Innovita Research Foundation

I.R.F. / Aging news / General / 04021001

Telomeres and Cancer
Posted on: February 10, 2004

Every day, we sacrifice many varied cell types such as granulocytes, keratinocytes, hepatocytes, and erythrocytes at the altar of organismal homeostasis. For the individual to thrive, lost cells must be constantly replaced, and recent evidence has identified significant capacities for repair and regeneration even in organs once thought to be postmitotic such as the pancreatic islet and the brain. Given this continuous cellular attrition, normal tissue function requires that the rate of cell loss be matched by the rate of renewal. Aging is hastened by forces that either accelerate cellular loss or retard tissue repair. When loss exceeds repair, ensuing cellular attrition eventually leads to a decline in organ function and ultimately failure. When restricted to specific organs, this condition would be expected to result in one of the many chronic degenerative diseases such as liver cirrhosis. If, however, the process operates across multiple organ systems, then this progressive multisystem functional compromise may manifest clinically as frailty, accelerated aging, and death. Elegant experiments from lower metazoans with postmitotic soma (e.g., Drosophila and Caenorhabditis elegans) have identified many of the pathways that influence the rate of cellular turnover and loss. These data have firmly established a link between the rate of cellular metabolism, the rate of production of unstable oxygen species, and longevity in these species. There appears little doubt that many of these conserved pathways (e.g., insulin-like growth factor-1 signaling) also influence the rate of mammalian metabolism and cellular decay, and therefore mammalian aging. As extensive tissue replacement does not occur in adults of these lower organisms, however, these model systems have been less helpful with regard to the genes that regulate tissue repair in the adult and thereby influence this aspect of aging. Therefore, one advantage of mammalian genetic systems is that they permit the investigation of the pathways responsible for the repair half of the aging equation.
To be sure, adult mammals require extensive proliferation and tissue replacement to survive. Even in the absence of pathology, the intestinal lining replaces itself entirely on a weekly basis, and the bone marrow produces trillions of new blood cells daily. An obvious cost of this massive and obligate proliferation, however, is that even under physiological conditions, it is presumed that stem cell genomes are showered with somatic mutations, some of which may target cancer relevant genes. In accord with this view is the remarkable observation that roughly 1% of neonatal cord blood collections contain significant numbers of myeloid clones harboring oncogenic fusions such as the AML-ETO fusion associated with acute leukemia; similarly, as many as one in three adults possess detectable IgH-BCL2 translocations, which are commonly associated with follicular lymphoma. As the prevalence of these cancers is far lower in the general population, it would appear that potent tumor suppressor mechanisms function to monitor and con-strain the growth and survival of these aspiring cancer cells. In humans, three principal and overlapping tumor suppressor barriers appear to be operative; they are represented by the p16INK4a-retinoblastoma protein (p16INK4a-Rb) pathway, the ARF-p53 pathway, and telomeres. The combined effect of these tumor suppressor mechanisms is to place a limit on the replicative life span of cells in the compartment capable of contributing to tissue regeneration (hence termed stem cells).

Tumor suppressor pathways engender senescence and apoptosis
A common endpoint for these major tumor suppressor mechanisms is senescence. This specialized form of terminal differentiation is induced by a variety of stimuli including alterations of telomere length and structure, some forms of DNA damage (for example, oxidative stress), and activation of certain oncogenes. Senescence differs from other physiologic forms of cell cycle arrest such as quiescence in two important ways. First, senescence in somatic cells is generally irreversible, barring the inactivation of p53 and/or Rb, which appear to be required for its maintenance in certain settings. Second, it is associated with distinctive molecular and morphologic alterations such as cellular flattening and increased adherence, a loss of c-fos induction to serum stimulation, an increased expression of plasminogen activator inhibitor (PAI) and the expression of senescence-associated β-galactosidase activity. Recently, senescence has been shown to correlate with the establishment of an unusual form of heterochromatin present in discrete nuclear foci, known as senescence-associated heterochromatic foci (SAHF). In aggregate, these data suggest that senescence results from the durable repression of promoters associated with growth control genes. This repression is enforced by the construction of stable, heterochromatin-like complexes, the formation of which is directed in part by hypophosphorylated Rb. Recent evidence suggests that senescence may differ qualitatively depending on the stimulus that leads to its establishment. Several groups have shown that either Rb or p53 inactivation could reverse the senescence in murine cells or certain types of human fibroblasts. Therefore, it seems clear that persistent p53 and Rb function are required to maintain certain forms of senescence.
In addition to senescence, it is worth noting that cancer-related stimuli such as oncogene activation, DNA damage, and telomere shortening can also induce an entirely distinct anticancer mechanism, namely apoptosis (Fig. 1).

