Abstract
Elevated expression of cyclooxygenase 2 (COX2 (PTGS2)) has been reported to occur in human ovarian cancer and to be associated with poor prognosis. We have previously demonstrated that COX2-derived prostaglandin E2 (PGE2) promotes human ovarian cancer cell invasion. We had also demonstrated that epidermal growth factor (EGF) induces human ovarian cancer cell invasion by downregulating the expression of E-cadherin through various signaling pathways. However, it remains unclear whether COX2 and PGE2 are involved in the EGF-induced downregulation of E-cadherin expression and cell invasion in human ovarian cancer cells. In this study, we showed that EGF treatment induces COX2 expression and PGE2 production in SKOV3 and OVCAR5 human ovarian cancer cell lines. Interestingly, COX2 is not required for the EGF-induced downregulation of E-cadherin expression. In addition, EGF treatment activates the phosphatidylinositol-3-kinase (PI3K)/Akt and cAMP response element-binding protein (CREB) signaling pathways, while only the PI3K/Akt pathway is involved in EGF-induced COX2 expression. Moreover, we also showed that EGF-induced cell invasion is attenuated by treatment with a selective COX2 inhibitor, NS-398, as well as PGE2 siRNA. This study demonstrates an important role for COX2 and its derivative, PGE2, in the mediation of the effects of EGF on human ovarian cancer cell invasion.
Introduction
Epithelial ovarian cancer accounts for approximately 90% of all ovarian malignancies. It is the leading cause of gynecological cancer death in developed countries because of the majority of patients present with disseminated disease, for which the average 5-year survival rate is very low (Jemal et al. 2005, Landen et al. 2008). The expression of epidermal growth factor receptor (EGFR) is upregulated in many types of human cancers (Hanawa et al. 2006, Lee 2006, Viana-Pereira et al. 2008). The overexpression of EGFR in human ovarian cancer is associated with poor prognosis and disease progression (Bartlett et al. 1996, Niikura et al. 1997, Baselga 2002, Mendelsohn & Baselga 2003). Our recent studies have demonstrated that EGF induces human ovarian cancer cell invasion by downregulating the expression of the cell–cell adhesion molecule E-cadherin through various signaling pathways (Cheng et al. 2010, 2012a,b, 2013a,b,c).
Cyclooxygenase (COX) is a key enzyme that catalyzes the conversion of arachidonic acid into prostaglandins (PGs). PGE2 is an important PG that acts in an autocrine/paracrine manner to regulate various physiological and pathological functions (Nakanishi & Rosenberg 2013). Two isoforms of COX, COX1 and COX2, have been described. The expression of COX1 (PTGS1) occurs constitutively in most tissues and COX1 maintains homeostasis, while the expression of COX2 (PTGS2) is induced by various stimuli, including cytokines, mitogens, and hypoxia (Smith et al. 1996, Kniss 1999). There is increasing evidence that the expression of COX2 is elevated in several types of human cancers, including skin, liver, breast, and lung cancers (Soslow et al. 2000, Bae et al. 2001, Shirahama & Sakakura 2001). Previous studies have shown that COX2-specific inhibitors can reduce the growth of liver, breast, and colorectal tumors (Steinbach et al. 2000, Bae et al. 2001, Howe & Dannenberg 2003), which indicates that COX2 is involved in tumor progression. Elevated expression of COX2 has been detected in malignant ovarian tumors (Li et al. 2004), and high levels of COX2 expression are correlated with poor prognosis and overall survival in human ovarian cancer (Ali-Fehmi et al. 2003, Lee et al. 2013).
Treatment with a combination of COX2 inhibitors and EGFR inhibitors inhibits the progression of various human cancers (Dannenberg et al. 2005, Zhang et al. 2005, Half et al. 2007). However, whether COX2 is directly involved in EGF-induced tumor progression remains unclear. Many growth factors and cytokines induce the expression of COX2 (Tsatsanis et al. 2006). Previous studies have shown that EGF strongly induces the expression of COX2 in human glioma cell lines and lung adenocarcinoma cells (Xu & Shu 2007, Cao et al. 2011). We have previously shown that gonadotropins induce ovarian cancer cell invasion and that this effect is mediated by COX-derived PGE2 production (Lau et al. 2010). However, whether the expression of COX2 can be induced by EGF treatment in human ovarian cancer cells is unknown.
The aim of this study was to determine whether COX2 mediates EGF-induced cell invasion in human ovarian cancer cells. Our results reveal that EGF treatment increases COX2 expression and PGE2 production in SKOV3 and OVCAR5 human ovarian cancer cell lines. However, COX2 is not involved in the EGF-induced downregulation of E-cadherin expression. EGF-induced COX2 expression is mediated by the phosphatidylinositol-3-kinase (PI3K)/Akt signaling pathway. Moreover, inhibition of COX2 and PGE2 attenuates EGF-induced cancer cell invasion. These results demonstrate that COX2 and PGE2 are involved in EGF-induced human ovarian cancer cell invasion.
Materials and methods
Cell culture
The SKOV3 human ovarian cancer cell line was obtained from the American Type Culture Collection (Manassas, VA, USA). The OVCAR5 ovarian cancer cell line was kindly provided by Dr T C Hamilton (Fox Chase Cancer Center, Philadelphia, PA, USA). The cells were grown in a 1:1 (v/v) mixture of M199/MCDB105 medium (Sigma–Aldrich) supplemented with 10% fetal bovine serum (FBS; Hyclone Laboratories, Inc., Logan, UT, USA). The cultures were maintained at 37 °C in a humidified 5% CO2 atmosphere.
