BRCA1-mediated G2/M cell cycle arrest requires ERK1/2 kinase activation

Article metrics

Abstract

Germline mutations in the BRCA1 gene are associated with an increased susceptibility to the development of breast and ovarian cancers. Evidence suggests that BRCA1 protein plays a key role in mediating DNA damage-induced checkpoint responses. Several studies have shown that ectopic expression of BRCA1 in human cells can trigger cellular responses similar to those induced by DNA damage, including G2/M cell cycle arrest and apoptosis. While the effects of ectopic BRCA1 expression on the G2/M transition and apoptosis have been extensively studied, the factors that dictate the balance between these two responses remain poorly understood. We have recently shown that ectopic expression of BRCA1 in MCF-7 human breast cancer cells resulted in activation of extracellular signal-regulated protein kinase 1 and 2 (ERK1/2) and G2/M cell cycle arrest. Furthermore, inhibition of BRCA1-induced ERK1/2 activation using mitogen-activated protein kinase kinase 1 and 2 (MEK1/2)-specific inhibitors resulted in increased apoptosis, suggesting a potential role of ERK1/2 kinases in BRCA1-mediated G2/M checkpoint response. In this study, we assessed the role of ERK1/2 kinases in the regulation of BRCA1-mediated G2/M cell cycle arrest. Results indicate that BRCA1-induced G2/M cell cycle arrest and ERK1/2 activation correlate with changes in the level and/or activity of several key regulators of the G2/M checkpoint, including activation of Chk1 and Wee1 kinases, induction of 14-3-3, and down-regulation of Cdc25C. Furthermore, inhibition of ERK1/2 kinases using MEK1/2-specific inhibitors results in a marked attenuation of the BRCA1-induced G2/M arrest. Biochemical studies established that ERK1/2 inhibition abolished the effects of BRCA1 on components of the G2/M checkpoint, including regulation of Cdc25C expression and activation of Wee1 and Chk1 kinases. These results implicate a critical role of ERK1/2 signaling in the regulation of BRCA1 function on controlling the G2/M checkpoint responses.

Introduction

The BRCA1 gene encodes a protein of 1864 amino-acid residues that is primarily located in the nucleus (Miki et al., 1994). Germline mutations in the BRCA1 gene have been implicated in the increased susceptibility to the development of breast and ovarian cancers (Yang and Lippman, 1999). Although the exact biological function of BRCA1 remains to be defined, a role for BRCA1 in transcriptional regulation, cell cycle checkpoint control, chromosome segregation, and DNA damage repair have been suggested (Venkitaraman, 2002). Consistent with its multiple cellular functions, BRCA1 protein has been shown to interact with many cellular proteins. The N-terminus of BRCA1 contains a ring finger domain that interacts with several proteins including BAP1 (ubiquitin hydrolase BRCA1-associated protein 1) and BARD1 (BRCA1-associated ring domain 1); the middle portion of BRCA1 contains two nuclear localization signals and domains that can bind c-Myc, p53, pRB, and the DNA repair proteins RAD50 and RAD51; and the C-terminus of BRCA1 includes two BRCT domains that associate with a large number of proteins including BRCA2, CtIP, MSH2, p300/CBP, p53, pRB, RNA Pol II, and RNA helicase A (Deng and Brodie, 2000; Wang et al., 2000).

To maintain integrity of the genome, cells must be able to inhibit cell cycle progression in response to DNA damage. This is ensured by the existence of checkpoints, mechanisms that arrest the cell cycle after activation in response to DNA damage (Qin and Li, 2003). Upon DNA damage, mammalian cells can arrest at G1, S, or G2/M phase of the cell cycle, depending on the phase in which the damage is sensed, allowing the cells sufficient time to accurately repair damaged DNA (Qin and Li, 2003).

The G2/M checkpoint is controlled by the Cdc2/Cyclin B complex, whose activity is required for entry into mitosis (Smits and Medema, 2001). Several studies point to the Tyr15 residue of Cdc2 as the main target of the G2/M DNA damage checkpoint. Tyr15 is situated in the ATP-binding site of Cdc2 and phosphorylation of this residue interferes with phosphate transfer to a bound substrate, thus inhibiting Cdc2 kinase activity (Atherton-Fessler et al., 1993). Previous studies indicate that Cdc2-Tyr15 phosphorylation is maintained during radiation-induced G2/M arrest in both yeast and mammalian cells (Kharbanda et al., 1994; O'Connell et al., 1997; Rhind et al., 1997). Introduction of the Cdc2-Y15F mutant, which cannot be phosphorylated at residue 15, attenuates DNA damage-induced G2/M checkpoint arrest in yeast as well as in Hela human cervical cancer cells (Jin et al., 1996; Rhind et al., 1997). Phosphorylation of Cdc2-Tyr15 is accomplished by two members of the Wee1 kinase family: Wee1, which phosphorylates Cdc2 at Try15, and Myt1, which phosphorylates Cdc2 at Thr14 and, to a lesser extent, Tyr15 (Lundgren et al., 1991; Parker et al., 1992). Dephosphorylation of Cdc2-Tyr15 involves Cdc25C, a dual specific phosphatase, whose activity is essential for entry into mitosis (Dunphy, 1994). Furthermore, two additional regulators of Cdc25C are involved in the regulation of DNA damage-induced G2/M checkpoint arrest. These include 14-3-3, which binds Cdc25C and inhibits the import of Cdc25C into the nucleus, and Chk1 kinase, which phosphorylates Cdc25C at Ser216 and creates a consensus binding site in Cdc25C for 14-3-3 (Peng et al., 1997; Sanchez et al., 1997; Graves et al., 2001).

Accumulating evidence has strongly implicated BRCA1 as a key mediator of the DNA damage checkpoint responses, especially the S and G2/M checkpoints (Venkitaraman, 2002). In response to DNA damage, BRCA1 is rapidly phosphorylated by several DNA damage-activated nuclear kinases including ATM/ATR and Chk2 (Cortez et al., 1999; Chen, 2000; Lee et al., 2000; Tibbetts et al., 2000). BRCA1 is necessary for effective S and G2/M checkpoints in DNA damage response (Xu et al., 2001, 2002b), and mouse embryonic fibroblasts (MEFs) carrying a targeted homologous deletion of exon 11 in BRCA1 have defective G2/M checkpoint in response to radiation treatment (Xu et al., 1999). Other studies have shown that expression of exogenous BRCA1 in human cancer cell lines can trigger a cell cycle checkpoint response similar to that induced by DNA damage (Harkin et al., 1999; MacLachlan et al., 2000, 2002; Thangaraju et al., 2000; Yan et al., 2002). Furthermore, studies by Yarden et al. (2002) demonstrated that activation of Chk1 kinase and G2/M arrest following DNA damage requires BRCA1. While a role for BRCA1 in mediating DNA damage checkpoint responses has been strongly suggested, the precise mechanism by which BRCA1 performs this role is still poorly understood.

Signaling mediated by mitogen-activated protein kinases (MAPKs) has been shown to play critical roles in the regulation of cell proliferation and in the response of cells to DNA damage. On the basis of their sequence similarities and the nature of their upstream activators, MAPKs are grouped into three subfamilies: ERK1/2, JNK/SAPK, and p38. ERK1/2 kinases are primarily activated in response to mitogens and activation normally leads to the induction of cyclin D1 and the initiation of cell cycle progression (Robinson and Cobb, 1997). In contrast, JNK/SAPK and p38 kinases are primarily activated by stress signals and activation is generally associated with inhibition of cell growth and/or reduced survival (Robinson and Cobb, 1997).

ERK1/2 and JNK/SAPK have both been implicated in the cellular response to BRCA1 expression in cells. Studies by Harkin et al. (1999) and Thangaraju et al. (2000) implicated JNK kinase activation in the apoptotic response to BRCA1 expression in human U2OS osteosarcoma cells, while a previous study from our laboratory demonstrated that ectopic expression of BRCA1 in MCF-7 human breast cancer cells resulted in a concomitant induction of both JNK and ERK1/2 kinases and was associated with a G2/M cell cycle arrest (Yan et al., 2002). Furthermore, inhibition of ERK1/2 kinases resulted in an induction of BRCA1-dependent apoptosis in MCF-7 cells, suggesting a central role for ERK1/2 kinases in determining the cellular response to BRCA1 (Yan et al., 2002).

In this study, we have examined in detail the effects of ERK1/2 activation on BRCA1-induced G2/M cell cycle arrest. Our results provide evidence showing that ERK1/2 signaling predominantly affect the G2/M cell cycle checkpoint response mediated by BRCA1 expression, suggesting a critical regulation of ERK1/2 signaling on BRCA1 function on the control of G2/M checkpoint.

