Differential regulation of the aggressive phenotype of inflammatory breast cancer cells by prostanoid receptors EP3 and EP4

Authors

  • Fredika M. Robertson PhD,

    Corresponding author
    1. Department of Experimental Therapeutics, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    2. Department of Breast Medical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    3. The Morgan Welch Inflammatory Breast Cancer Research Program and Clinic, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    • Department of Experimental Therapeutics, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Unit 422, Houston, TX 77030
    Search for more papers by this author
    • Fax: (713) 792-8089

  • Ann-Marie Simeone PhD,

    1. Department of Biochemistry and Molecular Biology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    Search for more papers by this author
  • Anthony Lucci MD,

    1. The Morgan Welch Inflammatory Breast Cancer Research Program and Clinic, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    2. Department of Surgical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    Search for more papers by this author
  • John S. McMurray PhD,

    1. Department of Experimental Therapeutics, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    Search for more papers by this author
  • Sukhen Ghosh PhD,

    1. Department of Experimental Therapeutics, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    Search for more papers by this author
  • Massimo Cristofanilli MD

    1. Department of Experimental Therapeutics, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    2. Department of Breast Medical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    3. The Morgan Welch Inflammatory Breast Cancer Research Program and Clinic, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    Current affiliation:
    1. Department of Medical Oncology, Fox Chase Cancer Center, Philadelphia, Pennsylvania
    Search for more papers by this author

  • The articles in this supplement were presented at the First International Inflammatory Breast Cancer Conference, Houston, Texas, December 5-7, 2008.

Abstract

BACKGROUND:

Although inflammatory breast cancer (IBC) is recognized as the most lethal variant of locally advanced breast cancer, few molecular signatures of IBC have been identified that can be used as targets to develop therapeutics that effectively inhibit the aggressive phenotype displayed by IBC tumors.

METHODS:

Real-time polymerase chain reaction analysis, Western blot analysis, modified Boyden chamber invasion assays, vasculogenic mimicry (VM) assays, and gelatin zymography were used in the current studies. Agonists and antagonists of the prostanoid receptors EP3 and EP4 and of EP4 short-hairpin RNA (shRNA) knockdown approaches were used as tools to assess the role of prostanoid receptors EP3 and EP4 in the regulation of specific biologic activities of IBC cells.

RESULTS:

The current studies revealed that the IBC breast cancer cell lines SUM149 and SUM190 express high levels of cyclooxygenase-2 messenger RNA and protein, produce abundant levels of prostaglandin E2, and produce both EP3 and EP4 receptor proteins. Studies using the EP4 antagonist GW627368X and shRNA molecular knockdown approaches revealed a role for EP4 in regulating invasion of IBC cells. EP3, but not EP4, regulated the ability of SUM149 cells to undergo VM, which is the ability to form capillary-like structures, a characteristic exhibited by very aggressive tumor types. Inhibition of VM by sulprostone was associated with an inhibition of matrix metalloprotease-2 (MMP-2) enzyme activity.

CONCLUSIONS:

The prostanoid receptors EP3 and EP4 differentially regulate activities exhibited by IBC cells that have been associated with the aggressive phenotype of this lethal variant of breast cancer. Whereas EP4 regulates invasion, EP3 regulates VM and the associated increased MMP-2 enzyme activity. Cancer 2010;116(11 suppl):2806–14. © 2010 American Cancer Society.

