Introduction
Ovarian cancer has the highest mortality rate of all gynecological malignancies all over the world, which accounts for approximately 3% of all female cancers [1]. Most of patients were found already at an advanced stage when diagnosed with ovarian cancer. The current standard treatment involves primary cytoreductive surgery followed by a platinum agent (cisplatin or carboplatin) - based chemotherapy for patients with advanced ovarian cancer. Nevertheless, the high mortality rate of ovarian cancer patients is due to treatment failure in the setting of recurrent/progressive disease that is unresponsive to chemotherapies based on CDDP [2]. Thus, understanding the molecular mechanism underlying CDDP chemoresistance of ovarian cancer may lead to improved clinical outcomes.
EMT is a cellular process during which epithelial cells lose their features, gain mesenchymal properties, and become motile and invasive [3]. The process of EMT involves loss of epithelial markers such as E-cadherin and β-catenin, and gain in the expression of mesenchymal markers such as N-cadherin, Vimentin and Snail, etc. Increasing evidence suggests a direct molecular and phenotypic correlation between acquisition of EMT characteristics and chemoresistance in tumor cells [4,5]. In previous studies, it has been reported that EMT plays a role in the chemoresistance of human tumor cells in contrast to conventional therapeutics in chemotherapeutic drugs such as 5-FU and paclitaxel or molecular target drugs such as epidermal growth factor receptor (EGFR)-targeted agents [6,7,8]. Understanding the involvement of EMT in CDDP chemoresistance and the underlying mechanisms is of great interest to develop therapeutic avenues for the treatment of CDDP-resistant patients with advanced ovarian cancer. FOXC2 is an important member of Forkhead box (Fox) transcription factors family which is an evolutionarily highly conserved helix transcription factor with a DNA binding domain [9]. Recent studies have shown that FOXC2 is necessary for embryonic development and many other key physiological as well as pathological processes, such as osteoblast differentiation, lymphangiogenesis, angiogenesis and cancer [10,11,12,13]. Meanwhile, the mechanisms regulating FOXC2 transcriptional activity mainly include phosphorylation, SUMOylation, etc [14,15]. In our previous study, we have shown that FOXC2 promotes the CDDP resistance of ovarian cancer cells by reducing CDDP-inducing apoptosis [16]. However, whether FOXC2 participates in regulation of EMT phenotype in CDDP-resistant ovarian cancer cells is unclear and remains to be further explored.
Here, we set out to investigate the involvement of FOXC2 in the acquisition of EMT and invasive characteristics of CDDP-resistant ovarian cancer cells and the underlying molecular mechanism. We showed that knockdown of FOXC2 could reverse EMT and invasive characteristics in the CDDP-resistant ovarian cancer cell line (SKOV3/CDDP), while upregulation of FOXC2 could induce the reverse effects in the parental ovarian cancer cell line (SKOV3). Besides, we reported that activation of ERK or AKT/GSK-3β signaling cascade was required for FOXC2-promoting EMT and invasive characteristics in CDDP-resistant ovarian cancer cells. Thus, our findings suggest that FOXC2 may play a key role in the development of CDDP resistance in ovarian cancer cells through induction of EMT.
Materials and Methods
Cell culture
The CDDP-resistant and parental human ovarian cancer cell lines (SKOV3/CDDP and SKOV3) were purchased from Xinyu Biotechnology Co. Ltd (Shanghai, China). All cell lines were cultured in RPMI 1640 (GIBCO-BRL) medium supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin, and 100 μg/ml streptomycin in humidified air at 37°C with 5% CO2. CDDP were purchased from Sigma-Aldrich (USA).
3-(4,5-Dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) assay
The single-cell suspensions were prepared and dispersed in 96-well plates. After incubation for 48h with the CDDP compounds (Sigma, MO, USA), the 0.5 mg/mL of MTT solution was added. Following incubation for 4h, the medium was discarded and 150 μL/well of dimethyl sulfoxide (Sigma-Aldrich) was added. The absorbance was measured at 490 nm using a microplate reader.
