Background
Breast cancer is the most common malignancy and a major cause of death among women in the Western world [
1]. Many anticancer agents, including 5-fluorouracil, cyclophosphamide, and monoclonal antibodies such as trastuzumab, have shown efficacy in extending the survival of breast cancer patients; however, the mechanisms by which these agents inhibit breast cancer progression are not clearly understood. Although many promising anticancer agents have been developed and show potential in preclinical trials, classic chemotherapeutic agents such as doxorubicin are still widely used in patients [
2].
A major problem with the use of chemotherapy to treat many cancers (including breast cancer) is intrinsic or acquired drug resistance, which results in disease recurrence and metastasis. Recent results from several laboratories have investigated the mechanism by which breast cancer cells become resistant to doxorubicin, as well as the molecular profile of breast cancer cells that are resistant to doxorubicin [
3,
4]. Bcl-xl is responsible for acquisition of resistance to chemotherapeutic agents such as doxorubicin, leading to decreased apoptosis and increased survival of breast cancer cells [
5,
6]. Furthermore, recent evidence has suggested that the ability of tumor cells to acquire an aggressive phenotype may result from accumulation of genetic alterations conferred by extended survival [
7,
8].
Cox-2 is involved in the inflammatory response and its expression is commonly upregulated in human cancers; therefore, Cox-2 has been suggested to play a major role in tumorigenesis [
9,
10]. Recent studies have reported that Cox-2 plays a key role as a regulator of chemotherapy resistance in cancer. Cox-2 expression has been reported to be indicative of an aggressive breast cancer phenotype that is resistant to doxorubicin [
11]. For example, drug-resistant cell lines that overexpress P-glycoprotein 170 (MDR1/Pgp170) also have significantly upregulated Cox-2 expression, indicating a strong correlation between Cox-2 expression and resistance to chemotherapy in breast cancer cell lines [
12]. In addition, selective inhibition of Cox-2 suppresses the invasion activity of oral squamous cells through downregulation of a matrix metalloproteinase-2 (MMP-2)-activating mechanism [
13]. Cox-2 overexpression in human breast cancer cells enhances their motility and invasiveness [
14]. Furthermore, Cox-2 overexpression in human breast cancers correlates with several clinical parameters that are characteristic of aggressive breast disease [
15,
16]. Inhibitors that are selective for Cox-2 have been developed as anti-inflammatory agents and also show effective anticancer properties in breast cancer patients at risk for disease recurrence. Furthermore, inhibition of Cox-2 has a significant effect on the drug resistance and metastatic potential of cancer cells [
17]. Knocking down Cox-2 using small interfering RNA (siRNA) or Cox-2 inhibitors suppresses cell growth and invasion and enhances the chemosensitivity of cancers, including breast cancer [
18‐
20].
Several lines of evidence have suggested that metastasis may be enhanced by an ability to resist apoptosis and highly metastatic cancer cells exhibit greater survival ability and resistance to apoptosis than poorly metastatic cells [
21,
22]. Therefore, cancer cells may acquire invasive and metastatic properties during the process of becoming resistant, a mechanism that remains poorly understood. To identify genes associated with the invasive and metastatic activities of drug-resistant cells, we analyzed changes in gene expression in doxorubicin-resistant MCF-7 breast cancer cells (MCF-7/DOX) that we established using DNA array analysis. We observed invasive activities related to high expression of Cox-2 in MCF-7/DOX cells.
Having identified Cox-2 as an important regulator of the invasiveness of MCF-7/DOX cells, we next asked which upstream pathway modulates the expression of Cox-2 and how the invasive activities increased doxorubicin-resistant cancer in this study.
Methods
Animals, cells, and materials
Female 6-week-old Balb/c nude mice were purchased from Charles River Laboratories (Wilmington, MA, USA). The human breast cancer cell lines MDA-MB-231, MCF-7, and T-47D were obtained from the American Type Culture Collection (Manassas, VA, USA). MCF-7/DOX cells were derived from MCF-7 cells by continuous culture in the presence of doxorubicin (Sigma-Aldrich, St. Louis, MO, USA) for more than 3 months. Exposure of MCF-7 cells to stepwise increasing concentrations (0.1-1 μM) of doxorubicin resulted in the selection of doxorubicin-resistant MCF-7/DOX cells. Exposure to doxorubicin was terminated 4 days prior to the experiments. Cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and penicillin/streptomycin (Invitrogen, Grand Island, NY, USA). Cell culture inserts incorporating polyethylene terephthalate membranes (6.4-mm diameter, 8- μm pore size) and 24-well plates for invasion assays were purchased from Costar (Cambridge, MA, USA).
