Caspasemediated Apoptosis And Cell Growth Inhibition In Tumour Expansion

The relevance of caspase-mediated apoptosis in tumour development has been emphasised recently by the finding that enhanced expression of caspase-2 leads to a reversion of ras-induced transformation in NIH3T3 cells (Hiwasa and Nakagawara, 1998). Hiwasa and Nakagawara (1998) transfected caspase cDNAs into c-Ha-ras-transformed cells. Enhanced expression of caspase-2 alone, but not caspases 1 and 2 together, resulted in a reduction in the ability of the transfected cell to grow in soft agar. This was accompanied by ras protein degradation, suggesting that the apparent reversion of the transformed phenotype was due to the degradation of the transforming gene product by caspase-2. However, Hiwasa and Nakagawara (1998) have not presented any evidence relating to the tumorigenicity of the ras-transformed cells, or any data on possible inhibition of tumorigenicity by the transfection of caspase-2 cDNA into these transformed cells. It is essential to recognise that cell transformation is a process quite distinct from tumorigenicity. Whether caspase-induced apoptosis occurring in the developing tumour can control the growth of the tumour is still an open question. As stated elsewhere, apoptosis subserves an important function in maintaining the kinetics of cell expansion in metastatic growth. By inference, there is no reason why a similar role may not be played by apoptosis in the control of primary tumour growth, or, to put it more precisely, primary tumour expansion. The apoptosis-mediated restriction of a cell population can be differentiated from control by inhibition of growth, although these might not represent totally independent mechanisms. This has been demonstrated in the case of TGF-p, which influences apoptosis as well as cell growth. However, in WEHI 231 cells, the caspase inhibitor BD-fmk can effectively inhibit TGFp-induced apoptosis but not counteract TGFp-mediated inhibition of cell growth (T.L. Brown et al. 1998). This apparent dichotomy of mechanisms is obvious also from the pathways of radiation-induced apoptosis. Ionising radiation induces the activation of caspase-3, which is followed by the induction of apoptosis. Ionising radiation induces rapid apoptosis in human lymphoblast cells expressing wild-type p53, but in cells where p53 is mutated or abrogated by viral oncoprotein, apoptosis is delayed as well as reduced (Yu and Little, 1998). In this instance apoptosis seems to occur in conjunction with growth control mediated by wild-type p53. On the other hand, caspases have been shown to cleave p53-induced cyclin-dependent kinase inhibitor p21waf1/ciP1 and induce cells to undergo apoptosis (Y.K. Zhang et al. 1999). In the latter case, apoptosis seems to depend on proliferation. In contrast with p53-mediated apoptosis, VD3 has been reported to induce apoptosis of certain breast cancer cells independently of both p53 and caspases. This report is based on the observation that VD3 produces growth arrest and apoptosis of both MCF7 cells, which are p53 positive, and T47D cells that are p53-negative. Furthermore, this apoptotic induction is not inhibited by inhibitors of caspase, whereas TNF- or staurosporine-induced apoptosis is inhibited (Mathiasen et al. 1999).