Inflammatory breast cancer (IBC) is the most aggressive subtype of locally advanced breast cancer with the worst prognosis and the shortest overall survival of any variant of this disease.1 IBC commonly presents as a skin rash caused by the invasion of IBC tumor cells into dermal lymphatics. Because of this peculiar pattern of presentation, IBC commonly is misdiagnosed, which delays the correct diagnosis. IBC patients often present with late-stage disease (stage IIIB or stage IV) at the time of the first accurate diagnosis. Because IBC does not present commonly as a lump, imaging modalities like as magnetic resonance imaging and positron emission tomography, rather than mammography and ultrasound, are required for accurate diagnosis and staging. Although the standard of care for patients with IBC for the past 30 years has been multimodality treatment combining chemotherapy, surgery, and radiation, there has been no change in the very low overall survival rate of 2.9 years for patients diagnosed with IBC during this time period.2 Studies performed over the past decade have defined few molecular characteristics of IBC tumors, including the association between increased expression of RhoC guanosine triphosphatase3 and loss of the tumor suppressor CCN6/Wnt-induced secreted protein 3 (WISP-3).4 Other studies have documented the increased expression of genes in IBC tumors, including the cyclooxygenase-2 (Cox-2) enzyme,5 which produces prostaglandin E2 (PGE2). On the basis of observations described in the current studies that the Cox-2 gene is highly up-regulated in the SUM190 and SUM149 cell lines, the objective of the current studies was to define the role of the distinct prostanoid (EP) receptors, EP1, EP2, EP3, and EP4, which mediate the biologic activities of Cox-2 and PGE2. By using EP receptor agonists and antagonists as well as molecular knockdown approaches, the effects of inhibition of these receptors was evaluated for their potential as therapeutic agents to block activities associated with the aggressive phenotype of IBC.

MATERIALS AND METHODS

Cell Lines and Conditions

The human MCF-7 and MDA-MB-231 breast cancer cell lines were obtained from American Type Cell Culture (Manassas, Va). MCF-7 is an estrogen receptor (ER)-positive breast cancer cell line and MDA-MB-231 cells were derived from a patient with triple-negative, basal-like breast cancer. MCF-7 and MDA-MB-231 cells were cultured in Dulbecco modified Eagles medium (DMEM) (DMEM/F12; Invitrogen, Carlsbad, Calif) supplemented with 10% heat-inactivated fetal bovine serum (FBS) (Invitrogen) at 37°C under 5% CO2 in a humidified incubator. The SUM149 and SUM190 cell lines were provided by Dr. Stephen Ethier (Asterand, Detroit, Mich) and were cultured in Ham F12 Nutrient Mixture (Invitrogen) supplemented with 10% FBS, insulin (1 mg/mL; Sigma-Aldrich, St. Louis, Mo), and hydrocortisone (1 mg/mL; Sigma-Aldrich). Cell lines were cultured at 37°C under 5% CO2 in a humidified incubator.

Reagents

Hydrogen peroxide (30%), Triton X-100, and the monoclonal β-actin antibody were purchased from Sigma-Aldrich. The Universal Vectastain Elite ABC Kit, hematoxylin, 3-amino-9-ethyl-carbazole, Aqua-Mount, and rabbit and mouse immunoglobuliln G (IgG) antibodies were purchased from Vector Laboratories (Burlingame, Calif). Matrigel was purchased from BD Biosciences (Bedford, Mass). Hema-3 was purchased from Fisher Scientific (Middleton, Va). Gelatin zymogram gels, nonreducing sample buffer, 10x zymogram renaturing buffer, 10x zymogram development buffer, Coomassie blue-R250, and Coomassie blue destaining solution were purchased from Bio-Rad Laboratories (Hercules, Calif). PGE2, arachidonic acid, PGE1 alcohol, butaprost, sulprostone, COX-2 monoclonal antibody, and EP3 and EP4 receptor polyclonal antibodies were purchased from Cayman Chemical (Ann Arbor, Mich). AH6809 and GW627368X were synthesized as described previously.6, 7 EP4 primers were purchased from Sigma-Genosys (St. Louis, Mo). Stock solutions of PGE2 (10 mM), arachidonic acid (10 mM), PGE1 alcohol (1 mM), butaprost (1 mM), AH6809 (10 mM), GW627368X (10 mM), and sulprostone (10 mM) were prepared in dimethyl sulfoxide (DMSO) and stored at −20°C. All reagents were diluted in culture medium to the indicated final concentration.