Results

Ectopic BRCA1 expression in MCF-7 cells induces G2/M cell cycle arrest and ERK1/2 activation

To investigate the effect of BRCA1 expression on cell growth, MCF-7 human breast cancer cells were infected with a recombinant adenovirus containing full-length BRCA1 cDNA (Ad.BRCA1) that was previously generated in our laboratory (Yan et al., 2002). For comparison, uninfected MCF-7 cells and MCF-7 cells infected with a control viral vector (Ad.Control), an empty vector that shares the same backbone as Ad.BRCA1, were also included in the same experiment. Following incubation at 37°C for 24 h, infected and uninfected cells were collected and analysed for BRCA1 expression and cell cycle progression. Western blot analysis using an anti-BRCA1 antibody showed high levels of BRCA1 protein expression in Ad.BRCA1-infected MCF-7 cells (Figure 1a). Low levels of BRCA1 protein was detected in uninfected (see Supplementary Figure 1) and Ad.Control-infected MCF-7 cells with longer exposure time (data not shown). Cell cycle analysis data showed that at 24 h following infection of MCF-7 cells with Ad.BRCA1, there was over a four-fold increase in 4N-DNA content cell population, indicative of G2/M phase of the cell cycle, compared to Ad.Control-infected cells (Figure 1b).

Figure 1
figure1

Expression of exogenous BRCA1 induces G2/M cell cycle arrest and ERK1/2 phosphorylation in MCF-7 cells. (a and b) Uninfected MCF-7 cells or cells infected with Ad.BRCA1 or Ad.Control at 50 PFU/cell were incubated at 37°C for 24 h, and then harvested and washed with PBS. The cell samples were divided into two portions for the following analysis: (a) One half of the samples were lysed and 100 μg of protein lysates was analysed for BRCA1 expression by Western blot using an anti-BRCA1-specific antibody (BRCA1). Protein loading was confirmed by monitoring the levels of Actin by immunoblotting using an anti-actin antibody (Actin). (b) The other half of the samples were fixed, stained with PI. PI-stained cells were gated on the basis of their DNA content, the location and amount of cells with 4N-DNA content, indicative of G2/M phase of the cell cycle, are indicated. (c) MCF-7 cells were infected with Ad.BRCA1 at the indicated doses for 24 h and the resulting cell samples were analysed for the level of BRCA1 protein and ERK1/2 phosphorylation by Western blotting, and for DNA contents by flow cytometry. (d) MCF-7 cells were infected with Ad.BRCA1 at 50 PFU/cell at 37°C for the hours indicated. The cells were harvested, lysed, and the levels of BRCA1, ERK1/2, Cdc2, Wee1, and Cdc25C determined by immunoblotting with appropriate specific antibodies. To assess the levels of phosphorylation of ERK1/2 (pERK1/2) and Cdc2-Tyr15 (Cdc2-Tyr15), 100 μg of protein lysates were separated by SDS–PAGE and analysed by immunoblotting using appropriate phosphorylation-specific antibodies. To determine the Wee1 kinase activity, Wee1 protein was immunoprecipitated from 500 μg of cell lysates and assayed for kinase activity using GST-Cdc2 substrate as described in the Materials and methods (Wee1 activity). The amount of Wee1 protein in the immunoprecipitate as well as in the cell lysates was determined by Western blot using an anti-Wee1-specific antibody (Wee1 IP Western and Wee1). To confirm equal loading of protein lysates, duplicate protein samples were probed with an anti-Actin antibody (Actin)

To confirm the correlation between BRCA1 expression and induction of ERK1/2 phosphorylation and G2/M arrest in MCF-7 cells, cells were infected with increasing doses of Ad.BRCA1 and incubated for 24 h prior to analysis. As shown in Figure 1c, Western blot analysis indicated that exposure of MCF-7 cells to Ad.BRCA1 virus resulted in a dose-dependent increase in both BRCA1 protein expression and ERK1/2 phosphorylation in MCF-7 cells (Figure 1c, upper panel). Furthermore, DNA-content analysis showed that the induction of G2/M cell cycle arrest following infection of MCF-7 cells with Ad.BRCA1 was correlated with the level of BRCA1 protein expression in cells (Figure 1c).

To examine the effect of BRCA1 expression on ERK1/2 phosphorylation, MCF-7 cells were infected with Ad.BRCA1 or Ad.Control (50 PFU/cell) for 0, 16, and 24 h and cell lysates were analysed for ERK1/2 phosphorylation by immunoblotting using a phosphorylation-specific antibody. As shown in Figure 1d, infection of MCF-7 cells with Ad.BRCA1 resulted in a time-dependent increase in BRCA1 protein levels and concomitant induction in ERK1/2 phosphorylation (Figure 1d, BRCA1 and pERK1/2). In contrast, infection of MCF-7 cell with Ad. Control virus had no effect on the level of ERK1/2 phosphorylation (data not shown).

To assess the influence of BRCA1 expression on the regulation of Cdc2 activity, Ad.BRCA1-infected MCF-7 cells were analysed for the level of Cdc2-Tyr15 phosphorylation as well as the level and/or activity of Wee1 kinase and Cdc25C phosphatase, two key enzymes that control the phosphorylation of Cdc2-Tyr15. The results indicated that induction of BRCA1 expression was associated with an increase in phosphorylation of Cdc2-Tyr15 beginning within 16 h following infection with Ad.BRCA1. By 24 h after Ad.BRCA1 infection, there was a 10-fold increase in Cdc2-Tyr15 phosphorylation in MCF-7 cells (Figure 1d, Cdc2-Tyr15). Consistent with the effect on phosphorylation of Cdc2-Tyr15, ectopic BRCA1 expression resulted in a marked induction of Wee1 kinase activity in the absence of change in levels of total Wee1 protein (Figure 1d, Wee1 activity and Wee1). In addition, the BRCA1 expression in MCF-7 cells also resulted in a significant decrease in Cdc25C protein levels. At 24 h after infection with Ad.BRCA1, there was a 65% reduction in Cdc25C protein levels in MCF-7 cells (Figure 1d, Cdc25C).

Thus, the induction of ERK1/2 kinases by BRCA1 expression in MCF-7 cells is correlated with an activation of Wee1 kinase, a decrease in Cdc25C protein levels and concomitant increase in Cdc2-Tyr15 phosphorylation. The net effect of the results is a G2/M arrest following BRCA1 expression.

ERK1/2 activation by BRCA1 is required for BRCA1-mediated G2/M cell cycle arrest

Previous studies from our laboratory (Yan et al., 2002) as well as results from the present study indicated that expression of exogenous BRCA1 in MCF-7 breast cancer cells is associated with an activation of ERK1/2 kinases and a G2/M cell cycle arrest (Figure 1). To determine the possible role of ERK1/2 kinase activation in the regulation of BRCA1-mediated G2/M arrest, we examined the effects of ERK1/2 kinases on BRCA1-induced G2/M arrest in MCF-7 cells using MEK1/2-specific inhibitors U0126 and PD98058 (Favata et al., 1998; Davies et al., 2000) as well as a dominant negative MEK1 (Alessi et al., 1994; Cowley et al., 1994).

We first studied the effect of ERK1/2 inhibition by U0126 on BRCA1-mediated G2/M cell cycle checkpoint activation. To determine the effective dose of U0126 for ERK1/2 inhibition, MCF-7 cells were infected with Ad.BRCA1 (50 PFU/cell) and then incubated in the presence of increasing concentrations of U0126 for 24 h. As shown in Figure 2a, ERK1/2 phosphorylation was inhibited by 60% in Ad.BRCA1-infected MCF-7 cells incubated with 10 μ M U0126 relative to Ad.BRCA1-infected cells incubated in the absence of drug (Figure 2a, lane 3 vs lane 1). Increasing the concentration of U0126 to 50 μ M resulted in near complete inhibition of ERK1/2 in Ad.BRCA1-infected MCF-7 cells (Figure 2a, lane 4). Inhibition of ERK1/2 phosphorylation by incubation with 50 μ M U0126 was greater than that produced by serum starvation of uninfected MCF-7 cells (Figure 2a, lane 4 vs lane 6), a condition known to down-regulate ERK1/2 phosphorylation (Robinson and Cobb, 1997). A similar effect of U0126 on ERK1/2 phosphorylation was observed in Ad.Control virus-infected MCF-7 cells (data not shown).