Transfection of plasmids and stable selection
The plasmid vector (pMD/FOXC2) expressing open-reading frame of FOXC2 was purchased from Sino Biological Inc (Beijing, China). Small hairpin RNA (shRNA) oligonucleotides targeting FOXC2 (shFOXC2, 5'-CCACACGTTTGCAACCCAA-3') and a negative control oligonucleotide (shcontrol, 5'-ACGTGACACGTTCGGAGAA-3') were subcloned into pSilencer4.1-CMVneo vector and the recombinant plasmids were named pS/shFOXC2 and pS/shcontrol, respectively. Transfection was performed using Lipofectamine TM 2000 (Invitrogen, USA) according to the manufacturer's instruction. At 48h post-transfection, G418 (800 μg/ml) was added to select stable transfectants and individual clones were maintained in a medium containing G418 (150 μg/ml). The stably transfected cells were named SKOV3/CDPP/shFOXC2 (or SKOV3/CDDP/shcontrol) and SKOV3/FOXC2 (or SKOV3/control), respectively.
Quantitative real-time PCR (qRT-PCR) assay
Total RNA from cells was isolated using the Trizol reagent (Invitrogen, USA) according to the manufacturer's instruction as described previously [17]. Reverse transcript (RT) was carried out with the SuperScript First-Strand Synthesis System for RT-PCR (Invitrogen, USA), and quantitative real-time PCR was carried out using the SYBR Green reporter. The primers used for PCR were as follows: FOXC2 forward 5'-CCTACCTGAGCGAGCAGAAT-3'; reverse 5'-ACCTTGACGAAGCACTCGTT-3'; GAPDH forward 5'-GCACCGTCAAGGCTGAGAAC-3'; reverse 5'-TGGTGAAGACGCCAGTGGA-3'. The data were normalized to the geometric mean of housekeeping gene GAPDH and calculated as 2-ΔΔCT method.
Western blot assay
Western blot assay was performed with anti-FOXC2 (Bethyl Laboratories, TX, USA), anti-phosphorylated AKT (p-AKT) (Ser473), phosphorylated ERK 1/2 (p-ERK) or phosphorylated GSK-3β (Cell Signaling, CA, USA), anti-total Akt or ERK 1/2, anti-E-cadherin, anti-N-cadherin, anti-Vimentin and anti-Snail (Santa Cruz Biotechnology, USA) as described previously [18]. Anti-GAPDH monoclonal antibody (Santa Cruze Biotechnology, CA, USA) was used as an internal control.
Scratch wound healing assay
The stably transfected cells were collected and implanted in the 35 mm culture dishes. When the cells grown to 80% confluence, a sterilized tip was used to draw a line with the same width on the bottom of 35 mm dishes. Photos were taken at 0 or 48h after the wounding.
Matrigel transwell invasion assay
Invasion assay was determined using Matrigel invasion chambers (BD Bioscience, San Diego, CA, USA). After incubation for 20 h, the upper surfaces of the Transwell chambers were scraped with cotton swabs, and the migrated and invaded cells were fixed with 4% paraformaldehyde, and then stained with Giemsa solution. The stained cells were photographed and counted under a light microscope in five randomly-selected fields.
Cell attachment and detachment assay
Cell attachment and detachment assays were performed as described previously [19]. Briefly, for attachment assay, the stably transfected cells were seeded in 24-well plates. Unattached cells were removed after 1 h incubation, and the attached cells were counted after trypsinization. For cell detachment assay, after 24 h incubation, the cells were incubated with 0.05 % trypsin for 3 min to detach the cells. Then, the culture medium was added to inactivate the trypsin and the detached cells were collected. The remaining cells were incubated with 0.25 % trypsin to detach and counted. The data were presented as a percentage of the attached or detached cells to total cells.
Statistical analysis
All statistical analyses were performed using the SPSS 17.0 statistical software. Experimental data were expressed as the mean±SD of at least three independent experiments. Statistical analyses were carried out using one-way ANOVA and Student's t test. Differences between groups were considered significant at P<0.05.