Whatever pathway apoptosis might take, it is inevitable that tumour growth and subsequent processes should be shaped by balancing forces of apoptosis and cell cycle control factors. Nonetheless, it is worthwhile to note here that ICE-like protease expression has been reported to correlate with progression and prognosis of neuro-blastoma. The frequency of expression of ICE mRNA was markedly reduced in advanced-stage neuroblastomas as compared with early-stage tumours (Ikeda et al. 1997). The expression of caspase-3 in normal gastric mucosa, gastric adenomas and adenocarcinomas has been studied in some detail by Hoshi et al. (1998). They found that caspase-3 expression decreased from a high level (42% cells staining for caspase) in nonneoplastic gastric mucosa to a lower level (33%) in adenomas, and to a still lower level (17%) in adenocarcinomas. These differences were statistically significant in spite of the large standard deviations of the mean. The caspase-3 positivity, of the three groups, correlated directly with apoptotic indices determined by the TUNEL method and inversely with proliferative indices provided by Ki67 labelling. These data are consistent with the view that a loss of apoptosis-mediated control over cell turnover is an important feature of tumour growth. Conversely, there is an implicit suggestion that ICE-like proteases may be involved in apoptosis-mediated regression of tumours, in the demonstration by Ikeda et al. (1997) that ICE protease staining could be co-localised in the nucleus with DNA fragmentation. In this context, one should also take note of the recent report that caspase-3 activity was found to increase in colonic carcinomas and adenomas as compared with normal mucosa (Leonardos et al. 1999). Obviously, these observations may serve to confirm that the expression of the caspase is related more to the degree of apoptosis taking place in the tumour than to the degree of tumour progression. Donoghue et al. (1999) reported a difference in the pattern of caspase-3 distribution in B-cell diffuse large cell lymphoma. In immunohistochemistry, caspase-3 showed a diffuse distribution in the cytoplasm or a punctate or spotty pattern. The diffused pattern appeared to relate to poor prognosis, and the punctate pattern was associated with complete response to therapy. The authors have stated further that where the percentage of caspase-3-expressing cells was low, prognosis was poor. Not only are some of these findings not compatible per se, but the study seems to raise more questions than it successfully answers. It should be conceded, however, that it would be unhelpful to attribute all apoptotic activity to caspases. Donoghue et al. (1999) found no correlation between apoptosis and caspase expression in the cells, but the degree of apoptosis was associated with poor prognosis. A finding of potential significance is that the caspase distribution pattern might be significant in terms of enzyme activity. Whether the punctate distribution might reflects sequestration of the enzyme is worth further investigation.

It should be recognised, nevertheless, that caspase expression may reflect cancer progression and that caspases may actively promote metastatic deposition by a more direct route. Thus, a high proportion of in situ and invasive (58 and 90%, respectively) carcinomas of the breast stain strongly for caspase-3. Caspases-6 and -8 also are expressed at high levels more frequently in carcinomas than in hyperplasia. The enhanced expression correlated well with apoptotic indices in the samples. Furthermore, enhanced apoptosis was associated with poor prognosis (Vakkala et al. 1999a, 1999b). The formation of metastatic lesions depends on a cascade of events; prominent among them is the invasion of the vascular and endothelial systems by cells of the primary tumour. The entry into the circulatory system has been attributed to an active process of transmigration or diapedesis across the endothelial layer, as well as to the inherent structural defects often found in the endothelium. Recently, Kebers et al. (1998) found that several breast cancer cell lines, among them MCF7, MDA-MB231, T47D, and HT1080, induced a four-fold increase in apoptosis of human umbilical vein endothelial cells (HUVEC), with an attendant enhancement of caspase-3 activity. The interaction of MCF7 cells with HUVEC caused a transient increase in intracellular calcium levels (Lewalle et al. 1998), which, presumably, may have led to caspase activation. The induction of apoptosis required cell-cell contact, because media conditioned by the growth of these cells were ineffective. Kebers et al. (1998) have further observed that lymphocytes do not induce apoptosis, suggesting that the apoptosis of endothelial cells might constitute a specific mechanism in the diapedesis of tumour cells. Whether caspase-mediated apoptosis of endothelial cells occurs in vivo is yet to be demonstrated.

The outcome of tumour progression has often been assessed in relation to a single given variable as a prognostic factor. As noted above, in the context of caspases we have a paradoxical situation that in both caspase-positive and caspase-negative circumstances some relationship is noticed with tumour progression. As observed earlier, tumour growth and progression are a net outcome of the balancing forces of apoptosis and cell cycle control factors and cell proliferation. Perhaps it would be more rewarding to reexamine the question of caspases in tumour progression in this light. For this viewpoint, the recent findings of Volm and Koomagi (2000) are most encouraging. They have examined the relevance not only of caspase-3 but also of c-myc expression in the prognosis of non-SCLC. Volm and Koomagi (2000) have reported that caspase-3-negative patients had a median survival time of 41 weeks as compared with 79 weeks for caspase-3-positive patients. They then looked at c-myc expression and its influence on prognosis. Patients who were c-myc negative had a median survival time of 89 weeks, whereas the median for c-myc-positive patients was 43 weeks. Not only did these two factors correlate inversely with survival, but, myc-/caspase+ patients showed a median survival time of 102 weeks as compared with only 22 weeks for myc+/caspase- patients. This clearly makes the point that more than one factor might influence a given cellular feature and thereby determine the outcome of the disease. The study also serves to further emphasise that it would be unhelpful to try to evaluate a single prognostic factor, while other factors might be present that would impinge on the direction of cellular changes.