Figure 2
figure2

Activation of ERK1/2 kinases by BRCA1 is required for BRCA1-induced G2/M cell cycle arrest. (a) MCF-7 cells were infected with Ad.BRCA1 at 50 PFU/cell and incubated in the presence of U0126 at the indicated doses for 24 h. For serum starvation experiment, log-phase growing cells were washed once with DMEM and then incubated in the medium containing DMEM and 0.1% serum for 24 h. The treated cells were collected, lysed, and the level of phosphorylated-ERK1/2 (pERK1/2) as well as the level of total ERK1/2 (ERK1/2) in each sample was determined by immunoblotting with relevant specific antibodies. (b) Cells were infected with Ad.Control (open bars) or Ad.BRCA1 (solid bars) at 50 PFU/cell and incubated for 24 h in the presence of U0126 at the indicated doses. The resulting cells were harvested, stained with PI and analysed for DNA contents by flow cytometry as described above. Percentage of cells with 4N-DNA content, indicative of G2/M phase of the cell cycle, is shown as the mean±s.d. of quadruplicate samples. (c) For treatment of PD98059, cells were infected with Ad.Control or Ad.BRCA1 (50 PFU/cell) and then incubated in medium containing 100 μ M PD98059 for 24 h prior to analysis (hatched bars). For experiments involving expression of dominant negative MEK1 (MEK1(dn)), cells were infected first with Ad.MEK1(dn) at 100 PFU/cell for 15 h, followed by infection with Ad.Control or Ad.BRCA1 at 50 PFU/cell for additional 24 h (open bars). As a control, a set of duplicate cell samples was infected with Ad.Control or Ad.BRCA1 without additional treatment (solid bars). The resulting cells were analysed for DNA content by FACS and for ERK1/2 phosphorylation by immunoblotting as described above. The percentage of G2 phase cells shown represents the average of two independent experiments. (d) MCF-7 cells were infected with Ad.Control or Ad.BRCA1 (50 PFU/cells) followed by incubation for 24 h in medium containing 0.1% DMSO (solid bars), 50 μ M U0126 (hatched bars), or 1 μ M wortmannin (open bars). Cell samples were analysed for DNA content and for ERK1/2 phosphorylation. The percentage of G2 phase cells shown represents the average of two independent experiments. (e) Cells were infected with or without Ad.BRCA1 at 30 PFU/cell for 24 h. Half of the Ad.BRCA1-infected and -uninfected MCF-7 cells were then either exposed to 10-Gy γ-irradiation or, as a control, remained untreated. The cells were then incubated for an additional 24 h and analysed by immunoblotting for levels of BRCA1 protein, total ERK1/2, and phosphorylated-ERK1/2. (f) Cells infected with or without Ad.BRCA1 at 30 PFU/cell for 24 h were treated with or without γ-irradiation at 10 Gy (solid bars). When U0126 was included in the experiment, U0126 (50 μ M) was either preincubated with uninfected MCF-7 cells for 2 h prior to irradiation treatment or added after 2 h following Ad.BRCA1 infection (open bars). Treated and control cells were analysed for DNA contents by flow cytometry. The results depict the percentage of cells with 4N-DNA content (G2 phase cells) and represent the average of two independent experiments

To assess the effects of ERK1/2 inhibition by U0126 on BRCA1-mediated G2/M arrest response, MCF-7 cells were infected with Ad.Control or Ad.BRCA1 (50 PFU/cell) and then cultured for 24 h in the presence of increasing concentrations of U0126. The resulting cells were analysed for DNA contents by fluorescence-activated cell sorting (FACS) as described in Materials and methods. The results showed that the 4N-DNA content cell population (G2/M phase of the cell cycle) in BRCA1-infected cells (Figure 2b, 0 μ M, solid bar) was 4.75-fold higher than that present in the Ad.Control-infected cells (Figure 2b, 0 μ M, open bar), indicating an induction of G2/M cell cycle arrest in Ad.BRCA1-infected cells. Furthermore, incubation of Ad.BRCA1-infected MCF-7 cells with U0126 resulted in a marked attenuation of BRCA1-induced G2/M arrest in a dose-dependent manner (Figure 2b, solid bars). Incubation with 10 μ M U0126 inhibited the BRCA1-induced G2/M arrest by approximately 50%, while incubation with 50 μ M U0126 diminished the BRCA1-induced G2/M cell cycle arrest by 80% (Figure 2b, solid bars). No further inhibitory effect on BRCA1-induced G2/M arrest was seen when higher doses of U0126 were used (>50 μ M) (data not shown). Interestingly, incubation of Ad.Control-infected MCF-7 cells with U0126 had no detectable effect on the population of cells containing 4N-DNA (Figure 2b, open bars). Moreover, the reduction of BRCA1-induced G2/M arrest in MCF-7 cells by U0126 treatment is directly correlated with the relative inhibition of ERK1/2 (Figure 2a and b).

To confirm the effect of ERK1/2 inhibition by U0126 on BRCA1-mediated G2/M arrest, we examined two additional ERK1/2 signaling specific inhibitors, PD98059 and dominant negative MEK1, for their effects on BRCA1-induced G2/M cell cycle arrest. To test the effect of PD98059, MCF-7 cells were infected with Ad.Control or Ad.BRCA1 (50 PFU/cell) and incubated for 24 h in medium containing 100 μ M PD98059 or, as a control, 0.1% DMSO. To assess the influence of dominant negative MEK1 on the response of MCF-7 cells to BRCA1 expression, we utilized an adenoviral vector expressing a dominant negative MEK1 [Ad.MEK1(dn)] (Alessi et al., 1994; Cowley et al., 1994). MCF-7 cells were initially infected at 100 PFU/cell with Ad.MEK1(dn) for 15 h, followed by exposure to 50 PFU/cell of Ad.BRCA1 or Ad.Control for additional 24 h. The cells were then analysed for the incidence of G2/M cell cycle arrest and the level of ERK1/2 phosphorylation as described under Materials and methods. The results showed that exposure of Ad.BRCA1-infected MCF-7 cells to either PD98059 or Ad.MEK1(dn) resulted in approximate 80–90% inhibition in ERK1/2 phosphorylation as determined by immunoblotting using a phosphorylation-specific antibody (Figure 2c, inset). Moreover, treatment of Ad.BRCA1-infected cells with PD98059 (Ad.BRCA1, hatched bar) or dominant negative MEK1 (Ad.BRCA1, open bar) caused a significant diminution of BRCA1-induced G2/M arrest (Figure 2c). Compared to the cells exposed to only Ad.BRCA1 (Ad.BRCA1, solid bar), Ad.BRCA1-infected cells treated with PD98059 (Ad.BRCA1, hatched bar) and dominant negative MEK1 (Ad.BRCA1, open bar) had a 76 and 74% reduction, respectively, in BRCA1-induced G2/M arrest (Figure 2c). In contrast, incubation of PD98059 (Ad.Control, hatched bar) or expression of dominant negative MEK1 (Ad.Control, open bar) did not cause any noticeable change in Ad.Control-infected MCF-7 cells with the 4N-DNA content cell population (G2/M phase) relative to that present in Ad.Control-infected cells in the absence of any additional treatment (Ad.Control, solid bar) (Figure 2c).

We also evaluated the effects of wortmannin, a specific inhibitor of phosphoinositide kinase 3 (PI3-kinase) (Chung et al., 1994) widely used in studies of cell growth regulation (Dent et al., 2003a), on BRCA1-induced G2/M arrest in MCF-7 cells. To examine the effect of wortmannin on G2/M cell cycle arrest, MCF-7 cells were infected with Ad.BRCA1 or Ad.Control (50 PFU/cell) and then incubated for 24 h in the medium containing 1 μ M wortmannin or, as a control, 0.1% DMSO. The percentage of cells with 4N-DNA content was measured by FACS analysis as described above. The dose of wortmannin used in this study (1 μ M) was found to maximally inhibit insulin-induced activation of AKT/PKB (direct downstream target of PI3 kinase) in MCF-7 cells (>85%) (data not shown) (Scott et al., 1998). As a control, a parallel set of infected cells was incubated in the presence of 50 μ M U0126. As shown in Figure 2d, BRCA1 expression in MCF-7 cells resulted in a three-fold increase in cell population with 4N-DNA content, indicative of a G2/M arrest (Figure 2d, Ad.BRCA1, solid bar). Consistent with the results described above, incubation of Ad.BRCA1-infected MCF-7 cells with U0126 resulted in 83% attenuation in BRCA1-resulted G2/M arrest (Figure 2d, Ad.BRCA1, hatched bar). In contrast, incubation of Ad.BRCA1-infected cells with 1 μ M Wortmannin, a specific inhibitor of PI3/AKT signaling, resulted in only a 12% decrease in BRCA1-induced G2/M arrest (Figure 2d, Ad.BRCA1, open bar). Incubation of Ad.Control-infected cells with either U0126 (Figure 2d, Ad.Control, hatched bar) or wortmannin (Figure 2d, Ad.Control, open bar) had little, if any, effect on 4N-DNA content cell population compared to untreated Ad.Control-infected cells (Figure 2d, Ad.Control, solid bar).

