The role of tyrosine kinase Etk/Bmx in EGF-induced apoptosis of MDA-MB-468 breast cancer cells


Etk/Bmx, a member of the Tec family of tyrosine kinases, mediates various signaling pathways and confers several cellular functions. In the present study, we have explored the functional role of Etk in mediating EGF-induced apoptosis, using MDA-MB-468 cell line as a model. We first demonstrated that EGF treatment induces Etk tyrosine phosphorylation in both HeLa and MDA-MB-468 cells. Overexpression of Etk by recombinant adenovirus in MDA-MB-468 cells potentiates the extent of EGF-induced cell apoptosis. The observed Etk-enhanced MDA-MB-468 cell apoptosis is associated with the Stat1 activation, as demonstrated by electrophoresis mobility shift assays and reporter gene assays. By contrast, a kinase domain deletion mutant EtkΔK, functioning as a dominant-negative mutant, ameliorates EGF-induced Stat1 activation and apoptosis in MDA-MB-468 cells. To explore whether the activated Etk alone is sufficient for inducing apoptosis, a conditionally activated Etk (ΔEtk-ER), a chimeric fusion protein of PH domain-truncated Etk and ligand-binding domain of estrogen receptor, was introduced into MDA-MB-468 cells. Upon β-estradiol ligand activation, the ΔEtk-ER could stimulate Stat1 activity and confer cell apoptosis independent of EGF treatment. Taken together, our findings indicate that Etk is a downstream signaling molecule of EGF receptor and suggest that Etk activation is essential for transducing the EGF-induced apoptotic signaling.


It is well recognized that EGF and TGFα are among the most potent growth factors (Wells, 1999; Jorissen et al., 2003; Suhardja and Hoffman, 2003), which, by engaging the EGF receptor (EGFR), transmits signals for growth, survival, and, in some cases, motility (Misra and Pizzo, 1998; Price et al., 1999; Tsang and Crowe, 1999; Yart et al., 2001). Overexpression of EGF receptor is common among human epithelial malignancies (Merlino et al., 1985; Derynck et al., 1987; Costa et al., 1988; Sainsbury et al., 1988; Kikuchi et al., 1990; Ibrahim et al., 1997). Amplification and mutation of the EGFR locus have also been found in a variety of tumors including glioblastoma (Xu et al., 1984; Wong et al., 1987; Ro et al., 1988; Ekstrand et al., 1991; Garcia de Palazzo et al., 1993). Yet, paradoxically, in some EGF-receptor-overexpressing cancer cells exemplified by breast cancer cell line MDA-MB-468 and vulvar carcinoma-derived cell line A431, EGF treatment induces growth arrest and apoptosis (Armstrong et al., 1994; Gulli et al., 1996). The signals that trigger apoptosis are likely to be different from those involved in other EGF responses. For example, the PI3 kinase pathway and Ras-Raf-MAPK kinase pathways activated by EGFR have been shown to mediate cell survival and mitogenic responses (Misra and Pizzo, 1998; Tsang and Crowe, 1999; Yart et al., 2001). PI-3 kinase and phospholipase C-dependent pathways were reported to mediate EGF-induced cell migration (Price et al., 1999). Activation of c-Src and Stat3 by EGF was shown to be associated with cellular transformation in certain cell types (Maa et al., 1995; Berclaz et al., 2001). Little is known about the pathway leading to apoptosis. Fu and coworkers provided some insights into this process. These authors showed that EGF-induced growth arrest and apoptosis correlate with the activation of Stat1, with consequent activation of p21/WAF1 and caspase 1 (Chin et al., 1997; Thomas et al., 1999). In addition, Darnell and colleagues demonstrated that a dominant-negative mutant of Stat1 inhibits EGF-induced growth arrest in A431 cells, suggesting the essential role of activated Stat1 in EGF-evoked arrest of growth (Bromberg et al., 1998). How Stat1 is activated by overexpressed EGFR, however, is not clear. In this report, we provide evidence that tyrosine kinase Etk/BMX is involved in the EGF-induced activation of Stat1 and apoptosis.

Etk/Bmx is a member of the Tec family nonreceptor tyrosine kinases, including Btk, Itk, Tec, and Txk (reviewed in Qiu and Kung, 2000; Lewis et al., 2001; Smith et al., 2001). Members of this family share an amino-terminal pleckstrin homology (PH) domain, a Tec homology (TH) domain, a SH3 domain, a SH2 domain, and a catalytic domain. The expression of Tec family kinases has been primarily identified in the hematopoietic cells. Btk is preferentially expressed in B cells and is necessary for B-cell development (Kerner et al., 1995; Khan et al., 1995). Mutations in the Btk gene result in human X-linked agammaglobulinemia and murine X-linked immunodeficiency (Tsukada et al., 1993; Satterthwaite et al., 1998; Vihinen et al., 2000). Likewise, Itk is predominately expressed in T cells and is crucial in T-cell development and T-cell receptor-initiated signaling cascade (Bunnell et al., 2000; Czar et al., 2001; August et al., 2002; Miller and Berg, 2002). Tec is expressed in most hematopoietic cells, and is involved in a variety of signal pathways including interleukin-3, stem-cell factor, and granulocyte colony-stimulating factor (Tang et al., 1994; Mano et al., 1995; Matsuda et al., 1995). Etk/Bmx was initially identified in bone marrow and subsequently in epithelial cells, fibroblast, and endothelial cells (Tamagnone et al., 1994; Robinson et al., 1996; Qiu et al., 1998; Tsai et al., 2000). Previous studies showed that Etk plays a pivotal role in IL-6-induced neuroendocrine differentiation and neuropeptide-induced androgen-independent growth of prostate cancer cells (Qiu et al., 1998; Lee et al., 2001). In prostate cancer cells, Etk can be activated by PI-3 kinase and protects LNCaP from radiation-induced apoptosis (Qiu et al., 1998; Xue et al., 1999). Interestingly, upon caspase cleavage, the constitutively activated Etk becomes proapoptotic (Wu et al., 2001). Etk has also been shown to mediate Src-induced transformation of epithelial cells and fibroblasts through activation of Stat3, and to transform the breast cancer cells via activation of Pak1 (Tsai et al., 2000; Bagheri-Yarmand et al., 2001). In addition, Etk was also reported to be activated by IL-3 and GM-CSF in mast cells and by VEGF and angiopoietin in endothelial cells (Ekman et al., 2000; Rajantie et al., 2001; Chau et al., 2002). Most recently, Etk has been found to play a critical role in the cell migration of endothelial cells in the signaling of integrin and TNF receptor type 2 (Chen et al., 2001; Pan et al., 2002). Thus, Etk seems to be an integrator of several signals with diverse outcomes.

