Mutagenesis, Vol. 17, No. 6, 539-550,
November 2002
© 2002 UK Environmental Mutagen Society/Oxford University Press
The significance of telomerase activation and cellular immortalization in human cancer
Brunel Institute of Cancer Genetics and Pharmacogenomics, Faculty of Life Sciences, Brunel University, Kingston Lane, Uxbridge, Middlesex UB8 3PH, UK
Email: robert.newbold{at}brunel.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionTelomerase: the...Mortality barriers to human...Recent progress in generating...Regulation of telomerase in...Origins of telomerase in...Rodent cell senescence and...Importance of telomerase...Alternative mechanisms of...Telomerase and cell immortality...ConclusionsReferences |
|---|
Repression of telomerase in the somatic tissues of humans, and probably other long-lived mammals, appears to have evolved as a powerful protective barrier against cancer. Immortalization in vitro of normal human cells that lack telomerase involves the reactivation of telomerase or, rarely, an alternative (ALT) mechanism for maintaining telomeres. Inactivation of the effectors of replicative senescence, i.e. genes encoding one or more elements of the p16/pRB and/or ARF/p53/p21 anti-proliferative pathways, is required for telomerase depression leading to immortalization. Regulation of telomerase in normal human cells is mediated primarily by transcriptional repression of hTERT, the gene encoding the catalytic subunit of telomerase. Rodent cells do not possess stringent controls on telomerase activity in the soma and this explains why they are so readily immortalized and transformed in culture compared with their human counterparts. Because active telomerase has been found to exist in the proliferative compartments of self-renewing tissues, it is not yet clear whether the telomerase present in 90% of human cancers exists as a consequence of selection of pre-existing telomerase-positive cells during carcinogenesis or through induction of hTERT expression in cells in which it is normally tightly repressed. In support of the latter, chromosome transfer techniques have revealed the presence of genes on normal human chromosomes that are able to extinguish hTERT transcription in cancer cells and induce them to undergo senescence. It is clear that telomerase is obligatory for continuous tumour cell proliferation, clonal evolution and malignant progression. Telomerase therefore represents an attractive target at which to aim new anti-cancer drugs. Results with a variety of telomerase inhibitory strategies in human cancer cells have confirmed that its functional inactivation results in progressive telomere shortening, leading to growth arrest and/or cell death through apoptosis. Promising candidate small molecule inhibitors are beginning to emerge that will form the basis for anti-telomerase drug development.
|
Introduction |
|---|
|
TopAbstractIntroductionTelomerase: the...Mortality barriers to human...Recent progress in generating...Regulation of telomerase in...Origins of telomerase in...Rodent cell senescence and...Importance of telomerase...Alternative mechanisms of...Telomerase and cell immortality...ConclusionsReferences |
|---|
The primary objective of fundamental cancer research is to identify highly specific molecular and cellular features of cancer cells that can be exploited as therapeutic targets. Early observations on the replicative potential of normal human cells explanted and grown as monolayers in tissue culture revealed that they possess an intrinsically programmed limit (known as the ‘Hayflick limit’) to their capacity for proliferation, usually after a substantial healthy period of growth (Hayflick and Moorhead, 1961
; Hayflick, 1965). In sharp contrast, cell cultures derived from the disaggregation of human cancer tissues, once established in vitro, are often immortal. Before the advent of sophisticated molecular biological techniques it was not possible to demonstrate convincingly that cellular immortalization is invariably associated with human carcinogenesis and it could be easily argued that the phenomenon represents a cell culture artefact. Furthermore, many human carcinomas (e.g. those developing from breast and prostate epithelia) were (and remain) very difficult to establish in culture, preventing any meaningful assessment of the proliferative potential of their constituent malignant cells.
In a series of studies in the early 1980s with cultures of normal diploid rodent fibroblasts, experimental evidence was obtained that an immortal phenotype could be induced as a rare event following treatment with powerful chemical carcinogens (Newbold et al., 1982). Only after immortalization were these cells, either spontaneously or after additional carcinogen treatment, able to undergo progression to a malignant phenotype. Moreover, when the first human oncogenes had been cloned (e.g. Ha-rasV12) these observations were extended to show that such oncogenes could exert their powerful oncogenic effects only if transfected into cells that had previously been immortalized (Newbold and Overell, 1983). Surprisingly, transfection of normal primary hamster dermal fibroblasts with Ha-rasV12 was found to induce premature senescence rather than malignant transformation, an observation that has recently been reproduced in primary human fibroblasts (Serrano et al., 1997). Independent and simultaneous studies showed that rat fibroblasts could be fully transformed by co-transfection of ras with either a myc oncogene (Land et al., 1983) or the gene encoding the large-T antigen of simian virus (SV) 40 (Ruley, 1983) suggesting that combinations of oncogenes may be needed for malignant transformation.
In one of these investigations (Newbold et al., 1982) the susceptibility of human cells to immortalization was compared with that of rodent cells under similar conditions. Human dermal fibroblasts (HDFs) proved totally refractory to immortalization by carcinogen treatment, even after repeated exposure to some of the most powerful known mammalian cell mutagens or clastogens. However, variant HDFs displaying the anchorage-independent phenotype (a good marker for malignancy in fibroblasts) could be readily induced by exposure to carcinogens. Transplanted into athymic mice such variants produced small tumours whose growth was shown to be limited by the finite intrinsic capacity retained by these cells (leading to replicative senescence), both in culture and in in vivo tumorigenesis studies.
The above early observations on the cell biology of immortalization and malignant transformation were consistent with the notion that immortalization represents an absolute prerequisite for the clonal evolution of neoplastic populations and, in particular, the acquisition of tumorigenic properties (Newbold, 1985a,b). They also provided direct evidence for the multi-step nature of the transformation process and highlighted the exceptional refractoriness of human cells to immortalization, suggesting that replicative senescence in human cells may have evolved as a tumour suppressive mechanism (reviewed by Reddel, 2000). However, the persuasiveness of such thinking was seriously blunted by a lack of information concerning the molecular mechanisms of human cell senescence and immortalization and by the prevailing view at the time that immortalization was an in vitro phenomenon unrelated to cancer development, since cancers were thought to be derived from stem cells which were assumed to be intrinsically immortal (Pierce, 1977).