Thus, in contrast to the effect of ERK1/2 inhibition on BRCA1-mediated activation of G2/M checkpoint, inhibition of PI3-AKT signaling has little influence on BRCA1-induced G2/M arrest in MCF-7 breast cancer cells.

ERK1/2 activation by γ-irradiation is also needed for irradiation-induced G2/M cell cycle arrest

It has been previously observed that ionizing irradiation induces a G2/M cell cycle arrest and that this treatment is also associated with activation of ERK1/2 kinases (Smits and Medema, 2001; Dent et al., 2003b). To test whether irradiation treatment in MCF-7 cells also causes G2/M arrest and ERK1/2 activation, log-phase growing MCF-7 cells were either left untreated or exposed to 10-Gy irradiation. For comparison, parallel cultures of MCF-7 cells were infected with 30 PFU/cell Ad.BRCA1 for 24 h and then treated with or without irradiation. Each of the groups of cells was subsequently incubated for additional 24 h prior to analysis. The results shown in Figure 2e indicated that infection of MCF-7 cells with Ad.BRCA1 or treatment of MCF-7 cells with 10-Gy irradiation resulted in activation of ERK1/2 kinases. Furthermore, as shown in Figure 2f, infection with Ad.BRCA1 or treatment with 10-Gy irradiation resulted in a 2.0- and 2.7-fold increase in 4N-DNA content cell population, respectively, compared to control MCF-7 cells (Figure 2f, solid bars). The combination treatment with Ad.BRCA1 infection and 10-Gy irradiation resulted in an additional 50% increase in G2/M-DNA content cell population compared to MCF-7 cells receiving only the radiation treatment (Figure 2f, solid bars). Thus, treatment of MCF-7 cells with γ-irradiation causes G2/M cell cycle arrest and ERK1/2 activation similar to that inducted following Ad.BRCA1 infection.

We also examined the effect of ERK1/2 activation on irradiation-induced G2/M arrest in MCF-7 cells. In this experiment, MCF-7 cells were preincubated with 50 μ M U0126 for 2 h, to inhibit ERK1/2 activity, followed by exposure to 10-Gy irradiation. For comparison, MCF-7 cells were infected with Ad.BRCA1 (30 PFU/cell) for 24 h in the presence of 50 μ M U0126 and then also receiving identical irradiation treatment. Following irradiation, the cells were cultured for additional 24 h and subjected to analysis. As shown in Figure 2f, treatment of MCF-7 cells with MEK1/2 inhibitor U0126, which results in nearly complete inhibition of ERK1/2 phosphorylation (data not shown), caused a marked diminution in G2/M arrest induced by ectopic BRCA1 expression, 10-Gy irradiation, or the combination treatment with both Ad.BRCA1 infection and irradiation (Figure 2f, open bars). These results indicate that ERK1/2 activation is necessary for irradiation-mediated G2/M checkpoint activation as that in BRCA1-induced G2/M checkpoint activation in MCF-7 cells.

ERK1/2 inhibition abolishes BRCA1-induced Tyr-15 phosphorylation of Cdc2

In order to determine the role of ERK1/2 signaling on the regulation of Cdc2-Tyr15 phosphorylation in Ad.BRCA1-infected MCF-7 cells, cells were infected with Ad.BRCA1 or Ad.Control virus and incubated in the presence or absence of 50 μ M U0126. As shown in Figure 3a, Western blot analysis using an anti-phospho-Cdc2-Tyr15 specific antibody demonstrated a marked increase in phosphorylation of Cdc2-Tyr15 in Ad.BRCA1-infected MCF-7 cells compared to the Ad.Control-infected cells (Figure 3a, lane 3 vs lane 1). Furthermore, treatment of the Ad.BRCA1-infected cells with 50 μ M U0126 diminished the BRCA1-induced phosphorylation of Cdc2-Tyr15 to levels similar to that present in Ad.Control-infected cells (Figure 3a, lane 4 vs lane 1). In contrast, treatment of Ad.Control-infected MCF-7 cells with U0126 had little effect on the basal level of phospho-Cdc2-Tyr15 (Figure 3a, lanes 1–2).

Figure 3
figure3

The effects of BRCA1 and ERK1/2 kinases on the regulators of G2/M checkpoint. (a) MCF-7 cells were infected with Ad.Control (lanes 1–2) or Ad.BRCA1 (lanes 3–4) at 50 PFU/cell and incubated in medium containing either 50 μ M U0126 (lanes 2 and 4) or, as a control, 0.1% DMSO (lanes 1 and 3). After 24 h incubation, the cells were collected and analysed for Cdc2-Tyr15 phosphorylation (Cdc2-Tyr15) by immunoblotting using a phosphorylation-specific antibody. Total Cdc2 protein levels (Cdc2) in the lysates were measured by immunoblotting using an anti-Cdc2-specific antibody. (b) MCF-7 cells were treated as described above. Wee1 and Chk1 kinases were immunoprecipitated and assayed for kinase activities using GST-Cdc2 and GST-Cdc25C substrate, respectively (Wee1 activity and Chk1 activity) as described in Materials and methods. The amounts of Wee1 and Chk1 present in the immunoprecipitates (Wee1 IP Western and Chk1 IP western) and the levels of Wee1 and Chk1 protein in the cell lysates (Wee1 Western and Chk1 Western) were assessed by immunoblotting. Equal protein loading was confirmed by measuring Actin levels on the immunoblots using an anti-Actin antibody (Actin). (c) Cells were infected with Ad.BRCA1 (lanes 2–4) or Ad.Control (lanes 1 and 5–6) at 50 PFU/cell and cultured for 24 h in the presence of 100 μ M PD98059 (lanes 3 and 5), expression of dominant negative MEK1 (lanes 4 and 6), or as a control 0.1% DMSO (lanes 1 and 2). The resulting cell samples were analysed for kinase activities of Wee1 and Chk1 as described above (Wee1 Activity and Chk1 Activity). The amounts of Wee1 and Chk1 protein presented in the immunoprecipitates were determined by immunoblotting using specific antibodies (Wee1 IP Western and Chk1 IP Western)

We next examined the effect of ERK1/2 inhibition on BRCA1-induced activation of Wee1 kinase. As shown in Figure 3b, basal Wee1 kinase activity was undetectable in both uninfected MCF-7 cells (data not shown) and in Ad.Control-infected MCF-7 cells (Figure 3b, lane 1). Infection of MCF-7 cells with Ad.BRCA1 resulted in a marked increase in Wee1 kinase activity (Figure 3b, lane 3 vs lane 1). Moreover, treatment of Ad.BRCA1-infected cells with U0126 diminished the BRCA1-induced activation of Wee1 kinase by 95% (Figure 3b, lane 4 vs lane 3). Thus, ERK1/2 activation by BRCA1 is apparently required for Wee1 kinase activation. These changes in Wee1 kinase activity were observed without detectable alterations in the steady-state level of the Wee1 protein, implicating post-translational activation of Wee1 kinase either directly or indirectly by ERK1/2 kinases in response to BRCA1 expression (Figure 3b).

Previous studies by Yarden et al. (2002) demonstrated that BRCA1-induced G2/M arrest is also associated with activation of Chk1 kinase. To determine whether Chk1 kinase activity is also regulated by ERK1/2 activation, we monitored Chk1 kinase activities in MCF-7 cells infected with Ad.BRCA1 and Ad.Control virus and incubated in the presence or absence of 50 μ M U0126. As shown in Figure 3b, ectopic expression of BRCA1 resulted in a nine-fold induction in Chk1 kinase activity relative to that present in Ad.Control-infected cells (lanes 1 and 3). Furthermore, treatment of Ad.BRCA1-infected cells with 50 μ M U0126 completely inhibited the BRCA1-induced activation of Chk1 kinase (Figure 3b, Chk1 activity, lane 4 vs lane 3). In addition, treatment with 50 μ M U0126 also inhibited basal Chk1 kinase activity detected in Ad.Control-infected cells (Figure 3b, Chk1 activity, lanes 1–2). These changes in Chk1 kinase activity were not associated with alterations in the level of Chk1 protein (Figure 3b, Chk1 Western). Thus, BRCA1-induced activation of ERK1/2 kinases is apparently also required for the activation of Chk1 kinase activity.