We obtained MTT [3-(4,5-dimethylthazol-2-yl)-2,5-diphenyltetrazolium bromide] from Sigma-Aldrich. The phosphoinositide 3-kinase (PI3K) inhibitor LY294002 and mitogen-activated protein kinase (MAPK) inhibitor U0126 were purchased from Calbiochem-Novabiochem (La Jolla, CA, USA). Sulprostone, 17-phenyl trinor Prostaglandin E2 (17-PT-PGE2), Prostaglandin E2 (PGE2), and the Cox-2 inhibitor NS398 were purchased from Cayman Chemical (Ann Arbor, MI, USA). Epidermal growth factor (EGF) was purchased from R&D Systems Inc. (Minneapolis, MN, USA). Gefitinib was purchased from Biaffin GmbH & Co KG (Kassel, Germany). Gelatin, fibrinogen, and plasminogen were obtained from Sigma-Aldrich. Antibodies against ERK1, Cox-2, and actin were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Antibodies against the EGF receptor (EGFR), pAkt, Akt, phosphorylated extracellular signal regulated kinase 1/2 (pERK1/2), pEGFR, poly(ADP-ribose) polymerase (PARP), and tubulin were purchased from Cell Signaling Technology (Danvers, MA, USA). Doxorubicin was purchased from Sigma-Aldrich and dissolved in phosphate-buffered saline (PBS) at various concentrations to establish dose responses. Synthetic siRNAs targeting EGFR, prostaglandin E receptor 1 (EP1), and EP3 were purchased from Bioneer (Seoul, Korea) and have the following sequences: EGFR (5'-GGCACGAGUAACAAGCUCA-3'); EP1 (5'-GUCGGUAUCAUGGUGGUGU-3'); and EP3 (5'-GUCAUCGUCGUGUACCUGU-3').
MTT assay
The inhibitory effect of doxorubicin, the Cox-2 inhibitor NS398, and the PI3K inhibitor LY294002 on growth of MCF-7 and MCF-7/DOX cell lines was determined using the MTT assay. Cells were plated onto 96-well plates (3 × 103 cells/well) and cultured in medium with or without various concentrations of doxorubicin, NS398, LY294002, gefitinib, and U0126. The cells then were grown for an additional total incubation of 24 or 72 h.
Flow cytometry assay
Cells were harvested, washed, fixed with paraformaldehyde and 70% ethanol, and stained using an APO-BRDU kit (Biovision, Inc., Mountain View, CA, USA) according to the manufacturer's protocol. Flow cytometric analysis was performed using a BD FACS Calibur flow cytometer (BD Biosciences, San Jose, CA, USA) equipped with a 488-nm argon-ion laser. Approximately 10,000 events (cells) were evaluated for each sample.
Western blot analysis
Total cell extracts were prepared from human breast cancer cells treated with various drugs as indicated. Preparation of whole-cell lysates, protein quantification, gel electrophoresis, and Western blotting were performed as described elsewhere [
23]. Protein concentrations were measured using the bicinchoninic acid protein assay (Pierce Biotechnology, Rockford, IL, USA), as described in the manufacturer's protocol. Equivalent amounts of protein from cell lysates or conditioned media (CM) from each treatment group were resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblotted with primary antibodies. Bands were detected using ECL Western blotting detection reagents from GE Healthcare (Chalfont St. Giles, United Kingdom).
Invasion assays
In vitro invasion assays were performed as described elsewhere [
24]. Briefly, CM obtained by culturing Wi38 fibroblasts for 18 h in DMEM with 10% FBS was placed into the lower chamber of each well as a chemoattractant. The upper chamber contained 1 × 10
5 MCF-7/DOX cells incubated in media alone or in the presence of NS398, LY294002, gefitinib, or U0126 for 18 h. Cells were fixed and stained with hematoxylin and eosin. Filters were cut out and mounted on glass slides for cell counting. Cells from the entire membrane field were counted. All experiments were repeated at least three times.
siRNA transfection
MCF-7/DOX cells were transfected with siRNA using Lipofectamine RNAiMAX (Invitrogen, Carlsbad, CA, USA). The siRNA-transfected cells were incubated for 48 h and harvested for Western blot analysis.