Results
Acquisition of EMT phenotype in the CDDP-resistant ovarian cancer cell line (SKOV3/CDDP)
The CDDP-resistant ovarian cancer cell line SKOV3/CDDP was developed from parental ovarian cell line SKOV3, and MTT assay indicated that the IC50 of CDDP to SKOV3/CDDP cell line (43.26μM) was significantly higher than that to SKOV3 cell line (7.35μM; Fig. 1A), suggesting that the SKOV3/CDDP cell line showed a 5.9-fold higher resistance to CDDP than the SKOV3 cell line (P<0.01; Fig. 1B). Then, we observed the morphological changes in SKOV3/CDDP cell line. It was observed that SKOV3/CDDP cell line shows loss of cell polarity and increased intercellular separation signifying loss of intercellular adhesion, suggesting that SKOV3/CDDP cell line was shown to be morphologically distinct from its parental cell line SKOV3 (Fig. 1C). To further determine the induction of EMT in SKOV3/CDDP cells, we detected the expression of epithelial markers such as E-cadherin, and mesenchymal markers such as N-cadherin, Vimentin, Snail by Western blot assays (Fig. 1D). The expression of E-cadherin protein was significantly downregulated in SKOV3/CDDP cell line, in comparison with the parental SKOV3 cell line. In contrast, the expression of N-cadherin, Vimentin and Snail in SKOV3/CDDP cell line was significantly upregulated. The induction of EMT has been reported to correlate with aggressive features of tumor cells, including migration, invasion, attachment and detachment. By scratch-wound healing assay, it was observed that SKOV3/CDDP cell line showed a faster closing of scratch wounds than SKOV3 cell line (P<0.05; Fig. 1E). In addition, Matrigel transwell assay indicated that SKOV3/CDDP cell line showed about 2.81-fold increase in the number of cells invading through a membrane compared with SKOV3 cell line (P<0.01; Fig. 1F). Also, compared with parental SKOV3 cell line, SKOV3/CDDP cell line showed the increased capacity of attachment and detachment (Fig. 1G). These results clearly suggest that CDDP-resistant ovarian cancer cells acquired the EMT phenotype and invasive characteristics.
Fig. 1
CDDP-resistant ovarian cancer cell line (SKOV3/CDDP) showed the acquired EMT and aggressive characteristics. A. MTT analysis of growth in CDDP-resistant ovarian cancer cell line (SKOV3/CDDP) and its parental cell line (SKOV3) at different concentrations of CDDP. B. The SKOV3/CDDP cell line showed a 5.9-fold higher resistance to CDDP than the SKOV3 cell line (P<0.05). C. Morphologies of SKOV3/CDDP and parental SKOV3 cells by microscopy. SKOV3/CDDP cells exhibited loss of cell polarity causing a spindle-cell morphology and increased formation of pseudopodia (arrow), whereas parental SKOV3 cells displayed an epithelioid appearance. D. Western blot detection of the expression epithelial protein markers (E-cadherin) and mesenchymal protein markers (N-cadherin, Vimentin and Snail). GAPDH was used as an internal control. E. Scratch wound healing assay of SKOV3/CDDP and SKOV3 cells. At 48 h after wounding, quantification of wound closure was done. The data present the mean distance of cell migration to the wound area at 48 h after wounding in three independent wound sites per group. (F) Matrigel transwell invasion assay of SKOV3/CDDP and SKOV3 cells. Cells in five random fields of view were counted and expressed as the average number of cells per field of view. (G) Cell attachment and detachment assays of SKOV3/CDDP and SKOV3 cells. All values represent the average of three independent experiments (mean±SD). *P<0.05 and **P<0.01, compared with mock cells.