To confirm the effect of ERK1/2 inhibition on activation of Wee1 and Chk1 kinases by BRCA1, we used two additional specific inhibitors of ERK1/2 signaling, PD98059 and dominant negative MEK1. In studies employing PD98059, MCF-7 cells were infected with Ad.BRCA1 or Ad.Control (50 pfu/cell) and then incubated in the presence of 100 μ M PD98059 for 24 h. In experiments involving dominant negative MEK1, cells were first infected with Ad.MEK1(dn) at 100 PFU/cell for 15 h, and then infected the cells with Ad.BRCA1 or Ad.Control (50 PFU/cell) and incubated for an additional 24 h. The cells were then analysed for Wee1 and Chk1 kinase activities as described in the Materials and methods. As shown in Figure 3c, treatment of Ad.BRCA1-infected cells with either PD98059 or dominant negative MEK1 resulted in more than an 80% reduction in BRCA1-induced Wee1 kinase activity (Figure 3c, Wee1 activity, lanes 2–4). Similarly, BRCA1-induced Chk1 kinase activity was nearly completely inhibited by both PD98059 and dominant negative MEK1 (Figure 3c, Chk1 kinase activity, lanes 2–4). In contrast, neither PD98059 treatment nor expression of dominant negative MEK1 had any noticeable effect on the activities of both Wee1 and Chk1 kinases in Ad.Control-infected cells (Figure 3c, Wee1 activity and Chk1 activity, lanes 1 and 5–6).

Collectively, these biochemical data provide strong evidence supporting the hypothesis that ERK1/2 kinases play an essential role in BRCA1-dependent G2/M arrest, which involves ERK1/2-dependent activation of the Wee1 and Chk1 kinases, and phosphorylation of Cdc2-Tyr15.

ERK1/2 activation is required for the destabilization of Cdc25C by BRCA1

We also examined the effects of ERK1/2 inhibition on Cdc25C protein levels. As shown in Figure 4a, 24 h following infection with Ad.BRCA1, the level of Cdc25C protein was decreased 70% relative to the level present in Ad.Control-infected cells (Figure 4a, lane 3 vs lane 1). Furthermore, exposure of Ad.BRCA1-infected MCF-7 cells to 50 μ M U0126 completely restored Cdc25C protein expression to the level present in Ad.Control-infected cells (Figure 4a, lane 4 vs lane 1). Thus, BRCA1-induced activation of ERK1/2 kinases either directly or indirectly regulates the level of Cdc25C protein. Although Ad.BRCA1 infection induces 14-3-3 protein levels in MCF-7 cells, ERK1/2 inhibition does not inhibit the induction of 14-3-3 by BRCA1 (data not shown). Thus, the restoration of Cdc25C protein expression through the inhibition of ERK1/2 kinases does not involve changes in 14-3-3 expression.

Figure 4
figure4

The effects of BRCA1 and ERK1/2 kinases on protein stability of Cdc25C. (a) Cells were treated as described in Figure 3a. Resulting cells were lysed and Cdc25C protein levels determined by immunoblotting (Cdc25C). The equal protein loadings were confirmed by measuring Actin protein levels in all lanes with immunoblotting (Actin). (b) MCF-7 cells were treated as described in Figure 3c. The cell samples were lysed and analysed for protein levels of Cdc25C by immunoblotting using a specific antibody (Cdc25C). The equal loading of protein lysates were confirmed by immunoblotting using an anti-Actin antibody (Actin). (c) Cells were infected with Ad.BRCA1 or Ad.Control at 50 PFU/cell and cultured in the presence of U0126 at the indicated doses for 24 h. Total RNA was isolated, and 20 μg samples were analysed for Cdc25C mRNA levels by Northern blot hybridization using a Cdc25C probe (upper panel, Cdc25C). The same blot was also hybridized with a radiolabeled probe for glyceraldehyde-3-phosphate dehydrogenase (upper panel, GAPDH) to ensure equal loading of mRNA samples. To confirm Northern blot analytical results, Cdc25C mRNA levels of all cell samples were also analysed by RT–PCR (bottom panel, Cdc25C) using a pair of Cdc25C primers as described in Materials and methods. As control, Actin mRNA levels in those RNA samples were determined by RT–PCR analysis in parallel (Actin). (d) Cells were infected with Ad.BRCA1 (upper panel) or Ad.Control (bottom panel) at 50 PFU/cell for 24 h and then treated with cycloheximide (CHX) at 10 μg/ml. At 1 h following CHX addition, U0126 (50 μ M) or, as a control, DMSO (0.1%) was added and the cells were incubated at 37°C for the additional indicated times. The treated cells were collected and examined by Western blot analysis for Cdc25C protein levels (Cdc25C). To confirm equal protein loading of all lanes, the same blots were reprobed with an anti-actin antibody (Actin). (e) MCF-7 cells were infected with either Ad.Control (lanes 1 and 4) or Ad.BRCA1 (lanes 2–3 and 5) at 50 PFU/cells and incubated at 37°C in the presence of DMSO (0.1%) (lanes 1–2), 50 μ M U0126 (lane 3), or 20 μ M lactacystin (lanes 4–5) for 24 h. The resulted cell samples were lysed and analysed for protein levels of Cdc25C with immunoblotting using an anti-Cdc25C antibody (Cdc25C). The protein loadings of all samples were assured by reprobing the same blot for Actin protein levels (Actin)

To confirm the role of ERK1/2 on the regulation of Cdc25C protein expression, we also examined the effects of two additional specific inhibitors of ERK1/2 signaling, PD98059 and dominant negative MEK1, on Cdc25C protein expression in MCF-7 cells. The results shown in Figure 4b indicate that both PD98059 treatment and dominant negative MEK1 expression in Ad.BRCA1-infected cells markedly diminished BRCA1-induced decrease in Cdc25C protein expression. The observed effects by these two treatments are similar to that obtained in Ad.BRCA1-infected cells incubated with U0126. Thus, the down regulation of Cdc25C induced by BRCA1 overexpression apparently requires ERK1/2 activation.

To investigate the mechanisms involved in regulating Cdc25C protein expression by BRCA1, we studied the potential effect of ectopic BRCA1 expression as well as ERK1/2 inhibition on Cdc25C mRNA expression in MCF-7 cells following incubation of Ad.BRCA1- or Ad.Control-infected cells in the presence of increasing doses of U0126 for 24 h. RNA was extracted and analysed for Cdc25C mRNA expression using Northern blot hybridization and reverse transcriptase–polymerase chain reaction (RT–PCR) as described in Materials and methods. The results shown in Figure 4c indicate that Ad.BRCA1 infection alone had no effect on Cdc25C mRNA (lane 4 vs lane 1). Furthermore, incubation of Ad.BRCA1- or Ad.Control-infected MCF-7 cells in presence of U0126 also had no detectable effect on Cdc25C mRNA levels present in MCF-7 cells as determined by either Northern blot analysis (Figure 4c, upper panel) or by RT–PCR (Figure 4c, bottom panel). Thus, the down regulation of Cdc25C protein expression by BRCA1 overexpression and the reversal of this effect by ERK1/2 kinase inhibition apparently involve post-transcriptional regulatory mechanisms.

We therefore examined the influence of ectopic BRCA1 expression and ERK1/2 activation on Cdc25C protein stability. For this study, MCF-7 cells were first infected with either Ad.BRCA1 or Ad.Control at 50 PFU/cell. Following 24 h incubation at 37°C, virus-infected cells were then treated with 10 μg/ml cycloheximide for 1 h (MacLachlan et al., 2002), a concentration shown to inhibit >90% of protein synthesis in MCF-7 cells, as determined by incorporation of [35S]methionine into trichloroacetic acid-precipitatable material (data not shown). Ad.BRCA1- or Ad.Control-infected cells were then incubated in the presence or absence of 50 μ M U0126 for the times indicated and analysed for Cdc25C protein levels by immunoblotting. Results shown in Figure 4d indicate that the half-life of Cdc25C protein in Ad.Control-infected cells (bottom panel) was approximately 12 h, which was similar to that observed in uninfected MCF-7 cells (data not shown). Furthermore, exposure of Ad.Control-infected cells to 50 μ M U0126 did not result in any noticeable change in Cdc25C protein stability compared to untreated Ad.Control-infected cells (Figure 4d, bottom panel, Cdc25C). In contrast, in Ad.BRCA1-infected MCF-7 cells, the half-life of Cdc25C protein was reduced to less than 6 h, an effect that was completely inhibited by the treatment of cells with 50 μ M U0126 (Figure 4d, upper panel, Cdc25C). Thus, BRCA1 overexpression leads to a destabilization of Cdc25C protein, an effect that apparently requires ERK1/2 activation.

To examine the specificity of BRCA1-induced change in Cdc25C protein stability, the half-life of p21WAF1/CIP1 protein was also examined in parallel. Our results showed that the half-life of p21WAF1/CIP1 was approximate 1 h in Ad.BRCA1-infected cells. In contrast to the results obtained with Cdc25C, the protein half-life of p21WAF1/CIP1 was unchanged by the treatment of Ad.BRCA1-infected cells with 50 μ M U0126 (data not shown). These studies indicate that BRCA1-induced activation of ERK1/2 specifically enhances Cdc25C degradation and has no influence on p21WAF1/CIP1 protein stability.