Real-Time Polymerase Chain Reaction Arrays

Cox-2 gene expression was evaluated using RT2 Profiler polymerase chain reaction (PCR) arrays (SABioscience, Frederick, Md). RNA was isolated from subconfluent SUM149, SUM190, MCF-7, and MDA-MB-231 breast tumor cells using Trizol reagent (Invitrogen). Complementary DNA (cDNA) was isolated using the RT2 First Strand Kit and then was added to the RT SYBR Green quantitative PCR Master Mix (Applied Biosystems, Foster City, Calif). This mixture was then aliquoted into each well of 96-well PCR array plates that contained predispensed, gene-specific primer sets. Thermal cycling was performed using an ABI 7300 real-time PCR system (Applied Biosystems). The threshold cycle for each well was calculated using SuperArray software (SABioscience). Five housekeeping genes with an additional 3 quality controls are re included in each array.

Western Blot Analysis of Cox-2, and EP Receptors

Western blot analyses were performed on protein lysates that were obtained from exponentially growing SUM149, SUM190, MCF-7, and MDA-MB-231 breast cancer cells. Cell pellets were lysed in 1% NP-40 lysis buffer. Equal amounts of protein were separated using 12% polyacrylamide gels (Bio-Rad) for COX-2 or 15% polyacrylamide gels (Bio-Rad) for EP3 and EP4. Proteins were then transferred to nitrocellulose membranes (GE Healthcare, Piscataway, NJ), blocked for nonspecific binding, and then probed with monoclonal COX-2 antibody or polyclonal EP3 and EP4 antibodies followed by incubation with antimouse or antirabbit IgG horseradish peroxidase-conjugated secondary antibodies (GE Healthcare). Levels of COX-2, EP3, and EP4 receptor protein expression were normalized to β-actin, which was used as a loading control. Protein bands were observed by enhanced chemiluminescence (GE Healthcare). Experiments were repeated 3 times, and representative Western blots are shown.

PGE2 Production By Human Breast Cancer Cells Measured by Enzyme Immunoassay

Cells were plated at 1.0 × 106 cells per well in 100-mm dishes in 10 mL of medium; and, 24 hours later, cells were placed in serum-free medium. At 24 hours, media was added to human breast tumor cells that either did or did not contain arachidonic acid (10 μM). After a 30-minute incubation, culture medium was isolated, centrifuged, and concentrated using spin columns with 10-kD cutoff filters (Millipore, Bedford, Mass). The amount of PGE2 in the concentrated supernatants was determined using a commercial enzyme immunoassay kit (Cayman Chemical, Ann Arbor, Mich) according to the manufacturer's instructions. PGE2 concentrations were normalized to cell numbers and are expressed as pg/mL/106 cells. The amount of PGE2 produced in the presence of arachidonic acid is designated as the amount of exogenous PGE2, and the amount of PGE2 produced in the absence of arachidonic acid represents that amount of endogenous PGE2 produced by human breast tumor cell lines. The experiments were performed in triplicate and were repeated twice.

Effects of EP Receptor Agonists and Antagonists on Invasion of an Artificial Basement Membrane by Human Breast Cancer Cells

The effects of PGE2, the EP2 agonist butaprost, the EP4 agonist PGE1 alcohol, the EP1/EP2 antagonist AH6809, the EP3 agonist sulprostone, and the selective EP4 antagonist GW627368X on in vitro invasion of SUM149 IBC cells through a Matrigel basement membrane compared with MDA-MB-231 breast cancer cells was determined based on the number of cells that invaded through transwell inserts coated with the artificial basement membrane Matrigel. Neither MCF-7 cells nor SUM190 cells were used in these assays, because they have negligible invasive activity. SUM149 and MDA-MB-231 cells were trypsinized, resuspended in serum-supplemented medium, and counted. Cells were then washed 3 times with serum-free medium. Plates (6-well) compatible with transwell inserts with 8 μm pore-size polycarbonate filters (Fisher Scientific) were coated with Matrigel in cold serum-free DMEM/F12 at a final concentration of 0.7 mg/mL and held at room temperature for 40 minutes. Cells (in 500 μL serum-free medium) were added into the transwell inserts and incubated for 72 hours in the absence or presence of PGE2 (1 μM), the EP2 agonist butaprost (0.1 μM, 1 μM, and 10 μM), the EP4 agonist, PGE1 alcohol (0.1 μM, 1 μM, and 10 μM), the EP1/EP2 antagonist AH6809 (0.1 μM, 1 μM, and 10 μM), the EP3 agonist sulprostone (0.1 μM, 1 μM, and 10 μM), or the EP4 antagonist, GW627368X (0.1 μM, 1 μM, and 10 μM). For a control, 10% FBS was used to evaluate the baseline extent of invasion of the different cell lines. After incubation, noninvading cells on the upper surface of the filter were removed with cotton swabs. Cells that had passed through the pores onto the lower side of the filter were fixed, stained with Hema-3 stain (Fisher Scientific), and quantified. The experiments were performed in triplicate and were repeated twice.

