γ-Secretase-Dependent Cleavage of E-Cadherin by Staurosporine in Breast Cancer Cells

Abstract

E-cadherin is a transmembrane protein that serves as a cell adhesion molecule component of the adherens junction. We previously showed that cadmium induced γ-secretase-dependent E-cadherin cleavage via oxidative stress. In this study, we report that staurosporine (STS)-induced apoptosis induces caspase-2 and/or -8-dependent E-cadherin cleavage. STS increased γ-secretase-dependent cleavage of E-cadherin in breast cancer cells through caspase activation. The ability of the γ-secretase inhibitor DAPT and the caspase inhibitor zVAD-FMK to block E-cadherin cleavage provided support for these results. The cleavage of E-cadherin was blocked by caspase-2 and -8 inhibitors. Immunofluorescence analysis confirmed that, along with the disappearance of E-cadherin staining at the cell surface, the E-cadherin cytoplasmic domain accumulated in the cytosol. In the presence of an inhibitor of γ-secretase or caspase, the cleavage of E-cadherin was partially blocked. Our findings suggest that activation of caspase-2/-8 stimulated the disruption of cadherin-mediated cell–cell contacts in apoptotic cells via γ-secretase activation.

Keywords::

• breast cancer
• apoptosis
• E-cadherin
• γ-secretase
• caspase

INTRODUCTION

Apoptosis is essential both for multicellular development and maintaining cellular homeostasis. All common apoptotic pathways require the participation of caspases, proteases with specificity for short amino acid motifs terminating in a critical aspartate, adjacent to which cleavage occurs. Staurosporine (STS), a potent protein kinase C inhibitor with a broad spectrum of activity (Ruegg & Burgess, 1989), induces apoptosis via a mitochondrion-mediated pathway (Yang et al., 1997). In this pathway, mitochondrion-generated reactive oxygen species (ROS) cause oxidative stress in a variety of cells, including breast cancer T47D cells (Gouaze et al., 2002; Seleznev et al., 2006), and cytochrome c is released. A similar, but slower appearance of apoptosis has been reported in STS-treated MCF-7 cells lacking caspase-3 (Tang et al., 2000).

Morphological changes observed during apoptosis result in part from effects on cell–cell contacts. The cadherin–catenin adhesion complex represents one of the major adhesive systems in multiple epithelial tissues. Mutations in E-cadherin have been identified in a number of carcinomas and invasive tumor cells (Nollet et al., 1999). Several reports have shown that cadherin is also targeted during apoptosis (Steinhusen et al., 2001; Vallorosi et al., 2000). E-cadherin-positive breast cell lines were less sensitive to STS than E-cadherin-negative ones (Wang et al., 2009). Recently, we have shown that cadmium increases the cleavage of E-cadherin by γ-secretase in T47D cells (Park et al., 2008). Although caspase activation was also observed in cadmium-treated T47D cells, caspase inhibition through treatment with zVAD failed to block the cleavage of E-cadherin by γ-secretase in T47D cells. Given that caspase-2/-8 induced γ-secretase activation in H4 neuroglioma cells (Chae et al., 2010), the role of caspase in γ-secretase-dependent E-cadherin processing in T47D cells is unclear. Thus, major questions regarding the molecular mechanisms involved in E-cadherin cleavage in T47D breast cancer cells during apoptosis remain to be answered.

Here, we demonstrate that caspase-2 and/or caspase-8 contribute to STS-mediated E-cadherin processing in breast cancer cells by activating γ-secretase. STS treatment resulted in the generation of E-cadherin cleavage products by caspase-2 and/or caspase-8 via the actions of γ-secretase. Our results suggest that caspase-2/-8 activation reduces E-cadherin levels at the cell surface in apoptotic cells via γ-secretase activation.