We further examined whether Cdc25C protein degradation induced by BRCA1 expression is proteasome dependent. MCF7 cells were infected with Ad.BRCA1 or Ad.Control (50 PFU/cell) and then incubated in the presence of 20 μ M lactacyctin, a proteasome-specific inhibitor (Chen et al., 2002). For comparison, parallel sets of virus-infected cells were incubated in medium containing either 0.1% DMSO vehicle solution (Figure 4e, lanes 1–2) or 50 μ M U0126 (Figure 4e, lane 3). Following 24 h incubation, the resulting cell samples were collected and analysed for Cdc25C protein levels by immunoblotting. As shown in Figure 4e, treatment of Ad.BRCA1-infected cells with lactacystin completely inhibited BRCA1-induced destabilization of the Cdc25C protein (Figure 4e, lane 5 vs lane 2). The level of Cdc25C protein detected in those cells was similar to that present in Ad.Control-infected cells treated with lactacystin (Figure 4e, lane 5 vs lane 4) and slightly higher than the Cdc25C level in Ad.Control-infected cells incubated without lactacystin (Figure 4e, lanes 4–5 vs lane 1).

In summary, the results described above suggest that BRCA1 expression in MCF-7 cells results in a proteasome-directed degradation of Cdc25C protein, and that this effect requires functional ERK1/2 kinases.

Discussion

To maintain the integrity of the genome, cells must have properly regulated DNA damage checkpoints. Activation of these checkpoints induces cell cycle arrest that allows time for repair of damaged DNA prior to cell cycle progression. BRCA1 has been suggested to play a critical role in mediating G2/M checkpoint cellular response following DNA damage. G2/M transition of the cell cycle is controlled by the activity of Cyclin B/Cdc2 complex, whose activation is absolutely required for cells to cross the G2/M border of the cell cycle. Previous studies have indicated that DNA damage-induced cellular response involves phosphorylation of Tyr15 residue of Cdc2, resulting in inhibition of CyclinB/Cdc2 activity and subsequent G2/M arrest (Smits and Medema, 2001). Previous studies from our laboratory and others have also shown that increased BRCA1 expression results in G2/M arrest and is associated with alterations in the activities and/or protein levels of Chk1, Cdc25C, and 14-3-3, the key regulators of Cdc2 (Aprelikova et al., 2001; Yan et al., 2002; Yarden et al., 2002). However, the precise molecular mechanism by which BRCA1 regulates this G2/M checkpoint signaling cascade remains unclear.

In this study, we investigated the specific effects of ERK1/2 signaling on the BRCA1-mediated G2/M arrest, and found that ERK1/2 activation by ectopic BRCA1 expression is necessary for BRCA1-mediated activation of G2/M arrest in MCF-7 cells. Furthermore, inhibition of ERK1/2 kinases specifically blocks the effects of BRCA1 on members of the G2/M signaling cascade, which include down regulation of the Cdc25C protein and activation of the Wee1 and Chk1 kinases. These results indicate that ERK1/2 signaling is necessary for BRCA1-mediated G2/M arrest and involves multiple pathways that affect the activity of Cdc2 kinase. In contrast to its effect on the G2/M signaling cascade regulating Cdc2-Tyr15 phosphorylation, ERK1/2 inhibition had no influence on the levels of Cyclin B1 and 14-3-3 in Ad.BRCA1-infected MCF-7 cells (data not shown).

Previous studies have indicated that wild-type BRCA1 expression is necessary for effective G2/M checkpoint in cells following radiation treatment (Xu et al., 2001; Yarden et al., 2002). Results in this study provide evidence that ERK1/2 signaling is apparently involved in the activation of downstream G2/M checkpoint signaling cascade following ectopic BRCA1 expression as well as in response to γ-irradiation. Although numerous reports have shown that DNA damage stimuli including γ-irradiation can induce G2/M cell cycle arrest as well as ERK1/2 activation (Smits and Medema, 2001; Xu et al., 2002a; Dent et al., 2003a, 2003b; Kawabe, 2004), the relationship between the two events has not yet been studied. In the present study, we found that ERK1/2 kinases are activated in MCF-7 cells following γ-irradiation as well as in response to BRCA1 overexpression. In addition, inhibition of ERK1/2 kinases using MEK1/2-specific inhibitor completely abolished G2/M arrest following γ-irradiation and/or Ad.BRCA1 infection of MCF-7 cells. Furthermore, although γ-irradiation treatment caused G2/M arrest and induction in ERK1/2 phosphorylation in MCF-7 cells, treatment of these cells with γ-irradiation alone did not produce any noticeable effect on BRCA1 protein level (Figure 2e, lane 1 vs 2). Previously studies have shown that BRCA1 is rapidly phosphorylated in response to radiation-induced DNA damage by several nuclear kinases including ATM/ATR and Chk2, and that phosphorylated BRCA1 complexes with several proteins involved in cell cycle checkpoint control as well as DNA-damage repair (Chen, 2000; Deng and Brodie, 2000; Gatei et al., 2000; Paull et al., 2001). Additional studies are needed to identify the role of ERK1/2 kinases in G2/M checkpoint activation following γ-irradiation and/or BRCA1 overexpression.

Cdc25C, a dual specific phosphatase, is the key enzyme that triggers the activation of Cyclin B/Cdc2 by dephosphorylating Cdc2 at Thr14 and Tyr15 (Dunphy, 1994). The overall level of Cdc25C protein is generally constant throughout the cell cycle and the cellular function of Cdc25C is mainly regulated through phosphorylation, which affects both the enzyme activity and the subcellular localization of the protein (Smits and Medema, 2001). Upon DNA damage, Cdc25C is phosphorylated by Chk1 kinases at Ser216 residue, which generates consensus binding sites for 14-3-3 proteins and sequesters Cdc25C protein in the cytoplasm (Peng et al., 1997; Sanchez et al., 1997; Yarden et al., 2002). A previous study by Yarden et al. (2002) showed that expression of wild-type BRCA1 in HCC1937 breast cancer cells induces cytoplasmic localization of Cdc25C through activation of Chk1 kinase. In the present study, we confirm the previous finding that BRCA1 expression induces activation of Chk1 kinase activity. In addition, in this study, we also identified that the mechanism of BRCA1-induced activation of Chk1 requires ERK1/2 activation, as inhibition of ERK1/2 activation abrogates the activation of Chk1 kinase by BRCA1. Thus, BRCA1-induced activation of ERK1/2 is required for the subsequent activation of Chk1.

Yarden et al. (2002) also showed previously that BRCA1 expression in HCC1937 breast cancer cells reduces Cdc25C protein levels. Consistent with this previous finding, results in this study indicate that BRCA1 expression also down regulates Cdc25C protein levels in MCF-7 cells. Furthermore, studies of Cdc25C mRNA expression as well as Cdc25C protein stability indicate that BRCA1-induced down regulation of Cdc25C protein involves post-translational control machineries and is proteasome dependent. More importantly, this effect of BRCA1 on regulation of Cdc25C degradation apparently involves BRCA1-induced ERK1/2 activation, since inhibition of ERK1/2 activity completely abrogates the down regulation of the Cdc25C protein level by BRCA1.

In summery, the results of this study provide evidence that BRCA1-induced G2/M arrest in MCF-7 cells involves regulation of Cdc2 activity through three distinct mechanisms, which include BRCA1-induced activation of Chk1 and Wee1 kinases, and BRCA1-induced down regulation of Cdc25C protein levels. Each of these mechanisms results in a decrease in Cdc2 kinase activity. Of particular interest is the finding that the BRCA1-induced activation of ERK1/2 is implicated in each of these mechanisms of Cdc2 regulation. Thus, studies in this report indicate that activation of ERK1/2 plays a central role, either directly or indirectly, in the induction of G2/M cell cycle arrest by BRCA1 through alterations in Cdc25C protein stability and the activation of Chk1 and Wee1 kinases. Although the ERK1/2 signaling pathway is known to be activated primarily by mitogen stimulation, accumulating studies have also shown that ionizing radiation can efficiently induce ERK1/2 signaling as well (Dent et al., 1999; Hagan et al., 2000; Dent et al., 2003a, 2003b). While the induction of G2/M arrest by irradiation is well documented and previous studies have shown that BRCA1 expression is necessary for radiation-induced G2/M arrest (Yarden et al., 2002), the potential role of ERK1/2 signaling in radiation-induced cell cycle arrest has not yet been defined. Additional studies will be needed to identify the specific mechanisms of BRCA1-induced activation of ERK1/2 as well as the precise role of ERK1/2 on the regulation of Wee1 and Chk1 kinases, and Cdc25C protein stability.

Materials and methods

Cell culture and drug treatment

The human breast cancer cell line MCF-7 was obtained from ATCC (Manassas, VA, USA). Cells were maintained in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum in an atmosphere of 5% CO2. For ERK1/2 inhibition studies, log-phase cells (1 × 106 cells per 100-mm dish) were incubated in medium containing U0126 or PD98059 (CALBIOCHEM, San Diego, CA, USA) dissolved in DMSO. Control incubation without drug but containing the same amount of DMSO (final concentration 0.1%) was performed in parallel. For experiments involving adenoviral vectors, drug treatment was started 2 h following initial exposure to adenoviral vectors unless indicated elsewhere. For experiments involving irradiation treatment, exponentially growing cells were treated with or without irradiation at 10 Gy and then incubated for 24 h at 37°C. For comparison, a set of cell samples infected with 30 PFU/cell Ad.BRCA1 for 24 h were also treated with or without γ-irradiation (10 -Gy) and incubated in parallel. When U0126 was included in the experiment, U0126 (50 μ M) was either preincubated with uninfected MCF-7 cells for 2 h prior to γ-irradiation or added after 2 h following Ad.BRCA1 infection. Following incubation, cells were analysed by Western blotting for levels of BRCA1 and ERK1/2 phosphorylation and by flow cytometry for DNA content. For serum starvation, the medium was removed from log-phase growing cells, the cells were washed once with serum-free DMEM and then incubated in medium containing 0.1% serum. The cells were maintained under serum starvation condition for 24 h prior to further analysis. For proteasome inhibitor studies, Ad.Control- or Ad.BRCA1-infected cells were incubated in medium containing 20 μ M lactacyctin (dissolved in DMSO) (Santa Cruz Biotechnology, Santa Cruz, CA, USA) as described previously (Chen et al., 2002). As a control, duplicate cell samples were incubated in medium containing the same amount of vehicle solution DMSO (0.1%).

Adenoviral vectors and adenoviral infections

Ad.BRCA1 and Ad.Control adenoviral vectors are nonreplicative E1/E3-defective recombinant adenoviruses carrying a wild-type BRCA1 cDNA or an empty cassette, respectively. Construction of the adenoviruses and the viral infection procedures have been described previously (Yan et al., 2002). Recombinant adenovirus Ad.MEK1(dn) was obtained from Dr J Han (The Scripps Research Institute, La Jolla, CA, USA). In Ad.MEK1(dn) vector, the MEK1 cDNA has been altered and two crucial serine residues located in the catalytic domain (Ser-217 and Ser-221) were replaced by alanines. The resulting MEK1 mutant has dominant negative activity and can be used to block the activation of ERK1/2 by wild-type MEK1 (Alessi et al., 1994; Cowley et al., 1994).

Log-phase cells were infected at 50 PFU/cell with either Ad.BRCA1 or Ad.Control for 24 h prior to analysis. In experiments involving two adenoviruses, Ad.MEK1(dn) was always transferred to the cells first (at 100 PFU/cell), 15 h prior to the second infection with the Ad.BRCA1 virus. In these experiments, samples were then collected at the indicated time points following the second infection.

Cell cycle analysis

Cells were harvested, washed with phosphate-buffered saline (PBS), and fixed in 70% ethanol. Fixed cells were stained with propidium iodide (PI) and analysed for DNA content by measuring the intensity of the fluorescence produced by PI using FACSCalibur instrument (Beckon Dickinson, Mountain View, CA, USA) as recommended by the manufacturer.

Antibodies and recombinant proteins

All antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA) unless indicated elsewhere. These include monoclonal antibodies specific for BRCA1 (D-9), Cdc2 (17), Cdc25C (H-6), Chk1 (G-4), Chk2 (B-4), and 14-3-3 (H-8); and affinity-purified rabbit polyclonal antibodies specific for BRCA1 (H-100), Cdc25C (C-20), Chk1 (FL-476), Chk2 (H-300), and Wee1 (C-20). Antibody against phospho-ERK1/2 (P-ERK1/2) was a mouse monoclonal IgG that binds to a short amino-acid sequence of ERK1 and ERK2 containing phosphorylated Tyr-204. Antibody C-14-G is a goat polyclonal IgG that recognize full-length polypeptides of ERK2 and to a lesser extent ERK1. Antibody P-Cdc2 (Tyr15) is a goat polyclonal IgG that specifically recognizes the amino-acid sequence containing phosphorylated Tyr15 of Cdc2. Antibody SKB1 is a mouse monoclonal IgG recognizing the PH domain of Akt protein of human origin (Upstate biotechnology, Lake Placid, NY, USA). To confirm that equal amounts of protein were loaded, an affinity-purified anti-Actin goat IgG (I-19) was used for quantitating Actin protein levels on all immunoblots.

The recombinant proteins used are: Cdc2 protein, substrate for Wee1 kinase assay, was purified as a glutathione S-transferase fusion protein containing full-length human Cdc2 (Santa Cruz Biotechnology, Santa Cruz, CA, USA); Cdc25C protein, substrate for Chk1 kinase assay, was a glutathione S-transferase fusion containing residues 200–256 of human Cdc25C (kindly provided by Dr Y Pommier (NIH)); glutathione S-transferase protein was used as a negative control substrate in all kinase assays and was prepared according to the standard procedure suggested by the manufacturer (Pharmacia, Piscataway, NJ, USA). AKT/SGK Specific Peptide, substrate for AKT kinase assay, contains sequence of RPRAATF that is related to the sequence surrounding the phosphorylation site of glycogen synthase kinase-3 (Upstate biotechnology, Lake Placid, NY, USA).

Immunoblotting, Immunoprecipitation, and kinase assays

Preparation of cell lysates, immunoblotting, immunoprecipitation were performed as described previously (Yan et al., 1997; Yan and Mumby, 1999) except that cell lysis buffer for immunoprecipitation was 50 mM Tris-HCL, pH 7.5; 1% Triton X-100; 0.5 mM Na3VO4; 50 mM sodium fluoride; 5 mM sodium pyrophosphate; 10 mM sodium 2-glycerophosphate; 0.1 mM PMSF; 1 μg/ml aprotinin; 1 μg/ml pepstatin; 1 μg/ml leupeptin; 1 μ M microcystin; 1 mM DTT; 1 mM EDTA, and 1 mM EGTA. Kinase assays for Chk1 and Wee1 were performed as described previously (McGowan and Russell, 1993; McGowan and Russell, 1995; Yarden et al., 2002). Each kinase assay used 500 μg of cell extract for immunoprecipitating kinase, and 2 μg of relative protein substrate. Kinase reaction was carried out by incubation for 30 min at 30°C and terminated by addition of 30 μl of 6 × Laemmli SDS sample buffer (Yan and Mumby, 1999). Substrate phosphorylation (Wee1 and Chk1 kinase assay) was analysed by SDS–polyacrylamide gel electrophoresis and autoradiography. AKT Kinase assay and analysis of substrate peptide phosphorylation were performed according to the manufacturer's instructions (Upstate biotechnology, Lake Placid, NY, USA).

To quantitate protein levels on Western blot, specific protein signals were visualized by chemiluminescence exposed to X-ray film and then scanned using EPSON Perfection 1200XPHOTO scanner. The intensity of those individual signals was analysed using NIH Image1.60 analytical program.

Northern blot and RT–PCR

Logarithmically growing MCF-7 cells were infected with Ad.BRCA1 or Ad.Control at 50 PFU/cell and incubated at 37°C for 24 h in the presence of U0126 at the doses of 0, 10, and 50 μ M. Total RNA was isolated using the RNA isolation system (Invitrogen, San Diego, CA, USA). A 20 μg sample of the RNA was separated by electrophoresis on a 1.0% agarose gel containing 7% formaldehyde, and transferred onto a nylon membrane. The membrane was hybridized with a 32P-labeled Cdc25C full-length cDNA probe. To confirm equal loading of RNA samples, the membrane was stripped and rehybridized with a 32P-labeled glyceraldehyde-3-phosphate dehydrogenase cDNA fragment.

To confirm the results obtained from the Northern blot analysis described above, duplicate cell samples were analysed in parallel by RT–PCR using the SuperSpript™ One-Step RT–PCR system (Invitrogen, San Diego, CA, USA). The primers used for analysis of Cdc25C expression were: IndexTerm5′-CGATGCCAGAGAGAACTTGAAC-3′ and IndexTerm5′-TGGATAGCGACAATCAATGAC-3′. As a control, we also analysed Actin RNA levels in samples using the primers IndexTerm5′-CACGAAACTACCTTCAACTCC-3′ and IndexTerm5′-CAAATAAAGCCATGCCAATCTC-3′.