We report here that Etk is activated by EGF treatment of breast cancer cell MDA-MB468, and is involved in EGF-induced apoptosis. Overexpression of Etk sensitizes cells toward EGF-induced apoptosis, while expression of a dominant-negative Etk blocks such as a process. We also found that Etk expression activates Stat1 activity, which in turn activates the expression of p21, accounting for the growth arrest and apoptosis induced by EGF in this cell type. Our studies provide a mechanism whereby overexpressed EGF receptor induces apoptosis.


EGF stimulates tyrosine phosphorylation of Etk

To investigate whether Etk is a downstream signaling molecule of EGFR in vivo, both HeLa cells and MDA-MD-468 cells were treated with 100 ng/ml EGF for various time periods as indicated, and assayed for Etk tyrosine phosphorylation, a hallmark of Etk activation. Cells were harvested and then equal amount of cell lysates was subjected to Western blot analyses with an antibody against tyrosine-phosphorylated Etk. Upon EGF treatment, the tyrosine phosphorylation of endogenous Etk in HeLa cells increased within 5 min and rapidly declined back to the basal level at 30 min post-treatment (Figure 1a). Likewise, EGF treatment also elicited a rapid Etk activation, except that the duration of Etk activation was sustained at least up to 120 min (Figure 1b). To further substantiate this prolonged EGF-induced Etk activation, MDA-MB-468 cells were infected with a recombinant adenovirus expressing T7-tagged Etk (Ad-Etk) or a control virus (Ad-vec). After 48 h, cells were stimulated with EGF and followed by the analyses of Etk activation. Similar to endogenous Etk, ectopically expressed T7-tagged Etk was persistently activated up to 120 min, as evidenced by increased tyrosine phosphorylation levels, in EGF-treated cells (Figure 1c). Together, these findings demonstrate that Etk is activated in cells in response to EGF treatment and this activation is not a cell-type-specific event. However, the duration of the Etk activation was more pronounced in MDA-MB-468 cells, expressing high levels of EGF receptor, than that in the HeLa cells, expressing low levels of EGF receptor (Filmus et al., 1985).

Figure 1

EGF stimulates tyrosine phosphorylation of Etk in HeLa cells and MDA-MB-468 cells. Confluent cultures of HeLa cells or MDA-MB-468 cells were stimulated with 100 ng/ml EGF for the indicated times. A measure of 40 μg of cell lysates from each sample was subjected to Western blot analysis with anti-phospho-Etk antibody (Cell Signaling), anti-Etk antibody (Cell Signaling) or antiactin antibody (Sigma). (a, b) represent the results of Western blot from HeLa cell lysates and MDA-MB-468 cell lysates, respectively. (c) EGF induces the phosphorylation of ectopically expressed Etk. MDA-MB-468 cells were infected with 10 MOI of the T7-tagged Etk recombinant adenovirus. After 48 h, cells were treated with EGF 100 ng/ml for the indicated period, lysed, and subjected to immunoprecipitation with anti-T7 antibody, followed by Western blot analysis with anti-T7 antibody and antiphosphotyrosine antibody (4G10, Upstate Biotech). (d) Activation of Etk by EGF stimulation could be blocked by specific inhibitors against EGFR, Src, and PI3-kinase. MDA-MB-468 cells were infected with 10 MOI of the T7-tagged Etk recombinant adenovirus. After 48 h, cells were pretreated with 5 μ M AG1498, 2 μ M PP2 or 10 μ M LY294002 for 30 min, followed by EGF stimulation for another 30 min. Cells were then harvested and subjected to immunoprecipitation with anti-T7 antibody and immunoblotting with anti-T7 and antiphosphotyrosine antibodies

To further demonstrate that the activation of the Etk induced by EGF is EGFR-dependent, an EGFR-specific inhibitor, AG1498, was included in our experiments. As expected, the EGF-induced phosphorylation of the Etk was markedly abrogated in AG1498-treated MDA-MB-468 cells (Figure 1d, lane 3 versus lane 4), suggesting an essential role of the EGFR in Etk activation. Etk was previously reported to be activated by PI3-kinase and/or Src kinase (Qiu et al., 1998; Tsai et al., 2000). To test whether these kinases are also involved in the EGF/EGFR-mediated Etk activation, MDA-MB-468 cells were treated with an Src kinase inhibitor, PP2, or a PI3-kinase inhibitor, LY294002, prior to the EGF stimulation. As shown in Figure 1d, both inhibitors could attenuate the level of Etk tyrosine phosphorylation, indicating that both Src and PI3-kinases partially mediate the Etk activation in response to EGF treatment.

Etk potentiates EGF-induced apoptosis of MDA-MB-468 cells

It has been shown that EGF stimulation induces apoptosis in MDA-MB-468 cells (Armstrong et al., 1994). Since EGF-induced Etk activation exhibited a sustained manner in MDA-MB-468 cells, we next examined whether Etk is involved in EGF-elicited apoptosis in MDA-MB-468 cells. To test this possibility, MDA-MB-468 cells were infected with a recombinant adenovirus expressing T7-tagged Etk (Ad-Etk) or a control virus (Ad-vec). After 48 h, cells were stimulated with EGF and followed by the analyses of cell apoptosis. EGF-treated MDA-MB-468 cells displayed markedly morphological changes, reflecting adherent cell apoptosis, such as cell condensation, rounding up, and detachment from the culture dishes (Figure 2a). Particularly, the extent of cell detachment was more prominent in EGF-treated and Ad-Etk-infected MDA-MB-468 cells. This is not due to the adenovirus infection per se, since the same titer of the vector-infected cells displayed a lower level of cell detachment comparable to mock-infected cells. The extent of cells undergoing apoptosis was further quantified by the propidium iodide (PI) staining followed by the FACS analysis of the sub-G1 fraction. As shown in Figure 2b, the amount of EGF-induced apoptosis was significantly increased to 30% in cells overexpressing Etk, compared to 10% in mock and vector-infected cells. These findings suggest that the overexpression of Etk could potentiate the EGF-induced MDA-MB-468 apoptosis. Intriguingly, in the absence of EGF-treatment, Etk overexpression alone was sufficient to confer cell apoptosis to a level comparable to that observed in EGF-treated control cells. This is probably due to the activation of Etk kinase activity by its overexpression (Figure 1c, lane 2), leading to the activation of the apoptotic signaling pathway(s) in MDA-MB-468 cells.