Reverse transcription-polymerase chain reaction (RT-PCR)
Total RNA was isolated from MCF-7/DOX cells using Trizol reagent (Invitrogen). cDNAs synthesized from 1 μg of total RNA were used as templates in a 50- μl reaction using the TaqMan RT reagents according to the manufacturer's protocol (Applied Biosystems, Foster City, CA). RT-PCR was performed to amplify genes using a cDNA template corresponding to gene-specific primer sets. The primer sequences used are as follows: EP1 (forward 5'-TCGGCCTCCACCTTCTTTGGC-3' and reverse 5'-CTGGCGCAGTAGGATGTACAC-3'), EP2 (forward 5'-GTCATGTTCTCGGCCGGGGTG-3' and reverse 5'-GAGGACTGAACGCATTAGTCT-3'), EP3 (forward 5'- CGCCGGGAGAGCAAGCGCAAG-3' and reverse 5'-GATGCGGCCCCACTGGGCACTGGA-3'), EP4 (forward 5'-ATCTTACTCATTGCCACC-3' and reverse 5'-TCTATTGCTTTACTGAGCAC-3'), urokinase-type plasminogen activator (uPA; forward 5'-CCAATTAGGAAGTGTAAGCAGC-3' and reverse 5'-GCCAAGAAAGGGACATCTATG-3'), MMP-2 (forward 5'-TCGCCCATCATCAAGTTC-3' and reverse 5'- GTGATCTGGTTCTTGTCC-3'), MMP-9 (forward 5'-AACCAATCTCACCGACAG-3' and reverse 5'-CAAAGGCGTCGTCAATCA-3'), β-actin (forward 5'-GTGGGGCGCCCCAGGCACCA-3' and reverse 5'-CTCCTTAATGTCACGCACGATTTC-3').
To avoid amplifying genomic DNA, gene primers were chosen from different exons. PCR was performed in a total reaction volume of 25 μl that contained 2 μl of cDNA solution and 0.2 μM of sense and antisense primers. The RT-PCR exponential phase was determined on cycles 28-33 to allow quantitative comparisons among the cDNAs amplified from identical reactions. The amplification products (8 μL) were resolved on a 2% agarose gel, stained with ethidium bromide, and visualized on a transilluminator and photographed.
MCF-7 and MCF-7/DOX cells (3 × 106 cells in 100 μL PBS) were injected into the tail vein of Balb/c nude mice (MCF-7, n = 8; MCF-7/DOX, n = 9). Three months after injection, the animals were killed by CO2 inhalation and their lungs were excised. Lung tumor formation was observed and tumor nodules were counted under a dissecting microscope. All animal experiment procedures were approved by the Institutional Animal Care and Use Committee in Korea National Cancer Center.
Gelatin and fibrinogen/plasminogen zymography
The proteolytic activity of MMP-2, MMP-9, and uPA in CM was analyzed by substrate-gel electrophoresis [
25] using SDS-PAGE gels containing 0.2% (m/v) gelatin or 0.12% (m/v) fibrinogen and plasminogen (0.01 NIH unit/mL). CM from each treatment group was concentrated using an Amicon Ultra-4 centrifugal device (Millipore, Bedford, MA, USA) and loaded onto gels. After electrophoresis, the gels were washed with 2.5% Triton X-100 and incubated overnight in zymogram incubation buffer (50 mM Tris-HCl, 0.15 M NaCl, 10 mM CaCl
2, and 0.02% NaN
3) at 37°C. Clear bands indicative of gelatinolytic activity were visualized by staining the gels with Coomassie blue.
Gene expression analyses from whole genome
Total RNA was isolated and purified from MCF-7 and MCF-7/DOX cells using the TRIzol reagent (Invitrogen) and RNease Mini kit (QIAGEN). Of those, 500 ng RNA was biotinylated and amplified using the Illumina TotalPrep RNA Amplification Kit (Ambion, TX) according to the manufacturer's instructions. The cRNA yield was measured using RiboGreen RNA quantitation kit (Invitrogen), and 750 ng of the cRNA sample was hybridized on a human HT-12 expression bead chip (Illumina, CA) for profiling 48,804 transcripts per sample. Bead chips were stained with streptavidin and scanned using an Illumina BeadArray Reader. BeadStudio V3 was used to quantile-normalize the data.
To find doxorubicin-resistant phenotype associated genes, we applied expression data to search and include genes with significant difference in expression levels between MCF-7 and MCF-7/DOX. Gene sets with 2-fold or more difference in mRNA level and
p value cutoff (0.05) are presented in Table
1.
Table 1
Differentially requlated genes in MCF-7/DOX cells
EGFR
| Epidermal growth factor receptor | 19.079 |
TWIST1
| Twist transcription factor | 8.661 |
MMP9
| Matrix metallopeptidase 9 | 5.890 |
PLAU
| Plasminogen activator, urokinase | 159.280 |
TIMP3
| TIMP metallopeptidase inhibitor3 | 0.020 |
CDH2
| N-cadherin | 94.160 |
CDH1
| E-cadherin | 0.001 |
Statistical analysis
The effect of doxorubicin or NS398 on breast cancer cell proliferation was analyzed using one-way ANOVA followed by Turkey's multiple test (SPSS version 10; SPSS, Chicago, IL, USA). The data of in vitro cancer cell invasion and tumor incidence in the mice were analyzed using Student's t-test (Microsoft Excel software version 2007; Microsoft Corporation).
