Although telomeres are short in primary prostate cancers, this gives no information regarding when telomere shortening occurs. It could be argued, for instance, that the observed telomere shortening is merely a reflection of the large amount of cell division that takes place during the process of tumor expansion, in which case, it might have little if anything to do with the actual initiating events causing the disease. For telomere shortening to have a causal role in prostate carcinogenesis one would expect to find evidence of it at the earliest recognizable, pre-invasive stage of malignant transformation, which, in the prostate, is high-grade PIN (HGPIN) (76). In fact, use of the telomere FISH method to probe HGPIN has revealed that telomeres are abnormally short in the majority of HGPIN foci, with significant telomere shortening found in 93% of HGPIN lesions examined in radical prostatectomy specimens (n = 30) (77). In addition, telomere shortening was also found in HGPIN lesions in needle biopsy samples from 20 patients without histopathological evidence of concurrent prostate cancer. Notably, in this study, telomere shortening in HGPIN foci was restricted to the luminal secretory epithelial cells only, although the underlying basal epithelial cells and surrounding stromal cells displayed normal telomere lengths (Fig. 2). In a separate independent study, Vukovic et al. reported significant shortening in 63% of HGPIN overall (n = 30), with a higher rate of telomere shortening (80%) reported for foci situated near (within 2 mm) adenocarcinoma within the same tissue sample (78).
The level of telomere shortening observed in PIN likely exceeds that which would normally elicit a cell cycle checkpoint response, thus, implying that this key tumor suppression system is defective in PIN lesions. Similar levels of telomere shortening have been correlated with markers of CIN in other organ systems (79-82). One candidate element of an abrogated checkpoint response in PIN is 14-3-3a sigma. This protein is normally induced by p53 in response to DNA damage and leads to G2/ M cell cycle arrest (83). Lodygin et al. found the promoter of this gene to be hypermethylated in prostate cancer cell lines and in 100% of primary prostate cancers (n = 41). Decreased protein expression was also observed in 91% of prostate cancers, whereas normal, atrophic, and BPH areas showed high levels of expression in both luminal and basal epithelial cells. Interestingly, 14-3-3a sigma protein expression levels were also low in the luminal cells of PIN lesions, whereas basal cells retained robust expression, matching the pattern of telomere lengths observed in PIN (77,84). Because the majority of PIN lesions are not thought to progress to invasive cancers, intact telomere-based replica-tive senescence or apoptosis checkpoints may represent a critical bottleneck restraining the outgrowth of most PIN lesions.
9. TELOMERASE ACTIVITY RE-STABILIZES CHROMOSOMES AND ALLOWS UNLIMITED REPLICATION
Although dysfunctional telomeres may help initiate cancer formation, if left unchecked, continued telomere shortening in premalignant lesions and cancers would cause increasing levels of genetic instability, ultimately becoming lethal to the tumor. Cancer cells overcome this problem by re-stabilizing their telomeres, primarily through activation of the enzyme telomerase, a specialized reverse transcriptase that adds back telomere DNA repeats to chromosome ends (85). Telomerase provides at least two critical functions to the tumor cell; namely, quelling CIN and supplying the capacity for unlimited replication (immortalization) (86,87). Research has revealed that telomere length maintenance seems to be a necessary step for human cells to become malignant; confirming the long-held belief that cellular immortalization is a prerequisite for carcinogenesis (88,89). In keeping with this, at least 85% of human epithelial cancers have telomerase activity (90-92).
Human telomerase is minimally composed of two essential subunits, a catalytic protein subunit (hTERT) and an RNA subunit (hTR or hTERC) that contains the template for telomere repeat addition (93-98). As mentioned in Section 5, telomerase activity is stringently repressed in most normal human somatic cells (99,100). There are notable exceptions, however, including activated lymphocytes, subpopulations of highly proliferative tissues (such as the hematopoietic system), and prolif-erative zones in renewal tissues (such as the intestinal crypts and the basal layer of the skin and cervix), plus hormonally responsive cells of the female breast and endometrium (101-109). Notably, the activity detected in these normal tissues is typically much weaker than that found in malignant tumors of the same tissue of origin. Whereas the RNA subunit, hTR, displays a widespread expression pattern, expression of the catalytic subunit, hTERT, seems to be the rate-limiting factor for telomerase activity in many (e.g., ovary, skin, and cervix) but not all (e.g., kidney, colon, and endometrium) tissues (56,93,94,96,98,110-119).
Although telomerase activity seems to be the preferred way cancer cells stabilize their telomeres, approx 15% of human cancer cases lack detectable telomerase activity. At least a subset of these, particularly certain tumors of mesenchymal origin, maintain their telomeres via a telomerase-inde-pendent pathway known as alternative lengthening of telomeres (ALT), the details of which are still being uncovered (120-124).
A consideration of telomere biology, as outlined, leads to the conclusion that telomeres can play opposing roles in the development of cancer. On the one hand, cellular responses to telomere shortening caused by cell division place strict limits on the proliferative capacity of a given cell lineage, thus,
Fig. 3. Telomere shortening can either promote or suppress tumorigenesis. If tumor-suppressive telomere length-sensitive cell cycle checkpoints are intact, then telomere shortening results in cell cycle exit before the telomeres become critically short and unstable. Checkpoint abrogation allows continued cell cycling with concomitant telomere shortening, eventually leading to chromosomal instability caused by telomere dysfunction.
serving to prevent the expansion of potentially malignant cells. On the other hand, critically short, dysfunctional telomeres can promote carcinogenesis by initiating CIN (50,51). The deciding factor in determining which of these opposing outcomes wins out seems to be the status of the cellular checkpoint pathways that monitor telomere length (Fig. 3). Thus, telomerase-deficient mouse models support a tumor suppressive role for telomere shortening only if cellular checkpoints are intact; particularly the p53 tumor suppressor pathway. Research on mice doubly deficient for telomerase and p53 has revealed a key role for dysfunctional telomeres in the development of epithelial cancers in the setting of defective checkpoints, and murine tumors arising in a background of dysfunctional telomeres display the types of complex karyotypic abnormalities typically observed in human carcinomas (125,126).
Was this article helpful?