Figure 2

Etk potentiates EGF-induced apoptosis. Overexpression of Etk potentiates EGF-induced cell apoptosis. MDA-MB-468 cells were infected without or with Ad-vec or Ad-Etk, followed by EGF 100 ng/ml treatment for 48 h. The culturing cells were then observed using a digital camera mounted on a microscope with × 100 magnification (a). The detached and attached cells were further harvested for apoptosis analysis. Cells were first fixed with ethanol and followed by the staining of fluorescent PI dye. The resulting stained cells were analysed by FACS to measure cells in the subG1 population. The percentage of the apoptotic cells represents the mean value±standard deviation from four independent experiments (b)

Etk potentiates EGF-induced Stat1 activation

Previously, it has been demonstrated that the induction of Stat1 activity and p21 expression is essential and indispensable for EGF-induced MDA-MB-468 cell apoptosis (Chin et al., 1997; Thomas et al., 1999). Hence, we next explored the possibility that the effect of Etk on the EGF-induced apoptosis is mediated through Stat1 activation. To test this possibility, we have performed electrophoresis mobility shift assays with a 32P-labeled SIE oligonucleotide to assay for Stat1 activation in the respective nuclear extract prepared from cells expressing Etk in the absence or presence of EGF stimulation. As a control, EGF treatment apparently induced the DNA–protein complex formation, including Stat3–Stat3 homodimer, Stat1–Stat3 heterodimer, and Stat1–Stat1 homodimer, in both mock and vector-infected cells (Figure 3a, lanes 2–5). Among these complexes, the formation of Stat1–Stat3 and Stat1–Stat1 dimers was robustly increased in Etk-overexpressing cells (lane 7), suggesting that Etk is capable of enhancing EGF-induced Stat1 activation. To further substantiate this notion, we have performed the Stat1-responsive reporter gene assays. An approximate 25-fold induction of Stat1-mediated reporter activation was detected in Etk-overexpressing cells by EGF treatment, whereas a ninefold induction was observed in the control cells (Figure 3b, left panel, lanes 2 and 4). In addition, overexpression of Etk alone was sufficient to activate the Stat1 reporter (lane 3) to the level comparable to EGF-treated control cells. These results confirm that Etk induces Stat1 activity and promotes the EGF effect on Stat1-mediated transcriptional activation. A similar scenario was also observed using p21 promoter linked to a luciferase reporter (Figure 3b, right panel). Taken together, these findings indicate that the effect of EGF on cell apoptosis could be mediated by Etk via the Stat1 activation and p21 induction.

Figure 3

Etk potentiates EGF-induced Stat1 activation. (a) Etk enhances the Stat1 activity stimulated by EGF. MDA-MB-468 cells were infected with Ad-vec or Ad-Etk for 2 days, and either left untreated or treated with EGF (100ng/ml) for 30 min. Nuclear extracts prepared from these cells were incubated with 32P-labeled SIE probe and subjected to gel mobility shift assay (see the experimental procedures detailed in text). Arrows indicate the position of specific protein–DNA complexes and free probe. Some nonspecific complexes are indicated by asterisks. (b) Etk potentiates the effect of EGF on the Stat1-mediated and p21 promoters. Cells were transfected with 1 μg of p3xLy6E-Luc or p21-Luc reporter, in the presence of either 1 μg of pcDNA3 empty vector or pCMV-T7-Etk. After 2 days, cells were treated with EGF for 8 h and then subjected to luciferase reporter assay (see the experimental procedures detailed in text). The reporter activity is represented as mean±s.d., based on three independent transfection experiments

A dominant-negative mutant of Etk inhibits EGF-induced Stat1 activation and cell apoptosis

To further demonstrate the necessity of Etk in EGF-induced Stat1 activation and apoptosis, we investigated the effect by a dominant-negative mutant of Etk on EGF-dependent events. We first constructed a kinase deletion mutant of Etk, EtkΔK (Figure 4a), and tested its efficacy in functioning as a dominant-negative mutant in Stat1-mediated reporter gene activation. As shown in Figure 4b, the wild-type Etk potentiated Stat1 activity, in a dose-dependent manner, in response to EGF treatment (lanes 4, 6, 8, and 10 versus lane 2). By contrast, increased amount of EtkΔK proportionally inhibited EGF-induced Stat1 transactivation (Figure 4b, lanes 12, 14, 16, and 18 versus lane 2), indicating that EtkΔK functions as an Etk dominant-negative mutant. We next tested whether this mutant could attenuate EGF-induced apoptosis using the recombinant Ad-EtkΔK. As expected, the EGF-induced MDA-MB-468 apoptosis was drastically reduced in EtkΔK-overexpressing cells (Figure 4c). Altogether, these observations suggest that Etk plays an essential role in mediating apoptotic pathway activation in EGF-treated MDA-MB-468 cells.