METHODS

Materials

Staurosporine, γ-secretase inhibitors (DAPT), zVAD-fmk and other caspase inhibitors were purchased from Calbiochem (La Jolla, CA). E-cadherin antibodies were obtained from BD Transduction (San Jose, CA). PARP antibodies (64D11) were from Cell Signaling Technology (Beverly, MA). Alamar Blue was obtained from Invitrogen (Camarillo, CA).

Cell culture

Human breast cancer T47D and MCF7 cells were grown in RPMI-1640 medium, supplemented with 10% fetal bovine serum, penicillin and streptomycin in a 5% CO₂ atmosphere at 37°C. Human epithelial carcinoma A431 cells were maintained in DMEM supplemented with 10% fetal bovine serum, penicillin and streptomycin in a 5% CO₂ atmosphere at 37°C.

Assays of cell viability

The Alamar Blue (Park et al., 2008) was measured using fluorescence-based kits from Biosource International (Camarillo, CA). Alamar Blue was measured at excitation and emission wavelengths of 555 and 600 nm, respectively. Cell viability was expressed as a percentage of the control and mean ± standard error for multiple wells tested in at least three separate experiments.

Cell lysates and Western blot analysis

Cells in plates were washed with phosphate-buffered saline (PBS) and lysed with lysis buffer. The lysates were cleared by centrifugation (15 000 rpm, 5 minutes), and protein concentrations were determined using the bicinchoninic acid (BCA) method, before being stored at − 80°C. Equivalent amounts of protein were separated on NuPAGE (4–12%, Invitrogen) gels and transferred to PVDF membranes. Membranes were blocked in 5% fat-free milk in TBST and incubated with the primary antibody overnight at 4°C. The immune complexes were then detected using the enhanced chemiluminescence (ECL) system (Amersham).

Immunocytochemistry and antibodies

The cells were fixed with 4% paraformaldehyde (PFA) in PBS for 15 minutes, permeabilized with blocking buffer (1% goat serum, 0.1% BSA in PBS) with 0.2% Triton X-100 at room temperature for 10 minutes, and blocked in blocking buffer for 1 hour at room temperature. They were then stained with primary antibodies in blocking buffer at 4°C overnight. The cells were incubated with Alexa 546- and 488-conjugated goat antimouse and goat anti-rabbit antibodies (Molecular Probes), washed in PBS and then photographed using a fluorescence microscope. Images were assembled using Photoshop (Adobe).

Statistical analysis

Data were expressed as mean ± SE. Data were analyzed by ANOVA followed by Duncan post hoc test. The level of significance was set at p < 0.05.

RESULTS

Staurosporine induces cytotoxicity in T47D breast cancer cells

To investigate the effect of staurosporine (STS) on the viability of T47D cells, an alamarBlue assay was performed using cells exposed to various concentrations of STS for 9 or 24 hours. Alamar Blue monitors metabolic activity/proliferation as a measure of cell viability. STS reduced cell viability in a dose-dependent manner (Figure 1A). Based on the alamarBlue assay results, DAPI (4′,6-diamidino-2-phenylindole) staining was conducted to investigate whether STS induced apoptosis. Treatment of cells with STS at a concentration of 1 or 25 μM caused nuclear morphological changes. Although T47D cells treated with 1 μM STS for 9 hours did not show a decrease in cytotoxicity, nuclear morphological changes were induced (Figure 1B).

Figure 1. Effect of staurosporine (STS) on T47D cell viability. (A) After treatment with various concentrations of STS for 9 or 24 hours, cytotoxicity was assessed in an alamarBlue assay. Data are presented as the mean ± standard error (n = 3). *p < 0.05, versus untreated control cells. (B) STS-treated T47D cells stained with DAPI.

STS-mediated apoptosis induced E-cadherin cleavage in breast cancer cells

To study the involvement of γ-secretase in E-cadherin processing during STS-mediated apoptosis in breast cancer cells, we first examined the effect of STS on E-cadherin proteolysis in the breast carcinoma cell lines T47D and MCF-7. STS enhanced the proteolysis of E-cadherin in T47D cells (Figure 2A). Treatment of T47D cells with STS resulted in the time-dependent generation of a C-terminal fragment of E-cadherin (E-cad/CTF2). When applied to cells for 24 hours, STS induced the dose-dependent cleavage of E-cadherin (Figure 2B).

