Cancer Research AACR Conference on Molecular Diagnostics - 2008 Tumor Immunology: New Perspectives
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Eibl, G.
Right arrow Articles by Hines, O. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Eibl, G.
Right arrow Articles by Hines, O. J.
[Cancer Research 65, 982-990, February 1, 2005]
© 2005 American Association for Cancer Research

Experimental Therapeutics, Molecular Targets, and Chemical Biology

Growth Stimulation of COX-2–Negative Pancreatic Cancer by a Selective COX-2 Inhibitor

Guido Eibl1, Yasunori Takata2, Laszlo G. Boros3, Joey Liu2, Yuji Okada1, Howard A. Reber1 and Oscar J. Hines1

1 Section of Gastrointestinal Surgery, 2 Division of Endocrinology, Diabetes and Hypertension, and 3 Harbor-University of California at Los Angeles Research and Education Institute, David Geffen School of Medicine at University of California at Los Angeles, Los Angeles, California

Requests for reprints: Guido Eibl, Section of Gastrointestinal Surgery, David Geffen School of Medicine at University of California at Los Angeles 72-215 CHS, 10833 LeConte Avenue, Los Angeles, CA 90095-6904. Phone: 310-794-9577; Fax: 310-206-2472; E-mail: Geibl{at}mednet.ucla.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Cyclooxygenase 2 (COX-2) inhibitors are promising antiangiogenic agents in several preclinical models. The aim of the present study was to evaluate the effect of selective COX-2 inhibitors on vascular endothelial growth factor (VEGF) production in vitro and angiogenesis and growth of pancreatic cancer in vivo, focusing on putative differences between COX-2–negative and COX-2–positive tumors. VEGF production and angiogenesis in vitro were determined by ELISA and endothelial cell migration assay. To determine whether the effect of COX-2 inhibitors was mediated by peroxisome proliferator–activated receptor {gamma} (PPAR-{gamma}), we used a dominant-negative PPAR-{gamma} and a pharmacologic inhibitor. In vitro findings were validated in a pancreatic cancer animal model. Microvessel density was assessed by CD31 immunostaining. Intratumoral prostaglandin and VEGF levels were measured by mass spectroscopy and ELISA. Selective COX-2 inhibitors had a concentration-dependent effect on VEGF production in vitro. Higher concentrations increased VEGF levels and stimulated angiogenesis by activating PPAR-{gamma}. In vivo, nimesulide increased VEGF production by cancer cells in COX-2–positive and COX-2–negative pancreatic tumors. In COX-2–negative pancreatic cancer, this effect was associated with an increase in angiogenesis and growth. In COX-2–positive pancreatic cancer, the nimesulide-induced increase of VEGF production by the cancer cells was offset by a decrease in VEGF production by the nonmalignant cell types leading to reduced tumor angiogenesis and growth. Selective COX-2 inhibitors had opposite effects on growth and angiogenesis in pancreatic cancer depending on COX-2 expression. These findings imply that assessing the COX-2 profile of the pancreatic tumor is mandatory before initiating therapy with a selective COX-2 inhibitor.

Key Words: COX-2 inhibitor • pancreatic cancer • PPAR-{gamma} • VEGF • angiogenesis


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Soon after the solution of the cyclooxygenase (COX) crystal structures in the early to mid-1990s (1, 2), an unparalleled structure-based drug development program was started by several pharmaceutical companies, ultimately culminating in the introduction of selective COX-2 inhibitors into the market. These drugs were and for the most part are still heralded as the ultimate weapon against a multitude of human diseases, including cancer, and are currently being tested in several clinical trials as antitumor agents. Targeting COX-2 provides an intriguing, multilayered opportunity to fight cancer, as COX-2 has been causally linked to the suppression of the host immune system, to conferring resistance to apoptosis, to stimulate cancer cell growth and invasion, and to promote angiogenesis (3–6). Most antitumor effects of COX-2 inhibitors are mediated by reducing prostaglandin levels. However, some effects of selective COX-2 inhibitors cannot be explained by simple inhibition of the COX-2 enzyme, as these drugs can also provoke responses in COX-2–negative cells (7–14). Because the pathways activated or inhibited by COX-2 inhibitors in COX-2–lacking cells are largely unknown, the phenotypic changes in these cells are unpredictable.

One putative COX-independent pathway includes peroxisome proliferator–activated receptor {gamma} (PPAR-{gamma}; ref. 15), which belongs to the nuclear hormone receptor superfamily and functions as a ligand-activated transcription factor (16). Recently, a possible role of PPAR-{gamma} in human malignancies (prostate, breast, gastric, lung, and pancreatic cancer) has been implicated (17–21). Although the majority of the published studies imply an antitumor effect of PPAR-{gamma} ligands, some authors have observed an opposite effect with PPAR-{gamma} ligands promoting the development of colon and breast tumors in vivo (22–24). The effect of PPAR-{gamma} activation on angiogenesis is also controversial in the literature. PPAR-{gamma} ligands have been associated with either inhibiting or stimulating angiogenesis (25, 26).