Effects of EP4 Short-Hairpin RNA Knockdown on Invasion of SUM149 Cells

The short-hairpin RNA (shRNA) constructs against PTGER4 (EP4) were obtained from OriGene (Rockville, Md). A pRS plasmid containing EP4 shRNA and the negative control for the shRNA (empty vector) under the control of the mammalian U6 promoter was used to generate the stable transfectants. To obtain optimal EP4 knockdown, 2 different EP4 constructs designated as T1340309 (TTAAGTGTCTCACTAAAGCATGAAAT GTG) and T1340312 (GCGCTGCTCCGCATGCAC CGCCAGTTCAT) were used in these transfection studies. Negative control shRNA and the T1240309 or the T1340312 EP4 shRNA vectors were added to OptiMEM media. Lipofectamine 2000 was mixed with OptiMEM medium (Invitrogen), then these 2 solutions were mixed together and incubated for 20 minutes. The mixture of Lipofectamine, OptiMEM, and EP4 shRNA constructs or the vector control shRNA was then added by single drops to SUM149 cells. At 24 hours after transfection, the medium was replaced, and the cells were cultured for an additional 72 hours. Transfected SUM149 cells were transferred to Ham F-12 Nutrient Mixture supplemented with 10% FBS, insulin (1 mg/mL), and hydrocortisone (1 mg/mL; Sigma-Aldrich) with the addition of 2 μg/mL puromycin and were grown for 4 weeks with limited dilution performed so that each puromycin-resistant colony was derived from a single cell. Colonies were then isolated using cloning rings and were grown in 6-well plates. Western blot analysis was used to confirm knockdown of EP4. SUM149/clone 1 is used to designate a clonally derived cell line that contains stable transfection of EP4 shRNA and has knockdown of EP4 receptor protein as assessed by Western blot analysis. SUM149/Vector 5 is designated as the clonally derived cell line that has stable integration of the scrambled shRNA vector plasmid as determined by Western blot analysis to confirm the presence of EP4 receptor protein, which is similar to the nontransfected control SUM149 cells. Studies were then performed to evaluate the effect of EP4 shRNA knockdown on invasion by SUM149 IBC cells using the methods described above.

Effects of the EP3 Agonist Sulprostone on Vasculogenic Mimicry Exhibited by Breast Cancer Cells

MDA-MB-231 and SUM149 cells (4 × 104) were added to the top of 24-well plates coated with Matrigel (300 μL) in serum-free DMEM/F12 medium. Cells were treated with sulprostone or GW627368X at concentrations of either 0.1 μM, 1 μM, or 10 μM or with DMSO vehicle. To highlight the matrix-associated vascular channels that were formed, cells were stained with periodic acid-Schiff (PAS) reagent. Representative photographs were taken at 24 hours at ×10 magnification. Each experiment was repeated 3 times.

Effects of Sulprostone on Matrix Metalloprotease-2 Activity Assessed By Zymography

Supernatant fluids collected from MDA-MB-231 and SUM149 cells that had undergone vasculogenic mimicry (VM) were collected, centrifuged, and concentrated using spin columns with 10-kDa cutoff filters. Aliquots of 20 μL of conditioned medium was mixed (1:1) with nonreducing sample buffer, incubated at 37°C for 15 minutes, and then applied to a gelatin zymogram gel. After electrophoresis, gels were incubated for 3 hours in zymogram renaturing buffer, followed by an overnight incubation in zymogram development buffer at 37°C. Gels were stained with Coomassie blue-R250 for 3 hours and then destained. Matrix metalloprotease-2 (MMP-2) (gelatinase) activity was visible as clear bands against the dark-blue background, indicating proteolysis of the substrate protein, gelatin.

