This study shows that in Barrett's adenocarcinoma SEG 1 cells, siRNAs targeting human telomerase can be efficiently delivered and results in a rapid inhibition of telomerase activity. The inhibition of telomerase activity is associated with: 1) a complete erosion of telomeres of majority of chromosomes and growth arrest of the cancer cells without affecting the normal cells from stomach and intestine; 2) induction of both senescence and apoptosis; and 3) up-regulation of key regulatory genes involved in cell cycle arrest and apoptosis.
RNA interference has emerged as a method for effective and specific inhibition of gene expression both in vitro and in vivo [18, 25]. This study shows that a mixture of siRNAs directed against two different regions of telomerase produced near complete silencing of telomerase function along with suppression of the protein. The mixture of siRNAs in nanomolar concentration led to 77 ± 3% inhibition of telomerase activity within 24 hours and near complete inhibition in 72 hours.
Inhibition of telomerase activity was associated with a marked reduction (30% of the control) in median telomere length. The effect of RNAi-mediated inhibition of telomerase activity on telomere length has not been reported before. However, other inhibitors of telomerase have been shown to produce similar median reduction in telomere length. The estimation of median telomere length, however, does not provide information on telomere length on individual chromosomes that possess telomere of different lengths. Loss of all telomeres in a small number of chromosomes may trigger a DNA damage response when median telomere length is not critically reduced. In order to determine loss of telomeres on individual chromosomes, we hybridized metaphase chromosome spreads of treated and untreated cells with Cy3-PNA (C3TA2)3 probe. We found that in siRNA treated cells, 60% of chromosomes had no telomere signal, indicating complete erosion of telomeres on these chromosomes. These data suggest that the siRNAs produced potent inhibition of telomerase activity and loss of telomeres on majority of the chromosomes.
SiRNA mediated telomerase inhibition also resulted in cell death of >80% cancer cells in four weeks. The long delay of four weeks that was required for induction of cell death is consistent with the fact that loss of telomerase caused a relatively gradual telomere shortening. A telomere length below critical ~2 kbp , has been reported to result in growth arrest . The growth arrest was associated with replicative senescence and apoptosis.
Telomere shortening induces senescence  which is permanent and irreversible cell cycle arrest in G1 phase thereby halting growth. Senescent cells express β-galactosidase and develop a characteristic flat and vacuole-rich cytoplasmic morphology, but remain metabolically active and are eliminated by cell necrosis and phagocytosis . However, more recently senescent cells have been shown also to undergo p73 mediated apoptotic cell death in cancer cells. Telomerase inhibitors have been reported to cause replicative senescence in cancer cells . Telomerase gene suppression by siRNAs induced β-galactosidase expression in 41 ± 6% and typical senescent phenotype in a subset (10 ± 4%) of the cells.
Telomerase inhibition with the siRNA mixture also induced apoptosis in 86 ± 5% of the cancer cells. Apoptosis was confirmed by both the annexin V labeling and PARP cleavage. Since molecular suppression of telomerase by gene silencing using siRNA is highly selective, these studies show that apoptosis is specifically due to telomerase inhibition. Although, chemical inhibitors of telomerase have also been reported to cause apoptosis , it was not clear whether they produced apoptosis specifically due to telomerase suppression or due to their unrelated side effects. However, induction of both senescence and apoptosis has also been observed following specific inhibition of telomerase by oligo-nucleotides targeting RNA component of telomerase.6 It has been suggested that telomerase inhibitors may cause apoptosis due to severe and precipitous telomere shortening that may be associated with chromosomal rearrangement and DNA fusion. This condition termed 'crisis' may be associated with activation of apoptotic pathways resulting in apoptotic cell death of the cancer cells.
Signaling pathway of telomere shortening-induced replicative senescence and apoptosis in the cancer cells is not fully understood. Telomere shortening may be recognized as DNA damage. This is supported by the fact that in the adenocarcinoma SEG 1 cells, telomerase inhibition was associated with over expression of gene for HR 23B, an early marker of DNA damage .