The majority of human pancreatic cancers overexpress COX-2. However, 10% to 40% of pancreatic tumors are considered to be COX-2 negative (27–33). The effect of selective COX-2 inhibitors on the subset of COX-2–negative pancreatic cancer is poorly understood. As the COX-2 profile of the tumor is usually not known before treatment with selective COX-2 inhibitors, the definition of COX-2–independent pathways activated and/or inhibited by COX-2 inhibitors and the resulting phenotypic changes bears tremendous clinical importance. In this context, the aim of the present study was to evaluate the therapeutic effects of selective COX-2 inhibitors on pancreatic cancer angiogenesis and growth, with special emphasis on the putative difference between COX-2–positive and COX-2–negative tumors as well as on the possible involvement of PPAR-{gamma}.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Reagents
The PPAR-{gamma} ligands 15-PGJ2 and troglitazone, the irreversible PPAR-{gamma} antagonist GW9662, the selective COX-2 inhibitors nimesulide and DuP697, the rabbit polyclonal antibodies against COX-2 and PPAR-{gamma} were purchased from Cayman Chemicals (Ann Arbor, MI). The mouse monoclonal ß-actin antibody was purchased from Sigma Chemical Co. (St. Louis, MO) and rat anti-mouse CD31 antibody was obtained from BD Biosciences Pharmingen (San Diego, CA). Gel shift oligonucleotides containing a consensus and mutant PPAR-{gamma} response element (PPRE) and the PPAR-{gamma} and RxR{alpha} antibodies for supershift assays were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The full-length PPAR-{gamma} was kindly provided by Alex Elbrecht (Merck Research Laboratories, Rahway, NJ; ref. 34). pCMV-Luc and prACO.ßneo were kindly provided by Syngenta CTL Cell Biology Group (Macclesfield, Cheshire, United Kingdom) and pCMX-mPPARg was a generous gift from R.M. Evans (Howard Hughes Medical Institute, The Salk Institute for Biological Studies, San Diego, CA). Other reagents were purchased from common commercial sources.

Cell Culture
The human pancreatic cancer cell lines BxPC-3 and MIA PaCa-2, human umbilical vein endothelial cells, and the embryonic mouse fibroblast cell line NIH/3T3 were obtained from the American Type Culture Collection (Rockville, MD). BxPC-3 cells were grown in RPMI 1640 containing 10% fetal bovine serum, penicillin G (100 units/mL), and streptomycin (100 µg/mL) in a humidified atmosphere of 5% CO2 at 37°C. MIA PaCa-2 cells were cultured in DMEM containing 10% fetal bovine serum, penicillin G (100 units/mL), and streptomycin (100 µg/mL) in a humidified atmosphere of 10% CO2 at 37°C. All experiments were carried out using cancer cell lines between passages 7 and 12. Human umbilical vein endothelial cells were grown in endothelial cell basal medium supplemented with 0.1% human epidermal growth factor, 0.1% hydrocortisone, 0.1% gentamicin, 0.1% amphotericin-B, 0.4% bovine brain extract, and 2% fetal bovine serum (Cambrex Bio Science Walkersville, Walkersville, MD) in a humidified atmosphere of 5% CO2 at 37°C. NIH/3T3 cells were grown in DMEM supplemented with 10% BCS, penicillin G (100 units/mL), and streptomycin (100 µg/mL) in 5% CO2 at 37°C.

VEGF and PGE2 Production In vitro
Vascular endothelial growth factor (VEGF) and PGE2 levels in the culture medium were determined as described previously (35). Briefly, cells were incubated with the indicated inhibitors and ligands for 24 hours in serum-free medium and VEGF and PGE2 levels in the culture medium were measured using the Quantikine Human VEGF Immunoassay (R&D Systems, Inc., Minneapolis, MN) and the PGE2 competitive enzyme immunoassay (Cayman Chemical), respectively, as recommended by the manufacturer. Protein levels were normalized to cell number.

Endothelial Cell Migration
Human endothelial cell migration was examined in a 24-well cell culture insert system (Becton Dickinson Labware, Franklin Lakes, NJ) as described previously (35). Briefly, pancreatic cancer cells (4 x 105) were seeded into each well of a 24-well companion TC plate in complete cell culture medium. After serum starvation for 24 hours, cells were challenged for 24 hours with nimesulide. Human umbilical vein endothelial cells (5 x 104) were then added to the cell culture inserts containing gelatin-coated polyethylene terephthalate track-etched membranes (8-µm pore size, 1 x 105 pores/cm2) and allowed to migrate for 6 hours. Migrated cells on the lower membrane surface were fixed in ice-cold methanol, stained using the Diff-Quick Stain Set (Dade Behring, Newark, DE) and counted.

Dominant-Negative PPAR-{gamma}
A dominant-negative form of PPAR-{gamma} was constructed by mutating the Leu468 and Glu471 of the full-length PPAR-{gamma} into Ala. Mutations at these sites create a dominant-negative form of PPAR-{gamma} (36). This dominant-negative PPAR-{gamma} was subcloned into recombinant type 5 adenovirus and designated as Adx-D/N-PPAR-{gamma}. Recombinant type 5 adenovirus expressing the LacZ gene was generated similarly and used as a control vector (Adx-LacZ). The human pancreatic cancer cell line MIA PaCa-2 was infected with 75 plaque forming units per cell in DMEM containing 0.4% fetal bovine serum.

PPAR-{gamma} Protein Expression
PPAR-{gamma} protein expression was determined as described previously (17). Briefly, total cell lysates were fractionated on 8% SDS-PAGE and transferred to nitrocellulose membrane. PPAR-{gamma} protein was detected using a rabbit polyclonal PPAR-{gamma} antibody (Cayman Chemical) and a horseradish peroxidase-conjugated anti-rabbit immunoglobulin as a secondary antibody (Amersham Pharmacia Biotech, Piscataway, NJ). Immunoreactive bands were visualized with the enhanced chemiluminescence system (Amersham Pharmacia Biotech).

PPAR-{gamma} Transcriptional Activity
NIH/3T3 cells were transfected with the luciferase vector pCMV-Luc and the reporter plasmid prACO.ßneo, which contains two functional PPREs of the acyl-CoA oxidase promoter (length: –1,273 to +20 of the promoter) and the PPAR-{gamma} expression plasmid pCMX-mPPARg. In some experiments, transfected cells were additionally infected with either Adx-D/N-PPAR-{gamma} or Adx-LacZ for 24 hours or preincubated with GW9662 for 3 hours before challenged with the indicated ligands and inhibitors for 4 hours. Cell lysates were prepared with the Reporter Lysis Buffer (Promega Co., Madison, WI). ß-Galactosidase activity was measured using the ß-Galactosidase Enzyme Assay System (Promega) and normalized to luciferase activity measured with the Luciferase Assay System (Promega).