Statistical Analyses

Two-tailed Student t tests were performed to compare the amount of PGE2 produced by the different human breast tumor cell types in the presence and absence of arachidonic acid. Two-tailed Student t tests were performed to evaluate the number of invaded cells between untreated cells and those treated with PGE2, butaprost, PGE1 alcohol, AH6809, and GW627368X and to compare the effect of transfection of EP4 shRNA on the invasion of SUM149 tumor cells. A P value of <.05 was considered statistically significant.

RESULTS

Comparative Analysis of Cox-2 Messenger RNA In Breast Cancer Cells

Both the SUM149 and SUM190 IBC breast tumor cells expressed high levels of the Cox-2 gene. The levels of Cox-2 were 2500-fold and 11,000-fold increased in SUM149 cells (Fig. 1A, open bar) and SUM190 IBC cells (Fig. 1A, hatched bar), respectively, compared with the triple-negative, basal-like MDA-MB-231 breast cancer cells, which express very low levels of Cox-2 messenger RNA (mRNA) or the PTGS-2 null MCF-7 breast cancer cell line, which served as the negative control for these studies.

Figure 1.

Cyclooxygenase-2 (Cox-2) messenger RNA (mRNA) protein and prostanoid receptor (EP) protein production by breast cancer cells is shown. (A) The SUM190 and SUM149 inflammatory breast cancer (IBC) tumor cell lines expressed Cox-2 at high levels compared with the Cox-2 mRNA levels expressed by MCF-7 and MDA-MB-231 breast cancer cells. (B) Western blot analysis of Cox-2, EP3, and EP4 receptors demonstrated that SUM149 and SUM190 IBC cells produced high levels of Cox-2 protein. All cell lines produced both EP3 and EP4 receptor proteins, but greater levels of EP4 protein were produced by SUM190 and SUM149 cells compared with the levels produced by MCF-7 and MDA-MB-231 breast tumor cells. (C) Endogenous prostaglandin E2 (PGE2) production in both SUM190 and SUM149 cells was significantly greater (P < .05) than its production in either MCF-7 or MDA-MB-231 breast cancer cells. The addition of arachidonic acid significantly increased PGE2 production by SUM190 and SUM149 IBC cells.

Western Blot Analysis of Cox-2, EP3, and EP4 Receptor Proteins in Breast Cancer Cells

Both SUM149 and SUM190 IBC breast cancer cells produced high levels of Cox-2 protein compared with the low/undetectable levels of Cox-2 protein produced by the 2 non-IBC breast cancer cells (Fig. 1B). All cell lines that we examined produced EP3 and EP4 receptor proteins. The IBC cell lines produced greater levels of protein for both of these receptors, with higher levels of EP4 than of EP3 produced by both SUM149 cells and SUM190 cells (Fig. 1B).

PGE2 Production by Breast Cancer Cells

Both SUM190 cells and SUM149 cells produced significantly greater amounts of PGE2 (P < .05) compared with either MCF-7 cells or MDA-MB-231 cells, and the SUM190 cells (Fig. 1C,) produced more PGE2 compared with the SUM149 cells. Addition of the Cox-2 enzymatic substrate arachidonic acid to IBC cells stimulated a significantly increased production of PGE2.

