunctional transcriptional regulator that causes cells to arrest growth or die when stressed or damaged (2). Studies of p53 have provided mechanistic insights into many biological processes, including cell cycle control, cellular senescence, apoptosis, DNA repair, and transcription. They also have offered a wealth of knowledge about how organisms with renewable tissues suppress the development of cancer. What more can p53 tell us about biology?

In the 3 January 2002 issue of the journal Nature (3), Tyner, Venkatachalam, and colleagues report on a remarkable new facet of p53 action that hints at how and why some aging phenotypes arise. Tyner et al. took advantage of cultured embryonic stem cells in which a serendipitous recombination event deleted part of a p53 gene. The researchers used these cells to produce mice that carry one mutant (m) and one wild-type (+) p53 allele. p53+/m mice appeared to be identical to wild-type mice for about a year after birth. By 2 years of age, however, the p53+/m mice had a strikingly lower incidence of cancer: <6% versus >45% in p53+/+ mice! The mutant allele is thought to produce a truncated protein that stabilizes wild-type p53 and augments its ability to activate transcription. Thus, p53 is apparently hyperactive in p53+/m mice. It is not surprising, then, that these mice are highly resistant to spontaneous tumorigenesis.

Cancer is a major cause of death in mice. What then becomes of p53+/m mice? Are they healthier and longer-lived than normal mice? Not at all! Despite the paucity of cancer, p53+/m mice prematurely developed several phenotypes associated with aging. These included osteoporosis, muscle and skin atrophy, impaired wound healing and stress tolerance, hematopoietic stem cell depletion, and reduced body mass. In contrast, some aging phenotypes did not appear prematurely in p53+/m mice; for example, intestinal atrophy, skin ulcers, joint diseases, and cataracts. Finally, median and maximum life-spans were ~20% shorter in p53+/m mice, as compared to wild-type mice (96 versus 118 weeks, and 136 versus 164 weeks), even though p53+/m mice typically died with no obvious signs of disease (the cause of death was often difficult to determine). Together, the results suggest that, surprisingly, a boosted version of p53 can trigger selected aspects of premature aging, even though it reduces cancer as expected.

There are two caveats to consider in light of the new data. First, mutant p53 protein was undetectable in p53+/m cells, although the mutant transcript was detected in vivo and produced the expected protein in vitro. Presumably, mutant protein is expressed at levels that are very low, yet sufficient to enhance wild-type p53 activity. Still, it is possible that the p53+/m phenotypes are not due to the mutant p53 allele. Second, the recombination event deleted p53 exons 1 through 6, as well as an undefined upstream region. Thus, the observed p53+/m phenotypes could result from the loss of a gene or genes upstream of the p53 gene. However, additional results from Tyner et al. argue against this possibility. They engineered transgenic mice that carried ~20 copies of a gene (pL53) that encoded a temperature-sensitive (ts) version of p53. At 37°C, some of the ts protein is presumed to adopt a wild-type conformation, and the transgenes almost certainly are expressed to varying extents in different tissues. Thus, at least some cells in these transgenic mice are presumed to have a modest increase in p53 activity. Consistent with the hypothesis that hyperactive p53 causes the p53+/m phenotypes, pL53 mice were modestly protected from cancer and showed some, although not all, of the premature aging phenotypes shown by p53+/m mice.

But caveats aside, the remarkable findings of Tyner et al. suggest that hyperactive p53 causes premature aging in certain mammalian tissues. This observation raises a host of questions and possibilities. For example, why might hyperactive p53 cause premature aging? One possibility is that p53 is antagonistically pleiotropic, a concept that stems from the evolutionary hypothesis of aging. The theory of antagonistic pleiotropy predicts that some genes that were selected to ensure the health of young organisms (by preventing cancer, for example) can have unselected deleterious effects in aged organisms. Thus, although p53 was selected to suppress tumorigenesis, it might have harmful effects that escaped natural selection because they manifest late in life (see Williams Classic Paper). As such, p53 may be optimized to balance the suppression of tumorigenesis against the appearance of selected aging phenotypes (Fig. 1). This hypothesis could explain why hyperactive p53 causes premature aging, despite conferring extraordinary protection against cancer. It also raises the possibility that normal p53 contributes to aging, and suggests that some aging phenotypes might be delayed in p53-/- or p53+/- mice, if it were possible to prevent them from dying of cancer.