PPAR-{gamma}Binding Activity
Electrophoretic mobility shift assays and supershift assays were done as described previously (37). Briefly, double-stranded oligonucleotides for consensus and mutant PPRE (Santa Cruz Biotechnology) were end-labeled with [{gamma}-32P] ATP using T4 polynucleotide kinase. Nuclear extracts were isolated from MIA PaCa-2 cells incubated with the indicated ligands or inhibitors for 4 hours and from s.c. MIA PaCa-2 tumors treated with the indicated compounds for 4 weeks using a Nuclear Extraction Kit (Panomics, Redwood City, CA). For competition and supershift assays, nuclear extracts were preincubated at room temperature for 45 minutes with cold consensus or mutant PPRE (x200 excess) or antibodies against PPAR-{gamma} (Santa Cruz Biotechnology) and RxR{alpha} (Santa Cruz Biotechnology) before addition of the labeled probe.

Pancreatic Cancer Animal Models
Animal studies were approved by the Chancellor's Animal Research Committee of the University of California, Los Angeles in accordance with the NIH Guide for the Care and Use of Laboratory Animals.

S.c. Model. Pancreatic cancer cells (2 x 106) were s.c. injected into the flanks of nude mice. After the tumor reached a size of ~2 x 2 mm animals were randomly allocated to receive either control vehicle (n = 8), low-dose nimesulide (150 µg/kg, n = 8), or high-dose nimesulide (1.5 mg/kg, n = 8). Drugs were prepared fresh each day and injected i.p. in a total volume of 50 µL for 2 weeks. In another series, animals were treated for 4 weeks with either control vehicle (n = 6) or high-dose nimesulide (1.5 mg/kg i.p.) in the absence (n = 6) or presence (n = 6) of the PPAR-{gamma} inhibitor GW9662 (0.3 mg/kg by daily peritumoral injection).

Orthotopic Model. Pancreatic tumors grown s.c. were harvested and tumor pieces (1 mm3) were transplanted into the tail of the pancreas of recipient nude mice (38). Animals were randomized and treated with either daily i.p. injections of nimesulide (1.5 mg/kg n = 8) or control vehicle (n = 8) for 4 weeks. At the end of the study period blood samples and tumors were harvested. Plasma was obtained by centrifugation. Tumor volume was assessed using the formula for a hemi-ellipsoid or ellipsoid for the s.c. or orthotopic tumor, respectively. For histologic evaluation, the tumor was either fixed in formalin and embedded in paraffin or frozen in 2-methyl-butane/dry ice and embedded in optimum cutting temperature.

Measurement of Nimesulide Levels in Animal Plasma
Nimesulide extraction was done using the single step organic extraction method with 1mL dichloromethane added to 100 µL of each plasma sample. The chromatographic separation and quantitation were achieved using a Hewlett-Packard high-performance liquid chromatography instrument (HPLC 1100, Palo Alto, CA). The mobile phase consisted of acetonitrile (100% v/v) flowing through a Phenomenex Kromasil (5 µm, 100 A, C18 x 250 x 2 mm, 5 µm beads) column at 0.2 mL/min. Nimesulide peaks were monitored using the UV detector at 230 nm wavelength against the 360 nm reference wavelength with 16 and 100 bandwidths, respectively. The obtained chromatograms were analyzed using Quantity Browser of the Xcalibur software package.

COX-2 Immunohistochemistry
Paraffin-embedded tissue samples were sectioned (4 µm), deparaffinized in xylene, and rehydrated in a graded ethanol series. Sections were subjected to heat-induced epitope retrieval using boiling citrate buffer (pH 6.0). Endogenous peroxidase activity was quenched by 0.3% H2O2 and unspecific binding sites were blocked with 5% bovine serum albumin. Immunostaining was done with a COX-2 antibody (5 µg/mL in blocking buffer) for 1 hour and a biotinylated secondary antibody (Vector Laboratories, Burlingame, CA) at 5 µg/mL in blocking buffer for 1 hour. Antibody binding sites were finally visualized by avidin-biotin peroxidase complex solution (Vectastain Elite ABC, Vector Laboratories) and 3,3'-diaminobenzidine (Sigma Chemical) as a chromogen. Sections were counterstained with Hematoxyllin QS (Vector Laboratories). As a control, nonspecific antibodies at a concentration corresponding to that of the COX-2 antibody were used to stain all sections using the identical protocol.

Microvessel Density
Cryostat sections (5 µm) were fixed in acetone and blocked with 2.5% bovine serum albumin for 1 hour. Immunostaining was done with a rat anti-mouse CD31 monoclonal antibody (Pharmingen) at 1:75 dilution or a rabbit anti-human Von Willebrand Factor antibody (DAKO Co., Carpintera, CA), which strongly cross-reacts with the murine isoform, at 1:200 for 2 hours and a biotinylated secondary goat anti-rat or goat anti-rabbit antibody, respectively, (Pierce Biotechnology, Inc., Rockford, IL) at 1:200 dilution for 30 minutes. Slides were then immersed in 0.3% H2O2 and antibody binding sites were visualized with the Vectastain Elite ABC (Vector Laboratories) and 3,3'-diaminobenzidine (Sigma Chemical) as a chromogen. Sections were counterstained with 1% methyl green (Sigma Chemical). Rat IgG1 was used as isotype-matched negative control. Images were obtained randomly on a Nikon Diaphot 300 microscope equipped with a Toshiba 3CCD camera. The number of vessels per area was determined by using ImagePro 5.0 software (Media Cybernetics, Silver Spring, MD).

Intratumoral PGE2 Levels
Intratumoral PGE2 levels were analyzed on a Finnegan LCQ Deca (Hiauea, FL) ion trap mass spectrometer connected to a Hewlett-Packard (HP1100) high-pressure liquid chromatography instrument as described elsewhere (39). Flow rate was 0.3 mL/min using a water-acetonitrile gradient, which started at 35% acetonitrile, linearly increased to 100% in 35 minutes, and held for 10 minutes. Parent ion and fragments were monitored in the normal mass range of 95.00 to 360.00.