Discussion
Chemotherapy plays an important role in the treatment of breast cancer; however, long-term treatment often results in chemoresistance, leading to disease recurrence and metastasis [
18‐
20]. To study the molecular mechanisms underlying invasive and metastatic activities in drug-resistant cancer cells, we generated the doxorubicin-resistant MCF-7 breast cancer cell line MCF-7/DOX. We found that MCF-7/DOX breast cancer cells displayed enhanced metastatic and invasive behavior both in
in vitro cell invasion assays and
in vivo in a mouse lung tumor model. We demonstrated that invasiveness of MCF-7/DOX cells resulted from Cox-2 activation, which was induced by either the EGFR-activated PI3K/Akt or MAPK pathway. Inhibiting either Cox-2 or the PI3K/Akt pathway effectively inhibited the invasiveness of MCF-7/DOX cells.
Cox-2 was coexpressed with EGFR in human colorectal cancer and bronchial adenocarcinomas [
35,
36] and induced in a human glioma cell line [
31]. We investigated the mechanisms by which EGFR signaling regulates Cox-2 expression. The EGFR pathway controls several pathways, including the PI3K/Akt and MAPK pathways [
32]. Our data showed that, in MCF-7/DOX cells, Cox-2 expression was regulated by both the PI3K/Akt and Ras/Raf/MAPK pathways through EGFR signaling. Western blot analysis showed that, in MCF-7/DOX and MDA-MB-231 cells, Cox-2 expression was reduced when EGFR expression was blocked by an EGFR-specific siRNA. In addition, the EGFR inhibitor gefitinib significantly suppressed EGF-induced Cox-2 expression and invasion of MCF-7/DOX cells. These data provide evidence that Cox-2 expression induced by the EGFR pathway is associated with invasiveness of MCF-7/DOX cells.
PGE
2, the major end product of Cox-2 activation, is also known to activate EGFR through various pathways [
37]. Therefore to clarify whether PGE
2 signaling through EPs promotes the PI3K/Akt or MAPK pathway-mediated invasion primarily, we evaluated the effect of EP1- or EP3-specific agonists or EP inhibitor on the PI3K/Akt or MAPK pathways, but we found that PGE
2 signaling through EPs didn't affect PI3K/Akt and MAPK pathway in the MCF-7/DOX cells (data not shown).
Recent studies have shown that Cox-2 mRNA and protein expression in several cancer cell lines are regulated by the insulin-like growth factor (IGF)-1R/PI3K and nuclear factor-kappa B/nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor pathways [
38,
39]. In addition to the PI3K/Akt pathway, the Ras/Raf/MAPK pathway is also a downstream transducer of IGF-1R signaling. The IGF-1R signaling pathway plays a major role in cell proliferation, apoptosis, invasion, and angiogenesis. Moreover, IGF-1R has been shown to upregulate Cox-2 mRNA expression and PGE
2 synthesis in cancer cells [
40]. Although we found that IGF-1R expression was neither increased nor constitutively activated in MCF-7/DOX cells, activation of the IGF-1R pathway may still contribute to Cox-2 expression and our efforts are ongoing to determine any further possibility.
Treating cells with EGF also increased pAkt and pERK1/2 expression in MCF-7/DOX cells. To investigate the role of the PI3K/Akt pathway in Cox-2 expression, we studied the effect of the PI3K/Akt inhibitor LY294002 on EGF-induced pAkt and Cox-2 expression. Western blot analysis showed that LY294002 dramatically suppressed pAkt activation and Cox-2 expression induced by EGF in MCF-7/DOX cells.
Because Cox-2 exerts its effects by producing PGE
2, which binds to specific EP receptors [
34], we investigated the role of specific EP receptors in Cox-2-mediated invasion of MCF-7/DOX cells. PGE
2 treatment induced expression of the EP1 and EP3 receptors, suggesting that these two receptors are likely involved in the invasiveness by MCF-7/DOX cells. Both EP1 and EP3 receptors played an important role in Cox-2 induced invasion of MCF-7/DOX cells. We showed that selective inhibition of EP1 and EP3 using siRNAs inhibited PGE
2-induced invasion of MCF-7/DOX cells, as well as expression of MMP-2 and MMP-9. A previous study showed increased Cox-2 expression in patients with poorly differentiated breast cancer and poor clinical outcomes for patients treated with doxorubicin [
11]. However, the expression pattern of EP receptors has never been studied in breast cancer. Therefore, our findings are the first to suggest a pivotal role for the EP1 and EP3 receptors in doxorubicin-resistant breast cancer cells.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
All authors participated in design of the study. JHK performed the experimental work and wrote the manuscript. KHS, KCJ, SK, CC, and CHL contributed to data analysis and interpretation. SHO conceived of the study, participated in the experimental design, and helped to draft the manuscript. All authors read and approved the final manuscript.