Among the several substrates of caspases is the membrane protein called fodrin. The erythroid homologue of fodrin is known as spectrin, with which fodrin shares substantial amino acid sequence homology. Fodrin is attributed with the function of maintaining the structural integrity of the plasma membrane. Fodrin and spectrin form a major component of the skeletal network that underlies the plasma membrane (Levine and Willard, 1981; W.J. Nelson et al. 1990; Bennett and Lambert, 1991; Bennett and Gilligan, 1993). Fodrin (alias spectrin) isoforms are also found in the membranes of the Golgi apparatus (Devarajan et al. 1996, 1997; Beck et al. 1994, 1997; Godi et al. 1998; Fath et al. 1997; Stankewich et al. 1998), lysosomes (Hoock et al. 1997), and intracellular vesicles (Malchiodi-Albedi et al. 1993; Stankewich et al. 1998). Fodrin is an actin-binding protein; therefore, two further putative functions should also be considered. One of these is a presumptive involvement in the process of signal transduction, because fodrin has been found to be able to inhibit phospho-lipases A2 (PLA2), C, and D (Lukowski et al. 1996, 1998). In comparison, other cytoskeletal proteins such as actin and vimentin are far less efficient than fodrin (Lukowski et al. 1996). Phospholipases are closely associated with the generation of DAG and IP3 from PIP2. These are involved in the activation of downstream pathways of signal transduction that, in turn, involve the activation of appropriate protein kinases, such as PKC, and Ca2+ release from intracellular stores.

Another potentially important function that can be attributed to fodrin is in influencing cell adhesion and migration. Spectrin is required for neurite extension in neuroblastoma cells. Sihag et al. (1996) were able to inhibit neurite extension in NE2a/dl neuroblastoma cells with an antibody directed against the N-terminal domain of spectrin. The latter is known to interact with actin. The protein called ankyrin mediates the link-up between membrane adhesion proteins and the spectrin cytoskeleton. The expression of some of these adhesion molecules has been shown to cause cell aggregation in which process both ankyrin and spectrin are recruited to the foci of adhesion (Dubreuil et al. 1996).

In wound healing of corneal epithelium, fodrin becomes redistributed from its subplasma membrane location to a cytoplasmic location. This redistribution occurs soon after wounding of the epithelium. A similar redistribution occurs in response to PMA treatment and is inhibited by PKC inhibitors (Amino et al. 1995). Presumably these events are related to the cell migration that follows, although there is little direct evidence linking these.

As stated above, fodrin (spectrin) is a substrate for caspases. Apoptosis induced by several different pathways has been shown to be accompanied by proteolysis of fodrin. Inhibition of apoptosis also results in the inhibition of fodrin proteolysis (Martin et al. 1995). This proteolysis seems to be produced by ICE/ced-3 proteases (Cryns et al. 1996; Vanags et al. 1996). Kouchi et al. (1997) believe that calpains are not associated with the cleavage of the 240-kDa a subunit of fodrin in the apoptosis of rat thymocytes both in vivo and in vitro. However, Porn-Ares et al. (1998) do implicate calpains. Fodrin occurs in a variety of cell types including keratinocytes, chromaffin cells, and renal epithelium, and in a variety of epithelial and fibroblast cell lines. Fodrin shows a homogeneous cytoplasmic and a discontinuous membrane distribution in benign melanocytic tumours, whereas normal mel-anocytes at the basal layer of the epidermis only faintly stain for fodrin at the plasma membrane. Overall, neoplastic cells show greater amounts of fodrin than their nonneoplastic counterparts (Tuominen et al. 1996). This observation has been confirmed in an immunohistochemical study of a variety of adenocarcinomas and squamous cell carcinomas by Sormunen et al. (1997). However, malignant melanomas contain subpopulations that do not express fodrin (Tuominen et al. 1996). This does not detract from any putative relationship between fodrin expression and malignancy, because malignant tumours are notoriously heterogeneous with respect to a large spectrum of cellular characteristics. Although much work needs to be done in this area, already there are clear indications that caspase-mediated alterations in fodrin expression and function might be involved in biological processes (e.g., apoptosis, cell adhesion, motility, and modulation of cell shape) that are inherent features of tumour development, dissemination, and metastasis.

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook

Post a comment