Intratumoral VEGF Levels
Tumor extracts were prepared by homogenizing and sonicating non-necrotic areas of pancreatic tumors in anti-protease lysis buffer (Roche Applied Science, Indianapolis, IN). Tumor homogenates were cleared by centrifugation and filtration. Mouse VEGF levels were determined using the mouse VEGF DuoSet ELISA Development System (R&D Systems). Briefly, a 96-well microplate was precoated with goat anti-mouse VEGF (0.4 µg/mL) overnight and mouse VEGF levels in the tumor homogenates were determined using a biotinylated goat anti-mouse VEGF (100 ng/mL) and Streptavidin-horseradish peroxidase. Human VEGF levels were measured using the Quantikine Human VEGF Immunoassay (R&D Systems) as recommended by the manufacturer. Mouse and human VEGF levels were normalized to total protein content.

Statistical Analysis
Data are presented as mean ±SD. Statistical comparisons were made using the Student's t test for paired observations or by one-way ANOVA with post hoc Bonferroni's correction for multiple comparisons. P < 0.05 was considered statistically significant.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Concentration-Dependent Effects of Selective COX-2 Inhibitors on VEGF Production in Pancreatic Cancer Cells. In previous experiments, we have found that PGE2 stimulates VEGF production in COX-2–positive pancreatic cancer cells in an autocrine fashion by binding to the EP2 receptor (35). Based on these findings, we hypothesized that blocking basal PGE2 production by selective COX-2 inhibitors decreases VEGF levels. Incubation of the COX-2–positive pancreatic cancer cell line BxPC-3 with the structurally unrelated selective COX-2 inhibitors nimesulide and DuP697 for 24 hours almost completely inhibited PGE2 production (Fig. 1A and B, insets). PGE2 levels in COX-2–negative MIA PaCa-2 cells were barely detectable and did not change upon treatment with selective COX-2 inhibitors (data not shown). The COX-2 inhibitor–induced decrease in PGE2 production in COX-2–positive pancreatic cancer cells correlated with a reduction of VEGF levels in the culture medium (Fig. 1A and B). However, this effect was seen only with lower concentrations of the COX-2 inhibitors (≤10 µmol/L). At higher concentrations, starting at around 30 µmol/L (data not shown) and being maximal at 100 µmol/L, both inhibitors markedly increased VEGF levels in both COX-2–positive and COX-2–negative pancreatic cancer cells (Fig. 1A and B), suggesting a COX-2–independent effect.



View larger version (16K):
[in this window]
[in a new window]
  		Figure 1. Concentration-dependent effects of selective COX-2 inhibitors on VEGF production and angiogenesis in vitro. COX-2–positive BxPC-3 and COX-2–negative MIA PaCa-2 cells were treated with the selective COX-2 inhibitors nimesulide (A) and DuP697 (B) before assaying the cell culture supernatant for VEGF. The COX-2 inhibitory effect of nimesulide and DuP697 was confirmed by measuring PGE2 levels in the supernatant of BxPC-3 cells (insets). The effect of nimesulide on endothelial cell migration was determined by a double-chamber assay (C) and expressed as a fold-increase over control vehicle. VEGF and PGE2 measurements were done on duplicate samples of three independent experiments. The endothelial cell migration assay was repeated twice. Columns, mean value; bars, ±SD. *, P < 0.05 versus 0 µmol/L.

 
To delineate whether the COX-2 inhibitor–induced increase in VEGF bears biological relevance, we used an endothelial cell migration assay. Incubation of COX-2–positive BxPC-3 but not COX-2–negative MIA PaCa-2 cells with nimesulide at concentrations ≤10 µmol/L slightly reduced endothelial cell migration, whereas treatment of both COX-2–positive and COX-2–negative pancreatic cancer cells with nimesulide at 100 µmol/L significantly stimulated endothelial cell migration (Fig. 1C).

The Role of PPAR-{gamma}in Mediating the Effect of High-Dose Nimesulide on VEGF Production. To investigate whether activation of the nuclear receptor PPAR-{gamma} affects VEGF levels in pancreatic cancer cells, we used the endogenous and exogenous PPAR-{gamma} ligands 15-PGJ2 and troglitazone, respectively. Both ligands dose-dependently stimulated VEGF production in COX-2–positive and COX-2–negative pancreatic cancer cells (Fig. 2A and B). To evaluate whether the effect of high-dose nimesulide on VEGF levels was mediated by PPAR-{gamma}, we constructed a dominant-negative PPAR-{gamma} mutant. This PPAR-{gamma} mutant retains ligand and DNA binding but exhibits markedly reduced transactivation due to impaired coactivator recruitment (36). MIA PaCa-2 cells were infected with adenoviruses encoding for the mutant PPAR-{gamma}. Successful infection was confirmed by Western blotting, which showed marked overexpression of PPAR-{gamma} in Adx-D/N-PPAR-{gamma} infected cells (Fig. 3A). The compound mutant almost completely abrogated the effect of nimesulide and 15-PGJ2 on VEGF levels (Fig. 3B). We confirmed these findings using the irreversible PPAR-{gamma} inhibitor GW9662, which dose-dependently attenuated the effect of nimesulide and 15-PGJ2 on VEGF production in MIA PaCa-2 cells (Fig. 3C and D). To determine whether high-dose nimesulide enhances PPAR-{gamma} binding activity, we did gel shift assays using consensus and mutant PPREs. Treatment of MIA PaCa-2 cells with nimesulide (100 µmol/L), DuP697 (100 µmol/L), or troglitazone (10 µmol/L) for 4 hours stimulated binding of nuclear complexes to the consensus PPRE (Fig. 4A). Supershift assays confirmed the presence of PPAR-{gamma} and RxR{alpha} in the bound complexes. PPAR-{gamma} binding activity was inhibited by cocompetition using unlabeled wild-type oligonucleotides but was unaffected by unlabeled mutant PPREs. No PPAR-{gamma} binding to radiolabeled mutant PPRE was detectable (Fig. 4A).