Effects of EP3 Receptor Agonist and EP4 Antagonist on Invasion of a Basement Membrane Matrix by Breast Cancer Cells

Neither the MCF-7 cells nor the SUM190 cells exhibited a robust invasive response; therefore, only SUM149 cells and MDA-MB-231 cells were analyzed in the current studies. SUM149 invasion was increased significantly by PGE2 (1 μM) and by the EP4 agonist PGE1 alcohol (0.1 μM, 1 μM, and 10 μM; P < .05) (Fig. 2A). The selective EP2 agonist butaprost and the EP1/EP2 mixed antagonist AH6809 had no effect on invasion by either SUM149 cells or MDA-MB-231 cells (data not shown). Both the EP3 agonist sulprostone and the EP4 antagonist GW627368X significantly (P < .05) decreased the number of SUM149 cells that exhibited invasion at all concentrations tested (0.1-10 μM) (Fig. 2B,D). The EP4 antagonist GW627368X inhibited invasion by SUM149 cells by 66%, and the EP3 agonist sulprostone inhibited invasion of SUM149 cells by approximately 40%. Knockdown approaches to assess the role of EP4 in SUM149 invasion revealed that a stable clone of SUM149 cells containing EP4 shRNA (designated as SUM149/Clone 1) exhibited significantly inhibited invasion, which was diminished by 95% of the SUM149 cells containing a scrambled shRNA vector that was used as the control cell line (Fig. 2C). Neither PGE2 nor PGE1-alcohol altered invasion by MDA-MB-231 cells. Although the EP4 antagonist GW627368X inhibited invasion of MDA-MB-231 cells at concentrations of 1 μM and 10 μM, the EP3 agonist sulprostone had no effect on invasion by these cells at any concentration examined. At the cell densities that were used for the invasion assay, the EP4 antagonist GW627368X and the EP3 agonist sulprostone did not induce greater than 15% growth inhibition of SUM149 cells, demonstrating that EP receptor-mediated suppression of invasion was not caused by nonspecific growth inhibition of these breast cancer cells by either compound.

Figure 2.

The effects of prostanoid receptor (EP) agonists and antagonists on invasion of breast cancer cells are shown. (A) The effects of prostaglandin E2 (PGE2) and EP4 agonist PGE1-alcohol on invasion were studied. PGE2 and the EP4 agonist PGE1-alcohol induced significantly increased invasion of SUM149 cells but had no effect on the invasion of MDA-MB-231 cells. (B) The effect of the EP4 antagonist GW627368X is illustrated on invasion. GW627368X induced a significant dose-dependent inhibition of invasion beginning at the lowest concentration of 0.1 μM compared with the inhibition of invasion by MDA-MB-231 breast cancer cells by GW627368X at higher concentrations. (C) The effect of EP4 short-hairpin RNA (shRNA) on invasion is illustrated. The presence of EP4 shRNA in SUM149/Clone 1 abolished 95% of invasion. (D) The effect of sulprostone on invasion is illustrated. Sulprostone inhibited invasion by SUM149 cells in a dose-dependent manner starting at 0.1 μM with no effect of this agent on invasion by MDA-MB-231 cells.

Effects of EP3 Agonist Sulprostone and EP4 Antagonist GW627368X on VM by Breast Cancer Cells

Tumor cells derived from very aggressive tumors display phenotypic plasticity, which includes their ability to form matrix-rich capillary-like networks when placed in 3-dimensional culture in the absence of endothelial cells and fibroblasts, in a process defined as VM.8, 9 We observed that SUM149 and MDA-MB-231 breast cancer cells, but not SUM190 or MCF-7 cells, undergo VM when placed onto a Matrigel matrix (Fig. 3A,B). Phase-contrast microscopy (Fig. 3A) and PAS staining (Fig. 3B) were used to directly image the capillary-like structures that have closed ends formed by aggressive breast cancer cells. The effects of agents on the formation of these closed structures can be evaluated to provide a quantitative analysis of the relative inhibitory effects of different agents. By using this approach, we observed that VM was inhibited by the EP3 agonist sulprostone but was not altered by the EP4 antagonist GW627368X (Fig. 3C). In contrast, neither sulprostone nor GW627368X had any effect on VM exhibited by MDA-MB-231 cells (Fig. 3A-C) (data not shown for GGW627368X). Because MMP-2 activity reportedly is increased after cells undergoing VM, we evaluated the effects of the EP3 agonist sulprostone on MMP-2 activity, as assessed by gelatin zymography. Under control conditions, SUM149 cells do not have detectable MMP-2 activity; however MMP-2 activity is increased after SUM149 cells undergo VM, which effectively was inhibited by sulprostone (Fig. 3D). In contrast, MDA-MB-231 cells produced MMP-2 under control conditions, which was increased when these cells underwent VM and was not altered by the EP3 agonist sulprostone. These results are consistent with the lack of inhibition of VM by sulprostone reported in MDA-MB-231 non-IBC breast tumor cells, as described above.