Effects of FOXC2 knockdown on EMT, migration, invasion, attachment and detachment in SKOV3/CDDP cells
In our previous study, the expression of FOXC2 was observed to be significantly upregulated in SKOV3/CDDP cell line compared with parental SKOV3 cell line. Although we have previously shown that knockdown of FOXC2 could reverse the CDDP resistance of SKOV3/CDDP cells by enhancement of CDDP-inducing apoptosis, the effects of FOXC2 knockdown on EMT and invasive characteristics of SKOV3/CDDP cells needs to be further explored. First, qRT-PCR and Western blot assays confirmed again the knockdown of endogenous FOXC2 in SKOV3/CDDP cells stably transfected with pS/shFOXC2 vector (SKOV3/CDDP/shFOXC2), in comparison with SKOV3/CDDP cells stably transfected with pS/shcontrol vector (SKOV3/CDDP/shcontrol) or mock SKOV3/CDDP cells (Fig. 2A). Then, we analyzed the effects of FOXC2 knockdown on the expression of EMT-related proteins in SKOV3/CDDP cells. The increased expression of E-cadherin and the decreased expression of N-cadherin, Vimentin and Snail could be obviously observed in SKOV3/CDDP/shFOXC2 cells, in comparison with SKOV3/CDDP/shcontrol or mock SKOV3/CDDP cells (Fig. 2B). Further, Scratch wound healing and Matrigel transwell invasion assays indicated that FOXC2 knockdown could reduce the capacity of migration and invasion in SKOV3/CDDP cells (Fig. 2C-D). Meanwhile, FOXC2 knockdown could lead to the decreased capacity of attachment and detachment in SKOV3/CDDP cells (Fig. 2E). These results suggest that FOXC2 knockdown reverses EMT and inhibits migration, invasion, attachment and detachment in SKOV3/CDDP cells.
Fig. 2
Effects of FOXC2 knockdown on EMT phenotype and capacity of migration, invasion, attachment and detachment in SKOV3/CDDP cells. A. qRT-PCR and Western blot detection of FOXC2 mRNA and protein expression in mock SKOV3, SKOV3/CDDP/shcontrol and SKOV3/CDDP/shFOXC2 cells, respectively. GAPDH was used as an internal control. B. Western blot detection of the expression of E-cadherin, N-cadherin, Vimentin and Snail proteins in mock SKOV3, SKOV3/CDDP/shcontrol and SKOV3/CDDP/shFOXC2 cells, respectively. GAPDH was used as an internal control. C. Scratch wound healing assay of mock SKOV3/CDDP, SKOV3/CDDP/shcontrol and SKOV3/CDDP/shFOXC2 cells, respectively. (D) Matrigel transwell invasion assay of mock SKOV3/CDDP, SKOV3/CDDP/shcontrol and SKOV3/CDDP/shFOXC2 cells, respectively. (E) Cell attachment and detachment assays of mock SKOV3/CDDP, SKOV3/CDDP/shcontrol and SKOV3/CDDP/shFOXC2 cells, respectively. All values represent the average of three independent experiments (mean±SD). *P<0.05, **P<0.01 and NS, P>0.05, compared with mock cells.
Effects of FOXC2 upregulatin on EMT, migration, invasion, attachment and detachment in SKOV3 cells
Likewise, we previously testified that FOXC2 upregulation could decrease the CDDP sensitivity of parental SKOV3 cells. We further determined the effects of FOXC2 upregulation on EMT and invasive characteristics in SKOV3 cells. As shown in Fig. 3A, qRT-PCR and Western blot assays confirmed again the upregulation of endogenous FOXC2 in SKOV3 cells stably transfected with pMD/FOXC2 vector (SKOV3/FOXC2), in comparison with SKOV3 cells stably transfected with pMD/control vector (SKOV3/control) or mock SKOV3 cells. Also, the decreased expression of E-cadherin and the increased expression of N-cadherin, Vimentin and Snail could be observed in SKOV3/FOXC2 cells, in comparison with SKOV3/control or mock SKOV3 cells (Fig. 3B). In addition, it was observed that upregulation of FOXC2 could significantly enhance the capacity of migration and invasion in SKOV3 cells (Fig. 3C-D). And, the capacity of attachment and detachment could be significantly enhanced in SKOV3/FOXC2 cells, in comparison with control cells (Fig. 3E). Therefore, FOXC2 upregulation promotes EMT, migration, invasion, attachment and detachment in SKOV3 cells.