View larger version (16K):
[in this window]
[in a new window]
  		Figure 2. PPAR-{gamma} ligands dose-dependently increase VEGF levels in vitro. COX-2–positive BxPC-3 and COX-2–negative MIA PaCa-2 cells were treated with the PPAR-{gamma} ligands (A) 15-deoxy-{Delta}-12,14-prostaglandin J2 (15-PGJ2) and (B) troglitazone and VEGF levels in the culture medium were determined by ELISA. Experiments were carried out at least in triplicate using separate cultures. Columns, mean value; bars, ±SD. Asterisks between two bars, significance for both cell lines. *, P < 0.05 versus 0 µmol/L.

 


View larger version (25K):
[in this window]
[in a new window]
  		Figure 3. High-dose nimesulide increases VEGF levels through the nuclear receptor PPAR-{gamma} in vitro. COX-2–negative MIA PaCa-2 cells were infected with adenoviruses encoding for a dominant-negative PPAR-{gamma} receptor (Adx-D/N-PPAR{gamma}) or LacZ (Adx-LacZ) as a control. Successful infection was confirmed by Western blot. Reprobing for ß-actin served as loading control (A). MIA PaCa-2 cells were infected with Adx-D/N-PPAR{gamma} or Adx-LacZ before challenge with nimesulide or 15-PGJ2 for 24 hours. VEGF levels in the culture medium were determined by ELISA (B). *, P < 0.05 versus 0 µmol/L; #, P < 0.05 versus Adx-LacZ. MIA PaCa-2 cells were preincubated for 3 hours with incremental concentrations of GW9662 (GW) followed by nimesulide (N) or 15-PGJ2 for 24 hours and VEGF levels in the culture medium were measured by ELISA (C and D). *, P < 0.05 versus GW (0 µmol/L) + N (100 µmol/L) or versus GW (0 µmol/L) + 15-PGJ2 (10 µmol/L). All experiments were carried out at least in triplicate using separate cultures. Columns, mean value; bars, ±SD.

 


View larger version (18K):
[in this window]
[in a new window]
  		Figure 4. Selective COX-2 inhibitors stimulate PPAR-{gamma} binding and transcriptional activity in vitro. COX-2–negative MIA PaCa-2 cells were treated with troglitazone (10 µmol/L), nimesulide (100 µmol/L), or DuP697 (100 µmol/L) for 4 hours and PPAR-{gamma} binding activity to its consensus response element (PPRE) was determined by gel shift analysis. Specificity was determined by cold competition with consensus and mutant PPRE and by mutant labeled probe. Presence of PPAR-{gamma} and RxR{alpha} in the bound nuclear complexes was confirmed by supershift analysis (A). NIH/3T3 cells were transiently transfected with pCMX-mPPARg (PPAR-{gamma}), prACO.ßneo (PPRE), and pCMV-Luc. Effect of nimesulide (Nime) and 15-PGJ2 on PPAR-{gamma} transcriptional activity was determined by measuring ß-galactosidase activity and standardizing to luciferase (B). *, P < 0.05 versus Nime (0 µmol/L) or 15-PGJ2 (0 µmol/L). NIH/3T3 cells transfected with pCMX-mPPARg, prACO.ßneo, and pCMV-Luc were infected with Adx-D/N-PPAR{gamma} and Adx-LacZ or pretreated with GW9662. Effect of nimesulide (Nime) and 15-PGJ2 on PPAR-{gamma} transcriptional activity was determined by measuring ß-galactosidase activity and standardizing to luciferase (C). *, P < 0.05 versus Nime (100 µmol/L) alone; #, P < 0.05 versus 15-PGJ2 (10 µmol/L) alone. All experiments were carried out at least in triplicate using separate cultures and transfections. Columns, mean value; bars, ±SD.

 
Having shown PPAR-{gamma} binding activity, we then sought to investigate whether nimesulide activates PPAR-{gamma} transcriptional activity using reporter assays. NIH/3T3 cells, which lack wild-type PPAR-{gamma} (40), were transiently transfected with PPAR-{gamma} and a PPRE-containing reporter construct. 15-PGJ2 dose-dependently and nimesulide at 100 µmol/L stimulated PPAR-{gamma} transcriptional activity (Fig. 4B), which was inhibited by the dominant-negative PPAR-{gamma} and by preincubation with GW9662 (Fig. 4C). Lower concentrations of nimesulide (≤10 µmol/L) had no effect on PPAR-{gamma} transcriptional activity.

Opposite Effects of the Selective COX-2 Inhibitor Nimesulide on Pancreatic Cancer Growth and Angiogenesis In vivo. Based on these in vitro findings, we employed a s.c. and an orthotopic pancreatic cancer model in nude mice (38) to evaluate whether selective COX-2 inhibitors could also stimulate VEGF production by cancer cells in vivo and therefore promote angiogenesis and tumor growth. Animals were treated with daily injections of a low (150 µg/kg) or high (1.5 mg/kg) dose of nimesulide for 2 weeks (s.c. model) or 4 weeks (orthotopic model). Treatment of animals with nimesulide at 1.5 mg/kg for 4 weeks resulted in plasma concentrations of nimesulide of 32 ± 8 µmol/L as determined by high-performance liquid chromatography.