Figure 3.

Effects of the prostanoid receptor 3 (EP3) agonist sulprostone on vasculogenic mimicry (VM) and matrix metalloprotease-2 (MMP-2) activity by breast cancer cells are shown. (A,B) These are breast cancer cells undergoing VM. (A) Phase-contrast microscopy and (B) light microscopy of breast cancer cells stained with periodic acid-Schiff (PAS) reagent allow visualization of SUM149 and MDA-MB-231 breast cancer cells undergoing VM. The EP3 agonist sulprostone blocked formation of capillary-like tubular structures by SUM149 cells but not by MDA-MB-231 cells. (C) Quantification of changes in the number of closed channels in breast cancer cells undergoing VM is illustrated. An analysis of changes in the number of closed channels of the capillary-like structures formed by breast cancer cells undergoing VM demonstrated that the EP3 agonist sulprostone blocked the formation of capillary-like structures formed by SUM149 inflammatory breast cancer cells but had no effect on VM exhibited by MDA-MB-231 cells. (D) MMP-2 activity was assessed by gelatin zymography. SUM149 cells under control conditions produced no MMP-2 activity; however, VM in these breast cancer cells produced MMP-2, which is inhibited by the EP3 agonist sulprostone. In contrast, sulprostone had no effect on MMP-2 activity by MDA-MB-231 cells.

DISCUSSION

IBC is the most aggressive variant of locally advanced breast cancer.1 Although studies have identified several genes that are over expressed or lost in IBC tumors and in IBC cell lines, to date, few molecular signatures have been identified that have led to “druggable” targets that would allow the development of effective therapeutic agents for the treatment of IBC. Previous studies have reported that Cox-2 is elevated in IBC tumors.5 Our laboratory was the first to report the association of elevated Cox-2 mRNA and protein levels in breast tumors with evidence of invasion,10 Numerous studies have now validated Cox-2 as a critically important regulator of invasion, angiogenesis, and metastasis.11-14 Additional studies performed in our laboratory demonstrated that the presence of Cox-2 cDNA directly regulates proliferation, invasion, anchorage-independent growth in soft agar, and production of the vascular endothelial growth factor by breast tumor cells in vitro15 as well as in breast tumor xenografts in immunocompromised mice.16 Other studies have reported that Cox-2 drives the ability of cells to undergo VM.17, 18

Although numerous studies have reported that selective Cox-2 inhibitors have antitumor and antiangiogenesis activities and effectively block breast cancer progression,19, 20 these agents either have been removed from the market (in the case of rofecoxib) or have black box warnings (in the case of celecoxib) because of the risk of cardiovascular side effects.21, 22 This has prompted a search for other approaches to block Cox-2/PGE2–associated activities. One alternative is to inhibit binding of PGE2 to its G protein-coupled receptors, defined as the prostanoid receptors (EPs). There are 4 members of this receptor family designated EP1, EP2, EP3, and EP4. Binding of PGE2 to these distinct receptors stimulates different biologic activities of cells, depending on the cell type as well as the distinct EP receptor(s) that is activated.23-26 Although binding of PGE2 or the appropriate agonist to the EP1 receptor stimulates the release of intracellular calcium and signals through the protein kinase C pathway, binding of PGE2 to the EP2 and EP4 receptors stimulates increased intracellular cyclic adenosine monophosphate and adenylate cyclase production.23 In contrast to the stimulatory activities of EP1, EP2, and EP4, EP3 is a negative regulator of the effects of PGE2, and it has been suggested that EP3 plays a role in inhibiting the effects of PGE2 through binding to an inhibitory subunit of the G protein, Gi, resulting in inhibition of adenylate cyclase. Although it is clear that binding of PGE2 to the EP receptors stimulates signaling pathways that play important roles in regulating normal homeostatic functions and also have play roles in a variety of human diseases, the importance of these receptors to biologic activities of tumor cells depends in part on the profile of the individual EP receptor and on the species and the tissue type under study. The availability of EP receptor agonists and antagonists provides the opportunity to evaluate the effects of blockade of the individual EP receptors on specific parameters associated with tumor growth and invasion. In addition, knockdown approaches can be used to define the role(s) of the EP receptors in regulating specific activities of tumor cells.