Figure 4

A dominant-negative Etk inhibits EGF-induced Stat1 activation and apoptosis. (a) Schematic presentation of different Etk mutants assayed in mammalian transfection experiments. ΔEtk-ER represents the fusion protein between the ligand-binding domain of the ER and an Etk truncated protein lacking the first 68 amino-acid residues of the PH domain. (b) EtkΔK blocked EGF-induced Stat1 activation. MDA-MB-468 cells were transiently transfected with 1 μg of p3xLy6E-Luc reporter along with either pcDNA3 empty vector or increasing amount of pCMV-T7-Etk or pCMV-EtkΔk (0.1, 0.2, 0.5, and 1 μg). After 2 days, cells were treated with EGF 100 ng/ml for 8 h and then subjected to luciferase assay. (c) EtkΔK abrogated EGF-induced apoptosis. MDA-MB-468 cells were infected with Ad-vec, Ad-Etk, or Ad-EtkΔK, followed by EGF 100 ng/ml treatment for 48 h. Cells were then subjected to apoptosis analysis as described in Figure 2. Western blot showing the expression of T7-tagged Etk and T7-tagged EtkΔK in cells infected with Ad-Etk and EtkΔK, respectively, was analysed on a 10% SDS–polyacrylamide gel transferred to Immobilon-P membrane, and probed by the anti-T7 antibody

Conditionally activated Etk induces both Stat1 activation and cell apoptosis

Since the kinase domain of Etk is critical for EGF-evoked Stat1 activation and apoptosis, we wondered whether the activation of Etk kinase activity, in the absence of EGF stimulation, is sufficient to activate Stat1 and p21, leading to cell apoptosis. To test this possibility, we utilized an inducible Etk (ΔEtk-estrogen receptor (ER)), a chimeric fusion of a PH domain-truncated Etk linked to the ligand-binding domain of the ER, to activate Etk activity conditionally. Upon the β-estradiol (E2) treatment, the kinase activity of ΔEtk-ER was activated in a time-dependent fashion in MDA-MB-468 cells infected with adenovirus carrying ΔEtk-ER, as evidenced by Western blot analysis of the increased amount of tyrosine phosphorylation of ΔEtk-ER (Figure 5a). The inducible activation of ΔEtk-ER was in a sustained manner in MDA-MB-468 cells, as previously reported by Wen et al. (1999), in Pa-4 and A549 cells. We next assessed whether the ΔEtk-ER activation alone could stimulate gene expression, independent of EGF treatment, by electrophoresis mobility shift assays and transient transfection of reporter assays. Upon the E2 treatment, the extent of detected SIE–Stat1 complexes in nuclear extracts from cells infected with Ad-ΔEtk-ER was comparable to that observed in EGF-treated control cells (Figure 5b, lanes 5 and 8). The increased Stat1 activity was primarily due to the Etk activation, but not the E2 treatment, since the observed Stat1 activation was not detectable in the control cells treated with E2 (lane 3). Furthermore, ΔEtk-ER was capable of activating both Stat1 and p21 reporter gene activities in the presence of E2 (Figure 5c, lane 6 in both panels). In the presence of EGF, the activation of Stat1 in ΔEtk-ER-infected cells was further enhanced (Figure 5c, lane 8). This probably resulted from the activation of endogenous Etk (Figure 1) and/or additional potentiation of ΔEtk-ER upon EGF treatment (Figure 5c, bottom panel, lane 4). It should be noted that, in the absence of E2, we failed to detect the tyrosine-phosphorylated-ΔEtk-ER in EGF-treated cells (Figure 5c, lane 2, lower panel), suggesting that further activation of ΔEtk-ER by EGF-treatment is dependent on the activated state of ΔEtk-ER. It is possible that the initial activation by E2 renders the Etk in ΔEtk-ER accessible for further activation by EGF-evoked signaling. Taken together, these findings suggest that activation of Etk kinase by itself is sufficient for induction of Stat1 activity.

Figure 5

Activation of Etk kinase activity is sufficient to induce the Stat1 activity and cell apoptosis. (a) Conditional activation of Etk kinase activity in MDA-MB-468 cells. MDA-MB-468 cells were infected with recombinant adenovirus expressing ΔEtk-ER. After 48 h, cells were treated with 1 μ M β-estradiol (E2) for various time points as indicated. Equal amounts of total cell extracts were immunoprecipitated with an anti-ER antibody, followed by Western blot analysis with antiphosphotyrosine antibody (4G10) and anti-ER antibody. (b) Conditionally activated Etk induces Stat1 activity. Cells were infected with recombinant viruses of Ad-vec or Ad-ΔEtk-ER for 2 days, and either left untreated or treated with E2 for 6 h. For the control of EGF stimulation, Ad-vec-infected cells were treated with EGF (100 ng/ml) for 30 min. Nuclear extracts prepared from these cells were incubated with 32P-labeled SIE probe and subjected to gel mobility shift assay (see the experimental procedures detailed in text). The arrows indicate the position of specific protein–DNA complexes and free probe. Some nonspecific complex is indicated by an asterisk. To demonstrate the DNA–protein complexes containing Stat1 and Stat3, anti-Stat1 and anti-Stat3 antibodies were included in the preincubation, and the supershifted complexes are indicated by arrowheads. (c) Conditionally activated Etk stimulates Stat1-mediated and p21 promoters. Cells were transiently transfected with 1 μg of p3xLy6E-Luc or p21-Luc, in the presence of either 1 μg of pcDNA3 vector or pCMV-ΔEtk-ER expression construct. After 48 h, cells were treated with E2 or EGF for 8 h. One portion of cells was harvested for luciferase assay as described in Figure 3, and the other portion of cells was subjected to immunoprecipitation with anti-ER antibody, followed by immunoblotting with antiphosphotyrosine antibody and anti-ER antibody at the bottom panel. (d) Conditionally activated Etk induces cell apoptosis. Cells infected with Ad-vec or Ad–ΔEtk-ER were treated with a combination of EGF and E2 as indicated. Cells were harvested on days 1, 2 or 3, and subjected to apoptosis analysis as described in Figure 2

We next sought to determine whether conditionally activated ΔEtk-ER could induce MDA-MB-468 cell apoptosis. As expected, upon E2 stimulation, cells expressing ΔEtk-ER exhibited cell apoptosis to a level comparable to EGF-treated control cells (Figure 5d), suggesting that activation of Etk is sufficient for triggering MDA-MB-468 cell apoptosis. However, EGF treatment of activated ΔEtk-ER cells further promoted the apoptosis. Together, our findings clearly provide a functional role of Etk in mediating EGF-induced Stat1 activation and cell apoptosis in MDA-MB-468 breast cancer cells.