Immunohistochemistry with an antibody recognizing human and murine COX-2 revealed the presence of the COX-2 protein only in BxPC-3 tumors (Fig. 5A). Importantly, COX-2 was detected exclusively in cancer cells. The low dose of nimesulide had no effect on the growth of COX-2–positive and COX-2–negative tumors (data not shown). In contrast, the high dose of nimesulide decreased the growth of COX-2–positive pancreatic tumors by ~50%, but stimulated the growth of COX-2–negative pancreatic cancers >2-fold in both models (Fig. 5B, C, and D). To determine whether the growth-promoting effect of high-dose nimesulide on MIA PaCa-2 tumors in vivo was mediated by activating PPAR-{gamma}, we used the PPAR-{gamma} inhibitor GW9662 at a concentration which has been successfully used in recent studies (41). Peritumoral injection of GW9662 (0.3 mg/kg) significantly attenuated the effect of high-dose nimesulide on the growth of s.c. MIA PaCa-2 tumors (Fig. 5E). Gel shift analysis of tumor nuclear extracts showed enhanced DNA binding of PPAR-{gamma} in high-dose nimesulide-treated MIA PaCa-2 tumors (Fig. 5F). This effect was partially inhibited by GW9662, indicating that nimesulide at this concentration stimulates PPAR-{gamma} binding to its response elements in vivo.



View larger version (44K):
[in this window]
[in a new window]
  		Figure 5. Nimesulide decreases growth of COX-2–positive pancreatic cancers but stimulates growth of COX-2–negative pancreatic cancers in vivo through PPAR-{gamma}. COX-2 expression in BxPC-3 and MIA PaCa-2 tumors was determined by immunohistochemistry (A). Nimesulide shows opposite effects on the growth of BxPC-3 and MIA PaCa-2 tumors in the s.c. (B) or orthotopic model (C). *, P < 0.05 versus control. Representative images from s.c. COX-2–negative MIA PaCa-2 tumors treated with nimesulide or control vehicle. Marker, 10 mm (D). S.c. MIA PaCa-2 tumors were treated with nimesulide (nime) in the absence or presence of GW9662 (GW) for 4 weeks (E). *, P < 0.05 versus nimesulide. PPAR-{gamma} binding activity to a consensus PPRE in tumor nuclear extracts was determined by gel shift analysis. Specificity was confirmed by cold competition with consensus and mutant PPRE and by mutant labeled probe. Presence of PPAR-{gamma} and RxR{alpha} in the bound nuclear complexes was determined by supershift analysis. Lower arrow, specific bands; upper arrow, supershift bands (F). Columns, mean value; bars, ±SD.

 
To further characterize the in vivo effects of nimesulide, we measured intratumoral PGE2 levels. Liquid chromatography/mass spectroscopy analysis revealed higher levels of PGE2 in the BxPC-3 tumors than in the MIA PaCa-2 tumors (Fig. 6A). Nimesulide (1.5 mg/kg) reduced PGE2 levels in BxPC-3 tumors, whereas it had no effect in MIA PaCa-2 tumors (Fig. 6A), demonstrating that tissue concentrations of nimesulide sufficient for inhibiting COX-2 activity were achieved.



View larger version (31K):
[in this window]
[in a new window]
  		Figure 6. Effect of nimesulide on intratumoral PGE2 and VEGF levels and on tumor angiogenesis. Intratumoral PGE2 levels were measured by liquid chromatography/mass spectroscopy and standardized to PGE2 levels in control-vehicle treated BxPC-3 tumors (A). *, P < 0.05 versus control. Microvessel density in vehicle-treated and nimesulide (nime)–treated BxPC-3 and MIA PaCa-2 tumors was assessed by immunohistochemical staining of CD31 (B) and quantified using ImagePro 5.0 software (C). *, P < 0.05 versus control. Intratumoral human and mouse VEGF levels were measured using species-specific ELISA kits in BxPC-3 (D) and MIA PaCa-2 tumors (E) and standardized to total tumor protein content. The effect of nimesulide on overall intratumoral VEGF levels was determined by adding murine and human VEGF together (D and E; insets). *, P < 0.05 versus control. Columns, mean value; bars, ±SD. For VEGF and PGE2 quantification, each tumor sample was measured in triplicate.

 
Macroscopic evaluation of the tumors revealed a greater number and a more complex network of blood vessels on the surface of nimesulide-treated as compared with untreated COX-2–negative tumors (Fig. 5D). Microscopic analysis confirmed that nimesulide (1.5 mg/kg) increased the number of microvessels in COX-2–negative pancreatic tumors as determined by CD31 immunostaining. In contrast, the COX-2 inhibitor reduced microvessel density in COX-2–positive tumors (Fig. 6B and C). These findings were confirmed by staining the tumors for the Von Willebrand Factor, which showed similar changes in microvessel density (data not shown). Because our in vitro findings showed an effect of the COX-2 inhibitor on VEGF levels, we sought to evaluate whether the differences in microvessel density correlate to changes in intratumoral VEGF content. We separately measured human (generated by pancreatic cancer cells) and mouse (secreted by nonmalignant host cells) VEGF levels in tumor homogenates. Mouse VEGF levels were significantly lower in the COX-2–negative compared with COX-2–positive tumors. Nimesulide (1.5 mg/kg) decreased mouse VEGF levels only in the COX-2–positive pancreatic cancers. In contrast, nimesulide increased human VEGF in both COX-2–negative and COX-2–positive pancreatic tumors (Fig. 6D and E). Combining mouse and human VEGF levels, nimesulide (1.5 mg/kg) decreased total VEGF in COX-2–positive pancreatic cancers but increased total VEGF in COX-2–negative pancreatic cancers (Fig. 6D and E, insets).


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
In the present study, we found that two structurally unrelated selective COX-2 inhibitors had concentration-dependent effects on VEGF levels in the culture medium of pancreatic cancer cells. At lower concentrations (≤10 µmol/L), both inhibitors decreased VEGF levels only in COX-2–positive pancreatic cancer cells, an effect which is most likely to be caused by inhibiting the autocrine effects of COX-2–derived prostaglandins. This is in accordance to our previous experiments (35), in which PGE2 produced by COX-2–positive pancreatic cancer cells stimulated VEGF production in an autocrine fashion. These data are also consistent to Chu et al. (42), showing antiangiogenic properties of a selective COX-2 inhibitor associated with a reduced VEGF production. However, at higher concentrations, starting at around 30 µmol/L and being maximal at 100 µmol/L, both inhibitors increased VEGF levels in COX-2–positive and COX-2–negative pancreatic cancer cells, indicating possible COX-independent effects. At higher inhibitor concentrations, VEGF levels decreased again, presumably caused by the proapoptotic properties of the COX-2 inhibitors at these concentrations (7). In the present study, using COX-2 inhibitors at concentrations of maximal 100 µmol/L for ≤24 hours, no induction of apoptosis was detectable (data not shown).