Antibodies and reagents
The monoclonal anti-E-cadherin antibody was obtained from BD Biosciences (Mississauga, ON, Canada). The polyclonal anti-COX2 antibody was obtained from Abcam (Toronto, ON, Canada). The monoclonal anti-α-tubulin and polyclonal anti-EGFR antibodies were obtained from Santa Cruz Biotechnology. The polyclonal anti-phospho-AKT, anti-AKT, anti-phospho-CREB1, and anti-CREB1 antibodies were obtained from Cell Signaling Technology (Danvers, MA, USA). The HRP-conjugated goat anti-mouse IgG and goat anti-rabbit IgG were obtained from Bio-Rad Laboratories. Human EGF, LY294002, and AG1478 were obtained from Sigma. Wortmannin was obtained from Calbiochem (Gibbstown, NJ, USA). NS-398 was purchased from Cayman Chemical (Ann Arbor, MI, USA).
siRNA transfection
To knock down endogenous EGFR, CREB, AKT (AKT1), or PGE2, the cells were transfected with 50 nM ON-TARGETplus SMARTpool siRNA (Dharmacon Research Inc., Lafayette, CO, USA) using Lipofectamine RNAiMAX (Invitrogen). siCONTROL non-targeting siRNA (Dharmacon Research, Inc.) was used as a transfection control.
Western blot analyses
The cells were lysed in a lysis buffer (Cell Signaling Technology), and protein concentrations were determined using a DC protein assay kit with BSA as the standard (Bio-Rad Laboratories). Equal amounts of protein were separated by SDS–PAGE and transferred onto PVDF membranes. After blocking with Tris-buffered saline containing 5% non-fat dried milk for 1 h, the membranes were incubated overnight at 4 °C with primary antibodies followed by incubation with the HRP-conjugated secondary antibodies. Immunoreactive bands were detected using an ECL substrate (Pierce, Rockford, IL, USA).
Real-time quantitative PCR
Total RNA was extracted using TRIzol (Invitrogen) according to the manufacturer's instructions. RT was carried out with 3 μg RNA, random primers, and M-MLV reverse transcriptase (Promega). The following primers were used for SYBR Green real-time quantitative PCR (RT-qPCR): COX2, 5′-CCC TTG GGT GTC AAA GGT AA-3′ (sense) and 5′-GCC CTC GCT TAT GAT CTG TC-3′ (antisense); EGFR, 5′-GGT GCA GGA GAG GAG AAC TGC-3′ (sense) and 5′-GGT GGC ACC AAA GCT GTA TT-3′ (antisense); CREB1, 5′-AAA ACC AAC AAA TGA CAG TT-3′ (sense) and 5′-TGG ACT GTC TGC CCA TTG G-3′ (antisense); and GAPDH, 5′-GAG TCA ACG GAT TTG GTC GT-3′ (sense) and 5′-GAC AAG CTT CCC GTT CTC AG-3′ (antisense). RT-qPCR was carried out using the Applied Biosystems 7300 Real-Time PCR System (Perkin-Elmer, Woodbridge, ON, Canada) equipped with a 96-well optical reaction plate. All RT-qPCR results are presented as the mean of at least three independent experiments conducted in triplicate. The relative quantification of mRNA levels was performed by the comparative Ct method using GAPDH as the reference gene and the formula 2−ΔΔCt.
PGE2 ELISA
A human PGE2-specific ELISA was performed according to the manufacturer's instructions (Cayman Chemical, Ann Arbor, MI, USA). Culture media in which the treated cells were grown were collected, and the concentrations of PGE2 in the culture media were measured by ELISA. The concentrations of PGE2 were normalized to the protein concentrations in the cell lysates. The normalized PGE2 values of the treated cells are presented as relative values compared with those of the control cells.
Invasion assays
The invasion assays were performed in Boyden chambers as described previously, with minor modifications (Woo et al. 2007). Cell-culture inserts (24 wells, pore size 8 μm; BD Biosciences, Mississauga, ON, Canada) pre-coated with growth factor-reduced Matrigel (40 μl, 1 mg/ml; BD Biosciences) were used for the invasion assays. The cell-culture inserts were seeded with 1×105 cells in 250 μl of medium supplemented with 0.1% FBS. Medium supplemented with 10% FBS (750 μl) was added to the lower chamber and it served as a chemotactic agent. After incubation for 48 h, non-invading cells were removed from the upper side of the membrane and the cells on the lower side of the membrane were fixed with cold methanol and air-dried. The cell nuclei were stained with Hoechst 33258 and counted by epifluorescence microscopy using Northern Eclipse 6.0 Software (Empix Imaging, Mississauga, ON, Canada). Each individual experiment was performed in triplicate (i.e. three inserts), and five microscopic fields were counted per insert.
Statistical analysis
Results are presented as the mean±s.e.m. of at least three independent experiments. The results were analyzed by one-way ANOVA and Tukey's multiple comparison test using PRISM Software (PRISM, La Jolla, CA, USA). Significant differences were defined as P<0.05.