The science of cell immortalization and its relationship to cancer was revolutionized in 1994 when Kim and co-workers published a landmark paper showing that human cancer tissues possess an enzyme activity, not present in normal human cells or tissues, that could account for the immortality of cancer cells (Kim et al., 1994). The enzyme, telomerase, was known to maintain the DNA of structures at the ends of eukaryotic chromosomes, known as telomeres, through the synthesis of characteristic telomeric repeat sequences. Telomerase (telomere terminal transferase) had been described several years earlier in the ciliated single celled protozoan Tetrahymena (Greider and Blackburn, 1985, 1987) and was first observed in human cancer cells (in which it synthesizes telomeric hexanuceotide repeats, TTAGGG) four years later (Morin, 1989). Following the observation that a mutant yeast strain est1 (ever shorter telomeres) underwent a kind of senescence after a certain number of divisions (see below) several workers began to produce experimental evidence that the limited proliferative capacity of human cells may be due to the same mechanism (Greider, 1990; Harley et al., 1990). A biochemical assay for telomerase was developed and used to demonstrate that human cancer cells and immortalized cells commonly possess functional telomerase (Counter et al., 1994a,b). The definitive study of Kim et al. (1994), in which a large number of human cancer samples and cell lines, counterpart normal tissues and in vitro transformed human cells were examined, was contingent on the development of a new highly sensitive PCR-based telomerase assay known as TRAP (telomere repeat amplification protocol) by these authors. The close association between telomerase and cancer and/or viral immortalization obtained in this study provided the molecular underpinning required to launch an era of intensive investigation into the role of telomerase-mediated immortalization as a key event in human cancer. The remainder of this article focuses on developments in the field since 1995, particularly the most recent advances.
|
Telomerase: the ‘immortality enzyme’? |
|---|
|
TopAbstractIntroductionTelomerase: the...Mortality barriers to human...Recent progress in generating...Regulation of telomerase in...Origins of telomerase in...Rodent cell senescence and...Importance of telomerase...Alternative mechanisms of...Telomerase and cell immortality...ConclusionsReferences |
|---|
Harley (1991)
proposed that telomeres act as a mitotic clock in human cells, i.e. as a specific timing mechanism to limit the division potential of human cells in vitro and presumably in vivo. This idea was essentially a revival and an elaboration of that proposed many years before by Olovnikov (1973), who suggested that the linear DNA of eukaryotic chromosomes would shorten with each round of DNA replication and that this failure to maintain telomeric DNA somehow triggers the phenomenon of replicative senescence that characterizes the Hayflick limit. An inevitable consequence of semi- conservative DNA replication in eukaryotic cells, known as the end-replication problem, is that the free DNA ends of each chromosome are not duplicated completely by DNA polymerase. Consequently, the ends of human chromosomes lose up to 200 bp of DNA per cell division. Since the publication of the study by Kim and co-workers, a large number of laboratories have confirmed the three basic observations: that the vast majority of normal human tissues and primary cell lines lack telomerase activity, that in the absence of telomerase, telomeres progressively shorten in normal human cells with each division cycle and that the majority of human cancers have active telomerase and stable telomeres. Rigorous functional proof that telomerase is responsible for immortality of cancer cells and that lack of it triggers a cell division counting mechanism (namely telomere shortening) depended on the isolation of the genes encoding the elements of human telomerase.
Telomerase is a ribonucleoprotein complex that is made up of two key components: (i) an RNA template molecule containing a sequence complementary to the telomeric TTAGGG repeat; and (ii) the catalytic component, known as hTERT, which is a type of reverse transcriptase able to synthesize TTAGGG repeats from the RNA template. The human genes encoding these components have been cloned (Feng et al., 1995; Meyerson et al., 1997) and the availability of these genes enabled it to be shown that the primary mode of regulation (i.e. repression) of telomerase in human cells is through silencing of the hTERT gene via transcriptional repression (elaborated below). Transfection of a hTERT cDNA expression vector into human fibroblasts (Bodnar et al., 1998; Counter et al., 1998a,b) led to immortalization of the cells. Such cells had elongated telomeres, an apparently normal karyotype and expressed none of the usual markers of malignancy (Jiang et al., 1999; Morales et al., 1999), exciting interest that this immortalization procedure could be used to generate human cellular material with indefinite replicative capacity for diverse medical applications. Telomerase-positive cells expressing a mutant telomerase enzyme failed to undergo immortalization, further strengthening the connection between telomere maintenance and immortalization. Furthermore, disruption of telomere maintenance by a mutant hTERT acting in a dominant negative fashion restored limited lifespan in human cancer cells resulting in loss of of tumorigenicity (Hahn et al., 1999b; Herbert et al., 1999). The level of expression of hTERT mRNA reflected the levels of active telomerase (Counter et al., 1998a,b), suggesting that induction of hTERT is required, and is perhaps sufficient, for expression of telomerase activity in tumour cells.
The vast majority of studies carried out since these seminal observations have confirmed that the primary mode of repression and derepression of telomerase in normal human cells and in cancers occurs at the level of transcription of hTERT. However, whether derepression of the hTERT gene is the sole requirement for immortalization in all human cell types remained unresolved. Indeed, some workers have questioned this premise and provided evidence that, in cultured primary human keratinocytes and breast epithelial cells, inactivation of a second tumour suppressive pathway in addition to telomerase reconstitution is necessary for immortalization (discussed below).
|
Mortality barriers to human cell immortalization and effectors of replicative cellular senescence in human cells |
|---|
|
TopAbstractIntroductionTelomerase: the...Mortality barriers to human...Recent progress in generating...Regulation of telomerase in...Origins of telomerase in...Rodent cell senescence and...Importance of telomerase...Alternative mechanisms of...Telomerase and cell immortality...ConclusionsReferences |
|---|
Prior to the discovery that telomere shortening and telomerase activation are the underlying cause, of human cell replicative senescence and immortalization, respectively, work with DNA tumour virus ‘early’ genes showed that the senescence of human fibroblasts, which normally manifests itself as growth arrest of the cells after between 40 and 70 population doublings (p.d.), depending on donor age, can be blocked by transfection with viral early genes such as that encoding the SV40 large-T antigen or co-transfer of those encoding the HPV16 E6 and E7 transforming proteins (see for example Ozer, 2000
). These viral transforming genes inactivate the p53 and pRB tumour suppressive proteins via sequestration and/or enhanced degradation. Human fibroblasts transfected with these genes normally display a substantial increase (of ~20–30 p.d.) in their proliferative capacity rather than attaining immortality. The cells then enter a proliferative phase known as ‘crisis’ that is characterized by very slow cell turnover and cell death. Rare clones of more rapidly dividing cells emerge in these cultures at a frequency of <10-6, which thereafter remain immortal. SV40 gene expression is therefore necessary, but not sufficient, for immortalization, pointing to a two-stage mechanism. Wright et al.(1989) introduced the term M1 (mortality 1) for the proliferative barrier (senescence) overcome by p53/pRB inactivation and M2 for the crisis restriction.