To our knowledge, the current observations are the first to demonstrate that Cox-2, PGE2, and the EP3 and EP4 prostanoid receptors all are up-regulated in IBC cells and that less EP3 and EP4 receptor protein is produced by the ER-positive, slowly growing, minimally invasive MCF-7 breast cancer cells and by the triple-negative, rapidly proliferating, highly invasive, and metastatic MDA-MB-231 breast cancer cells. These results suggest that EP3 and EP4 are the primary EP receptors that are active on IBC cells and that there is no apparent role of either the EP1 or EP2 receptors in any biologic activity of the IBC cells that we examined. Previous studies,27-30 including our own,31 have demonstrated that Cox-2 is involved in regulating invasion and metastasis. The current studies confirm that the EP4 receptor is involved primarily in regulating the activities of IBC cells associated with migration, invasion, and metastasis.27-30 In contrast to studies that have elucidated the roles of EP4 in breast metastasis, significantly less is known about the specific role of the down-regulatory EP3 receptor in activities of tumor cells. The current studies reveal that the MCF-7 and MDA-MB-231 breast cancer cells, which either lack Cox-2 or have low Cox-2 expression, are relatively resistant to the inhibitory effects of the EP3 agonist sulprostone and the EP4 antagonist GW627368X on any parameter that we evaluated until concentrations of these agents are reached that are associated with off-target effects.

To our knowledge, these studies are the first to define a differential response of breast cancer cells to EP receptor agonists and antagonists with identification of the specific roles of EP4 and EP3 receptors in regulating invasion and VM, respectively, exhibited by IBC breast tumor cells. On the basis of the combination of pharmacologic and molecular knockdown approaches, the current studies reveal that the EP4 receptor pathway appears to have a significant and primary role in regulating the invasion exhibited by SUM149 IBC cells, which were derived from pleural effusion of an IBC patient. The current studies are also the first to our knowledge to report a role for EP3 in regulating the ability of IBC tumor cells to undergo VM, which has been associated with an aggressive phenotype of other tumors, including uveal melanoma and ovarian carcinoma. EP4 does not appear be involved in VM exhibited by SUM149 IBC cells. These studies also suggest that EP3 regulates MMP-2 activity during VM in IBC cells, which has not been reported previously.

Taken together, the current studies demonstrate that Cox-2 is up-regulated in IBC tumor cells, which is consistent with previous reports that IBC is up-regulated in IBC primary tumors.5 The current studies extend these previous observations and demonstrate that the EP3 and EP4 receptors are critically important in differential regulation of the aggressive phenotype of IBC cells characterized by their robust invasion and their ability to undergo VM with associated increased production of the proteolytic enzyme, MMP-2. Agents that directly or indirectly may target the signaling pathways activated after binding of PGE2 to the EP3 and EP4 receptors may be important for evaluating their effectiveness as potential therapeutic agents for inhibiting the aggressive phenotype of IBC.

CONFLICT OF INTEREST DISCLOSURES

This supplement was sponsored by the Houston Affiliate of Susan G. Komen for the Cure, the National Cancer Institute, and the State of Texas Rare and Aggressive Breast Cancer Research Program. The First International Inflammatory Breast Cancer Conference was supported in part by GlaxoSmithKline, Pfizer, Eli Lilly and Company, and Cardinal Health. Supported by American Airlines-Komen Foundation Promise grant KGO81287 (to F.M.R. and M.C.) and National Institutes of Health National Cancer Institute grant CA-128797-02 (to M.C.).

Ancillary