Results
EGF treatment induces COX2 expression
To determine whether EGF treatment can induce the expression of COX2 in human ovarian cancer cells, we treated two human ovarian cancer cell lines (SKOV3 and OVCAR5) with 100 ng/ml EGF for different periods of time and quantified the COX2 mRNA and COX2 protein levels. As shown in Fig. 1A, EGF treatment resulted in the upregulation of COX2 mRNA levels in both SKOV3 and OVCAR5 cells, with the most significant degree of upregulation occurring after 1 h of treatment. In addition, we also investigated the stimulatory effect of different concentrations of EGF (1, 10, 50, and 100 ng/ml) on COX2 mRNA levels. As shown in Fig. 1B, treatment with 50 and 100 ng/ml EGF significantly induced COX2 mRNA levels in SKOV3 cells, while treatment with only 100 ng/ml EGF could significantly induce COX2 mRNA levels in OVCAR5 cells. Moreover, western blot analyses revealed that treatment with EGF for 3 and 6 h significantly upregulated COX2 protein levels (Fig. 1C). To confirm that EGFR is required for EGF-induced COX2 expression, we blocked EGFR activation with the EGFR-specific inhibitor AG1478. RT-qPCR and western blot analyses revealed that AG1478 blocked the EGF-induced upregulation of COX2 mRNA and COX2 protein levels (Fig. 2A and B). We further confirmed the involvement of EGFR in EGF-induced COX2 expression by siRNA-mediated knockdown of EGFR. As shown in Fig. 2C and D, EGFR mRNA and EGFR protein levels were significantly downregulated in the presence of EGFR siRNA. In addition, the siRNA-mediated downregulation of EGFR expression attenuated the EGF-induced elevation of COX2 mRNA and COX2 protein levels. It has been shown that the binding of EGF to EGFR rapidly induces clustering and internalization of the ligand–receptor complexes, ultimately resulting in lysosomal degradation of both EGF and its receptor (Carpenter 1987). This process was supported by the data shown in Fig. 2D, which revealed that the expression of EGFR was downregulated in SKOV3 and OVCAR5 cells in response to EGF treatment.
EGF induces COX2 expression in SKOV3 and OVCAR5 cells. (A) Cells were treated with a vehicle control (Ctrl) or 100 ng/ml EGF. COX2 mRNA levels were determined at different time points by RT-qPCR. (B) Cells were treated with increasing concentrations of EGF (1, 10, 50, and 100 ng/ml) for 1 h. COX2 mRNA levels were determined by RT-qPCR. (C) The cells were treated with a vehicle control (Ctrl) or 100 ng/ml EGF for 1, 3, and 6 h. COX2 protein levels were determined by western blot analyses. Results are expressed as the mean±s.e.m. of at least three independent experiments. Values without a common letter are significantly different (P<0.05).
Citation: Endocrine-Related Cancer 21, 4; 10.1530/ERC-13-0450
EGFR is required for the EGF-induced upregulation of COX2 expression. (A and B) Cells were pretreated for 1 h with AG1478 (10 μM) and then treated with 100 ng/ml EGF for 1 h (mRNA) or 3 h (protein). COX2 mRNA (A) and COX2 protein (B) levels were determined by RT-qPCR and western blot analyses respectively. (C and D) Cells were transfected for 48 h with 50 nM control siRNA (si-Ctrl) or EGFR siRNA (si-EGFR) and then treated with EGF for 1 h (mRNA) or 3 h (protein). COX2 mRNA (C) and COX2 protein (D) levels were determined by RT-qPCR and western blot analyses respectively. Results are expressed as the mean±s.e.m. of at least three independent experiments. Values without a common letter are significantly different (P<0.05).
Citation: Endocrine-Related Cancer 21, 4; 10.1530/ERC-13-0450
COX2 is not required for EGF-induced downregulation of E-cadherin expression
We had previously shown that EGF induces human ovarian cancer cell invasion by downregulating E-cadherin expression through various signaling pathways (Cheng et al. 2010, 2012a,b, 2013a,b,c). However, it is not clear whether COX2 is involved in the EGF-induced downregulation of E-cadherin expression in human ovarian cancer cells. To investigate this, a selective COX2 inhibitor, NS-398, was used to block COX2 activity. Consistent with our previous studies, EGF downregulated E-cadherin mRNA levels in both SKOV3 and OVCAR5 cells. Interestingly, NS-398 treatment did not affect basal or EGF-downregulated E-cadherin mRNA levels (Fig. 3A). Western blot analyses revealed that NS-398 treatment also had no effect on basal and EGF-downregulated E-cadherin protein levels (Fig. 3B).
EGF-induced downregulation of E-cadherin expression does not require COX2. Cells were pretreated for 1 h with NS-398 (10 μM) and then treated with 100 ng/ml EGF for 24 h. E-cadherin mRNA (A) and E-cadherin protein (B) levels were determined by RT-qPCR and western blot analyses respectively. Results are expressed as the mean±s.e.m. of at least three independent experiments. Values without a common letter are significantly different (P<0.05).
Citation: Endocrine-Related Cancer 21, 4; 10.1530/ERC-13-0450
Active AKT is required for the induction of COX2 expression by EGF
We have previously shown that the PI3K/Akt signaling pathway is involved in gonadotropin-induced COX2 expression in human ovarian cancer cells (Lau et al. 2010). Therefore, we investigated whether the Akt signaling pathway is also involved in EGF-induced COX2 expression. Treatment with EGF increased the levels of phosphorylated AKT in both SKOV3 and OVCAR5 cells (Fig. 4A). To determine the involvement of PI3K/Akt signaling in EGF-induced upregulation of COX2 expression, two PI3K inhibitors, wortmannin and LY294002, were used. Treatment with wortmannin and LY294002 suppressed EGF-induced AKT phosphorylation in both cell lines (Supplementary Fig. 1, see section on supplementary data given at the end of this article). In addition, the EGF-induced upregulation of COX2 mRNA and COX2 protein levels was attenuated by co-treatment with wortmannin and LY294002 (Fig. 4B and C). To avoid off-target effects of the pharmacological inhibitors and further confirm that AKT is involved in EGF-induced COX2 expression, an siRNA-mediated knockdown approach was used to block AKT activity. As shown in Fig. 4D, EGF-induced COX2 protein levels were suppressed by the knockdown of AKT. Taken together, these results clearly indicate that the PI3K/Akt signaling pathway is required for EGF-induced COX2 expression in human ovarian cancer cells.