Fig. 1. DNA damage accumulates as the consequence of endogenous (telomere dysfunction, oxidative stress) or exogenous (oxidative stress, g-irradiation, UV light, and others) attacks. Damaged DNA activates checkpoint responses that are mediated by the p53 and p16-Rb pathways and that result in apoptosis or cellular senescence. If these events occur in stem/progenitor cells, tissue homeostasis is altered ‐ a phenomenon that might contribute to aging. If, instead, DNA mutations that inactivate these checkpoint pathways accumulate, then cancer can arise.

Apoptotic loss of progenitor cells in response to such stimuli has been clearly demonstrated in animal models; for example, mice with shortened, dysfunctional telomeres demonstrate increased apoptosis in germ cells of the testes and crypt cells of the intestine. In these systems, an increase in apoptosis correlates with tissue atrophy and other phenotypes associated with premature aging.
The role of p53 in mediating apoptosis is well documented and this activity seems its major anticancer function in certain animal models. Correspondingly, p53 loss greatly attenuates the apoptotic phenotype seen in proliferative organs in the setting of telomere dysfunction in animal models. While loss of p53 in these animals affords resistance to the effects of telomere dysfunction, these mice also demonstrate a marked increase in epithelial tumor formation, reinforcing the view that aging and cancer are closely linked in this model system. Therefore, telomere shortening and p53 activation modulate two potent anticancer mechanisms: senescence and apoptosis. While the molecular biology of this fate-decision is incompletely understood, the specific response appears to depend on many variables including cell type, genetic context, and proliferation state.

Telomeres, telomerase, and checkpoints
Telomeres are nucleoprotein complexes at the chromosome ends, which consist of many overhanging double-stranded TTAGGG repeats and associated telomere binding proteins. It has been that telomeres play a critical role in the maintenance of chromosomal integrity ‐ a fact that has since been confirmed in diverse model systems from yeast to plants to humans.
Several lines of evidence have established that telomeres adopt a complex secondary and tertiary structure that relies on DNA-DNA, DNA-protein, and protein-protein interactions. With regard to senescence, telomere structure appears at least as important to telomere function as absolute telomere length.
The progressive telomere shortening with each cell division eventually triggers an alteration in telomere structure. From the perspective of senescence and apoptosis, the most important result of this telomere dysfunction is that the deprotected telomere end becomes, for all intents, a double-strand break (DSB) of DNA. Classical DSBs potently induce p53, and telomere dysfunction in cultured human and mouse cells have been shown to induce p53-mediated senescence or other checkpoint responses depending on the cell type.
The relationship between telomere dysfunction and activation of the p16INK4a-Rb pathway is less clear. Therefore, like p53, p16INK4a appears to be induced by telomere-dependent and - independent stimuli. The accumulation of p16INK4a in many tissues is noted with aging in both mice and humans but whether this results from in vivo alterations of telomere structure remains to be established. Similarly, it is unclear if the major in vivo tumor suppressor function of p16INK4a in humans results as a response to telomere dysfunction or other stimuli.
In the absence of the p53- and p16INK4a-Rb-mediated checkpoints, cultured cells with dysfunctional telomeres continue to proliferate, entering a period of slow growth called "crisis" that is characterized by genomic instability. Clones emerging from crisis invariably either reactivate telomerase or the alternative lengthening of telomeres (ALT) mechanism. During crisis, the deprotected telomere ends in proliferating cells can be illegitimately fused through DNA repair mechanisms, ultimately leading to the generation of complex non-reciprocal translocations, a hallmark feature of adult solid tumors. In a tissue stem cell, telomere dysfunction would appear to have two outcomes depending on the integrity of checkpoint mechanisms: respond to checkpoint activation with senescence/apoptosis or proliferate and engender genomic instability. Evidence suggests that most cells facing this decision respond to checkpoints, producing a progressive diminution of stem cell reserve throughout the human lifespan. This decline in tissue regeneration capacity eventually manifests as aging. The rare cells that continue to proliferate in the setting of telomere dysfunction and emerge from crisis having accumulated additional, transforming mutations are equally problematic from the organismal perspective. These cells necessarily have incurred proliferative genetic lesions and inactivated tumor suppressor checkpoints, and appear well on their way down the path to full-fledged cancer.