Figure 2. E-cadherin was cleaved during staurosporine (STS)-induced apoptosis in T47D and MCF7 cells. T47D (A) and MCF7 cells (C) were treated for the indicated periods of time with 1 μM staurosporine (STS), lysed in RIPA and blotted with anti-E-cadherin antibody (C36) or anti-PARP antibody. (B) Dosedependent effect of STS in T47D cells treated for 24 hours. Actin was used as a loading control.

A previous study showed that E-cadherin degradation was minimal in caspase-3-deficient MCF-7 cells (Keller & Nigam, 2003). Nonetheless, treatment of MCF-7 cells with STS resulted in the generation of E-cad/CTF2 (Figure 2C). This fragment may not, however, have been produced by caspase-3, since MCF-7 cells were previously reported to be deficient in caspase-3 (Janicke et al., 1998). PARP cleavage was monitored as a direct measure of apoptosis. Treatment of MCF-7 cells with STS clearly induced PARP cleavage (Figure 2C). These observations suggest that apoptosis increased the generation of E-cad/CTF2 in a caspase-3-independent manner in T47D and MCF-7 cells.

γ-secretase-mediated cleavage of E-cadherin by STS in T47D cells is mediated by caspase-8

Next, we sought to characterize E-cad/CTF generation by STS-treated T47D cells. To test which type(s) of protease(s) were involved in the formation of E-cad/CTF2 in STS-treated T47D cells, inhibitor studies were performed. In the presence of the matrix metalloproteinase inhibitor GM6001, no significant inhibitory effect was detected in E-cad/CTF2 generation by STS (Figure 3A). The γ-secretase inhibitor DAPT completely blocked the generation of E-cad/CTF2 in STS-treated T47D cells (Figure 3A). These observations suggest that the generation of E-cad/CTF2 in STS-treated cells is due to PS1/ γ-secretase activity. The next question was whether caspases are involved in the generation of E-cad/CTF2 by STS via γ-secretase activation, since zVAD blocked the generation of E-cad/CTF2 by STS. To identify the caspases involved in STS-mediated E-cad/CTF2 generation, a panel of inhibitors with selectivity toward caspases-1, -2, -3, -5, -6, -8 and -9 was used to investigate specific inhibitory effects. The caspase-2 inhibitor z-VDVAD-fmk and the caspase-8 inhibitor z-IETD-fmk were effective in blocking γ-secretase-dependent E-cadherin cleavage in T47D cells (Figure 3B). However, the other caspase inhibitors, z-YVAD-fmk, z-DEVD-fmk, z-WEHD-fmk, z-VEID-fmk and Z-LEHD-fmk, which predominantly inhibit caspases-1, -3, -5, -6 and -9, respectively, did not block the cleavage of E-cadherin via γ-secretase (Figure 3B). Unexpectedly, z-YVAD-fmk, z-DEVD-fmk and Z-LEHD-fmk—inhibitors of caspases-1, -3, -5, -6 and -9, respectively—increased the cleavage of E-cadherin by γ-secretase (Figure 3B). To verify the activity of the caspase inhibitors, cleavage of the caspase substrate PARP was also examined. The induction of PARP cleavage by STS was inhibited by the inhibitors of caspases-1, -3, -5, -6 and -9 (Figure 3B). Collectively, our observations suggest that the γ-secretase-mediated E-cadherin cleavage in apoptotic T47D cells is mediated by caspases-2 and -8.