The Akt signaling pathway is involved in EGF-induced COX2 expression. (A) Cells were treated with a vehicle control (Ctrl) or 100 ng/ml EGF for 5, 10, or 30 min, and the levels of phosphorylated AKT were determined by western blot analyses. (B) Cells were pretreated for 1 h with wortmannin (1 μM) or LY294002 (10 μM) and then treated with 100 ng/ml EGF for 1 h (mRNA) or 3 h (protein). COX2 mRNA (B) and COX2 protein (C) levels were determined by RT-qPCR and western blot analyses respectively. (D) Cells were transfected for 48 h with 50 nM control siRNA (si-Ctrl) or AKT siRNA (si-AKT) and then treated with EGF for 3 h. COX2 protein levels were determined by western blot analyses. Results are expressed as the mean±s.e.m. of at least three independent experiments. Values without a common letter are significantly different (P<0.05).
Citation: Endocrine-Related Cancer 21, 4; 10.1530/ERC-13-0450
CREB does not mediate EGF-induced COX2 expression
The cAMP response element (CRE) has been identified as one of the central regulatory elements in the COX2 promoter region and cAMP response element-binding protein (CREB) is well known to be involved in the regulation of COX2 gene expression in a variety of cells (Tanabe & Tohnai 2002, Klein et al. 2007). Thus, we investigated the involvement of the CREB signaling pathway in EGF-induced COX2 expression. Treatment with EGF increased the levels of phosphorylated CREB in both SKOV3 and OVCAR5 cells (Fig. 5A). Interestingly, contrary to our expectations, siRNA-mediated knockdown of CREB did not affect EGF-induced COX2 mRNA and COX2 protein levels (Fig. 5B and C). These results indicate that CREB does not play a role in EGF-induced COX2 expression in human ovarian cancer cells.
CREB is not involved in EGF-induced COX2 expression. (A) Cells were treated with a vehicle control (Ctrl) or 100 ng/ml EGF for 5, 10, or 30 min, and the levels of phosphorylated CREB were determined by western blot analyses. (B and C) Cells were transfected for 48 h with 50 nM control siRNA (si-Ctrl) or CREB siRNA (si-CREB) and then treated with EGF for 1 h (mRNA) or 3 h (protein). COX2 mRNA (B) and COX2 protein (C) levels were determined by RT-qPCR and western blot analyses respectively. Results are expressed as the mean±s.e.m. of at least three independent experiments. Values without a common letter are significantly different (P<0.05).
Citation: Endocrine-Related Cancer 21, 4; 10.1530/ERC-13-0450
COX2 and PGE2 mediate EGF-induced cell invasion
PGE2 is the most common PG derived from COX2 (Nakanishi & Rosenberg 2013). To determine whether EGF-induced COX2 expression contributes to the production of PGE2, we measured PGE2 protein levels in the culture medium by ELISA after 1, 3, and 6 h of EGF treatment. As shown in Fig. 6A, treatment with EGF for 3 and 6 h significantly upregulated the production of PGE2 in both SKOV3 and OVCAR5 cells. Importantly, NS-398 treatment suppressed EGF-induced PGE2 production (Fig. 6B). To determine whether COX2 and PGE2 are involved in EGF-induced cell invasion, we performed a Matrigel-coated Transwell invasion assay. Consistent with results from previous studies, we found that EGF stimulates significant cell invasion in both SKOV3 and OVCAR5 cells. Treatment with NS-398 did not affect the basal levels of invasiveness, but attenuated the EGF-stimulated cell invasion (Fig. 6C). We have previously shown that PGE2 treatment stimulates human ovarian cancer cell invasion (Lau et al. 2010). To determine whether COX2-derived PGE2 also contributes to EGF-induced cell invasion, we used an siRNA-mediated knockdown approach to inhibit the expression of PGE2. As shown in Fig. 6D, treatment with PGE2 siRNA did not affect the basal levels of invasiveness, but attenuated the EGF-stimulated cell invasion. Taken together, these results indicate that COX2 and its derivative, PGE2, are involved in EGF-induced human ovarian cancer cell invasion.
COX2 and PGE2 are involved in EGF-induced cell invasion. (A and B) Cells were treated with 100 ng/ml EGF for 1, 3, and 6 h (A). Cells were pretreated for 1 h with NS-398 (10 μM) and then treated with 100 ng/ml EGF for 3 h (B). The PGE2 protein levels in the culture medium were determined by ELISA. (C and D) Cells were pretreated for 1 h with NS-398 (10 μM) (A) or the cells were transfected with 50 nM control siRNA (si-Ctrl) or PGE2 siRNA (si-PGE2) for 48 h (B). Following pretreatment or transfection, cells were treated with a vehicle control (Ctrl) or 100 ng/ml EGF and then seeded into Matrigel-coated Transwell inserts and cultured for an additional 48 h. Non-invading cells were removed from the upper side of the filter, and the nuclei of the invasive cells were stained with Hoechst 33258. The top panel shows representative fluorescence images obtained in the invasion assays. The scale bar represents 200 μm. The bottom panel summarizes the quantitative results, which are expressed as the mean±s.e.m. of at least three independent experiments. Values without a common letter are significantly different (P<0.05).