Telomere-mediated checkpoints prevent cancer but contribute to aging
The vast majority of cells exhibiting telomere dysfunction undergo senescence or apoptosis, and disparate lines of evidence have convincingly shown that these processes are significant barriers to cancer formation. Reactivation of telomerase is one of the most commonly observed features of cancer seen in greater than 80% of all human tumors. Telomerase expression greatly enhances the transformation of human cells in vitro and tumorigenesis is inhibited by telomere shortening in animal models of cancer. In aggregate, these data suggest that telomere-induced checkpoint activation is a major in vivo tumor suppressor mechanism.
The tumor-promoting effects of telomerase activity appear to be more complex than initially believed. Certainly, tumor clones must resolve the telomere length problem to traverse crisis, and therefore telomerase reactivation or the development of ALT is a prerequisite of malignant growth. For this reason, inhibitors of telomerase have been considered promising therapeutic candidates for novel antineoplastic agents. Several lines of evidence, however, also suggest that telomerase activity may contribute to tumorigenesis in a manner independent of its telomere lengthening effects. For example, telomerase expression has been shown, independent of telomere length, to produce resistance to the antiproliferative signals of TGF-β in cultured mammary cells lacking p16INK4a. Additionally, animals over expressing telomerase are more prone to neoplasia than littermate wild-type animals even though telomere length per se does not appear to limit tumor growth in these models. Lastly, transformed ALT+ murine clones demonstrate further enhanced growth in a mouse model of metastasis after transduction with telomerase. These disparate lines of evidence suggest that the catalytic activity of telomerase contributes to malignant growth by influencing a cellular feature in addition to absolute telomere length, but the precise mechanism of these effects has remained elusive.
In addition to preventing cancer, however, telomere-mediated checkpoints appear to contribute to aging, and, as stated, mice with shortened, dysfunctional telomeres exhibit many characteristics of premature aging. A consideration of ataxia telangectasia (AT) is particularly intriguing in this regard. Humans with deficiency of the AT tumor suppressor (ATM) (ataxiatelangiectasia mutated kinase), which plays a role in DNA damage signaling, develop a progeroid syndrome in the setting of premature telomere shortening.
Further direct genetic evidence for a role of telomere dysfunction in human aging comes from the recent discovery that germline mutations of the telomerase complex cause the progeroid syndrome dyskeratosis congenita. Also, telomere shortening has been shown to precede the development of overt cirrhosis in patients with chronic hepatitis of various etiologies. In addition, several studies have demonstrated a relationship between telomere length in peripheral blood leukocytes (PBL) and the onset of certain diseases associated with aging. Such studies in nonneoplastic diseases have shown that PBL telomere lengths can provide predictive information on the risk of developing atherosclerosis and on overall mortality. In aggregate, these data indicate that although telomerase plays a clear role in malignant progression, telomere-induced checkpoints also contribute to certain aspects of human aging.

Telomere dysfunction contributes to cancer
As telomere-mediated checkpoints are no doubt major barriers to malignant progression in most would-be cancer cells, it therefore appears somewhat paradoxical that telomere dysfunction may also fuel tumorigenesis. Nonetheless, recent cytogenetic and molecular studies have provided strong evidence that telomere dynamics can contribute to genomic instability, particularly early in tumorigenesis. In this model, a transient period of telomere dysfunction contributes to carcinogenesis by engendering large numbers of genome-wide changes. This occurs through breakage-fusion-bridge cycles that result from the formation of dicentric chromosomes after inappropriate fusions of deprotected telomeres. This rapid reshuffling of the genetic deck produces rare cells with a threshold number of relevant changes to become full-fledged cancer.
The model of telomere dysfunction-driven carcinogenesis matches with the timing of telomerase activation and appearance of genomic changes during various stages of epithelial tumorigenesis. Correspondingly, the measurement of telomerase activity in corresponding preneoplastic lesions has demonstrated a consistent pattern with the ploidy and cytogenetic changes. There is widespread chromosomal instability early in neoplastic progression at a time when telomerase activity is low. As these cancers progress and reactivate telomerase or ALT, genomic instability continues at a moderate rate, with further mutations presumably resulting from non-telomere-based mechanisms. Additionally, it has recently been shown that the presence of short telomeres in PBLs is associated with increased risk for the development of carcinomas of the head and neck, kidney, bladder, and lung. While these correlative results can be interpreted in several ways, substantiation of the utility of PBL telomere lengths as a reliable surrogate marker for neoplasia would prove invaluable in assessing patient risk of later developing an epithelial malignancy.