Figure 3. γ-secretase-mediated E-cadherin cleavage in cells exposed to staurosporine (STS) is due to caspase activation. (A) T47D cells were pre-incubated for 1 hour with DAPT (2.5 μM), GM6001 (2.5 μM), or zVAD (100 μM). Cells were treated with STS for 6 hours to induce apoptosis, and extracts were analyzed by Western blotting with anti-E-cadherin antibody (C36). Actin was used as a loading control. (B) T47D cells were pre-incubated for 1 hour with inhibitors of caspases-1, -3, -5, -6, -8 and -9. Cells were treated with STS for 6 hours to induce apoptosis, and extracts were analyzed by Western blotting with anti-E-cadherin antibody (C36) or anti-PARP antibody. (C) T47D cells were pre-incubated for 30 minutes with GM6001 (c), DAPT (d), zVAD (e), or z-IETD-fmk (f) and then treated for 6 hours with 1 μM STS. Cells were fixed in 4% paraformaldehyde and stained with anti-cytoplasmic E-cadherin antibody (C36).

STS removes E-cadherin from the cell surface

The STS-mediated cleavage of cytoplasmic E-cadherin domains proximal to the transmembrane region would be expected to release the E-cadherin CTF fragment into the cytosol. Immunofluorescence microscopy was used to determine whether DAPT or the caspase inhibitors maintained E-cadherin at the cell surface in STS-treated cells. E-cadherin was primarily located at the cell–cell contacts in T47D cells (Figure 3C). The staining for E-cadherin at the plasma membrane of T47D cells with an anti-E-cadherin antibody (C36) was reduced and became diffuse after the induction of apoptosis. STS disrupted cell–cell adhesion and reduced plasma membrane cytoplasmic E-cadherin staining in T47D cells. Significantly, more E-cadherin was localized in the cytoplasm in T47D cells after STS treatment (Figure 3C). Pre-incubation with the MMP inhibitor GM6001 did not delay the loss of E-cadherin staining at the cell membrane after STS treatment (Figure 3C). In contrast, pre-incubation with zVAD or a caspase-8 inhibitor significantly inhibited the loss of E-cadherin staining at the cell membrane (Figure 3C). Furthermore, DAPT treatment caused E-cadherin to be retained at cell–cell contacts (Figure 3C). Our observations suggest that treatment with caspase inhibitors (zVAD or a caspase-8 inhibitor) or a γ-secretase inhibitor (DAPT) limited E-cadherin cleavage and disassembly of the adherens junction in during STS-induced apoptosis.

STS reduced E-cad/FL levels via γ-secretase in A431 cells

Although STS increased E-cadherin cleavage in breast cancer cells, the alteration in the full-length E-cadherin (E-cad/FL) level was modest. Therefore, we examined changes in the E-cad/FL levels in A431 cells. STS enhanced the proteolysis of E-cadherin in A431 cells. Treatment with STS for 6 hours markedly reduced levels of E-cad/FL (Figure 4). The γ-secretase inhibitor DAPT completely blocked the generation of the E-cad/CTF2 by STS in A431 cells (Figure 4A). Finally, we investigated the changes in cell surface E-cadherin levels in cells treated with STS. The staining for E-cadherin at the cell surface of A431 cells with anti-E-cadherin antibody (C36) was reduced (Figure 4B).

Figure 4. Treatment with staurosporine (STS) induced the translocation of E-cadherin in A431 cells. (A) A431 cells were treated with STS for 6 hours, lysed in RIPA and blotted with anti-E-cadherin antibody (C36). (B) STS-treated cells were fixed in 4% paraformaldehyde and stained with anti-cytoplasmic E-cadherin antibody (C36).

Discussion

Previous studies have shown that cadmium-mediated apoptosis induced the disappearance of E-cadherin from the cell surface in breast cancer cells via a caspase-independent pathway (Park et al., 2008). In the present study, we showed that the disappearance of E-cadherin during STS-induced apoptosis was mediated by caspases. Specifically, we present evidence that γ-secretase-dependent E-cadherin cleavage occurs downstream of the activation of caspase-2 and/or caspase-8 by STS. We also demonstrated that caspase-8 inhibition blocked the disappearance of E-cadherin from the cell surface.