Citation: Endocrine-Related Cancer 21, 4; 10.1530/ERC-13-0450
Discussion
The overexpression of EGFR and COX2 in human ovarian cancer is correlated with poor prognosis and survival rates (Ali-Fehmi et al. 2003, Siwak et al. 2010, Lee et al. 2013). This indicates that EGF/EGFR and COX2 are important for the development and progression of human ovarian cancer. We have previously shown that EGF could induce ovarian cancer cell invasion through a complex signaling network (Cheng et al. 2010, 2012a,b, 2013a,b,c). However, it is not known whether COX2 mediates EGF-induced ovarian cancer cell invasion. In the present study, we showed that EGF induced COX2 expression and PGE2 production in two human ovarian cancer cell lines, SKOV3 and OVCAR5, through the activation of the PI3K/Akt signaling pathway. Moreover, inhibition of COX2 and PGE2 attenuated the ovarian cancer cell invasion induced by EGF. These results indicate that COX2 and its derivative, PGE2, are involved in EGF-induced cell invasion in human ovarian cancer cells.
COX2 is known to be involved in the regulation of E-cadherin expression in other human cancers (Dohadwala et al. 2006, Sitarz et al. 2009). Treatment of gastric cancer cells with celecoxib, a specific COX2 inhibitor, leads to the upregulation of E-cadherin expression at both the mRNA and protein levels (Sitarz et al. 2009). In human non-small-cell lung carcinoma cells, inhibition of COX2 by genetic or pharmacological methods increases E-cadherin expression (Dohadwala et al. 2006). In this study, we showed that treatment with the selective COX2 inhibitor NS-398 did not affect the EGF-induced downregulation of E-cadherin expression. These results indicate that the requirement for COX2 in the EGF-induced downregulation of E-cadherin expression may depend on the type of cancer.
PGE2 stimulates cell invasion in many cancers via an autocrine/paracrine mechanism (Dohadwala et al. 2002, Timoshenko et al. 2003, Lau et al. 2010, Paquette et al. 2011, Vo et al. 2013). We have previously shown that PGE2 acts in an autocrine/paracrine fashion in ovarian cancer cells to upregulate the expression of matrix metalloproteinase 2 (MMP2) and MMP9, which in turn contribute to gonadotropin-induced cell invasion (Lau et al. 2010). Interestingly, proteinase activity is required for EGF-induced ovarian cancer invasion (Ellerbroek et al. 1998). We have previously shown that the downregulation of E-cadherin expression mediates EGF-induced ovarian cancer cell invasion (Cheng et al. 2010, 2012a,b, 2013a,b,c). However, our previous results had indicated that although EGF-induced invasion is significantly reduced in E-cadherin-overexpressing ovarian cancer cells, it is not completely abolished (Cheng et al. 2010). This indicates the presence of additional, E-cadherin-independent, mechanisms for EGF-induced ovarian cancer cell invasion. It has been shown that additional mechanisms such as enhanced protease activity/secretion, changes in actin cytoskeleton, and enhanced motility are involved in EGF-induced cell invasion (Ellerbroek et al. 2001, Kassis et al. 2001). Because the inhibition of COX2 did not affect the EGF-induced downregulation of E-cadherin expression in ovarian cancer cells, it is possible that COX2-derived PGE2 may mediate EGF-induced cell invasion by regulating the expression of MMP in an autocrine/paracrine fashion.
EGF induces the expression of COX2 in different types of human cancers (Kulkarni et al. 2001, Xu & Shu 2007, Cao et al. 2011). However, the underlying molecular mechanism that mediates EGF-induced COX2 expression in human ovarian cancer is unknown. We and other groups have shown that PI3K/Akt signaling is required for gonadotropin- and insulin-like growth factor 1-induced COX2 expression, but it is unclear whether the same is true for EGF-induced COX2 expression (Cao et al. 2007, Lau et al. 2010). In human cervical cancer cells, EGF-induced COX2 expression is attenuated by treatment with either of two PI3K inhibitors: wortmannin and LY294002 (Kulkarni et al. 2001). In this study, we showed that treatment with wortmannin or LY294002 attenuated EGF-induced COX2 expression in ovarian cancer cells. We further confirmed the involvement of AKT in EGF-induced COX2 expression by siRNA-mediated knockdown of AKT, which could avoid the non-specific effects of pharmacological inhibition. Taken together, our results clearly indicate for the first time, to our knowledge, that PI3K/Akt signaling plays an important role in the mediation of EGF-induced COX2 expression in ovarian cancer cells.
It has been reported that the transcription factor CREB is involved in the regulation of COX2 expression by binding to the CRE in the COX2 promoter (Tanabe & Tohnai 2002, Ghosh et al. 2007, Klein et al. 2007, Sharma-Walia et al. 2010). However, it is not known whether CREB regulates EGF-induced COX2 expression in human ovarian cancer cells. In this study, we found that CREB is not involved in EGF-induced COX2 expression. These results indicate that other molecules mediate the induction of COX2 expression by EGF in ovarian cancer cells. NF-κB and c-Jun participate in LPS-induced COX2 expression in mouse macrophages (Kang et al. 2006). In rat insulinoma cells, CREB and the Ets family members Ets-1 and Elk-1 increase Cox2 promoter activity, while STAT1 inhibits Cox2 promoter activity (Zhang et al. 2007). In human monocytes, ESE1 binds to and increases the activity of the COX2 promoter (Grall et al. 2005). A recent study has shown that the FOXM1/Sp1 complex binds to the Sp1-binding site in the COX2 promoter to mediate EGF-induced COX2 expression in human glioma cells (Xu & Shu 2013). Further research is needed to determine the transcriptional regulatory mechanisms that mediate EGF-induced COX2 expression in ovarian cancer cells.