Fig. 3
Effects of FOXC2 upregulation on EMT phenotype and capacity of migration, invasion, attachment and detachment in SKOV3 cells. A. qRT-PCR and Western blot detection of FOXC2 mRNA and protein expression in mock SKOV3, SKOV3/control and SKOV3/FOXC2 cells, respectively. GAPDH was used as an internal control. B. Western blot detection of the expression of E-cadherin, N-cadherin, Vimentin and Snail proteins in mock SKOV3, SKOV3/control and SKOV3/FOXC2 cells, respectively. GAPDH was used as an internal control. C. Scratch wound healing assay of mock SKOV3, SKOV3/control and SKOV3/FOXC2 cells, respectively. (D) Matrigel transwell invasion assay of mock SKOV3, SKOV3/control and SKOV3/FOXC2 cells, respectively. (E) Cell attachment and detachment assays of mock SKOV3, SKOV3/ control and SKOV3/FOXC2 cells, respectively. All values represent the average of three independent experiments (mean±SD). *P<0.05, **P<0.01 and NS, P>0.05, compared with mock cells.
Activation of ERK or AKT/GSK-3β signaling pathways was involved in FOXC2-promoting EMT in CDDP-resistant ovarian cancer cells
It has been well documented that activation of ERK or AKT signaling mediates EMT induction in tumor cells by inhibiting GSK-3β, leading to the stabilization and nuclear localization of Snail, which transcriptionally inhibits E-cadherin. Previously, we showed that FOXC2 could activate MAPK and AKT signaling pathways to induce apoptosis resistance in ovarian cancer cells. Other studies also reported that FOXC2 could activate the ERK or Akt/GSK-3β/Snail pathway to induce EMT and enhance metastasis in colorectal cancer cells [20,21]. However, whether these signaling pathways mediate FOXC2-promoting EMT in ovarian cancer cells needs to be further explored. Here, the decreased expression of p-ERK, p-AKT and p-GSK-3β proteins could be observed in FOXC2-downregulated SKOV3/CDDP cells, but the expression of total ERK, and AKT and GSK-3β proteins showed no changes, in comparison with control cells (Fig. 4A). Contrastly, the expression of p-ERK, p-AKT and p-GSK-3β proteins was observed to be significantly upregulated in FOXC2-overexpressed SKOV3 cells, in comparison with control cells (Fig. 4B). Furthermore, combined treatment with Akt inhibitor (LY294002, 8.0 μM) plus MAPK/ERK kinase inhibitor (U0126, 10.0 μM) could partially reverse the changes of p-GSK-3β, E-cadherin, N-cadherin, Vimentin and Snail proteins in SKOV3 cells induced by FOXC2 upregulation (Fig. 4C). These results indicated that activation of ERK or AKT/GSK-3β signaling pathways mediates FOXC2-promoting EMT phenotype of CDDP-resistant ovarian cancer cells.
Fig. 4
Effects of FOXC2 on the expression of ERK or AKT signaling-related proteins in CDDP-resistant and parental ovarian cancer cells. (A) Western blot detection of the expression of p-ERK, total ERK, p-AKT, total AKT, p-GSK-3β and total GSK-3β proteins in mock SKOV3/CDDP, SKOV3/CDDP/sh-control, SKOV3/CDDP/shFOXC2 cells, respectively. B. Western blot detection of the expression of p-ERK, total ERK, p-AKT, total AKT, p-GSK-3β and total GSK-3β proteins in mock SKOV3, SKOV3/control, SKOV3/FOXC2 cells, respectively. C. Western blot detection of the expression of p-ERK, total ERK, p-AKT, total AKT, p-GSK-3β and total GSK-3β proteins in SKOV3/FOXC2 cells treated with PBS or combined with LY294002 (8.0 μM) plus U0126 (10.0 μM), respectively. GAPDH serves as an internal control. All values represent the average of three independent experiments (mean±SD). *P<0.05, **P<0.01 and NS, P>0.05, compared with mock cells.
Discussion
The experimental data from this study indicated FOXC2 as a promoter of EMT in human ovarian cancer cells. Knockdown of FOXC2 reversed EMT phenotype in CDDP-resistant SKOV3 cells, while upregulation of FOXC2 promoted EMT phenotype in parental SKOV3 cells. Furthermore, activation of EKR or AKT/GSK-3β signaling was testified to mediate FOXC2-promoting EMT in ovarian cancer cells. To the best of my knowledge, this is the first report that FOXC2 might regulate formation of EMT phenotype in ovarian cancer cells, which may be a mechanism of resistance to CDDP in ovarian cancer.