In the present study, we provide evidence that Etk plays an important role in mediating the EGF-elicited apoptotic signals. First, EGF stimulation rendered Etk tyrosine phosphorylation, suggesting that Etk is a downstream signaling molecule of EGFR. Secondly, overexpression of Etk led to cell death and enhanced the ability of EGF to induce apoptosis via the activation of Stat1 and the induction of p21 promoter activity. Thirdly, a dominant-negative Etk, EtkΔK, abrogated the afore-mentioned effects of EGF on gene regulation and apoptosis. Finally, conditionally activated Etk could mimic EGF effect on Stat1 activation and p21 promoter, leading to the cell apoptosis. To our knowledge, this is the first report in which activation of Etk is able to induce apoptosis in breast carcinoma cells.

The biological outcomes of the various signals, including growth, differentiation, and apoptosis, transmitted by the Tec family kinases, are in a cell context-dependent fashion (Hsueh and Scheuermann, 2000). There are a number of reports suggesting a role of Tec family kinases in growth and antiapoptosis. For example, Etk has been shown to be a critical mediator of cellular transformation through a cascade link of v-Src-Etk-Stat3 (Tsai et al., 2000) in fibroblast and epithelial cells. Overexpression of Etk stimulates the proliferation of MCF-7 breast cancer cells and protects prostate cancer cells from ionization- and thapsigargin-induced apoptosis (Xue et al., 1999; Bagheri-Yarmand et al., 2001). Targeted disruption of Btk has been shown to lead the B cells toward apoptosis (Anderson et al., 1996) and overexpression of Btk in chicken B cells suppresses the Fas-mediated apoptotic signal via the disruption of the FAS–FADD interaction. By contrast, Btk has also been demonstrated to induce apoptosis in HeLa cells (Islam et al., 2000) and to mediate the radiation-induced apoptosis in DT-40 lymphoma B-cells (Uckun et al., 1996). Our findings reported here, that Etk sensitizes MDA-MB-468 cells toward EGF-induced apoptosis, provide additional evidence that Tec family kinases are involved in proapoptotic functions. These data together suggest that Tec family kinases are bidirectional switches of cell growth and apoptosis. While the mechanism of this switch is presently unclear, we noted that Etk assumes two forms inside the cells: the wild-type form which requires upstream signals such as PI-3 kinase, FAK, PTPD1, and Src for its activation (Ekman et al., 2000; Jui et al., 2000; Tsai et al., 2000; Chen et al., 2001), and the caspase-8 truncated form which is constitutively active. We previously showed that the constitutively activated form induced apoptosis (Wu et al., 2001). It is possible that sustained active Etk, either by very strong upstream signals as overexpressed EGF receptor or by caspase cleavage, drives cells to the apoptosis pathway. The ΔEtk-ER used in this study mimics the caspase-activated Etk, and its activation alone is sufficient to induce apoptosis, consistent with this hypothesis. The fact that EGF treatment of ΔEtk-ER cells gave rise to superinduction of apoptosis suggests that the ΔPH-Etk can still respond to EGF receptor signals such as Src, consistent with the increased tyrosine phosphorylation of ΔPH-Etk upon EGF treatment.

As to how hyperactive Etk activity leads to apoptosis is not completely understood. We found that Etk's ability to induce apoptosis and to activate Stat activities (based on SIE-gel shift or SIE reporter assays) seemed to go hand-in-hand. Furthermore, dominant-negative kinase-dead EtkΔK, which negated EGF-induced apoptosis, also downmodulated SIE promoter. This observation is echoed by the finding that the kinase activity-defective mutant of Btk is compromised by its capacity in mediating HeLa cell apoptosis (Islam et al., 2000). Both studies suggest that the kinase activity is crucial for transducing the apoptotic signals to the downstream effectors. The downstream signaling pathway of Btk-mediated apoptosis in HeLa cells is through p38 MAPK activation (Islam et al., 2000), whereas that of Etk-mediated apoptosis in MDA-MB-468 cells is through Stat1 activation. Recently, p38 MAPK has been demonstrated to enhance Stat1-dependent gene transcription and apoptosis through serine 727 phosphorylation (Stephanou et al., 2001, 2002; Ramsauer et al., 2002). Thus, there is a likelihood that a common pathway is utilized. Further experiments are required to support or refute this notion.

In addition to MDA-MB-468 cells, EGFR overexpression leading to ligand-dependent apoptosis has also been observed in other cell types, including A431, 293, mesenchymal Rat-1 fibroblasts, U-87 MG, and C6 glial cells (Chin et al., 1997; Hognason et al., 2001). Furthermore, Hognason et al. (2001) have demonstrated that EGFR overexpression also induces a ligand-independent apoptosis in 293 EBNA cells, suggesting that apoptosis is a direct result of EGFR overexpression. Although the correlation between EGFR overexpression and cell apoptosis has been well established in these cell lines, the underlying mechanism for EGFR-induced apoptosis is largely unclear. Studies in MDA-MB-468 and A431 cells revealed that activation of Stat1 is important in EGF-induced cell growth arrest and apoptosis (Chin et al., 1997). In these cell types, expression of p21/WAF1 and caspase 1 is upregulated by Stat1-dependent manner, leading to cell growth and apoptosis (Thomas et al., 1999). Stat1-mediated apoptosis was abrogated in Stat1-deficient cell line, suggesting that activation of Stat1 pathway is essential for induction of apoptosis (Chin et al., 1997). Chin et al. previously suggested that the ability of overexpressed EGFR to activate Stat1 was critical in EGF-induced apoptosis. We wish to extend this hypothesis by suggesting that the sustained activation of Etk that leads to sustained Stat1 is the underlying cause of the EGF-induced apoptosis. Consistent with this notion is the finding that, in HeLa cells, the activation of Etk is not sustained and the cells do not undergo apoptosis.