View larger version (21K):
[in this window]
[in a new window]
Fig. 1. In organisms with renewable tissues, p53 activity may be optimally balanced to prevent the development of cancer as well as the premature occurence of aging phenotypes. In p53+/m mice, hyperactive p53 provides extraordinary protection against cancer at the expense of early development of selected aging phenotypes (cancer incidence is low and aging phenotypes are high). In p53-/- or p53+/- mice, a deficiency in p53 activity results in the early development of cancer (cancer incidence is high). If these mice did not die of cancer, they might show delayed aging in some tissues. The fact that this is only a hypothesis is depicted with a question mark.

How might p53 activity cause aging? p53 causes cell cycle arrest or apoptosis, depending on the type of cell, stress, or damage. The cell cycle arrest can either be transient, during which time damage is repaired, or permanent. The permanent arrest, termed cellular senescence, is crucial for preventing cancer but may itself be antagonistically pleiotropic, because senescent cells secrete factors that can disrupt tissue structure and function (4). Similarly, apoptosis is crucial for preventing cancer but may contribute to degenerative aging pathologies by irreversibly depleting cells that keep tissues intact and functional (5). Hyperactive p53 might cause dysfunctional or damaged cells to die or senesce, rather than transiently arrest their growth. Excessive apoptosis or senescence could explain many of the p53+/m phenotypes. However, because senescent cells are known to create a tissue environment that promotes tumorigenesis (4), p53-mediated apoptosis may be more important than cellular senescence in causing the p53+/m aging phenotypes.

Does normal p53 activity eventually lead to aging phenotypes in wild-type mice? Can the aging phenotypes seen in p53+/m mice arise by p53-independent mechanisms? And do other tumor suppressors prevent cancer at the cost of contributing to aging, and vice versa? In light of the findings of Tyner et al., some recent findings, discussed below, might provide a starting point for answering these questions.

(1) p21 is a p53-inducible cell cycle inhibitor. When an expression vector that generates levels of p21 similar to those induced by p53 is introduced into cells in culture, p21 stimulates (by unknown mechanisms) the expression of many secreted proteins (for example, inflammatory cytokines, {beta}-amyloid, and proteases) that have been implicated in aging (6). Interestingly, p21 expression can be induced by both p53-dependent and p53-independent mechanisms in cells. Taken together, these findings suggest that p21 might be at least partly responsible for the aging phenotypes that Tyner et al. attribute to hyperactive p53 and might also mediate aging phenotypes independent of p53. Not surprising, p21-/- mice are cancer-prone, although less so than p53-/- mice (7).

(2) The Sir2 protein is a nicotinamide adenine dinucleotide (NAD)-dependent histone deacetylase that maintains chromatin-mediated gene silencing. Overexpression of Sir2 increases yeast replicative life-span and extends the life-span of the postmitotic multicellular nematode Caenorhabditis elegans. Might Sir2 overexpression extend life-span in mammals? Perhaps not. A mammalian version of Sir2 (Sir2-alpha) was recently shown to negatively regulate p53 by deacetylating it (8, 9). If mammalian p53 is optimized to balance cancer and aging phenotypes, as the new work suggests, Sir2 overexpression might delay aging in some tissues but also increase cancer incidence, both owing to reduced p53 activity.

(3) A deficiency of WRN--a DNA helicase and exonuclease--causes Werner syndrome (WS) in humans (see "Of Hyperaging and Methuselah Genes"). WS individuals are cancer-prone but also prematurely display some of the aging phenotypes shown by p53+/m mice. WRN and p53 interact physically, and p53 suppresses WRN exonuclease activity (10). Might hyperactive p53 cause aging by oversuppressing WRN in some cells? WRN also appears to facilitate the induction of p53 by DNA damage (11) and p53-mediated apoptosis (12). Decreased induction of p53 by WRN might contribute to the increased incidence of cancer seen in WS patients. And a reduction in p53-mediated apoptosis might lead to an increase in p53-mediated senescence and thus explain some of the aging phenotypes in WS.

Mammals evolved multiple strategies to suppress the development of cancer. Thus, it is difficult to predict how subtle differences in p53 or its regulators might influence longevity in mammals. However, a polymorphism (Arg72) that predisposes p53 to degradation and is associated with increased cancer risk is as prevalent in centenarians as in young people (13). Do subtly reduced p53 levels contribute to the longevity of these centenarians (who presumably resist cancer via other tumor suppressor pathways)? Whatever the case, it now appears that cancer and aging may be linked by common processes and molecules, and we have p53 to thank for the insight.