In summary, our results reveal that EGF induces COX2 expression and PGE2 production in human ovarian cancer cells. They also reveal that COX2 expression is not involved in the EGF-induced downregulation of E-cadherin expression. Moreover, EGF-induced COX2 expression is mediated by the PI3K/Akt signaling pathway. Inhibition of COX2 as well as PGE2 attenuates EGF-stimulated cell invasion. This study provides important insights into the molecular mechanisms that mediate EGF-stimulated human ovarian cancer cell invasion.
Supplementary data
This is linked to the online version of the paper at http://dx.doi.org/10.1530/ERC-13-0450.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This study was supported by an operating grant from the Canadian Institutes of Health Research to P C K Leung. X Qiu is supported by China Scholarship Council.
Author contribution statement
X Qiu, J-C Cheng, and P C K Leung conceived and designed the experiments; X Qiu, J-C Cheng, and H-M Chang performed the experiments; X Qiu, J-C Cheng, and H-M Chang analyzed the data; X Qiu, J-C Cheng, and P C K Leung wrote the article. All authors read and approved the final manuscript.
References
Ali-Fehmi R, Che M, Khalifeh I, Malone JM, Morris R, Lawrence WD & Munkarah AR 2003 The effect of cyclooxygenase-2 expression on tumor vascularity in advanced stage ovarian serous carcinoma. Cancer 98 1423–1429. (doi:10.1002/cncr.11650).
Bae SH, Jung ES, Park YM, Kim BS, Kim BK, Kim DG & Ryu WS 2001 Expression of cyclooxygenase-2 (COX-2) in hepatocellular carcinoma and growth inhibition of hepatoma cell lines by a COX-2 inhibitor, NS-398. Clinical Cancer Research 7 1410–1418.
Bartlett JM, Langdon SP, Simpson BJ, Stewart M, Katsaros D, Sismondi P, Love S, Scott WN, Williams AR & Lessells AM et al. 1996 The prognostic value of epidermal growth factor receptor mRNA expression in primary ovarian cancer. British Journal of Cancer 73 301–306. (doi:10.1038/bjc.1996.53).
Baselga J 2002 Why the epidermal growth factor receptor? The rationale for cancer therapy. Oncologist 7 (Suppl 4) 2–8. (doi:10.1634/theoncologist.7-suppl_4-2).
Cao Z, Liu LZ, Dixon DA, Zheng JZ, Chandran B & Jiang BH 2007 Insulin-like growth factor-I induces cyclooxygenase-2 expression via PI3K, MAPK and PKC signaling pathways in human ovarian cancer cells. Cellular Signalling 19 1542–1553. (doi:10.1016/j.cellsig.2007.01.028).
Cao S, Yan Y, Zhang X, Zhang K, Liu C, Zhao G, Han J, Dong Q, Shen B & Wu A et al. 2011 EGF stimulates cyclooxygenase-2 expression through the STAT5 signaling pathway in human lung adenocarcinoma A549 cells. International Journal of Oncology 39 383–391. (doi:10.3892/ijo.2011.1053).
Carpenter G 1987 Receptors for epidermal growth factor and other polypeptide mitogens. Annual Review of Biochemistry 56 881–914. (doi:10.1146/annurev.bi.56.070187.004313).
Cheng JC, Klausen C & Leung PC 2010 Hydrogen peroxide mediates EGF-induced down-regulation of E-cadherin expression via p38 MAPK and snail in human ovarian cancer cells. Molecular Endocrinology 24 1569–1580. (doi:10.1210/me.2010-0034).
Cheng JC, Auersperg N & Leung PC 2012a EGF-induced EMT and invasiveness in serous borderline ovarian tumor cells: a possible step in the transition to low-grade serous carcinoma cells? PLoS ONE 7 e34071. (doi:10.1371/journal.pone.0034071).
Cheng JC, Chang HM & Leung PC 2012b Epidermal growth factor induces human oviductal epithelial cell invasion by down-regulating E-cadherin expression. Journal of Clinical Endocrinology and Metabolism 97 E1380–E1389. (doi:10.1210/jc.2011-2751).
Cheng JC, Chang HM & Leung PC 2013a Egr-1 mediates epidermal growth factor-induced downregulation of E-cadherin expression via Slug in human ovarian cancer cells. Oncogene 32 1041–1049. (doi:10.1038/onc.2012.127).
Cheng JC, Klausen C & Leung PC 2013b Hypoxia-inducible factor 1 alpha mediates epidermal growth factor-induced down-regulation of E-cadherin expression and cell invasion in human ovarian cancer cells. Cancer Letters 329 197–206. (doi:10.1016/j.canlet.2012.10.029).
Cheng JC, Qiu X, Chang HM & Leung PC 2013c HER2 mediates epidermal growth factor-induced down-regulation of E-cadherin in human ovarian cancer cells. Biochemical and Biophysical Research Communications 434 81–86. (doi:10.1016/j.bbrc.2013.03.062).
Dannenberg AJ, Lippman SM, Mann JR, Subbaramaiah K & DuBois RN 2005 Cyclooxygenase-2 and epidermal growth factor receptor: pharmacologic targets for chemoprevention. Journal of Clinical Oncology 23 254–266. (doi:10.1200/JCO.2005.09.112).
Dohadwala M, Batra RK, Luo J, Lin Y, Krysan K, Pold M, Sharma S & Dubinett SM 2002 Autocrine/paracrine prostaglandin E2 production by non-small cell lung cancer cells regulates matrix metalloproteinase-2 and CD44 in cyclooxygenase-2-dependent invasion. Journal of Biological Chemistry 277 50828–50833. (doi:10.1074/jbc.M210707200).