EMT is a unique process by which epithelial cells undergo remarkable morphologic changes characterized by a transition from epithelial cobblestone phenotype to mesenchymal phenotype leading to increased invasive characteristics of cells. Increasing evidence demonstrates that chemoresistance correlates with the acquisition of EMT-like phenotypic change of cancer cells. Many human chemoresistant tumor cells have been reported to acquire EMT phenotype, such as hepatocelluar cancer, pancreatic cancer, colorectal cancer, etc. For example, Wu et al. reported that gemcitabine-resistant HCC cells acquired EMT phenotype [22]. The same research group also reported that the PDGF-D/miR-106a/Twist1 pathway orchestrates EMT phenotype in gemcitabine-resistant hepatoma cells [23]. In addition, Uchibori et al. testified that 5-fluorouracil-resistant HCC cell lines had typical morphologic phenotypes of EMT, loss of cell-cell adhesion, spindle-shaped morphology and increased formation of pseudopodia [24]. In pancreatic cancer, Wang et al showed that acquisition of EMT phenotype of gemcitabine-resistant pancreatic cancer cells is linked with activation of the notch signaling pathway [25]. Wang and his colleagues also found that down-regulation of miR-223 reverses EMT phenotype in gemcitabine-resistant pancreatic cancer cells [26]. Meanwhile, they also testified that up-regulation of miR-200 and let-7 by natural agents [3,3'-diindolylmethane (DIM) or isoflavone] leads to the reversal of EMT phenotype in gemcitabine-resistant pancreatic cancer [27]. In other human cancers, oxaliplatin-resistant colorectal cancer cells acquired EMT phenotype and chemoresistance to 5-fluorouracil induces epithelial-mesenchymal transition via up-regulation of Snail in MCF7 human breast cancer cells [28,29]. Although the correlation between EMT and chemoresistance in tumor cells remains unclear, induction of EMT phenotype in chemoresistant tumor cells might represent a new potentially exciting research area into chemoresistance mechanisms. Therefore, targeting EMT may be a potential targeted therapeutic approach for overcoming chemoresistance toward the prevention of tumor progression and / or treatment of metastatic cancers.
Platinum agents-based chemotherapy has been considered to be the current standard treatment for patients with advanced ovarian cancer. However, the acquired chemoresistance leads to tumor recurrence of a sizable proportion of the patients after chemotherapy [30]. Recently, the correlations of EMT with CDDP resistance of ovarian cancer are increasingly reported. Miow et al. showed that epithelial-mesenchymal status renders differential responses to CDDP and EMT may be a contributing mechanism in ovarian cancer [31]. Also, Marchini et al. reported that resistance to platinum-based chemotherapy is associated with EMT in epithelial ovarian cancer [32]. Likewise, Kajiyama and his colleagues showed that chemoresistance to paclitaxel could induce EMT and enhance metastatic potential for epithelial ovarian carcinoma cells [33]. The roles of EMT in the acquired CDDP resistance of ovarian cancer are demonstrated, but the molecular mechanisms underlying EMT phenotype of CDDP-resistant ovarian cancer cells are not well understood. The FOX gene family encodes proteins which regulate the transcription of genes participating in many functions, such as development of various organs, regulation of senescence or proliferation, metabolic homeostasis and malignant transformation [34,35]. To date, human FOX gene family consists of at least 43 members, including FOXA1-3, FOXB1, FOXC1-2, FOXD1-6, etc [36]. Recently, it was reported that the polymorphism of 3'-UTR FOX genes is associated with an increased risk of human cancers and their nuclear translocation plays a role in tumor progression [37,38]. Meanwhile, some FOX genes were found to be post-transcriptionally regulated by microRNAs [39,40,41]. FOXC2, an important member of FOX family, has been reported to be overexpressed in a variety of human cancer, including colorectal cancer, esophageal cancer, gastric cancer, and so on [42,43,44]. The overexpression of FOXC2 has been found to be correlated with poor prognosis of patients