In recent years, an increasing number of studies have reported COX-independent effects of selective COX-2 inhibitors (7–14). However, as these findings are mostly single reports in specific cellular systems, a well-defined general COX-independent pathway activated and/or inhibited by selective COX-2 inhibitors has not emerged yet. Earlier studies have shown that nonselective COX inhibitors can activate the nuclear receptor PPAR-{gamma} (15). Intriguingly, the transcriptional activation was seen only at higher concentration of the inhibitors. Based on these findings and our own studies showing that pancreatic cancer cells express PPAR-{gamma} (17), we sought to determine whether the effect of selective COX-2 inhibitors (at higher concentrations) on VEGF levels in pancreatic cancer cells was indeed mediated by PPAR-{gamma}. The exact role of PPAR-{gamma} in tumor angiogenesis and growth is still ambiguous. Whereas some studies attribute PPAR-{gamma} ligands antiangiogenic and antiproliferative properties, others showed proangiogenic and growth-stimulating effects of PPAR-{gamma} activation. The overall effect of PPAR-{gamma} activation may depend on the pharmacologic properties of the specific ligand, the cellular equipment of coactivators and corepressors, the timely pattern of activation, and the presence of other cellular pathways interacting with PPAR-{gamma} (24, 43–45) .

In our study, two different PPAR-{gamma} ligands dose-dependently increased VEGF levels in pancreatic cancer cells, supporting the proangiogenic and protumorigenic role of PPAR-{gamma} activation. This observation seems to be contrary to our previous study, in which PPAR-{gamma} ligands inhibited proliferation of pancreatic cancer cells by induction of apoptosis (17), advocating the antitumorigenic properties of PPAR-{gamma} activation. However, the proapoptotic effects in vitro were clearly seen only at relatively high concentrations of the PPAR-{gamma} ligands (≥10 µmol/L). In the present study, concentrations of PPAR-{gamma} ligands (≤1 µmol/L) sufficient for increasing VEGF levels did not induce apoptosis (data not shown). Importantly, an increase in VEGF levels, neglecting autocrine pathways, confers protumorigenic properties dominantly in vivo and may not affect proliferation in vitro. Using a dominant-negative PPAR-{gamma} mutant and a pharmacologic PPAR-{gamma} inhibitor, we were able to show that the effect of high-dose COX-2 inhibitor on VEGF production was mediated by PPAR-{gamma}. Gel shift and reporter assays showed that the selective COX-2 inhibitor stimulated PPAR-{gamma} binding and transcriptional activity, respectively. However, only high concentrations of the selective COX-2 inhibitor increased PPAR-{gamma} transcriptional activity, an effect that agrees with earlier studies (15). The concentrations of the COX-2 inhibitor needed to stimulate PPAR-{gamma} transcriptional activity correlated thereby with the concentrations needed to increase VEGF levels, supporting the causal relationship. For the reporter studies, we used murine NIH/3T3 fibroblasts, which lack wild-type PPAR-{gamma} (40). This allowed us to clearly delineate the contribution of PPAR-{gamma} to the observed effect of the COX-2 inhibitor. However, the exact mechanism by which the COX-2 inhibitor activates PPAR-{gamma} still remains unknown. Nimesulide may activate PPAR-{gamma} indirectly through accumulation of the COX-2 substrate arachidonic acid (46). However, the finding that the dominant-negative receptor and GW9662 abrogate the effect of nimesulide also in COX-2–negative cells makes this an unlikely explanation.

Our in vitro findings raised the hypothesis, that a selective COX-2 inhibitor may also stimulate VEGF levels and as a result tumor growth of pancreatic cancer in vivo. Nimesulide at plasma concentration, which can also be achieved in humans (47), reduced angiogenesis and growth of the COX-2–positive pancreatic tumor but stimulated microvessel density and growth of the COX-2–negative pancreatic tumor. To our knowledge, this is the first report of a stimulatory effect of a selective COX-2 inhibitor on angiogenesis and growth of any human cancer. The nimesulide-induced decrease in microvessel density in COX-2–positive pancreatic cancer was accompanied by a reduction in total intratumoral VEGF. Exploiting the characteristics of our xenograft model, in which human cancer cells grow in a mouse host, further analysis showed that treatment with nimesulide increased cancer cell generated human VEGF in both COX-2–positive and COX-2–negative tumors: an effect which correlates to our in vitro finding. In COX-2–positive pancreatic cancer this effect was offset by a marked reduction in host cell generated murine VEGF, leading to a decrease in total VEGF and a reduction in the number of microvessels. Because nimesulide had no effect on murine VEGF in COX-2–negative pancreatic cancer, the increase of human VEGF dominated, which led to stimulated angiogenesis.

Our study strongly implies that PGE2 generated by COX-2 in pancreatic cancer cells is a driving force for angiogenesis in pancreatic carcinomas. This conclusion is supported by the following observations: (1) Using an antibody which recognizes murine and human COX-2 isoforms, immunohistochemical analysis detected COX-2 exclusively in pancreatic cancer cells; a notion which has also been described by others (30, 33). (2) PGE2 levels were significantly higher in COX-2–positive pancreatic tumors compared with COX-2–negative pancreatic tumors. (3) Expression of COX-2 and high intratumoral PGE2 levels positively correlated to the number of tumor microvessels. (4) Treatment with a selective COX-2 inhibitor decreased PGE2 levels only in COX-2–positive pancreatic cancer, which was accompanied by a reduction in angiogenesis. The importance of the COX-2/PGE2 system in tumor angiogenesis has also been recognized by others. PGE2 is capable of stimulating VEGF production in various cell types, which are all found in the tumor stroma, including fibroblasts, endothelial cells, and inflammatory cells (48–50). This highlights the potentially critical role of PGE2 as a paracrine mediator of VEGF stimulation and hence angiogenesis. In addition, in a sarcoma model Amano et al. (51) found that angiogenesis was attenuated in prostaglandin receptor knockout animals. Although we did not study the pathways by which PGE2 stimulates VEGF production, recent evidence suggests that hypoxia inducible factor-1 might be involved (52).