Dohadwala M, Yang SC, Luo J, Sharma S, Batra RK, Huang M, Lin Y, Goodglick L, Krysan K & Fishbein MC et al. 2006 Cyclooxygenase-2-dependent regulation of E-cadherin: prostaglandin E2 induces transcriptional repressors ZEB1 and snail in non-small cell lung cancer. Cancer Research 66 5338–5345. (doi:10.1158/0008-5472.CAN-05-3635).
Ellerbroek SM, Hudson LG & Stack MS 1998 Proteinase requirements of epidermal growth factor-induced ovarian cancer cell invasion. International Journal of Cancer 78 331–337. (doi:10.1002/(SICI)1097-0215(19981029)78:3<331::AID-IJC13>3.0.CO;2-9).
Ellerbroek SM, Halbleib JM, Benavidez M, Warmka JK, Wattenberg EV, Stack MS & Hudson LG 2001 Phosphatidylinositol 3-kinase activity in epidermal growth factor-stimulated matrix metalloproteinase-9 production and cell surface association. Cancer Research 61 1855–1861.
Ghosh R, Garcia GE, Crosby K, Inoue H, Thompson IM, Troyer DA & Kumar AP 2007 Regulation of Cox-2 by cyclic AMP response element binding protein in prostate cancer: potential role for nexrutine. Neoplasia 9 893–899. (doi:10.1593/neo.07502).
Grall FT, Prall WC, Wei W, Gu X, Cho JY, Choy BK, Zerbini LF, Inan MS, Goldring SR & Gravallese EM et al. 2005 The Ets transcription factor ESE-1 mediates induction of the COX-2 gene by LPS in monocytes. FEBS Journal 272 1676–1687. (doi:10.1111/j.1742-4658.2005.04592.x).
Half E, Sun Y & Sinicrope FA 2007 Anti-EGFR and ErbB-2 antibodies attenuate cyclooxygenase-2 expression and cooperatively inhibit survival of human colon cancer cells. Cancer Letters 251 237–246. (doi:10.1016/j.canlet.2006.11.020).
Hanawa M, Suzuki S, Dobashi Y, Yamane T, Kono K & Enomoto N 2006 EGFR protein overexpression and gene amplification in squamous cell carcinomas of the esophagus. International Journal of Cancer 118 1173–1180. (doi:10.1002/ijc.21454).
Howe LR & Dannenberg AJ 2003 COX-2 inhibitors for the prevention of breast cancer. Journal of Mammary Gland Biology and Neoplasia 8 31–43. (doi:10.1023/A:1025731204719).
Jemal A, Murray T, Ward E, Samuels A, Tiwari RC, Ghafoor A, Feuer EJ & Thun MJ 2005 Cancer statistics, 2005. CA: A Cancer Journal for Clinicians 55 10–30. (doi:10.3322/canjclin.55.1.10).
Kang YJ, Wingerd BA, Arakawa T & Smith WL 2006 Cyclooxygenase-2 gene transcription in a macrophage model of inflammation. Journal of Immunology 177 8111–8122. (doi:10.4049/jimmunol.177.11.8111).
Kassis J, Lauffenburger DA, Turner T & Wells A 2001 Tumor invasion as dysregulated cell motility. Seminars in Cancer Biology 11 105–117. (doi:10.1006/scbi.2000.0362).
Klein T, Shephard P, Kleinert H & Komhoff M 2007 Regulation of cyclooxygenase-2 expression by cyclic AMP. Biochimica et Biophysica Acta 1773 1605–1618. (doi:10.1016/j.bbamcr.2007.09.001).
Kniss DA 1999 Cyclooxygenases in reproductive medicine and biology. Journal of the Society for Gynecologic Investigation 6 285–292. (doi:10.1016/S1071-5576(99)00034-9).
Kulkarni S, Rader JS, Zhang F, Liapis H, Koki AT, Masferrer JL, Subbaramaiah K & Dannenberg AJ 2001 Cyclooxygenase-2 is overexpressed in human cervical cancer. Clinical Cancer Research 7 429–434.
Landen CN Jr, Birrer MJ & Sood AK 2008 Early events in the pathogenesis of epithelial ovarian cancer. Journal of Clinical Oncology 26 995–1005. (doi:10.1200/JCO.2006.07.9970).
Lau MT, Wong AS & Leung PC 2010 Gonadotropins induce tumor cell migration and invasion by increasing cyclooxygenases expression and prostaglandin E2 production in human ovarian cancer cells. Endocrinology 151 2985–2993. (doi:10.1210/en.2009-1318).
Lee SM 2006 Is EGFR expression important in non-small cell lung cancer? Thorax 61 98–99. (doi:10.1136/thx.2005.047936).
Lee JY, Myung SK & Song YS 2013 Prognostic role of cyclooxygenase-2 in epithelial ovarian cancer: a meta-analysis of observational studies. Gynecologic Oncology 129 613–619. (doi:10.1016/j.ygyno.2013.02.011).
Li S, Miner K, Fannin R, Carl Barrett J & Davis BJ 2004 Cyclooxygenase-1 and 2 in normal and malignant human ovarian epithelium. Gynecologic Oncology 92 622–627. (doi:10.1016/j.ygyno.2003.10.053).
Mendelsohn J & Baselga J 2003 Status of epidermal growth factor receptor antagonists in the biology and treatment of cancer. Journal of Clinical Oncology 21 2787–2799. (doi:10.1200/JCO.2003.01.504).
Nakanishi M & Rosenberg DW 2013 Multifaceted roles of PGE2 in inflammation and cancer. Seminars in Immunopathology 35 123–137. (doi:10.1007/s00281-012-0342-8).