January 9, 2002

Suggested ReadingBack to Top

  1. T. K. Maclahlan, R. D. Meng, W. S. El-Deiry. p53: Twenty years on. J. Cell. Physiol. 181, 371-374 (1999). [CrossRef][Medline]
  2. K. M. Ryan, A. C. Phillips, K. H. Voudsen. Regulation and function of the p53 tumor suppressor protein. Curr. Opin. Cell Biol. 13, 332-337 (2001). [CrossRef][Medline]
  3. S. D. Tyner, S. Venkatachalam, J. Choi, S. Jones, N. Ghebranious, H. Ingelmann, X. Lu, G. Soron, B. Cooper, C. Brayton, S. H. Park, T. Thompson, G. Karsenty, A. Bradley, L. Donehower. p53 mutant mice that display early aging-associated phenotypes. Nature, 415, 45-53 (2001).
  4. J. Campisi. Cellular senescence as a tumor suppressor mechanism. Trends Cell Biol. 11, S27-31 (2001). [CrossRef][Medline]
  5. A. M. Joaquin, S. Gollapudi. Functional decline in aging and disease: a role for apoptosis. J. Am. Geriatr. Soc. 4B9: 1234-40 (2001).
  6. B. D. Chang, K. Watanabe, E. V. Broude, J. Fang, J. C. Poole, T. V. Kalinichenko, I. B. Roninson. Effects of p21/Waf1/Cip1/Sdi1 on cellular gene expression: implications for carcinogenesis, senescence, and age-related diseases. Proc. Natl. Acad. Sci. U.S.A. 97, 4291-4296 (2000). [Abstract/Free Full Text]
  7. J. Martin-Caballero, J. M. Flores, P. Garcia-Palencia, M. Serrano. Tumor susceptibility of p21(Waf1/Cip1)-deficient mice. Cancer Res. 61, 6234-6238 (2001). [Abstract/Free Full Text]
  8. J. Luo, A. Y. Nikolaev, S. Imai, D. Chen, F. Su, A. Shiloh, L. Guarente, W. Gu. Negative control of p53 by Sir2-alpha promotes cell survival under stress. Cell 107, 137-148 (2001). [CrossRef][Medline]
  9. H. Vaziri, S. K. Dessain, E. N. Eaton, S. I. Imai, R. A. Frye, T. K. Pandita, L. Guarente, R. A. Weinberg. hSIR2 (SIRT1) functions as an NAD-dependent p53 deacetylase. Cell 107, 149-159 (2001). [CrossRef][Medline]
  10. R. M. Brosh, P. Karmakar, J. A. Sommers, Q. Yang, X. W. Wang, E. A. Spillare, C. C. Harris, V. A. Bohr. p53 modulates the exonuclease activity of Werner syndrome protein. J. Biol. Chem. 276, 35093-35102 (2001). [Abstract/Free Full Text]
  11. G. Blander, N. Zalle, J. F. M. Leal, R. L. Bar-Or, C. E. Yu, M. Oren. The Werner syndrome protein contributes to induction of p53 by DNA damage. FASEB J. 14, 2138-2140 (2000). [Free Full Text]
  12. E. A. Spillare, A. I. Robles, X. W. Wang, J. C. Shen, C. E. Yu, G. D. Schellenberg, C. C. Harris. p53-mediated apoptosis is attenuated in Werner syndrome cells. Genes Dev. 13, 1355-1360 (1999). [Abstract/Free Full Text]
  13. M. Bonafe, F. Olivieri, D. Mari, G. Baggio, R. Mattace, P. Sansoni, G. De Benedictis, M. De Luca, S. Bertolini, C. Barbi, D. Monti, C. Franceschi. p53 variants predisposing to cancer are present in healthy centenarians. Am. J. Hum. Genet. 64, 292-295 (1999).[CrossRef][Medline]

Absence of Cardiotrophin 1 Is Associated With Decreased Age-Dependent Arterial Stiffness and Increased Longevity in Mice.
N. Lopez-Andres, L. Calvier, C. Labat, R. Fay, J. Diez, A. Benetos, F. Zannad, P. Lacolley, and P. Rossignol (2013)
61, 120-129
   Abstract »    Full Text »    PDF »
Pathologies Associated with the p53 Response.
A. V. Gudkov and E. A. Komarova (2010)
Cold Spring Harb Perspect Biol 2, a001180
   Abstract »    Full Text »    PDF »
Oxidative Stress Increases the Number of Stress Granules in Senescent Cells and Triggers a Rapid Decrease in p21waf1/cip1 Translation.
X. J. Lian and I.-E. Gallouzi (2009)
J. Biol. Chem. 284, 8877-8887
   Abstract »    Full Text »    PDF »

Science of Aging Knowledge Environment. ISSN 1539-6150