and promotes proliferation, EMT and metastasis of tumor cells [45,46]. Previously, we have shown that FOXC2 could promote the CDDP resistant of ovarian cancer cells by reduction of CDDP-inducing apoptosis [16]. Another report from Liu and his colleagues indicated that FOXC2 was required for the maintenance of the mesenchymal phenotype after TGF-β1-induced EMT in human ovarian cancer cells [47]. However, whether FOXC2 regulates EMT phenotype of CDDP-resistant ovarian cancer cells requires to be further investigated. Here, we first showed that CDDP-resistant ovarian cancer cell line (SKOV3/CDDP) acquired the EMT phenotype and invasive characteristics. It was observed that shRNA-mediated FOXC2 knockdown could reduce the capacity of migration, invasion, attachment and detachment in SKOV3/CDDP cells. In contrast, upregulation of FOXC2 could increase those invasive capacities in parental SKOV3 cells. EMT is a key event in tumor metastasis, and the correlations of FOXC2 with EMT phenotype of tumor cells were also reported. For example, Li et al. reported that overexpression of forkhead Box C2 promotes tumor metastasis and indicates poor prognosis in colon cancer via regulating EMT [21]. Also, Zhou et al. showed that FOXC2 promotes chemoresistance in nasopharyngeal carcinomas via induction of EMT [48]. Here, the effects of FOXC2 on EMT of CDDP-resistant ovarian cancer cells were investigated by detecting the protein levels of EMT markers. FOXC2 knockdown increased the expression of an epithelial marker (E-cadherin) but decreased mesenchymal markers (N-cadherin and Vimentin), and Snail (EMT-related transcription factor), while FOXC2 upregulation could induce opposite effects on the expression of those proteins. Above data clearly suggest that FOXC2 is a promoter of EMT phenotype in CDDP-resistant ovarian cancer cells. Next, we will further investigate the underlying mechanism for FOXC2-promoting EMT. GSK-3β, a member of the GSK-3-binding protein family, has been recognized as the primary kinase involved in regulating the sub-celluar location and stability of Snail protein, which is generally considered to be a critical transcriptional regulator because of its direct suppressive effect on CDH1 (encoding E-cadherin) promoter [49]. It is well known that the activity of GSK-3β enzyme is inversely correlated with its Ser-9 phosphorylation level [50]. Many studies have shown that the activation of AKT or ERK increases the nuclear expression and transcriptional activity of Snail via by inhibitory phosphorylation of GSK-3β, thereby triggering tumor EMT and invasion [51,52]. Thus, it is plausible to hypothesize that FOXC2 may regulate the EMT-related proteins via AKT or ERK/GSK-3β signaling. In this study, FOXC2 knockdown could decrease the expression of phosphorylated ERK and AKT proteins and then induce inhibitory phosphorylation of GSK-3β, which led to the final activation of GSK-3β kinase. Likewise, FOXC2 upregulation could induce the adverse changes of those proteins, and importantly, combined treatment with Akt inhibitor plus MAPK/ERK kinase inhibitor could partially reverse the FOXC2 upregulation-inducing changes of those proteins. These results further confirmed that FOXC2 could promote formation of EMT phenotype and enhancement of invasive characteristics in CDDP-resistant ovarian cancer cells by activation of AKT or ERK/GSK-3β signaling. Further investigation is required to determine how FOXC2 activates these two signaling pathways and determine whether FOXC2 influences the in vivo metastatic processes of CDDP-resistant ovarian cancer cells.
Taken together, this study showed that FOXC2 may be a promoter of EMT phenotype in CDDP-resistant ovarian cancer cells by activation of ERK or AKT/GSK-3β signaling pathways. This FOXC2-mediated formation of EMT phenotype was identified as a molecular mechanism which regulates the CDDP resistance of ovarian cancer cells and will provide a potential strategy for the treatment of human ovarian cancers.
Acknowledgements
The authors would like to express their gratitude to every one of the Departement of Biochemistry and Molecular Biology participating in this work.
Disclosure Statement
None.


Get Permission