Recent studies revealed that Etk could be activated by a wide range of cell-surface receptors, through coupling with different signaling molecules. For example, PI3-kinase has been shown to be required for IL-6-mediated Etk activation in prostate cancer cells (Qiu et al., 1998). Both Src and FAK activities are necessary for Etk activation by neurotropic factor bombesin in LNCaP prostate cancer cells (Lee et al., 2001). Furthermore, integrin-engaged Etk activation in epithelial cells and endothelial cells has also been reported to be FAK dependent (Chen et al., 2001). Etk activated by these signaling molecules was thought to be via lipid–protein and protein–protein interactions. Presumably, binding of the lipid products of PI3-kinase to the PH domain of Etk recruits Etk to plasma membrane, where the active Src kinase is located in order to activate Etk through phosphorylation. Protein–protein interaction between FAK and the PH domain of Etk appears to disrupt the intramolecular interaction of Etk protein, leading to unfolding of the kinase domain and leaving Etk available for upstream regulators. Our results that both Src and PI3-kinase inhibitors partially blocked the Etk activation by EGF treatment (Figure 1d) suggest that the activation of the Etk by EGFR is at least through the signaling pathways mediated by PI3-kinase and Src kinase. These findings do not exclude the possibility that other signaling molecules such as FAK are also involved in the EGFR-Etk activation. For instance, FAK could form complexes with EGFR in the absence of EGF in A431 cells (Lu et al., 2001). David et al (Sieg et al., 2000) have demonstrated that FAK forms a complex with the activated EGFR and is required for EGF-stimulated cell motility. Currently, we are in the process of elucidating the possible pathways in the EGFR-mediated Etk activation in MDA-MB-468 cells.

In summary, we have demonstrated that Etk functions as an important signaling molecule in the process of regulating EGF-induced cell apoptosis of MDA-MB-468 breast cancer cells. More importantly, the conditionally activated Etk alone is sufficient to trigger the apoptosis of such breast cancer cells. Thus, our studies provide a potential strategy for developing an effective treatment on breast cancers overexpressing EGFR.

Materials and methods

Plasmid construction

T7-tagged Etk lacking the kinase domain and ΔEtk-ER were amplified by polymerase chain reaction with the template of pCMV-T7-Etk and pLNCX-ΔEtk-ER (Wen et al., 1999), respectively, and cloned into pcDNA3 (Invitrogen) mammalian expression plasmid, to generate pCMV-T7-EtkΔK and ΔEtk-ER.

Recombinant adenovirus construction

The recombinant adenoviruses expressing T7-tagged Etk wild type and mutants were constructed and produced according to the manufacturer's instructions for the adenovirus expression vector kit (TaKaRa). Briefly, the cDNA coding the T7-Etk was inserted into the SwaI site of pAxCAwt cosmid vector. The resulting cosmid was then cotransfected with the complexes of adenovirus genomic DNA and terminal protein into 293 cells to generate the recombinant virus. After confirmation, each recombinant virus was subjected to virus amplification and titration. In all, 10 MOI of each recombinant virus was used to infect MDA-MB-468 cells.

Cell culture, Western blot analysis, transient transfection, and luciferase assays

HeLa cells and human adenocarcinoma cell line MDA-MB-468 were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum. For endogenous Etk phosphorylation, cells were plated into 10-cm Petri dishes and then treated with EGF 100 ng/ml for different times, as indicated. A measure of 40 μg of each cell lysate was subjected to Western blot analysis with anti-phospho-Etk (specifically recognized phosphorylation of the Etk Y40) and anti-Etk antibodies (Cell Signaling). For ectopic Etk phosphorylation, 10 MOI of Ad-Etk recombinant adenovirus was added to MDA-MB-468 cells. After 48 h, cells were stimulated with EGF for different time courses and then harvested for immunoprecipitation and Western blot analysis. For the inhibitor experiments, cells expressing T7-tagged Etk were pretreated with AG1498 (Sigma), PP2 (Calbiochem) or LY294002 (Sigma) for 30 min, followed by EGF stimulation for 30 min, and then harvested for immunoprecipitation and Western blot analysis. All the transient transfection experiments were performed by using FuGENE 6 transfection reagent (Roche Molecular Biochemicals), as instructed by the manufacturer. Cells (2 × 105) were transfected with an expression vector of wild-type T7-tagged Etk or its derived mutants, along with the p3xLy6E-Luc or p21-Luc reporter construct (a gift from Dr S-Y Hsieh) as well as pRL-TK plasmid as an internal control for normalization of transfection efficiency. Following transfection, cells were cultured in a medium containing charcoal-stripped 0.05% (V/V) FBS for starvation. Cells were further treated with 100 ng/ml EGF in combination with β-estradiol (E2) (1 μ M), as indicated. Cell lysates were harvested and assayed for relative activity (Firefly-luciferase for the reporter and Renilla luciferase for the indicator) as instructed by the manufacturer (Parkard BioScience).

Flow-cytometric analysis of apoptosis

Cells (5 × 105) were grown in six-well plates for recombinant adenovirus infection. After 24 h, they were treated with or without EGF (100 ng/ml) for different times. For analysis of cell apoptosis, detached and attached cells were pelleted and resuspended in 70% EtOH for fixation. After fixation, the cells were stained by the flurorescent dye PI for 30 min, and further analysed by FACS (Becton Dickinson, Franklin Lakes, NJ, USA)

Electrophoretic mobility shift assay (EMSA)

Human SIE (5′-IndexTermAGCTTCATTTCCCGTAAATCCCTAAAGCT-3′) was synthesized and annealed to its corresponding oligonucleotide to form a double strand. EMSA was performed according to the procedure described by Vignais et al. (Vignais and Gilman, 1999). Briefly, DNA–protein binding reactions (20 μl) were performed by incubation of the nuclear extracts (6 μg) in a solution containing 13 mM HEPES (pH7.9), 65 mM NaCl, 0.5 mM EDTA, 2% Ficoll 400, 5% glycerol, 50 μg of poly (dI-dC)/ml, and 0.5 mM dithiothreitol for 10 min at room temperature, followed by an addition of 30-min incubation with 50 000 c.p.m. of 32P-labeled human SIE probe at room temperature. For supershift assay, the nuclear extracts were preincubated with anti-Stat1 (Upstate Biotechnology) or anti-Stat3 (Upstate Biotechnology) antibody for 20 min before addition of SIE probe. The DNA–protein complexes were separated on 5% nondenaturing polyacrylamide gel in 0.5 × TBE buffer. Gels were dried and subsequently detected by phosphorimager analysis (Bio-imaging Analyser BSA 1500, Fuji).



epithelial and endothelial tyrosine kinase


epidermal growth factor

PI3 kinase:

phosphoinositol-3 kinase




estrogen receptor


  1. Anderson JS, Teutsch M, Dong Z and Wortis HH . (1996). Proc. Natl. Acad. Sci. USA, 93, 10966–10971.

  2. Armstrong DK, Kaufmann SH, Ottaviano YL, Furuya Y, Buckley JA, Isaacs JT and Davidson NE . (1994). Cancer Res., 54, 5280–5283.

