In our previous studies [7, 14], we demonstrated that caspase-2 was significantly activated (up to 20-fold) along with other caspases (caspase-3, caspase-9 and caspase-8) during apoptosis induction by taxanes in some cancer cells. We have also shown that, at least in some cases, the mitochondrial pathway does not represent the main pathway of apoptosis induction by taxanes. Recently, we found that apoptosis was induced by taxane application in SK-BR-3 and MCF-7 breast cancer cells and that caspase-2 was also significantly activated in these cells. Furthermore, MCF-7 cells are without functional caspase-3, as we confirmed (data not shown). Thus it is reasonable to assume that caspase-2 could play an important role in apoptosis induction by taxanes in breast cancer cells.
In order to study the role of caspase-2 in apoptosis induction by taxanes in breast cancer cells, we employed a convenient model using SK-BR-3 and MCF-7 cells. SK-BR-3 cells have nonfunctional p53 and functional caspase-3 . On the other hand, MCF-7 cells have functional p53, but they do not have functional caspase-3 . The use of these two cell lines, with their opposite properties, could help to elucidate the role of caspase-2 in apoptosis induction by taxanes. It could particularly contribute to our knowledge concerning the relationship between caspase-2 activation and the activation of other caspases as well as the mechanism of caspase-2 activation itself.
We tested both a classical taxane paclitaxel and a novel (second-generation) taxane SB-T-1216. In our previous study with MDA-MB-435 and NCI/ADR-RES cell lines , SB-T-1216 was shown to be more effective than paclitaxel, particularly in NCI/ADR-RES cells resistant to paclitaxel. However, SB-T-1216 and paclitaxel seemed to use the same or similar mechanism of cell death induction [7, 14]. In this study, MCF-7 cells were shown to be a slightly more resistant to taxanes than SK-BR-3 cells but both cell lines showed similar sensitivity to paclitaxel and SB-T-1216 (see Figure 1). Furthermore, it seems that paclitaxel and SB-T-1216 use the same mechanism of apoptosis induction (see Figures 2, 3, 4, 5, 6 and “Results”).
Significant activation of caspase-2 in SK-BR-3 and MCF-7 cells after taxane application at death inducing concentrations (see Figure 2) was confirmed using several different methods (see Figure 3). It is known that the measurement of caspase-2 activity can be affected by the activity of caspase-3 . Thus, the measured activity of caspase-2 (see Figure 2) can be in fact a sum of caspase-2 and caspase-3 activities. It can explain the discrepancy in timing of measured caspase-2 activity (see Figure 2) and detected caspase-2 activation (see Figure 3). In MCF-7 cells, we only detected cleavage of procaspase-2 (see Figure 3A) under different conditions than procaspase-2 was detected. However, the problem of detecting the cleaved form of caspase-2 in MCF-7 cells could be a specific feature of these cells as described previously . With regard to other cell types, several studies have demonstrated caspase-2 activation in various types of cancer cells after apoptosis induction by taxanes [37, 38, 43].
Several studies with human prostate cancer cells, human melanoma cells and mouse embryonic fibroblasts have shown that caspase-2 is required for apoptosis induction by taxanes. These studies employed various techniques using siRNA, caspase-2-/- cells and specific caspase-2 inhibitors [21, 23, 44]. We also demonstrated that caspase-2 is required, at least partially, for apoptosis induction by taxanes in both studied breast cancer cell lines. The inhibition of caspase-2 expression using siRNA technique resulted in a significantly increased number of surviving cells following application of taxanes at death-inducing concentrations. MCF-7 cells, with inhibited caspase-2 expression, grew even after application of death-inducing concentrations of taxanes (see Figure 4). This effect could not be attributed to the stimulatory effect of siRNA application on cell proliferation (see Figure 4B).
In spite of the fact that several studies have shown the involvement of caspase 2 in apoptosis induction by various stimuli in breast cancer cells [27, 34, 35], the mechanism of caspase-2 involvement in apoptosis induction is not fully understood. Caspase-2 has been described as an apical caspase [21, 23] as well as a possible executioner caspase , in various types of cancer cells, together with its functions which are independent of apoptosis [46, 47]. In this study with SK-BR-3 and MCF-7 cells we demonstrated that, together with caspase-2, other caspases (caspase-8, caspase-9, caspase-3, caspase-7) were activated after application of taxanes.