An attractive explanation for the two stages observed in human fibroblast immortalization by DNA tumour virus early genes has emerged from our understanding of the role telomerase plays in the process. It is now thought that p53 plays a central role as an effector of senescence by recognizing critically shortened telomeric DNA as a free end (i.e. as would be encountered following a double-strand break) and in this way inducing proliferative arrest via negative cell cycle control through p21 (i.e. by DNA damage recognition systems) (Wynford-Thomas, 1999; Webley et al., 2000). The second target of large-T and other viral early proteins, pRB, is also a negative cell cycle regulator acting in part via sequestration of the transcription factor E2F and family members. These proteins function as activators of cellular genes involved in DNA synthesis and are required for progression through the G1/S transition and S phase of the cell cycle. It is thought that M1 occurs when telomeres become critically shortened leading to destabilization of the protective ‘cap’ at the telomere formed by a fold-back (lariat-type) structure known as a t-loop (Griffith et al., 1999); recent evidence suggests that it is the shortest telomere rather than average telomere length that is critical for cell viability (Hemann et al., 2001). Inactivation of p53 by large-T prevents the p53-mediated growth arrest signal, allowing the cells to continue to divide until M2, when telomeres become so shortened that chromosomal end–end fusions occur. The resulting dicentric chromosomes threaten cell viability either by blocking mitosis per se or via a fusion–breakage–bridge cycle that leads to the loss of essential chromosomal material. Rare immortal clones that emerge have usually (but not always, see below) activated telomerase, either through loss of hTERT transcriptional repressor genes (discussed below) during crisis or perhaps via an unknown epigenetic mechanism (Bechter et al., 2002). The requirement for both p53 and pRB inactivation for the telomere-mediated M1 proliferative restriction to be bypassed suggests cooperation between the two anti-proliferative pathways as effectors of replicative senescence.
The viral transforming gene studies, when interpreted in the light of the telomere hypothesis of cellular senescence, suggest a possible model for the kinetics of cell immortalization in human cancer development. In such a model, immortalization is an essential event in malignant transformation and the robust down-regulation of hTERT expression (and thereby telomerase) in human cells is assumed to have evolved as a protective mechanism against clonal evolution and neoplastic progression. Loss of p53 function would be expected to confer a selective advantage only when telomeres became critically shortened at the equivalent of M1 in somatic cells. This would be expected to trigger lifespan extension, massive chromosomal/genomic instability (a hallmark of cancer) and derepression of hTERT, as a rare event, by genetic/epigenetic changes. The resulting telomerase-positive immortal lineage would then be primed to undergo further clonal evolution leading to the acquisition of malignant characteristics.
The p16(INK4A) gene is a third tumour suppressor implicated as an effector of cellular senescence and thus as a barrier to carcinogenesis. The encoded suppressor protein (and other related INK4 family members) binds to cyclin-dependent kinases cdk-4 and cdk-6 (Serrano et al., 1993) and, by blocking their association with D-type cyclins, induces cell cycle arrest by preventing phosphorylation of pRB; p16 therefore lies upstream of pRB and loss of function of p16 is reported to be involved in immortalization as an alternative to inactivation of pRB. The INK4A (or CDKN2A) locus is unusual in that, through alternative splicing, it encodes two unrelated proteins in rodent and human cells by means of distinct but overlapping reading frames (Quelle et al., 1995). One is p16, the other is ARF (alternative reading frame), also known as p14 in humans and p19 in mice. Both proteins appear to be involved in cell senescence: p16 activates pRB proteins through inhibition of their phosphorylation by cdk-4 and cdk-6, while ARF activates p53. Thus, the INK4A locus regulates the two tumour suppressive pathways (Rb and p53) that are most commonly disrupted in a wide range of human malignancies. The relative importance of the p16 and ARF proteins in the senescence process has recently been reviewed by Bennett and Medrano (2002), who conclude that their roles differ substantially in different cell types and species. p16 is frequently silenced in a number of human cancers, often involving an epigenetic mechanism. In contrast, germline mutations in p16 appear to predispose specifically to malignant melanoma leading to the proposal that p16 silencing is central to melanocyte immortalization.
|
Recent progress in generating a malignant human cell in vitro |
|---|
|
TopAbstractIntroductionTelomerase: the...Mortality barriers to human...Recent progress in generating...Regulation of telomerase in...Origins of telomerase in...Rodent cell senescence and...Importance of telomerase...Alternative mechanisms of...Telomerase and cell immortality...ConclusionsReferences |
|---|
The availability of genetic reagents for manipulating each of the above tumour suppressive pathways and for reconstituting telomerase opened the door to the experimental generation of malignant human cells from their normal counterparts in culture. It was reasoned that such experiments would pinpoint the key events required for human cell transformation. To investigate the role of hTERT in the transformation of human cells, Hahn et al.(1999a)
co-expressed combinations of hTERT, the SV40 large-T oncoprotein and an oncogenic Ha-ras gene (Ha-rasV12) in both human diploid fibroblasts and epithelial cells (the latter derived from normal kidney and breast epithelium) and demonstrated that direct tumorigenic transformation of all the cell types could be achieved by these three genetic elements alone. In a more detailed investigation of the malignant transformation of human mammary epithelial cells by the same three genetic elements, co-transfected cells were found to form poorly differentiated carcinomas that infiltrated through adjacent tissue and, unusually, showed amplification of the c-myc oncogene (Elenbaas et al., 2001). Subsequent studies have shown that malignant transformation is enhanced by the inclusion of the gene encoding the SV40 small-t antigen, which appears to act by perturbation of protein phosphatase 2A (Hahn et al., 2002). While it is accepted that additional alterations would almost certainly be needed for tumour cells to metastasize, the observation that a defined set of genetic alterations can cooperate to achieve malignant transformation of human cells is a major advance in our understanding of the primary mechanisms operative in human cancer development.