Niikura H, Sasano H, Sato S & Yajima A 1997 Expression of epidermal growth factor-related proteins and epidermal growth factor receptor in common epithelial ovarian tumors. International Journal of Gynecological Pathology 16 60–68. (doi:10.1097/00004347-199701000-00010).
Paquette B, Therriault H, Desmarais G, Wagner R, Royer R & Bujold R 2011 Radiation-enhancement of MDA-MB-231 breast cancer cell invasion prevented by a cyclooxygenase-2 inhibitor. British Journal of Cancer 105 534–541. (doi:10.1038/bjc.2011.260).
Sharma-Walia N, George Paul A, Patel K, Chandran K, Ahmad W & Chandran B 2010 NFAT and CREB regulate Kaposi's sarcoma-associated herpesvirus-induced cyclooxygenase 2 (COX-2). Journal of Virology 84 12733–12753. (doi:10.1128/JVI.01065-10).
Shirahama T & Sakakura C 2001 Overexpression of cyclooxygenase-2 in squamous cell carcinoma of the urinary bladder. Clinical Cancer Research 7 558–561.
Sitarz R, Leguit RJ, de Leng WW, Morsink FH, Polkowski WP, Maciejewski R, Offerhaus GJ & Milne AN 2009 Cyclooxygenase-2 mediated regulation of E-cadherin occurs in conventional but not early-onset gastric cancer cell lines. Cellular Oncology 31 475–485. (doi:10.3233/CLO-2009-0496).
Siwak DR, Carey M, Hennessy BT, Nguyen CT, McGahren Murray MJ, Nolden L & Mills GB 2010 Targeting the epidermal growth factor receptor in epithelial ovarian cancer: current knowledge and future challenges. Journal of Oncology 2010 568938. (doi:10.1155/2010/568938).
Smith WL, Garavito RM & DeWitt DL 1996 Prostaglandin endoperoxide H synthases (cyclooxygenases)-1 and -2. Journal of Biological Chemistry 271 33157–33160. (doi:10.1074/jbc.271.52.33157).
Soslow RA, Dannenberg AJ, Rush D, Woerner BM, Khan KN, Masferrer J & Koki AT 2000 COX-2 is expressed in human pulmonary, colonic, and mammary tumors. Cancer 89 2637–2645. (doi:10.1002/1097-0142(20001215)89:12<2637::AID-CNCR17>3.0.CO;2-B).
Steinbach G, Lynch PM, Phillips RK, Wallace MH, Hawk E, Gordon GB, Wakabayashi N, Saunders B, Shen Y & Fujimura T et al. 2000 The effect of celecoxib, a cyclooxygenase-2 inhibitor, in familial adenomatous polyposis. New England Journal of Medicine 342 1946–1952. (doi:10.1056/NEJM200006293422603).
Tanabe T & Tohnai N 2002 Cyclooxygenase isozymes and their gene structures and expression. Prostaglandins & Other Lipid Mediators 68–69 95–114. (doi:10.1016/S0090-6980(02)00024-2).
Timoshenko AV, Xu G, Chakrabarti S, Lala PK & Chakraborty C 2003 Role of prostaglandin E2 receptors in migration of murine and human breast cancer cells. Experimental Cell Research 289 265–274. (doi:10.1016/S0014-4827(03)00269-6).
Tsatsanis C, Androulidaki A, Venihaki M & Margioris AN 2006 Signalling networks regulating cyclooxygenase-2. International Journal of Biochemistry & Cell Biology 38 1654–1661. (doi:10.1016/j.biocel.2006.03.021).
Viana-Pereira M, Lopes JM, Little S, Milanezi F, Basto D, Pardal F, Jones C & Reis RM 2008 Analysis of EGFR overexpression, EGFR gene amplification and the EGFRvIII mutation in Portuguese high-grade gliomas. Anticancer Research 28 913–920.
Vo BT, Morton D Jr, Komaragiri S, Millena AC, Leath C & Khan SA 2013 TGF-β effects on prostate cancer cell migration and invasion are mediated by PGE2 through activation of PI3K/AKT/mTOR pathway. Endocrinology 154 1768–1779. (doi:10.1210/en.2012-2074).
Woo MM, Salamanca CM, Minor A & Auersperg N 2007 An improved assay to quantitate the invasiveness of cells in modified Boyden chambers. In Vitro Cellular & Developmental Biology. Animal 43 7–9. (doi:10.1007/s11626-006-9002-4).
Xu K & Shu HK 2007 EGFR activation results in enhanced cyclooxygenase-2 expression through p38 mitogen-activated protein kinase-dependent activation of the Sp1/Sp3 transcription factors in human gliomas. Cancer Research 67 6121–6129. (doi:10.1158/0008-5472.CAN-07-0141).
Xu K & Shu HK 2013 Transcription factor interactions mediate EGF-dependent COX-2 expression. Molecular Cancer Research 11 875–886. (doi:10.1158/1541-7786.MCR-12-0706).
Zhang X, Chen ZG, Choe MS, Lin Y, Sun SY, Wieand HS, Shin HJ, Chen A, Khuri FR & Shin DM 2005 Tumor growth inhibition by simultaneously blocking epidermal growth factor receptor and cyclooxygenase-2 in a xenograft model. Clinical Cancer Research 11 6261–6269. (doi:10.1158/1078-0432.CCR-04-2102).
Zhang X, Zhang J, Yang X & Han X 2007 Several transcription factors regulate COX-2 gene expression in pancreatic β-cells. Molecular Biology Reports 34 199–206. (doi:10.1007/s11033-007-9085-3).