How are cancer and aging linked in vivo?
The aforementioned data suggest that senescence is induced in cultured cells by a variety of telomere-dependent and telomere-independent mechanisms. Likewise, the in vivo effects of inactivation of p53, p16INK4a and telomerase and their hyper function have been determined in mice, confirming the hypotheses that these mechanisms link cancer and aging. A problem, however, from the use of germline knockout animals and broadly expressing transgenics strains is that it is not necessarily straightforward to determine which effects are cell-autonomous and which are not.
An additional conceptual problem stems from the observation that the expression of p53 and telomere shortening do not strictly correlate with the onset of tissue aging. While undoubtedly telomere shortening is seen in certain progeroid syndromes and in states characterized by chronic hyper proliferation such as cirrhosis and myelodysplasia, it also appears that telomere length is heterogeneous in the human population, and shorter lengths do not always correlate with aging. A caveat to this conclusion, however, is that these studies necessarily have analyzed mean telomere length, although it appears that checkpoint activation may occur in response to the shortest telomeres in the cell. Nonetheless, while telomere shortening can be seen immediately preceding organ failure in states of high cellular turnover, it may not be an obligate feature of all normal physiologic aging. This finding points to the existence of telomere-independent causes of aging, as is almost certainly the case in normal mice. There are two important caveats to this sort of analysis. First, only surviving cells are considered in these molecular characterizations of aged tissues. Therefore, one would not expect to detect telomere shortening and p53 expression in cell types where these stimuli are proapoptotic. Nonetheless, apoptotic loss of stem cells may play an important role in organismal aging. Additionally, it is possible that alterations in telomere structure, which can potently induce senescence and apoptosis occur in vivo. Therefore, alterations in telomere dynamics may contribute to a decline in tissue function even in the absence of an overall decrease in telomere length. Alternatively, p53 or p16INK4a activation may be contributing to aging in a very restricted compartment such as tissue stem cells. In this model, expression of p16INK4a or p53 need not be detected throughout the tissue to exert its senescence-related effects.

Our new molecular understanding of cancer and aging is a major tool in this effort, for two reasons. First, we are beginning to develop molecular surrogate markers of aging and future cancer risk. Such markers will revolutionize medical advice regarding nutrition and wellness. Amazingly, physicians have very little hard evidence to support a given dose of vitamin supplements, exercise, or diet; and most wellness therapeutics of proven benefit are directed toward a measurable predisease state like hypercholesterolemia or hypertension. By examining markers such as telomere function or p53 / p16INK4a expression in a given tissue, however, we may be able to better predict the future onset of cancer and/or aging, and likewise determine the beneficial or harmful effects of a therapeutic intervention with regard to these surrogate endpoints. Secondly, we will be able to improve our understanding of environmental or lifestyle exposures that cause or contribute to aging. Likewise, we will be able to identify which tissue compartments are targeted by these exposures, allowing a delineation of the cell-autonomous and non-cell-autonomous effects. Whether such an understanding of the environmental triggers of aging will allow us to boost the maximum human lifespan is highly controversial, but certainly at the minimum such information would help extend the healthy life expectancy; free of cancer and other adverse consequences of aging.

Source 1: Norman E. Sharpless and Ronald A. DePinho, Telomeres, stem cells, senescence, and cancer J. Clin. Invest. 113:160-168 (2004).
Source 2: Pier Giuseppe Pelicci, Do tumor-suppressive mechanisms contribute to organism aging by inducing stem cell senescence? J. Clin. Invest. 113:4-7 (2004).
< Previous |  Next >