The immortalization of human mammary epithelial cells (HMECs), a clinically highly significant cell type in cancer research, has been studied in some detail in other laboratories (reviewed by Yaswen and Stampfer, 2002). HMEC cultures, commonly obtained from reduction mammoplasty tissue, normally proliferate for 15–30 p.d. in serum-containing medium, before undergoing a growth arrest indistinguishable from replicative senescence. In a serum-free medium developed specifically for HMECs (MCDB 170) culture lifespan is reduced to 10–20 p.d. A cell population arises spontaneously in some cultures that is capable of long-term growth (up to 100 p.d.). Independently, several groups (Brenner et al., 1998; Huschtscha et al., 1998; Kiyono et al., 1998) showed that self-selection generating longer lifespan cultures is invariably associated with the silencing of p16 by promoter methylation. Post-selection cells, which have stable p53, hit the next replicative barrier (which has been termed ‘agonescence’, to distinguish it from the crisis that occurs in human fibroblasts immortalized with DNA tumour viruses) (Yaswen and Stampfer, 2002) when their telomeres become critically shortened. Such cells accumulate chromosome aberrations, particularly telomeric fusions. However, agonescence differs from fibroblast crisis in that most of the cells possess long-term viability and immortal transformants have never been seen to arise spontaneously in several independent studies. Romanov et al. (2001) reported spontaneous escape from senescence (but not immortalization) of HMECs, with the resulting cells displaying the types of chromosome aberrations seen in early breast cancers. Post-selection HMECs are readily immortalized by hTERT reconstitution, indicating that short telomeres are the main impediment to growth in these cells. These stages in the immortalization process are known to apply to other epithelial cell types (Kiyono et al., 1998; Munro et al., 1999; Dickson et al., 2000).
The strict block to proliferation imposed by robust repression of the telomerase hTERT gene seems to be the primary barrier to clonal evolution of human breast epithelial and other human cells. Nevertheless, if p53 is experimentally inactivated in post-selection breast cell cultures, e.g. by transfection of dominant negative p53 mutants, activation of telomerase and conversion to an immortal phenotype is facilitated (Yaswen and Stampfer, 2002). From the limited information available, it appears that telomerase activation, at least in this particular model, is a gradual process rather than an abrupt step-change typical of a mutation, raising the possibility that an epigenetic mechanism underlies derepression of the hTERT gene.
|
Regulation of telomerase in normal human cells and in cancer |
|---|
|
TopAbstractIntroductionTelomerase: the...Mortality barriers to human...Recent progress in generating...Regulation of telomerase in...Origins of telomerase in...Rodent cell senescence and...Importance of telomerase...Alternative mechanisms of...Telomerase and cell immortality...ConclusionsReferences |
|---|
We have seen that telomerase is necessary for the maintenence of telomeres because of the so-called end-replication problem (and possibly other assaults on telomeric integrity such as oxidative damage). Lack of telomerase characterizes the majority of normal human cells in vivo, while in cancers the telomerase reverse transcriptase subunit (hTERT) is derepressed, thereby permitting unlimited cell proliferation. How then is hTERT regulated in normal cells and by what mechanism is it reactivated during human carcinogenesis?
Of the two components of the core telomerase enzyme the catalytic subunit, hTERT, is limiting; consequently, a considerable amount of effort has been devoted to understanding the transcriptional regulation of hTERT. Most telomerase-negative normal human cells lack detectable hTERT transcripts, while in telomerase-positive tumour cells 0.2–6 mRNA molecules have been detected by using sensitive RT–PCR-based techniques (Ducrest et al., 2002). The complete transcription unit of hTERT has been cloned and was found to consist of 16 exons and 15 introns spanning ~37 kb of genomic sequence (Wick et al., 1999). The 5' regulatory (promoter) sequences were identified using reporter constructs. By using the same approach, the promoter was shown to be inactive in normal human cells, but became active following immortalization (Cong et al., 1999), providing further evidence that transcriptional regulation of hTERT gene expression lies at the heart of replicative senescence and immortalization. Of the major factors binding to the core promoter, the c-Myc transcription factor (and oncoprotein) interacts with two evolutionarily conserved E-boxes (CACGTG) and is able to stimulate hTERT transcription and telomerase activity in hTERT-silenced human cells (probably in conjunction with the transcription factor SP1, which also binds to a cognate GC-box within the core promoter). More recent work (Kyo et al., 2000) showed that overexpression of Myc/Max dimer (again in conjunction with SP1 binding) led to activation of hTERT transcription, whereas Mad/Max acted as a repressor. The highly GC-rich content of the 5'-region extending into intron 1 constitutes a CpG island, suggesting that epigenetic mechanisms may play a key role in silencing hTERT transcription in normal cells and possibly therefore also during its reactivation in carcinogenesis. This notion is supported by studies using the demethylating agent 5-azacytidine and the histone deactetylase inhibitor trichostatin A (TSA), both of which were found to activate hTERT in human fibroblasts and lymphocytes (Takakura et al., 2001; Xu et al., 2001; Hou et al., 2002). Intriguingly, TSA had the opposite effect in prostate carcinoma cell lines in that it inhibited both hTERT mRNA expression and telomerase activity, leading to suppression of cell proliferation. The chromosomal location of hTERT, as the most telomeric gene on chromosome 5p (Bryce et al., 2000), has excited interest that it may be subject to telomere position effect modulation (Baur et al., 2001) involving chromatin remodelling of sub-telomeric sequences. Such a proposal has provoked speculation that inhibition of telomerase as telomeres lengthen could constitute an autoregulatory feedback mechanism for controlling telomere dynamics and telomerase expression. However, there is as yet no evidence that directly supports this hypothesis.
Controls working at the level of the 5' core promoter appear to be concerned with regulating hTERT during differentiation, the cell cycle or quiescence. However, at least in normal somatic human cells, the presence is indicated of an additional, far more stringent process for permanently silencing hTERT transcription. Somatic cell genetic analysis, involving microcell-mediated monochromosome transfer (MMCT), has provided evidence for the existence of one or more genes that repress telomerase in normal human cells. Thus far, human chromosomes 3, 4, 6, 7 and 10 have been shown to suppress telomerase when transferred by MMCT into certain tumour cell lines. For example, Ohmura et al. (1995) showed that normal chromosome 3 induced senescence on transfer into a human renal cell carcinoma cell line, with growth arrest manifesting after 23–43 p.d. This was associated with loss of telomerase activity as measured by TRAP and telomere shortening. A subsequent study confirmed that the repressive effect was due to down-regulation of hTERT. Powerful repression of telomerase activity by chromosome 3 was also reported in the early passage breast cancer cell line 21NT, by Cuthbert et al. (1999). By fine mapping deletions in the introduced chromosome in segregant hybrids (i.e. hybrids that carried a cytogenetically intact copy of normal chromosome 3 but had wild-type levels of telomerase) these authors proposed a tentative location for the telomerase repressor, at 3p14–21. Using the same MMCT approach, chromosomes 4, 6, 7 and 10 have been found to be repressors of telomerase activity in, respectively, HeLa cells (Backsch et al., 2001), the HPV-16-containing cervical cancer cell line SiHa (Steenbergen et al., 2001), a telomerase-positive human mesothelial cell line, MeT5A (Nakabayashi et al., 1999), and the hepatocellular carcinoma cell line Li7HM (Nishimoto et al., 2001). However, the genes responsible for the repressive activity of these chromosomes remain to be identified.