  3. August A, Fischer A, Hao S, Mueller C and Ragin M . (2002). Int. J. Biochem. Cell. Biol., 34, 1184–1189.

  4. Bagheri-Yarmand R, Mandal M, Taludker AH, Wang RA, Vadlamudi RK, Kung HJ and Kumar R . (2001). J. Biol. Chem., 276, 29403–29409.

  5. Berclaz G, Altermatt HJ, Rohrbach V, Siragusa A, Dreher E and Smith PD . (2001). Int. J. Oncol., 19, 1155–1160.

  6. Bromberg JF, Fan Z, Brown C, Mendelsohn J and Darnell Jr JE . (1998). Cell Growth Differ., 9, 505–512.

  7. Bunnell SC, Diehn M, Yaffe MB, Findell PR, Cantley LC and Berg LJ . (2000). J. Biol. Chem., 275, 2219–2230.

  8. Chau CH, Chen KY, Deng HT, Kim KJ, Hosoya K, Terasaki T, Shih HM and Ann DK . (2002). Oncogene, 21, 8817–8829.

  9. Chen R, Kim O, Li M, Xiong X, Guan JL, Kung HJ, Chen H, Shimizu Y and Qiu Y . (2001). Nat. Cell Biol., 3, 439–444.

  10. Chin YE, Kitagawa M, Kuida K, Flavell RA and Fu XY . (1997). Mol. Cell. Biol., 17, 5328–5337.

  11. Costa S, Stamm H, Almendral A, Ludwig H, Wyss R, Fabbro D, Ernst A, Takahashi A and Eppenberger U . (1988). Lancet, 2, 1258.

  12. Czar MJ, Debnath J, Schaeffer EM, Lewis CM and Schwartzberg PL . (2001). Biochem. Soc. Trans., 29, 863–867.

  13. Derynck R, Goeddel DV, Ullrich A, Gutterman JU, Williams RD, Bringman TS and Berger WH . (1987). Cancer Res., 47, 707–712.

  14. Ekman N, Arighi E, Rajantie I, Saharinen P, Ristimaki A, Silvennoinen O and Alitalo K . (2000). Oncogene, 19, 4151–4158.

  15. Ekstrand AJ, James CD, Cavenee WK, Seliger B, Pettersson RF and Collins VP . (1991). Cancer Res., 51, 2164–2172.

  16. Filmus J, Pollak MN, Cailleau R and Buick RN . (1985). Biochem. Biophys. Res. Commun., 128, 898–905.

  17. Garcia de Palazzo IE, Adams GP, Sundareshan P, Wong AJ, Testa JR, Bigner DD and Weiner LM . (1993). Cancer Res., 53, 3217–3220.

  18. Gulli LF, Palmer KC, Chen YQ and Reddy KB . (1996). Cell Growth Differ., 7, 173–178.

  19. Hognason T, Chatterjee S, Vartanian T, Ratan RR, Ernewein KM and Habib AA . (2001). FEBS Lett., 491, 9–15.

  20. Hsueh RC and Scheuermann RH . (2000). Adv. Immunol., 75, 283–316.

  21. Ibrahim SO, Vasstrand EN, Liavaag PG, Johannessen AC and Lillehaug JR . (1997). Anticancer Res., 17, 4539–4546.

  22. Islam TC, Branden LJ, Kohn DB, Islam KB and Smith CI . (2000). J. Gene Med., 2, 204–209.

  23. Jorissen RN, Walker F, Pouliot N, Garrett TP, Ward CW and Burgess AW . (2003). Exp. Cell Res., 284, 31–53.

  24. Jui HY, Tseng RJ, Wen X, Fang HI, Huang LM, Chen KY, Kung HJ, Ann DK and Shih HM . (2000). J. Biol. Chem., 275, 41124–41132.

  25. Kerner JD, Appleby MW, Mohr RN, Chien S, Rawlings DJ, Maliszewski CR, Witte ON and Perlmutter RM . (1995). Immunity, 3, 301–312.

  26. Khan WN, Alt FW, Gerstein RM, Malynn BA, Larsson I, Rathbun G, Davidson L, Muller S, Kantor AB, Herzenberg LA, Rosen FS and Sideras P . (1995). Immunity, 3, 283–299.

  27. Kikuchi A, Amagai M, Hayakawa K, Ueda M, Hirohashi S, Shimizu N and Nishikawa T . (1990). Br. J. Dermatol., 123, 49–58.

  28. Lee LF, Guan J, Qiu Y and Kung HJ . (2001). Mol. Cell. Biol., 21, 8385–8397.

  29. Lewis CM, Broussard C, Czar MJ and Schwartzberg PL . (2001). Curr. Opin. Immunol., 13, 317–325.

  30. Lu Z, Jiang G, Blume-Jensen P and Hunter T . (2001). Mol. Cell. Biol., 21, 4016–4031.

  31. Maa MC, Leu TH, McCarley DJ, Schatzman RC and Parsons SJ . (1995). Proc. Natl. Acad. Sci. USA, 92, 6981–6985.

  32. Mano H, Yamashita Y, Sato K, Yazaki Y and Hirai H . (1995). Blood, 85, 343–350.

  33. Matsuda T, Takahashi-Tezuka M, Fukada T, Okuyama Y, Fujitani Y, Tsukada S, Mano H, Hirai H, Witte ON and Hirano T . (1995). Blood, 85, 627–633.