Telomerase suppression and telomere shortening was also associated with up-regulation of genes for p73, p63 of the p53 gene family. Expression of p63 was also confirmed by western blot analysis. Although P53 is frequently mutated in cancer, p73 and p63 are rarely affected. Unlike p53, p73 and p63 exist in different isoforms that may have different targets and opposing effects on replicative senescence and apoptosis. For example, whereas full length form called p73α may induce cell cycle arrest and apoptosis, its truncated isoform AN-p73 actually inhibits cell cycle arrest and apotosis. p73 and p63, like p53, remain localized to the cell nucleus and exert their effects by regulating expression of other genes such as those that mediate cell cycle arrest and apoptosis. Although the precise regulation of p53 family members is not fully understood, it is clear that expressions of p53, p73 and p63 are differently regulated. P73 and p63 share many but not all target genes with p53. Furthermore, the genetic alterations associated with DNA damage are complex and involve genes other than those in the p53 family only that may also influence replicative senescence and apoptosis. For example, E2F1 may induce apoptosis through p53-independent mechanisms. These studies suggest that inhibition of telomerase causes loss of telomeres; loss of telomeres is sensed as DNA damaged by DNA damage sensor such as HR23B; this through other intermediary proteins may lead to up-regulation of p73, p63, E2F1, and mdm2 (see Additional file 1). It has also been shown that in response to DNA damage, E2F1 is stabilized which subsequently upregulates p73. Similarly mdm2 which binds with p53 to inhibit its activity, infact positively regulates p63 and probably also p73.
Up-regulation of p73, p63 and E2F1 may mediate replicative senescence. Cyclin dependent kinase inhibitor, p21 is a key mediator of cell cycle arrest and is a target for p53. Senescence associated with telomere shortening during ageing is accompanied by p53 mediated p21 upregulation. However, p73, p63 and E2F1 have also been shown to cause p53 independent up-regulation of p21. Therefore, it appears that telomerase suppression in SEG- 1 cancer cells leads to induction of p73, p63 and E2F1, which may up-regulate p21 leading to cell cycle arrest and senescence. In addition p16 and GAAD45 were also upregulated in the telomerase siRNA treated cells. Both p16 and GAAD45 are potent inhibitors of cell cycle and contribute to replicative senescence in the cancer cells
In our studies telomerase suppression in SEG1 cells was also associated with up-regulation of genes for FasL, Fas and caspase 8, suggesting that in these cells telomere shortening leads to activation of death receptor mediated apoptosis. P73 can upregulate several genes whose products participate in mitochondrial apoptotic pathway. We found upregulation of caspase recruitment domain containing protein 9 (CARD9) that has been implicated in apoptosis through interaction with Bcl10 and activation of NFKB . We also found that there was upregulation of genes for caspase7 and caspase 3 which are executioner caspases that are implicated in the cleavage of PARP in the final stages of apoptosis. Consistent with these data, a marked induction in the appearance of cleaved (89 kDa) PARP fragment (Figure 5B) was seen in telomerase siRNA treated adenocarcinoma cells. Western blot analysis also indicated a substantial upregulation of TRAIL, which induces apoptosis in cancer cells but not in normal cells.
The gene expression studies provide only an overview of changes in genes involved in the apoptotic pathways. Although for p63, FASL, and p16, the expression changes were confirmed by western blot analyses, further studies of changes in protein expression and activities would be required to fully define the signaling pathways.
In contrast to the effect on the cancer cells, telomerase inhibition did not affect normal somatic cells. This may be due to the fact that normal somatic cells do not express telomerase. Moreover, signaling pathway for these responses in the cancer cells involve p53 homologs p73a, p63a and others that are different from p53-mediated responses in normal cells. Further studies in telomerase expressing normal cells are needed to determine whether effects of the telomerase siRNA are selective for cancer cells.