Despite the current lack of a clear explanation for the apparent tumour-specific repression of telomerase by individual human chromosomes, an improved understanding of the effect of chromosome 3 on hTERT expression in 21NT breast cancer cells has emerged (Ducrest et al., 2001). These authors examined cellular hTERT RNA expression (by quantitative RT–PCR of immature intron-containing RNA) in 21NT cells. The cells used in this study had been transfected with a hTERT cDNA in order that they would retain immortality even after transfer of chromosome 3. Introduction of chromosome 3 into these modified cells by MMCT was accompanied by complete down-regulation of endogenous hTERT mRNA. However, chromosome 3 transfer did not reduce the expression of transfected green fluorescent protein (GFP) reporter constructs driven by up to 7.4 kb of non-coding DNA flanking the 5'-end of hTERT. Furthermore, chromosome 3 transfer had no effect on c-Myc, Mad1 or other c-Myc target genes. These results indicate that telomerase is regulated primarily at the level of hTERT transcription by complex mechanisms involving regulatory elements distant from the 5' regulatory region and that the candidate repressor on chromosome 3 does not modulate hTERT expression through c-Myc or its co-regulators.
Regulation downstream of transcription represents a fine tuning mechanism for hTERT regulation. In a recent review, Aisner et al. (2002) emphasized the significance of post-transcriptional mechanisms for regulating hTERT level in human cells, including alternative splicing of hTERT mRNA, transport to the nucleus, assembly of hTERT with other telomerase components, proper localization of the complex at the telomere and post-translational modification of hTERT protein by phosphorylation. However, while such mechanisms are undoubltedly important in the fine control of telomerase activity, the twin facts that (i) the vast majority of normal human cells that lack telomerase also have hTERT transcriptionally repressed and (ii) that active telomerase is reconstituted by hTERT cDNA transfection firmly places transcriptional control as the central mechanism of repression in the majority of circumstances.
|
Origins of telomerase in cancer cells: two competing hypotheses |
|---|
|
TopAbstractIntroductionTelomerase: the...Mortality barriers to human...Recent progress in generating...Regulation of telomerase in...Origins of telomerase in...Rodent cell senescence and...Importance of telomerase...Alternative mechanisms of...Telomerase and cell immortality...ConclusionsReferences |
|---|
The suggestion from chromosome transfer studies that genes exist in normal human cells that repress telomerase activity in cancer cells, via a highly specific effect on hTERT transcriptional silencing, is consistent with the idea that the telomerase activity found in the vast majority of cancers arises from mutational or epigenetic disruption of such repressor genes. In addition, the kinetics of viral immortalization of human fibroblasts, i.e. the emergence of a faster growing clone of cells as a rare event (~10-6) during crisis, M2, would, superficially at least, be consistent with a step change in a rare cell involving the mutational inactivation of a repressor (assuming some degree of selective advantage is conferred by mutation in a single allele). In addition, the fact that many human cancers, particularly carcinomas, have short telomeres maintained in steady-state, often by relatively low levels of telomerase, would suggest that telomerase had been activated only after a considerable period of telomere shortening had taken place in the originating (telomerase-negative) target cell.
There is, however, a second possibility for the origin of telomerase in cancers that still warrants serious investigation. It is that the majority of human cancers arise from stem cells or from transiently amplifying populations in renewing tissues, that already have active telomerase. The initial view that telomerase is present only in cancers and germ cells was subsequently found to be too simplistic. Using sensitive PCR-based assays for telomerase, relatively low levels of activity were subsequently found in the proliferative cells of certain self-renewing tissues, e.g. the bone marrow, trachea and bronchi, skin (basal layer) and gut (lower crypt) (reviewed in Forsyth et al., 2002). It has been suggested that such levels of telomerase are sufficient to slow down, but not prevent, telomere shortening during tissue renewal. The evidence for this comes from the observation that telomeres shorten in human skin and gut with increasing age, in spite of the presence of telomerase. Thus, although telomerase is silenced in almost all human organs at between 18 and 21 weeks of gestation (Forsyth et al., 2002), it is retained at low levels in rapidly dividing self-renewing tissues, presumably to offset the effects that rapid telomere loss might have on chromosomal stability and the potential initiation of cancer (discussed in detail below). The consequence of inhibiting telomerase in these normal tissues by anti-telomerase drugs used to control the proliferation of cancer cells is unclear, but needs further investigation.
The key question that remains unresolved is, therefore, whether the cells from which human cancers originate are in fact those cells that apparently have not robustly down-regulated hTERT or, alternatively, whether hTERT-repressed cells undergo hTERT derepression. If the former prevails, then the presence of telomerase in most cancers could be regarded as a process of ‘selection’ (Greaves, 1996) of pre-existing telomerase-positive cells with subsequent enhancement of activity (e.g. by further selection of clones with minor epigenetic changes) sufficient to maintain telomeres indefinitely. If the latter, telomerase activation would occur through a process of ‘induction’ (Shay and Wright, 1996), possibly resulting from a single or small number of genetic/epigenetic events inactivating a hTERT repressor gene(s). Of course, the two models are not necessarily mutually exclusive.
|
Rodent cell senescence and immortalization: do telomere-independent senescence mechanisms exist? |
|---|
|
TopAbstractIntroductionTelomerase: the...Mortality barriers to human...Recent progress in generating...Regulation of telomerase in...Origins of telomerase in...Rodent cell senescence and...Importance of telomerase...Alternative mechanisms of...Telomerase and cell immortality...ConclusionsReferences |
|---|
As discussed in the Introduction, the majority of studies aimed at transforming normal diploid cells in culture (e.g. by carcinogens and cloned oncogenes) were carried out using freshly explanted diploid rodent cells. This was simply because, after many attempts, human cells proved virtually impossible to transform due to the replicative senescence barrier and their intrinsic resistance to immortalization. How then does our contemporary understanding of telomerase regulation help us explain the increased transformability of rodent cells?