Although our study implies that the growth-inhibitory effect of nimesulide in COX-2–positive pancreatic cancer was primarily caused by a reduction of tumor angiogenesis, a direct antiproliferative effect on the cancer cells seems also plausible. However, despite the evidence in vitro showing induction of apoptosis in cancer cells treated with COX-2 inhibitors, we strongly favor the antiangiogenic explanation. Our preference is mainly based on the histologic examination of the tumors. Nimesulide-treated COX-2–positive pancreatic tumors had a prominent central necrotic area with a reduction in microvessels predominantly in the tumor periphery. This pattern indicates that the reduction in blood supply is the primary cause of cancer cell death. Furthermore, staining with the proliferation marker Ki-67 did not detect changes in pancreatic cancer cell proliferation in non-necrotic tumor regions (data not shown).

In conclusion, our studies provide the first evidence of growth-stimulating properties of selective COX-2 inhibitors in a well-defined subset of COX-2–negative pancreatic cancers. In COX-2–positive pancreatic carcinomas, the selective COX-2 inhibitor reduced angiogenesis and growth. Molecular profiling of the pancreatic tumor seems to be warranted, especially in the light of several ongoing clinical trials using selective COX-2 inhibitors as a therapeutic agent in pancreatic cancer patients.


   Acknowledgments
 
Grant support: NIH grant R01 CA104027 (G. Eibl) and Hirshberg Foundation for Pancreatic Cancer Research (H.A. Reber).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received 6/18/04. Revised 10/ 8/04. Accepted 11/18/04.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Kurumbail RG, Stevens AM, Gierse JK, et al. Structural basis for selective inhibition of cyclooxygenase-2 by anti-inflammatory agents. Nature 1996;384:644–8.[CrossRef][Medline]
  2. Picot D, Loll PJ, Garavito RM. The X-ray crystal structure of the membrane protein prostaglandin H2 synthase-1. Nature 1994;367:243–9.[CrossRef][Medline]
  3. Crosby CG, DuBois RN. The cyclooxygenase-2 pathway as a target for treatment or prevention of cancer. Expert Opin Emerg Drugs 2003;8:1–7.[CrossRef][Medline]
  4. Dannenberg AJ, Altorki NK, Boyle JO, et al. Cyclo-oxygenase 2: a pharmacological target for the prevention of cancer. Lancet Oncol 2001;2:544–51.[CrossRef][Medline]
  5. Subbaramaiah K, Dannenberg AJ. Cyclooxygenase 2: a molecular target for cancer prevention and treatment. Trends Pharmacol Sci 2003;24:96–102.[CrossRef][Medline]
  6. Trifan OC, Hla T. Cyclooxygenase-2 modulates cellular growth and promotes tumorigenesis. J Cell Mol Med 2003;7:207–22.[Medline]
  7. Eibl G, Reber HA, Wente MN, Hines OJ. The selective cyclooxygenase-2 inhibitor nimesulide induces apoptosis in pancreatic cancer cells independent of COX-2. Pancreas 2003;26:33–41.[CrossRef][Medline]
  8. Elder DJ, Halton DE, Hague A, Paraskeva C. Induction of apoptotic cell death in human colorectal carcinoma cell lines by a cyclooxygenase-2 (COX-2)-selective nonsteroidal anti-inflammatory drug: independence from COX-2 protein expression. Clin Cancer Res 1997;3:1679–83.[Abstract]
  9. Grosch S, Tegeder I, Niederberger E, Brautigam L, Geisslinger G. COX-2 independent induction of cell cycle arrest and apoptosis in colon cancer cells by the selective COX-2 inhibitor celecoxib. FASEB J 2001;15:2742–4.[Free Full Text]
  10. Han C, Leng J, Demetris AJ, Wu T. Cyclooxygenase-2 promotes human cholangiocarcinoma growth: evidence for cyclooxygenase-2-independent mechanism in celecoxib-mediated induction of p21waf1/cip1 and p27kip1 and cell cycle arrest. Cancer Res 2004;64:1369–76.[Abstract/Free Full Text]
  11. Kulp SK, Yang YT, Hung CC, et al. S. 3-phosphoinositide-dependent protein kinase-1/Akt signaling represents a major cyclooxygenase-2-independent target for celecoxib in prostate cancer cells. Cancer Res 2004;64:1444–51.[Abstract/Free Full Text]
  12. Liu G, Ma WY, Bode AM, Zhang Y, Dong Z. NS-398 and piroxicam suppress UVB-induced activator protein 1 activity by mechanisms independent of cyclooxygenase-2. J Biol Chem 2003;278:2124–30.[Abstract/Free Full Text]
  13. Minter HA, Eveson JW, Huntley S, Elder DJ, Hague A. The cyclooxygenase 2-selective inhibitor NS398 inhibits proliferation of oral carcinoma cell lines by mechanisms dependent and independent of reduced prostaglandin E2 synthesis. Clin Cancer Res 2003;9:1885–97.[Abstract/Free Full Text]
  14. Wick M, Hurteau G, Dessev C, et al. Peroxisome proliferator-activated receptor-{gamma} is a target of nonsteroidal anti-inflammatory drugs mediating cyclooxygenase-independent inhibition of lung cancer cell growth. Mol Pharmacol 2002;62:1207–14.[Abstract/Free Full Text]
  15. Lehmann JM, Lenhard JM, