Brk is a cytoplasmic nonreceptor tyrosine kinase that is overexpressed in breast tumors but undetectable in normal or benign mammary tissues. Brk promotes proliferation of human mammary epithelial cells and tumor growth in a mouse model, but the role of Brk in cell cycle regulation is not known. In this study, we describe the mechanism of Brk-induced deregulation of the cell cycle. We provide evidence that Brk antagonizes the transcriptional activity of the transcription factor FoxO family of proteins by inhibiting its nuclear localization. As a result, the cell cycle inhibitor p27, a FoxO target gene, is down-regulated. This event is accompanied by G1/S cell cycle progression of quiescent cells. As p27 is a key regulator of the G1/S cell cycle checkpoint, these data suggest that perturbation of p27 expression induced by Brk causes S phase entrance. Deregulation of the cell cycle is a key event in neoplasia, and thus, the mechanism presented here likely contributes to breast cancer development.
Breast tumor kinase (Brk, PTK6) is a cytoplasmic nonreceptor tyrosine kinase associated with breast tumors. Whereas normal mammary tissues or benign lesions express low or undetectable levels, ∼65% of breast tumors express Brk, with 27% of the tumors overexpressing Brk by at least 5-fold.1 Brk is also expressed in other cancers including metastatic melanomas, and colon and prostate tumors.2 Considered to be distantly related to the Src-family protein tyrosine kinases, Brk contains SH2, SH3, and tyrosine kinase domains, and is capable of autophosphorylation.3 Brk promotes proliferation of human mammary epithelial cells and tumor growth in a mouse model.4, 5 Studies show that Brk is co-overexpressed with and promotes the mitogenic signaling of epidermal growth factor receptors (EGFR).5, 6 Brk interacts with EGFR and ErbB3 and, as a consequence, sensitizes the mitogenic effect of EGF and activates the ErbB3/PI3-kinase/Akt pathway, respectively.6–8 A recent study shows that Brk and ErbB2 genes coamplify in human breast cancers and the interaction between Brk and ErbB2 enhances the intrinsic kinase activity of Brk and mitogenic signaling of ErbB2.5 Furthermore, Brk activates the Ras and Rac signaling pathways and insulin receptor substrate-4 (IRS-4), phosphorylates and activates signal transducer and activator of transcription (STAT) 3 and STAT5b, key mediators of cytokine and growth factor signaling, and activates the MAPK pathway.8–13 Together, these data suggest that complex mitogenic signaling pathways are involved in the regulation of cell proliferation by Brk.
Cell cycle deregulation plays an important role in neoplasia. Progression of the cell cycle from quiescence to mitosis is tightly regulated at multiple steps by cyclins and cyclin-dependent kinases (CDKs). CyclinD/CDK4/6 and CyclinE/CDK2 complexes activate the transcription factor E2F resulting in the expression of genes necessary for G1 to S phase progression. The kinase activity of CDKs is inhibited by CDK inhibitors (CDKIs), which bind to the cyclin/CDK complexes and block cell cycle progression.14 The CDKI p27(Kip1) (hereafter called p27), whose expression and activity are modulated by mitogenic signaling, connects the extracellular environment to cell cycle regulation. In the absence of mitogens, the cellular level and activity of p27 are upregulated leading to cell cycle arrest.15 In many types of cancer, p27 is deregulated through mechanisms, such as gene expression, proteolysis and functional inactivation.15 In this study, we investigated the role of Brk in the regulation of the cell cycle. We found that Brk downregulates the expression of p27 through a mechanism that involves the modulation of the subcellular localization of the transcription factor FoxO3a, a p27 transcription regulator. As a consequence, the cell cycle is deregulated and quiescent cells enter S-phase in a growth factor-independent manner. These data provide insight into the mechanism of cell cycle deregulation induced by Brk, which may contribute to the development of breast cancer.
Material and Methods
Cell culture, constructs and reagents
The breast cancer cell line MDA-MB-231 was purchased from American Type Culture Collection. The cells were maintained as described by the manufacturer. cDNAs encoding human Brk (generous gift from Dr. Todd Miller) were subcloned into pCGN vector containing HA-tag epitope. The plasmid was transfected into the cells utilizing lipofectamine-2000 (Invitrogen). siRNAs targeting Brk (ON-TARGET plus smart pool predesigned PTK6, # L003166-00), FoxO3a (ON-TARGET plus smart pool #L003007-00-0005) and control (ON-TARGET plus siControl nontarget pool, #D001810-10) were purchased from Dharmacon. The siRNAs were transfected into cells utilizing Dharmafect (Dharmacon). siRNA targeting p27 (Stealth Select RNAi™ siRNA, # 1299003) and control (Stealth Select RNAi™ siRNA negative control, # 12935-400) were purchased from Invitrogen and transfected into cells utilizing Lipofectamine 2000.
Cell viability was assayed using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) colorimetric assay according to the manufacturer's protocol (Roche Diagnostics). The cells were seeded in 96-well tissue culture dishes at a density of 1 × 104 cells per well. Eighteen hours after seeding, cells were transfected with plasmids or siRNAs, as indicated. Forty-eight hours after transfection, the cells were treated with MTT reagent I followed by 4 hr incubation at 37°C, subsequently, MTT reagent II was added and the incubation continued overnight. Cell number was determined by spectrophotometry utilizing an ELISA plate reader at 590 nm.
Flow cytometry and cell cycle analysis
The cells were labeled with 10 μM BrdU 45 min before trypsinization. The cell pellets were washed with PBS and fixed in 70% ethanol and stored at 4°C. Cells were collected by centrifugation. The cells were denatured with 2M HCl, neutralized with 0.1M sodium borate and incubated with anti-BrdU antibody (BD biosciences), followed by fluorescence-conjugated goat-anti mouse (Invitrogen). The cell pellets were resuspended in 10 μg/ml of propidium iodide containing RNase and the suspension was incubated in the dark at room temperature for 30 min. The cell suspension was analyzed for DNA content on a FACScan flow cytometer (Becton Dickinson). The percent of cells in different phases of the cell cycle was analyzed with CELLQUest software (Becton Dickinson)
DNA synthesis (BrdU incorporation)
Cells were serum-starved for 30 hr before incubation with BrdU (10 μM) for 10 hr. Subsequently, the cells were fixed with ethanol/acetic acid (95:5) and immunostained with BrdU antibody (Sigma), followed by Alexa-conjugated goat-anti mouse (Invitrogen) and DAPI (Sigma). In brief, after fixation, cells were rehydrated in PBS, washed with TBST (20 mM Tris, 150 mM NaCl and 0.1% Tween20) and denatured in 2N HCl. Subsequently, cells were neutralized in 0.1M sodium borate (pH 8.5) and rinsed with TBST before incubation with BrdU antibody for 1 hr, and secondary antibody for 1 hr. DAPI was added at the last 15 min. The image was captured by a Zeiss Axiovert 200M microscope (Zeiss).
Cells were lysed in lysis buffer containing 150 mM NaCl, 10 mM Tris-HCl (pH 7.4), 1% Triton X-100, 10% glycerol, 1 mM EDTA, 10 mM NaF, 1 mM sodium orthovanadate and protease inhibitors. The cell lysates were centrifuged at 14,000 rpm for 10 min and the supernatant was treated with 5× Laemlli buffer. The proteins were resolved by SDS-PAGE.
Total RNA from cells was isolated by TRIzol (Invitrogen) and purified by RNeasy Mini Kit and RNase-free DNase Set (Qiagen) according to the manufacturer's protocols. Reverse transcription was carried out using a SuperScript Preamplification Kit (Life Technologies) on 1 μg of total RNA aliquots. PCR was performed using capillary LightCycler (Roche, Indianapolis, IN) with the following p27 primers: 5′-TGG CCA GGA TTG CTA CAG TTG-3′ and 5′-CGC CGC GGA CAT CAT CTT-3′
Cells grown on coverslips were fixed for 1h with 3.7% formaldehyde. After fixation, the cells were washed with PBS, permeabilized with 0.1% Triton X-100, rinsed with PBS and blocked with 2% BSA before incubation with FoxO3a antibody for 1 hr, and Rhodamine-conjugated goat anti-rabbit (Invitrogen) for 1 hr. The images were captured using a Zeiss Axiovert 200M digital deconvolution microscope (Zeiss).
Cells were lysed in hypotonic buffer (10 mM HEPES pH7.4, 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, 2 mM DTT, 10 mM NaF, 1 mM sodium orthovanadate and protease inhibitors). Nuclei were sedimented by centrifugation (2,000 rpm, 10 min) and the cytoplasmic fraction was retained. The nuclei were extracted in the hypotonic buffer containing 1% NP40 and 400 mM NaCl. The nuclear and cytoplasm fractions were resolved by SDS-PAGE.
Brk promotes cell proliferation
We first assessed the role of Brk in the proliferation of MDA-MB-231, an estrogen receptor- and ErbB2-negative breast carcinoma cell line expressing detectable levels of endogenous Brk. The cells were transfected with siRNA targeting Brk (siBrk) to suppress the expression of Brk or scramble siRNA (siSc) as a control. Forty-eight hours after transfection, cell proliferation was analyzed by MTT assay. As shown in Figure 1a, suppression of Brk caused a significant decrease in cell proliferation. We did not observe cell death in the culture suggesting that this decrease was not due to loss of cell viability. Similarly, knockdown of endogenous Brk in the breast carcinoma cell lines T47D and SKBr3 resulted in a decrease in cell number of 60 and 50%, respectively, (data not shown) indicating that breast cancer cells require Brk for cell proliferation. We next examined whether elevated levels of Brk would enhance the cell proliferation rate. MDA-MB-231 cells were transfected with pCGN HA-tagged Brk (pCGNBrk), BrkK219M, a kinase defective mutant of Brk (pCGNBrk-KD), or empty vector. Forty-eight hours after transfection, the MTT assay was performed. Overexpression of Brk enhanced cell proliferation, whereas BrkK219M had no effect on cell proliferation (Fig. 1b). Together, these results indicate that Brk plays an important role in the proliferation of these cells and the mechanism of the regulation is dependent on the kinase activity of Brk.
Brk induces G1 to S phase progression
To maintain normal cell proliferation, progression through the cell cycle is tightly regulated at multiple steps. Growth factor stimulation transiently downregulates the activity of cell cycle inhibitors allowing cells to exit quiescence and enter the cell cycle. Several oncogenes induce cell cycle progression in the absence of mitogens by deregulating the cell cycle checkpoint machinery. Brk has previously been implicated in G1/S phase transition4 and has been implicated as an oncogene. Therefore, we investigated whether Brk was able to induce S-phase entry in a growth factor-independent manner. Bromodeoxyuridine (BrdU) incorporation assay was performed to monitor DNA synthesis. MDA-MB-231 cells were transfected with pCGNBrk or vector, and the cells were serum starved before BrdU addition. Subsequently, cells were fixed, immunostained and scored for cells that entered S phase. As demonstrated in Figures. 2a and 2b, 70% of cells transfected with Brk had incorporated BrdU, compared with 18% of control cells. This data suggests that Brk induces G1/S phase progression in a growth factor-independent manner. To confirm that Brk induces G1/S phase progression, FACS analysis was performed. Cells were transfected with pCGNBrk or control vector and serum starved for the indicated times. Subsequently, the cells were double labeled with BrdU and propidium iodide before flow cytometry. As shown in Figures 2c and 2d, the S-phase population of Brk-transfected cells is ∼2-fold higher than that of control cells, with Brk promoting G1 to S phase progression while control cells entered quiescence. These results support the immunostaining data that Brk induces cell cycle progression in the absence of serum. The population of Brk-transfected cells entering S-phase does, however, progressively decline over time (Figures 2c and 2d). By 48 h, less than 20% of Brk-expressing cells are in S-phase, while less than 5% of control cells were detected in S-phase (data not shown). Together, these data suggest that Brk plays a key role in S phase progression by promoting S phase entry and delaying S phase exit in a growth-factor independent manner.
Brk inhibits p27 expression
We next investigated the pathway that connects Brk to cell cycle regulation. The CDKI p27 plays a key role in the regulation of the G1/S cell cycle checkpoint and it is controlled by growth factor levels. Therefore, we investigated whether Brk regulates p27. MDA-MB-231 cells were transfected with pCGNBrk or vector, followed by serum-starvation to induce quiescence. Subsequently, p27 expression levels were observed by immunoblot analysis. Overexpression of Brk induced a significant decrease in p27 levels indicating that Brk down-regulated p27 (Figs. 3a and 3b). To further examine the effect of Brk on p27 expression in a physiological setting, p27 levels were examined in cells in which Brk was knocked down. siBrk was transfected into MDA-MB-231 cells followed by serum-starvation. Suppression of endogenous Brk led to an increase in the levels of p27 (Figs. 3c and 3d) demonstrating that Brk mediated p27 downregulation in a physiological setting. We next examined the effect of Brk on the mRNA levels of p27 by performing quantitative PCR. Cells were transfected with Brk or control vector, followed by serum starvation before RNA extraction and quantitative PCR analysis. As shown in Figure 3e, p27 mRNA levels were significantly lower in cells expressing Brk compared with control cells indicating that Brk downregulated p27 at the mRNA level. In addition to being regulated at the transcriptional level, p27 is also regulated by ubiquitin-mediated degradation. Previously it has been shown that Src induces phosphorylation of tyrosine 88 of p27 which promotes the SCF-Skp2-dependent degradation of p27.16 Therefore, we examined whether Brk had an effect on p27 degradation and phosphorylation. MDA-MB-231 cells were transfected with Brk or vector, followed by serum-starvation. The cells were treated with the proteasome inhibitor MG-132 for 6 hr before analyzing the levels of p27. Treatment with MG-132 did not alter the effect of Brk on p27 suggesting that Brk does not downregulate p27 by protein degradation (Figs. 3f and 3g). Furthermore, analysis of tyrosine phosphorylation in cells transfected with Brk revealed that Brk did not affect the tyrosine phosphorylation of p27 (Supporting Information Fig. S1). As tyrosine-88 is the principal phosphoacceptor site of p27,16 this result suggests that, unlike Src, Brk does not play a role in p27 phosphorylation.
Brk inhibits nuclear localization of FoxO3a
Expression of p27 is activated by the transcription factor forkhead boxO (FoxO) family, which is comprised of 4 members, FoxO1, FoxO3a, FoxO4 and FoxO6. The nuclear localization of FoxO proteins is pivotal to their transcriptional activities, and is negatively regulated by Akt-mediated protein phosphorylation.17 The phosphorylated FoxO proteins bind to the chaperone protein 14-3-3, which disrupts the association of FoxOs with DNA. Subsequently, 14-3-3-bound FoxO proteins are exported into the cytoplasm and, as a consequence, the proteins fail to activate gene expression. As Brk has been shown to activate the PI3-kinase/Akt pathway,6 we hypothesized that Brk downregulates p27 by inducing the nuclear exclusion of FoxO proteins. To test this, we examined the effect of Brk on FoxO subcellular localization. We used FoxO3a (FKHRL1) as it is a major FoxO family protein expressed in MDA-MB-231. pCGNBrk or vector was transfected into MDA-MB-231 cells and subsequently, the cells were serum starved. Following serum starvation, the cells were harvested and subjected to subcellular fractionation followed by SDS-PAGE and immunoblot analysis. Expression of Brk induced a marked reduction in the nuclear fraction of FoxO3a (Figs. 4a and 4b). However, we did not detect a significant increase of the cytoplasmic fraction of FoxO3a in the presence of Brk. As FoxO proteins are degraded in the cytoplasm, we predicted that this may be the reason that there was no significant change detected in levels of the cytoplasmic FoxO3a. To test this, we treated cells with the proteasome inhibitor MG-132. MDA-MB-231 cells were transfected with pCGNBrk or vector, followed by serum-starvation. The cells were treated with MG-132 for 3 hr before analyzing the levels of FoxO3a. Treatment with MG-132 led to elevated levels of cytoplasmic FoxO3a in both control and Brk transfected cells, indicating that FoxO3a is degraded in the cytosol and that Brk induces the sequestration of FoxO3a in the cytoplasm (Supporting Information Fig. S2). Brk had no effect on FoxO3a gene expression, as the mRNA levels of FoxO3a were similar in control and Brk-transfected cells (data not shown). To determine if Brk altered the subcellular localization of FoxO3a, MDA-MB-231 cells were cultured on glass-coverslips and cotransfected with CMV-GFP and pCGNBrk. Subsequently, cells were serum starved, fixed and stained with FoxO3a antibody. The localization of FoxO3a was observed by fluorescence microscopy. As demonstrated in Figure 4c, the majority of FoxO3a localized in the nuclei of control cells, however, in Brk-transfected cells, indicated by GFP expression, FoxO3a was excluded from the nucleus and present in the cytoplasm. This result supports the fractionation data that Brk induces nuclear exclusion of FoxO3a.
We next investigated whether the PI3-kinase/Akt pathway mediated the Brk-induced nuclear exclusion of FoxO3a. MDA-MB-231 cells were transfected with pCGNBrk or vector and serum starved. The cells were incubated with the PI3-kinase inhibitor LY294002 for 14 hr before subcellular fractionation. Treatment with LY294002 inhibited the antagonistic effect of Brk on FoxO3a localization in the nucleus suggesting that the regulation of FoxO3a localization by Brk is mediated by the PI3-kinase pathway (Fig. 4d and 4e).
p27 and FoxO3a play an important role in the regulation of cell proliferation by Brk
We next asked if Brk promotes cell proliferation through downregulation of the FoxO3a/p27 pathway. MDA-MB-231 cells were transfected with siSc, sip27, and/or siRNA targeting FoxO3a (siFoxO3a) together with pCGNBrk or vector, as indicated. Forty-eight hours after transfection, cells were subjected to the MTT assay. As shown in Figure 5a, expression of Brk led to a significant increase in cell proliferation, compared to control (siSc + vector). Knockdown of either p27 or FoxO3a enhanced cell proliferation but to a lesser extent than that of Brk expression, suggesting that there is an additional pathway(s) mediating Brk induction of cell proliferation. Codepletion of both p27 and FoxO3a produced an effect similar to depletion of either one alone suggesting a linear pathway of FoxO3a to p27. Furthermore, expression of Brk in cells depleted of either p27 or FoxO3a or codepleted of p27 and FoxO3a produced an effect on cell proliferation similar to Brk expression alone, indicating a linear signaling pathway from Brk to FoxO3a and subsequently to p27.
As our data indicates that the FoxO3a/p27 pathway is not the sole pathway responsible for Brk inducing cell proliferation, we assessed the contribution of FoxO3a and p27 in Brk-induced cell proliferation. MDA-MB-231 cells were cotransfected with siSc, sip27 or siFoxO3a together with pCGNBrk or vector, as indicated. Forty-eight hours after transfection, cells were subjected to the MTT assay. Expression of Brk led to a 2.3-fold increase in cell number compared with vector control (Fig. 5b). However, in a p27-knockdown background, the effect of Brk was reduced to 1.5-fold. Since the proliferative effect of Brk is markedly reduced in cells in which p27 was depleted, it indicates that p27 plays a significant role in the regulation of cell cycle by Brk. Corresponding to the p27-knockdown result, the proliferative effect of Brk was also decreased by 1.5-fold in cells in which FoxO3a was knocked down (Fig. 5b). These results indicate a significant role of both FoxO3a and p27 in the mitogenic signaling pathway of Brk. Taken together these data suggest that downregulation of the FoxO3a/p27 pathway plays an important role in the mechanism of Brk-induced cell proliferation.
Brk promotes cell proliferation and has been associated with multiple mitogenic pathways. However, the role of Brk in the cell cycle is still elusive. Here, we present a mechanism of Brk-induced cell cycle progression. We demonstrate that Brk downregulates the cell cycle inhibitor p27 by altering FoxO3a subcellular localization. p27 is a key component of the G1/S cell cycle checkpoint and downregulation of p27 promotes S phase entry. The progression of the cell from G1 to S phase, induced by Brk, is independent of growth factors suggesting a role for Brk in malignant transformation. Downregulation of p27 occurs through multiple mechanisms. One major mechanism is the activation of Skp2-mediated ubiquitination-dependent degradation.18 Surprisingly, our experiments indicated that Brk does not down-regulate p27 via the ubiquitin-proteasome pathway (Figs. 3f and 3g). Further study is needed to understand how Brk fails to induce p27 degradation. Our data indicates that FoxO3a appears to be a major component in p27 down-regulation by Brk, as the effect of Brk on p27 expression is significantly abrogated in cells transfected with siRNA targeting FoxO3a (Supporting Information Fig. S3). Furthermore, our results indicate that both FoxO3a and p27 are essential targets of Brk in promoting cell proliferation since the ability of Brk to affect cell proliferation is significantly reduced in FoxO3a or p27 knockdown cells. However, the effect of Brk on cell proliferation is not abolished by p27 suppression which indicates additional key factors in Brk signaling. Such factors may include the Ras/MAPK pathway and cyclinE/cdk2 activity, which have been shown to mediate the effect of Brk in enhancing ErbB2-induced cell proliferation and tumorigenesis in mice.5
The regulation of FoxO3a subcellular localization by Brk is inhibited by LY294002 indicating that the mechanism is mediated by the PI3-kinase/Akt pathway. This data is supported by previous reports demonstrating that Akt phosphorylates FoxO proteins, which promotes their binding to the adaptor protein 14-3-3.17 As a result, FoxOs are exported from the nucleus. The relationship between Brk and Akt needs further elucidation since, depending on the cellular context, Brk can activate or inhibit Akt activity.6, 19, 20 Additionally, the mechanism of Brk-induced activation of the PI3-kinase pathway in a growth factor-independent manner is complex and needs further investigation. A previous study showed that Brk elevated the levels of tyrosine phosphorylation of ErbB3 and induced association of ErbB3 with the p85 subunit of PI3-kinase in quiescent HB4a cells.6 This data suggests that the physical interaction of Brk with ErbB3 may induce a conformational change that is sufficient for ErbB3 to recruit PI3-kinase. However, ErbB3 may not be the sole mediator linking Brk to Akt activation since Brk also influences other PI3-kinase activators including Ras and EGFR pathways.5, 6, 9
Brk has multiple substrates which may impact the role of Brk in cancer. For example, Brk promotes cell proliferation upon interacting with EGF receptors and STAT3, but induces cell cycle arrest when bound to the polypyrimidine tract-binding (PTB) protein-associated splicing factor PSF.5, 12, 21 Therefore, it is essential to define the signaling cascade of Brk and understand the regulation of its substrate selection to control cancer development in cells overexpressing Brk.
Our findings indicate that Brk modulates the subcellular localization of FoxO proteins and down-regulates the expression of p27. FoxO proteins regulate the expression of genes involved in all phases of the cell cycle as well as in apoptosis, and their role is crucial in preventing neoplasia. Therefore, the mechanism presented here is likely to play an important role in cancer development.
The authors thank Dr. Laura Taylor for invaluable comments and insightful discussions; Ms. Mei-ling Chin and Mr. James Keller for technical expertise; and Dr. Todd Miller for the Brk genes.