Mouse, rat and Syrian hamster cells from trypsinized embryos, neonatal dermis or other tissues undergo a process that closely resembles replicative senescence of human cells (Newbold, 1997). Fibroblasts normally dominate such cultures and terminal division generally occurs much earlier with these cells (after ~10–25 p.d.) than with human fibroblasts (40–70 p.d.). In sharp contrast to human cells, rodent cells show a greater propensity to undergo immortalization in culture, either spontaneously or following a single exposure to a carcinogen (Newbold et al., 1982). Mouse cells appear to be more susceptible than rat in this respect, almost invariably immortalizing spontaneously, whereas in hamster cell cultures spontaneous immortalization is rare, but can be induced by treatment with various mutagens and carcinogens, for example polycyclic hydrocarbons, alkylating agents, non-genotoxic rodent carcinogens and ionizing radiation (Trott et al., 1995). The reason for the increased susceptibility of rodent cells to immortalization is explained by the fact that, unlike their human counterparts, primary rodent cell cultures possess active telomerase and maintain it throughout their culture lifespan (Carman et al., 1998; Greenberg et al., 1998; Russo et al., 1998). As a result, the chromosomes of these cells have long telomeres that do not shorten with time in culture. Tissues taken from rodents also have active telomerase and therefore show no evidence of the stringent controls on hTERT expression found in human counterparts
Telomere-mediated senescence does not therefore appear to exist as a barrier to immortalization and malignant transformation in rodent cells. Rodents, being physically much smaller and shorter lived, may not have needed to evolve the protection of a powerful telomerase repressive mechanism to prevent cancer. However, the fact that primary rodent cells, even those isolated from mice lacking the RNA component of telomerase (see below), still undergo a senescence-like process in vitro raises the question of the existence of telomere-independent ‘clocks’ that limit the proliferative capacity of somatic cells.
Ramirez et al.(2001) have recently argued persuasively that the loss of division potential of rodent cells in vitro is a cell culture artefact and therefore imply that immortalization of these cells does not reflect a genuine event that must occur in rodent carcinogenesis. They suggest that the cause of the ‘senescence’ seen in these cells is inadequate culture conditions leading to ‘stress’ and DNA damage. The primary effector of growth arrest in cultured mouse fibroblasts appears to be the ARF/p53 pathway, inactivation of which seems to be sufficient for immortalization (Carnero et al., 2000), whereas specific silencing of p16 leaving ARF intact indicates that p16 is not involved (Sharpless et al., 2001). ‘Stress’ and resulting DNA damage may, Ramirez and colleagues speculate, be the consequence of free radicals produced by amine oxidase in serum, since the immortal growth of diploid mouse cells in a serum-free medium has been demonstrated (Loo et al., 1987) and also of rat oligodendrocyte precursor cells cultured under conditions that inhibit differentiation (Tang et al., 2001).
A superficially analogous phenomenon of telomere-independent growth arrest is encountered in the immortalization of human epithelial cells, as discussed earlier in this article. The initial proliferative potential of human mammary epithelial cells in chemically defined media is ~20 p.d. This mortality barrier, which has been termed M0 (Foster et al., 1998), is mediated by p16 rather than ARF as the effector and raises the possibility of the existence of an additional tumour suppressive mechanism for counting cell divisions. However, Ramirez et al. (2001) provided experimental evidence against this explanation by showing that the proliferative capacity of cultured HMECs could be extended all the way to the telomere-dependent M1 barrier merely by growing the cells on feeder layers. Moreover, under these conditions HMECs could be immortalized directly by hTERT without the need to inactivate the p16/pRB pathway.
The above views dismissing the existence of telomere-independent ‘clocks’ that count cell divisions in mammalian somatic cells provide an all-encompassing model of replicative senescence centred on repression of telomerase and underline the importance of senescence as the critical cell lifespan barrier in humans (Parkinson et al., 2000). In this model, the telomere-dependent barrier to division potential is seen as the only true mechanism of replicative senescence, with all other limits to mammalian cell proliferative capacity in culture being artefacts (Wright and Shay, 2001).
An alternative view is also persuasive. Dividing cells in differentiating tissues are naturally programmed, as part of a commitment process, to enter terminal division. The process of differentiation can be assumed to be highly complex, depending on signalling either through contact with other cell types in the tissue ‘niche’ or through humoral factors that orchestrate unfolding programmes of gene expression and repression, culminating in the ultimately specialized (usually post-mitotic) terminally differentiated cell. This kind of process does in essence incorporate a cell counting mechanism that imparts a limit to the division capacity of the differentiating cell lineage. It is reasonable to suppose that such programmes would need to be bypassed during carcinogenesis and would therefore represent important tumour suppressive mechanisms.
The fact that normal human epithelial cells will, under certain culture conditions, reach the M1 (telomere-dependent) barrier can be accommodated in this alternative explanation if it is supposed that the observed M0 senescence in the absence of feeder cells does not necessarily indicate an artefact, but rather reflects a natural response of cells to being removed from their tissue niche (and as such could still be regarded as a kind of stress). This type of response would represent a genuine safeguard against cancer because, by definition, clonally evolving neoplastic cell lineages ultimately need to escape from niche constraints (e.g. by removal of the p16/pRB negative cell cycle regulator pathway) to undergo malignant transformation. Such populations would then come up against the telomere-mediated M1 replicative block as a second line safeguard. The fact that the p16/pRB pathway has been shown (Herman et al., 1995) to be functionally inactivated in many human cancers (in the case of p16 often by promoter methylation, e.g. in around one-third of breast tumours), together with the observation that p16 inactivation was observed in mildly dysplastic lung lesions adjacent to carcinomas (Belinsky et al., 1998), would tend to support the view that its loss of function represents an early and important event in the initiation of many cancers.
The same explanation can be applied to explain the ‘replicative senescence’ of rodent cells in culture. Having not retained or evolved the M1 mechanism it is somehow counter-intuitive to imagine that no cellular lifespan controls exist in rodents. It may be that the limitation to replicative potential of rodent cells in culture may be a genuine consequence of removal from their natural environment, a function that acts in vivo to prevent carcinogenesis. Thus, the ARF/p53 silencing needed for rodent cell immortalization would also be expected to be a prerequisite event in rodent carcinogenesis. Moreover, the hypothesis that DNA damage-mediated ‘stress’ is responsible for growth arrest of rodent cells is not wholly in accord with previous observations on Syrian hamster cells showing that treatment with some of the most powerful known chemical point mutagens and clastogens leads to the induction of rare immortal variants (with many lines being diploid) and does not reduce the proliferative lifespan (as measured in population doublings) of the majority of mortal cells (Newbold et al., 1982; Trott et al., 1995). Future work will no doubt clarify these important issues.
|
Importance of telomerase repression as a tumour suppressive mechanism: evidence from telomerase ‘knockout’ mouse studies |
|---|
|
TopAbstractIntroductionTelomerase: the...Mortality barriers to human...Recent progress in generating...Regulation of telomerase in...Origins of telomerase in...Rodent cell senescence and...Importance of telomerase...Alternative mechanisms of...Telomerase and cell immortality...ConclusionsReferences |
|---|
Evidence from work with rodent cell cultures and from observations that telomerase is expressed in many rodent tissues strongly points towards major differences between humans and rodents with respect to telomerase regulation in the soma. However, despite this, genetically engineered rodents have proved to be extremely valuable in studying both the consequences for the whole organism of telomere attrition due to lack of telomerase and also the modulating effects of telomerase inactivation and, conversely, its overexpression on tumour suppression and carcinogenesis, respectively.
Blasco et al. (1997) were the first to construct a telomerase ‘knockout’ mouse strain by deleting the gene encoding the telomerase RNA component (mTR) from the germline. Such mice possessed no detectable telomerase in any tissue but were viable for six generations, presumably reflecting the fact that the original mouse strain had long telomeres. Cells isolated from animals at the fourth generation onwards had no detectable telomeric repeats and displayed severe chromosomal abnormalities. Did cells from such mice behave as human cells with respect to resistance to immortalization? The answer was surprisingly that they did not. Telomerase-deficient cells could be readily immortalized in culture (albeit at a lower frequency than telomerase-positive mouse cells; Espejel and Blasco, 2002) and transformed by viral oncogenes to the point that they produced tumours in athymic mice following s.c. injection. An essential role for telomerase in highly proliferative mouse organs was demonstrated in a subsequent study (Lee et al., 1998) on late generation mTR knockout mice that revealed defective spermatogenesis and attenuated proliferation of cells in the bone marrow and spleen. Progressive loss of organismal viability, including atrophy of the small intestine, was described by Herrera et al. (1999). Using a double knockout mouse strain lacking p53 in addition to mTR, Chin et al. (1999) found that loss of p53 mitigated the effects of telomerase deficiency on cellular and organismal dysfunction. However, the loss of telomerase coupled with p53 deficiency were shown to cooperate in enhancing cell transformation. These observations led the authors to question the role of crisis as a tumour-suppressive mechanism, instead arguing that the genetic catastrophe ultimately observed under these conditions (in the absence of p53) promotes malignant transformation. In such telomerase knockout, mutant p53 mice there was also a shift in the spectrum of cancers towards epithelial tumours (carcinomas) and these displayed complex cytogenetic abnormalities reminiscent of adult human malignancies (Chang et al., 2001). The results suggest that fusion-bridge-breakage triggered by critically short telomeres (and mediated by the telomere-associated protein ku86) (Espejel et al., 2002) may be one of the key reasons for the genomic instability that characterizes human carcinomas.
In an extension of these extremely valuable studies, a double knockout mouse strain, in which the INK4A locus was deleted in addition to mTR, was employed to investigate the results of loss of telomere function in a cancer-prone background (Greenberg et al., 1999). These mice displayed significantly reduced tumour formation, further underlining the oncoprotective effect conferred by lack of telomerase. A separate study, however, described an increased frequency of spontaneous tumours in mTR knockout mice (Rudolph et al., 1999). Mice engineered to overexpress the catalytic subunit of telomerase in a specific tissue (basal keratinocytes), despite showing normal stratified epithelia, were found to be highly responsive to the mitogenic effects of phorbol ester tumour promoters and were more susceptible to the induction of malignant skin carcinomas by chemical carcinogens (Gonzales-Suarez et al., 2001). More recently, Artandi et al. (2002) generated transgenic mice that overexpressed telomerase in a wide variety of tissues and invasive mammary carcinomas developed spontaneously in a significant proportion of aged female mice. These latest data add yet more weight to the argument that somatic telomerase promotes the development of cancer.
The ease of immortalization of telomerase-deficient mouse cells confounded expectations and also raised concerns about the prevailing notion that telomerase repression in human cells represents an evolutionary anti-cancer strategy. Hande et al.(1999) shed some light on this conundrum by showing that, in immortal cells from telomerase-null mice, telomere length became stabilized in a chromosome-specific manner, indicating that a telomerase-independent telomere maintenance mechanism can be activated in mouse cells.
Collectively, results to date provide persuasive evidence that lack of telomerase protects against cancer whilst its expression promotes carcinogenesis. Very short telomeres, however, increase the propensity for chromosomal instability and this may, paradoxically, lead to telomerase reactivation and carcinogenesis in humans. The information available suggests that reactivation of telomerase is extremely rare in human cells, particularly in the presence of p53. As telomeres shorten in renewing tissues with ageing, the probability of widespread genomic instability would be expected to increase, which may partly explain the increase in cancer incidence with age. Thus, telomerase repression exerts its maximum effect during normal reproductive age when telomeres are long, i.e. when evolutionary selective pressures for survival are at their highest. As telomeres shorten with age or when unusual demands are placed on renewing tissues, lack of telomerase may have the opposite effect and increase the probability of telomerase reactivation, immortalization, clonal evolution and cancer.
|
Alternative mechanisms of telomere maintenance |
|---|
|
TopAbstractIntroductionTelomerase: the...Mortality barriers to human...Recent progress in generating...Regulation of telomerase in...Origins of telomerase in...Rodent cell senescence and...Importance of telomerase...Alternative mechanisms of...Telomerase and cell immortality...ConclusionsReferences |
|---|
From the results of the telomerase mTR knockout studies, it is clear that mouse cells have little difficulty in activating an alternative mechanism for telomere maintenance under the selective pressure that results from the absence of telomerase activity. Moreover, the fact that not all human tumour cells and in vitro immortalized (e.g. with SV40) cell lines have detectable telomerase activity (Kim et al., 1994
) points to the existence of similar mechanisms in humans. As discussed by Reddel (2000), some of these telomerase-negative cancers may genuinely not be immortalized, particularly those localized focal growths that have not undergone much in the way of clonal evolution. However, the majority have been shown to maintain the length of their telomeres by a mechanism that has become known as ALT (alternative lengthening of telomeres). ALT in human cells was first identified by Bryan et al. (1995).
Studies in various eukaryotes have revealed the existence of a telomerase-independent telomere maintenance pathway. Inactivation of telomerase in the yeast Saccharomyces cerevisisae resulted in senescence of the majority of the population (Lundblad and Szostak, 1989; Lundblad and Blackburn, 1993). Clones of immortal cells emerged, however, that were able to maintain their telomeres via a RAD52-dependent recombination pathway that involved amplification of sub-telomeric elements (see also McEachern and Blackburn, 1996). In these and subsequent studies it appeared that there were two different classes of telomerase-negative survivors: those that possessed amplified sub-telomeric sequences and short telomeric elements and those that had much longer telomeres made up of amplified telomeric (TTAGGG) repeats (Teng and Zakian, 1999). Analysis of telomere restriction fragments from human cells that rely on ALT for telomere maintenance revealed that they possess telomeric tracts that are extremely heterogeneous in length, ranging from undetectable to abnormally long (Bryan et al., 1995), which would also point towards a recombinational origin (see also Henson et al., 2002).
The incidence of ALT in human cancer cells and in in vitro immortalized human cell lines has been studied in some detail by Reddel and co-workers. In SV40-immortalized fibroblasts up to 50% utilized an ALT mechanism rather than telomerase. However, the available evidence indicates that epithelial and mesothelial cells lines are predominantly telomerase-positive (Perrem, 2001). The fact that the frequency of ALT cell lines is high in both DNA tumour virus-immortalized (and virtually 100% in immortalized Li Fraumeni) cells (Bryan et al., 1995) implicates p53 allele loss as part of the mechanism underlying ALT activation. In human tumours and tumour-derived cell lines, ALT was identified in ~7% of samples assayed in each case (Bryan et al., 1997). Interestingly, the majority of ALT-positive cell lines were sarcomas.
Somatic cell hybridization studies of the genetic basis of the ALT mechanism point to the existence of genes in normal human cells or telomerase-positive cells that act as repressors of the ALT pathway. Perrem et al.(1999) fused an ALT human cell line with normal cells and described hybrids that showed a rapid reduction in telomeric DNA and became senescent. Fusion of ALT cells with immortal telomerase-positive cells resulted in immortal hybrids with repressed ALT. The gene or genes responsible for ALT repression, which are presumably silenced during ALT activation, have still to be identified.
There is intense interest in the ALT pathway because it represents a potential stumbling block to the application of anti-telomerase approaches to cancer therapy. The key question still requiring resolution is whether and at what frequency do ALT variants arise when telomerase-positive human cancer cells are challenged with telomerase inhibitors. Some clues can be gained from in vitro experiments, as will now be described below.
|
Telomerase and cell immortality as an anti-cancer target |
|---|
|
TopAbstractIntroductionTelomerase: the...Mortality barriers to human...Recent progress in generating...Regulation of telomerase in...Origins of telomerase in...Rodent cell senescence and...Importance of telomerase...Alternative mechanisms of...Telomerase and cell immortality...ConclusionsReferences |
|---|
Because telomerase is found in ~90% of human cancers and is essential for the continued proliferation (and clonal evolution) of cancer cells, it represents probably one of the most exciting anti-cancer targets thus far discovered. The telomerase ribonucleoprotein complex is biochemically unique, since it possesses a functional RNA component as well as a catalytic protein subunit, and therefore would a priori be expected to be well suited to highly specific functional inhibition by small molecules. Telomerase inhibition has to date been accomplished in the laboratory by a range of experimental procedures, including the use of: (i) antisense oligonucleotides or hammerhead ribozymes directed against the RNA (hTR) component of telomerase, (ii) dominant negative mutant hTERT constructs, (iii) reverse transcriptase inhibitors aimed at blocking the catalytic function of hTERT, (iv) agents that stabilize and/or encourage the formation of four-stranded G-quadruplex structures by telomeric DNA (which inhibit telomerase function), (v) other small molecule inhibitors targeted at a variety of processes involved in telomere maintenance and (vi) the use of hTERT as a cancer-specific antigen in immunotherapy. As might be expected, there has been a large number of studies using these specific approaches. However, only a few of them have been shown to generate a response consistent with a highly specific effect on telomere dynamics (i.e. reduced telomerase activity, progressive telomere shortening, absence of non-specific toxicity and phenotypic lag followed by replicative senescence, or apoptosis, after a period of time that reflected initial mean telomere length).
Herbert et al.(1999) used an antisense oligodeoxynucleotide approach based on 2'-O-methyl-RNA to target telomerase hTR in immortalized human breast epithelial cells. Telomerase was inhibited and telomeres shortened progressively until cell death from apoptosis ensued. Significantly, no variants emerged in these experiments that had activated an ALT pathway for telomere maintenance under the extreme selective pressure generated by telomerase inhibition. In two independent studies, telomerase was experimentally inhibited in human cancer cells by transfecting them with dominant negative hTERT (DN-hTERT) constructs (encoding a catalytically inactive mutant hTERT that sequesters hTR). Zhang et al.(1999) observed telomerase inhibition and telomere shortening in A431 (epidermoid) tumour cells followed by abnormal mitoses, delayed growth arrest and apoptosis. In a more comprehensive study, Hahn et al.(1999b) introduced a DN-hTERT into a panel of human cancer cell lines and immortalized cells using a retroviral vector. All transfectants lacked detectable telomerase activity, showed progressive telomere shortening and ultimately arrested and died. A tumorigenic ovarian cancer cell line failed to form tumours in athymic mice after DN-hTERT transfection. These observations strongly suggest that telomerase inhibition reverts cancer cells into M2 crisis and that the resulting chromosome damage from fusion bridge breakage triggers apoptosis. Since this response was obtained in cells lacking functional p53 (as do on average at least 50% of human cancers), widespread applicability of hTERT inhibition as an anti-cancer strategy is indicated. Once more, no evidence of emergence of ALT variants was obtained nor were any such variants apparent in studies of telomerase-repressed human breast cancer cells following microcell transfer of a copy of normal human chromosome 3 (Cuthbert et al., 1999).
Targeting the tumour cell specificity of hTERT expression immunotherapeutically via an immunostimulation approach has also shown promise (reviewed in Vonderheide, 2002). Work in both human and murine model systems has shown that cytotoxic T lymphocytes can recognize peptides derived from TERT and kill a variety of TERT-positive malignant cells. Clinical trials are now underway to determine the applicability of hTERT immunotherapy for treating human cancer.
Small molecule inhibitors of telomerase thus far described include compounds that inhibit the action of telomerase at the telomere through the stabilization of G-quadruplex structures. Such tetraplex DNA conformations have been shown to form, thus far, only in vitro, from the G-rich regions that exist in the single-stranded telomere 3'-overhang. A wide variety of agents has been discovered that stabilize such structures and exert cytotoxic effects on cells in culture. The main problem with such studies is that telomere shortening was not demonstrated and cytotoxic effects could be interpreted as being non-specific (reviewed in White et al., 2001; see also Kelland, 2001). However, very recently a