  34. Merlino GT, Xu YH, Richert N, Clark AJ, Ishii S, Banks-Schlegel S and Pastan I . (1985). J. Clin. Invest., 75, 1077–1079.

  35. Miller AT and Berg LJ . (2002). Curr. Opin. Immunol., 14, 331–340.

  36. Misra UK and Pizzo SV . (1998). Cell Signal, 10, 441–445.

  37. Pan S, An P, Zhang R, He X, Yin G and Min W . (2002). Mol. Cell. Biol., 22, 7512–7523.

  38. Price JT, Tiganis T, Agarwal A, Djakiew D and Thompson EW . (1999). Cancer Res., 59, 5475–5478.

  39. Qiu Y and Kung HJ . (2000). Oncogene, 19, 5651–5661.

  40. Qiu Y, Robinson D, Pretlow TG and Kung HJ . (1998). Proc. Natl. Acad. Sci. USA, 95, 3644–3649.

  41. Rajantie I, Ekman N, Iljin K, Arighi E, Gunji Y, Kaukonen J, Palotie A, Dewerchin M, Carmeliet P and Alitalo K . (2001). Mol. Cell. Biol., 21, 4647–4655.

  42. Ramsauer K, Sadzak I, Porras A, Pilz A, Nebreda AR, Decker T and Kovarik P . (2002). Proc. Natl. Acad. Sci. USA, 99, 12859–12864.

  43. Ro J, North SM, Gallick GE, Hortobagyi GN, Gutterman JU and Blick M . (1988). Cancer Res., 48, 161–164.

  44. Robinson D, He F, Pretlow T and Kung HJ . (1996). Proc. Natl. Acad. Sci. USA, 93, 5958–5962.

  45. Sainsbury JR, Nicholson S, Angus B, Farndon JR, Malcolm AJ and Harris AL . (1988). Br. J. Cancer, 58, 458–460.

  46. Satterthwaite AB, Li Z and Witte ON . (1998). Semin. Immunol., 10, 309–316.

  47. Sieg DJ, Hauck CR, Ilic D, Klingbeil CK, Schaefer E, Damsky CH and Schlaepfer DD . (2000). Nat. Cell Biol., 2, 249–256.

  48. Smith CI, Islam TC, Mattsson PT, Mohamed AJ, Nore BF and Vihinen M . (2001). BioEssays, 23, 436–446.

  49. Stephanou A, Scarabelli TM, Brar BK, Nakanishi Y, Matsumura M, Knight RA and Latchman DS . (2001). J. Biol. Chem., 276, 28340–28347.

  50. Stephanou A, Scarabelli TM, Townsend PA, Bell R, Yellon D, Knight RA and Latchman DS . (2002). FASEB J., 16, 1841–1843.

  51. Suhardja A and Hoffman H . (2003). Microsc. Res. Tech., 60, 70–75.

  52. Tamagnone L, Lahtinen I, Mustonen T, Virtaneva K, Francis F, Muscatelli F, Alitalo R, Smith CI, Larsson C and Alitalo K . (1994). Oncogene, 9, 3683–3688.

  53. Tang B, Mano H, Yi T and Ihle JN . (1994). Mol. Cell. Biol., 14, 8432–8437.

  54. Thomas T, Balabhadrapathruni S, Gardner CR, Hong J, Faaland CA and Thomas TJ . (1999). J. Cell. Physiol., 179, 257–266.

  55. Tsai YT, Su YH, Fang SS, Huang TN, Qiu Y, Jou YS, Shih HM, Kung HJ and Chen RH . (2000). Mol. Cell. Biol., 20, 2043–2054.

  56. Tsang DK and Crowe DL . (1999). Int. J. Oncol., 15, 519–523.

  57. Tsukada S, Saffran DC, Rawlings DJ, Parolini O, Allen RC, Klisak I, Sparkes RS, Kubagawa H, Mohandas T, Quan S, Belmont JW, Cooper MD, Conley ME and Witte ON . (1993). Cell, 72, 279–290.

  58. Uckun FM, Waddick KG, Mahajan S, Jun X, Takata M, Bolen J and Kurosaki T . (1996). Science, 273, 1096–1100.

  59. Vignais ML and Gilman M . (1999). Mol. Cell. Biol., 19, 3727–3735.

  60. Vihinen M, Mattsson PT and Smith CI . (2000). Front Biosci., 5, D917–D928.

  61. Wells A . (1999). Int. J. Biochem. Cell Biol., 31, 637–643.

  62. Wen X, Lin HH, Shih HM, Kung HJ and Ann DK . (1999). J. Biol. Chem., 274, 38204–38210.

  63. Wong AJ, Bigner SH, Bigner DD, Kinzler KW, Hamilton SR and Vogelstein B . (1987). Proc. Natl. Acad. Sci. USA, 84, 6899–6903.

  64. Wu YM, Huang CL, Kung HJ and Huang CY . (2001). J. Biol. Chem., 276, 17672–17678.

  65. Xu YH, Richert N, Ito S, Merlino GT and Pastan I . (1984). Proc. Natl. Acad. Sci. USA, 81, 7308–7312.

  66. Xue LY, Qiu Y, He J, Kung HJ and Oleinick NL . (1999). Oncogene, 18, 3391–3398.

  67. Yart A, Laffargue M, Mayeux P, Chretien S, Peres C, Tonks N, Roche S, Payrastre B, Chap H and Raynal P . (2001). J. Biol. Chem., 276, 8856–8864.

Download references


This work was supported by the intramural funds of the National Health Research Institutes (to H-MS), and in part by the National Institute of Health Research Grants CA39207 and 91574 (to HJK) and R01 DE 10742 and DE14183 (to DKA).

Author information

Correspondence to Hsiu-Ming Shih.

Additional information

Supplementary Information accompanies the paper on Oncogene website (

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Chen, K., Huang, L., Kung, H. et al. The role of tyrosine kinase Etk/Bmx in EGF-induced apoptosis of MDA-MB-468 breast cancer cells. Oncogene 23, 1854–1862 (2004).

Download citation


  • Etk kinase
  • EGF-induced
  • apoptosis
  • Statl activation

Further reading