The activation of caspase-9, -3 and -7 was significantly affected in SK-BR-3 cells with inhibited caspase-2 expression after cell death induction by taxane application. While the cleavage of caspase-9 and caspase-3 significantly decreased, caspase-7 cleavage increased. In MCF-7 cells (without functional caspase-3) the cleavage of caspase-9 was nearly blocked and the cleavage of caspase-7 was significantly decreased. Cleavage of caspase-8 decreased slightly in SK-BR-3 cells and significantly in MCF-7 cells (see Figure 5). The substantial inhibitory effect of the inhibition of caspase-2 expression on the activation of caspase-8 and caspase-9, as well as the activation of executioner caspase-3 in SK-BR-3 cells and executioner caspase-7 in MCF-7 cells, supports a suggestion that caspase-2 functions as an apical caspase. The increase of caspase-7 activation together with the decrease of caspase-3 activation, due to the inhibition of caspase-2 expression, in SK-BR-3 cells seems rather obscure (see Figure 5). Some kind of compensatory mechanism could be involved.
Additional information concerning caspase-2 function could come from testing the cleavage of caspase-2 substrates. However, there are only a few known specific substrates of caspase-2 such as golgin-160 [31, 46]. In this study, we demonstrated golgin-160 cleavage after application of death-inducing taxane concentrations in both studied cell lines (see Figure 3B). Unfortunately, it was not very helpful in elucidating caspase-2 function without data connecting golgin-160 cleavage to other relevant events.
Caspase-2 has been found in the cytosol, Golgi complex, mitochondria and also in the nucleus of cells . Regarding SK-BR-3 and MCF-7 cells, we showed that caspase-2 seemed to be primary localized in the nucleus. However, caspase-2 did not colocalize with DNA, as demonstrated with mitotic cells. Treating the cells with death-inducing taxane concentrations seemed to lead to a redistribution of caspase-2 from the nucleus to the cytosol (see Figure 6). Caspase-2 was probably released from the nucleus because of the long-term disintegration of the nuclear envelope associated with the mitotic block after application of taxanes. The question, whether caspase-2 is activated within the nucleus or in the cytoplasm, has not been answered yet . In our case, long-term exposure of most of the cellular caspase-2 to the cytoplasmic environment, which can comprise caspase-2-activating capacity, could lead to caspase-2 activation. Thus, there could be a very simple explanation for caspase-2 activation in breast cancer cells after taxane application.
Caspase-2 activation is usually connected with PIDDosome formation and PIDD protein upregulation via the induction of expression by activated p53 [23–25]. We found significant p53 activation assessed by the induction of p21WAF1/CIP1 expression, after the application of death-inducing taxane concentration in MCF-7 cells with functional p53 . On the other hand, in SK-BR-3 cells without functional p53  we confirmed that there was no p53 activity (see Figure 7). In the next step, we assessed the effect of taxanes on PIDD expression. No significant effect of taxanes on PIDD upregulation was found in either SK-BR-3 or MCF-7 cells (see Figure 8). This means that p53 activation, and the subsequent PIDD upregulation, is not involved in caspase-2 activation.
Nevertheless, we detected some coimmunoprecipitation of RAIDD protein with both PIDD protein and caspase-2 in both cell lines 24 h after taxane application at the death-inducing concentrations. However, suprisingly the inhibition of RAIDD expression using siRNA technique did not affect the number of surviving SK-BR-3 as well as MCF-7 cells after taxane application at all. Therefore, we suggest that PIDDosome formation  does not represent the main platform for caspase-2 activation in breast cancer cells when apoptosis is induced by taxanes. Recently, other pathways of caspase-2 activation which circumvent PIDDosome formation have been reported , e.g. the activation of caspase-2 in DISC complex . Therefore other pathways of caspase-2 activation in cells treated with taxanes should be considered.