Apoptosis is a critical process controlling cancer development. Although there is growing evidence for the disappearance of E-cadherin during apoptosis in cancer cells, the mechanisms involved remain unclear. The cleavage of E-cadherin and redistribution of fragments via γ-secretase appear to occur as a direct result of apoptosis cascade activation. Despite evidence suggesting a connection between E-cadherin and apoptosis, there has, until now, been little evidence as to whether caspases affect E-cadherin processing. Our observations not only confirm the activation of γ-secretase during apoptosis, but also provide a possible molecular mechanism for how apoptosis induces E-cadherin cleavage via γ-secretase at cell-cell contacts. Interestingly, whereas several previous reports (Keller & Nigam, 2003; Steinhusen et al., 2001) showed that E-cadherin cleavage in response to stimuli including STS and antimycin/deoxyglucose did not occur in MCF-7 breast cancer cells (which lack caspase-3), we clearly demonstrated that STS-mediated apoptosis induced E-cadherin cleavage via γ-secretase in MCF-7 breast cancer cells. Our results clearly showed that E-cadherin cleavage during apoptosis was not due to caspase-3, in spite of the fact that previous studies suggested that caspase-3 may be involved in the cleavage of E-cadherin during apoptosis (Keller & Nigam, 2003; Steinhusen et al., 2001). Interestingly, we previously reported that the cleavage of E-cadherin in response to cadmium treatment was due not to caspase activation, but to ROS generation. However, we clearly demonstrated that caspase activation was important for the cleavage of amyloid precursor protein (APP), a substrate of γ-secretase (Chae et al., 2010). It is important to examine whether caspases are involved in the cleavage of E-cadherin during apoptosis in T47D cells.

Of the two major caspase activation pathways that have been described for apoptosis, one involves the activation of death domain receptors (e.g., TNF-R, Fas) and results in the activation of the apoptosis initiator caspase-8 (Muzio et al., 1996). Caspase-8 is implicated in apoptosis in breast cancer cells. It is noteworthy that disappearance of E-cadherin from the cell surface during apoptosis is due to caspase-8. Given its role as an initiator during apoptosis, diverse studies have examined polymorphisms in the caspase-8 gene in the context of breast cancer risk (Zhang et al., 2011). Although there is evidence for crosstalk between caspase pathways, such that caspase-8 can be activated by caspase-3 (Tang et al., 2000), our results suggest that γ-cleavage-dependent E-cadherin cleavage is due to caspase-2/-8. Recent studies showed that E-cadherin-positive breast cell lines were less sensitive to STS than E-cadherin-negative ones (Wang et al., 2009). Downregulation of E-cadherin expression is often observed in various tumors and support its role in suppressing cell growth. Tumor suppression is also achieved by beginning of cell death as well. Although E-cadherin expression in plasma membrane was significantly decreased in T47D cells by STS, the E-cad/FL level was not significantly decreased (Figure 3). In the present studies, we found that the E-cad/FL levels on T47D cells (Figure 2) were less sensitive to STS compared to those on E-cadherin-positive A431 cells (Figure 4). These observations suggest the possible sensitivity of breast cancer to STS due to E-cadherin cleavage. Although STS did not significantly alter E-cad/FL levels, STS markedly increased the disappearance of E-cadherin from the T47D cell surface. It seems that E-cadherin cleavage as a result of γ-secretase activation partly contributes to the disappearance of E-cadherin from the cell surface in STS-treated T47D cells.

In conclusion, we have provided evidence for caspase-mediated E-cadherin processing in breast cancer cells during apoptosis. Caspases-2 and -8 may be key regulators of E-cadherin that act via γ-secretase.

Acknowledgments

This study was supported by Korea National Institute of Health Intramural Research Grant (091 - 4845-300 - 210-13) to Dr. Y.H. Koh.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. The authors declare no competing financial interests.