Abbreviations

Ad:

adenovirus

CHX:

cycloheximide

ERK:

extracellular signal-regulated protein kinase

pERK:

phosphorylated-extracellular signal-regulated protein kinase

FACS:

fluorescence-activated cell sorting

JNK:

c-Jun N-terminal kinase

MAPK:

mitogen-activated protein kinase

MEK1/2:

mitogen-activated protein kinase kinase 1 and 2

PI:

propidium iodide

PI3-kinase:

phosphoinositide kinase 3

RT–PCR:

reverse transcriptase–polymerase chain reaction

References

  1. Alessi DR, Saito Y, Campbell DG, Cohen P, Sithanandam G, Rapp U, Ashworth A, Marshall CJ and Cowley S . (1994). EMBO J., 13, 1610–1619.

  2. Aprelikova O, Pace AJ, Fang B, Koller BH and Liu ET . (2001). J. Biol. Chem., 276, 25647–25650.

  3. Atherton-Fessler S, Parker LL, Geahlen RL and Piwnica-Worms H . (1993). Mol. Cell. Biol., 13, 1675–1685.

  4. Chen F, Zhang Z, Bower J, Lu Y, Leonard SS, Ding M, Castranova V, Piwnica-Worms H and Shi X . (2002). Proc. Natl. Acad. Sci. USA, 99, 1990–1995.

  5. Chen J . (2000). Cancer Res., 60, 5037–5039.

  6. Chung J, Grammer TC, Lemon KP, Kazlauskas A and Blenis J . (1994). Nature, 370, 71–75.

  7. Cortez D, Wang Y, Qin J and Elledge SJ . (1999). Science, 286, 1162–1166.

  8. Cowley S, Paterson H, Kemp P and Marshall CJ . (1994). Cell, 77, 841–852.

  9. Davies SP, Reddy H, Caivano M and Cohen P . (2000). Biochem. J., 351, 95–105.

  10. Deng CX and Brodie SG . (2000). BioEssays, 22, 728–737.

  11. Dent P, Reardon DB, Park JS, Bowers G, Logsdon C, Valerie K and Schmidt-Ullrich R . (1999). Mol. Biol. Cell., 10, 2493–2506.

  12. Dent P, Yacoub A, Contessa J, Caron R, Amorino G, Valerie K, Hagan MP, Grant S and Schmidt-Ullrich R . (2003a). Radiat. Res., 159, 283–300.

  13. Dent P, Yacoub A, Fisher PB, Hagan MP and Grant S . (2003b). Oncogene, 22, 5885–5896.

  14. Dunphy WG . (1994). Trends Cell Biol., 4, 202–207.

  15. Favata MF, Horiuchi KY, Manos EJ, Daulerio AJ, Stradley DA, Feeser WS, Van Dyk DE, Pitts WJ, Earl RA, Hobbs F, Copeland RA, Magolda RL, Scherle PA and Trzaskos JM . (1998). J. Biol. Chem., 273, 18623–18632.

  16. Gatei M, Scott SP, Filippovitch I, Soronika N, Lavin MF, Weber B and Khanna KK . (2000). Cancer Res., 60, 3299–3304.

  17. Graves PR, Lovly CM, Uy GL and Piwnica-Worms H . (2001). Oncogene, 20, 1839–1851.

  18. Hagan M, Wang L, Hanley JR, Park JS and Dent P . (2000). Radiat Res., 153, 371–383.

  19. Harkin DP, Bean JM, Miklos D, Song YH, Truong VB, Englert C, Christians FC, Ellisen LW, Maheswaran S, Oliner JD and Haber DA . (1999). Cell, 97, 575–586.

  20. Jin P, Gu Y and Morgan DO . (1996). J. Cell Biol., 134, 963–970.

  21. Kawabe T . (2004). Mol. Cancer Ther., 3, 513–519.

  22. Kharbanda S, Saleem A, Datta R, Yuan ZM, Weichselbaum R and Kufe D . (1994). Cancer Res., 54, 1412–1414.

  23. Lee JS, Collins KM, Brown AL, Lee CH and Chung JH . (2000). Nature, 404, 201–204.

  24. Lundgren K, Walworth N, Booher R, Dembski M, Kirschner M and Beach D . (1991). Cell, 64, 1111–1122.

  25. MacLachlan TK, Somasundaram K, Sgagias M, Shifman Y, Muschel RJ, Cowan KH and El-Deiry WS . (2000). J. Biol. Chem., 275, 2777–2785.

  26. MacLachlan TK, Takimoto R and El-Deiry WS . (2002). Mol. Cell. Biol., 22, 4280–4292.

  27. McGowan CH and Russell P . (1993). EMBO J., 12, 75–85.

  28. McGowan CH and Russell P . (1995). EMBO J., 14, 2166–2175.

  29. Miki Y, Swensen J, Shattuck-Eidens D, Futreal PA, Harshman K, Tavtigian S, Liu Q, Cochran C, Bennett LM and Ding W . (1994). Science, 266, 66–71.

  30. O'Connell MJ, Raleigh JM, Verkade HM and Nurse P . (1997). EMBO J., 16, 545–554.

  31. Parker LL, Atherton-Fessler S and Piwnica-Worms H . (1992). Proc. Natl. Acad. Sci. USA, 89, 2917–2921.

  32. Paull TT, Cortez D, Bowers B, Elledge SJ and Gellert M . (2001). Proc. Natl. Acad. Sci. USA, 98, 6086–6091.

  33. Peng CY, Graves PR, Thoma RS, Wu Z, Shaw AS and Piwnica-Worms H . (1997). Science, 277, 1501–1505.

  34. Qin J and Li L . (2003). Radiat Res., 159, 139–148.

  35. Rhind N, Furnari B and Russell P . (1997). Genes Dev., 11, 504–511.

  36. Robinson MJ and Cobb MH . (1997). Curr. Opin. Cell Biol., 9, 180–186.

  37. Sanchez Y, Wong C, Thoma RS, Richman R, Wu Z, Piwnica-Worms H and Elledge SJ . (1997). Science, 277, 1497–1501.

  38. Scott PH, Brunn GJ, Kohn AD, Roth RA and Lawrence Jr JC . (1998). Proc. Natl. Acad. Sci. USA, 95, 7772–7777.

  39. Smits VA and Medema RH . (2001). Biochim. Biophys. Acta, 1519, 1–12.

  40. Thangaraju M, Kaufmann SH and Couch FJ . (2000). J. Biol. Chem., 275, 33487–33496.

  41. Tibbetts RS, Cortez D, Brumbaugh KM, Scully R, Livingston D, Elledge SJ and Abraham RT . (2000). Genes Dev., 14, 2989–3002.

  42. Venkitaraman AR . (2002). Cell, 108, 171–182.

  43. Wang Y, Cortez D, Yazdi P, Neff N, Elledge SJ and Qin J . (2000). Genes Dev., 14, 927–939.

  44. Xu B, Kim S and Kastan MB . (2001). Mol. Cell. Biol., 21, 3445–3450.

  45. Xu B, Kim ST, Lim DS and Kastan MB . (2002a). Mol. Cell. Biol., 22, 1049–1059.

  46. Xu B, O'Donnell AH, Kim ST and Kastan MB . (2002b). Cancer Res., 62, 4588–4591.

  47. Xu X, Weaver Z, Linke SP, Li C, Gotay J, Wang XW, Harris CC, Ried T and Deng CX . (1999). Mol. Cell, 3, 389–395.

  48. Yan Y, Haas JP, Kim M, Sgagias MK and Cowan KH . (2002). J. Biol. Chem., 277, 33422–33430.

  49. Yan Y and Mumby MC . (1999). J. Biol. Chem., 274, 31917–31924.

  50. Yan Y, Shay JW, Wright WE and Mumby MC . (1997). J. Biol. Chem., 272, 15220–15226.

  51. Yang X and Lippman ME . (1999). Breast Cancer Res. Treat., 54, 1–10.

  52. Yarden RI, Pardo-Reoyo S, Sgagias M, Cowan KH and Brody LC . (2002). Nat. Genet., 30, 285–289.

Download references

Acknowledgements

We thank Dr J Han for providing Ad.MEK1(dn) adenoviral expression vector, Dr Y Pommier for GST-Cdc25C construct. We also thank Dr Charles A Kuzynski and Linda A Wilkie for assistance on the flow cytometry analysis.

Author information

Correspondence to Kenneth H Cowan.

Additional information

Supplementary Information accompanies the paper on Oncogene website (http://www.nature.com/onc)

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Yan, Y., Spieker, R., Kim, M. et al. BRCA1-mediated G2/M cell cycle arrest requires ERK1/2 kinase activation. Oncogene 24, 3285–3296 (2005) doi:10.1038/sj.onc.1208492

Download citation

Keywords

  • BRCA1
  • Cdc2
  • Cdc25C
  • Chk1
  • ERK1/2 and Wee1

Further reading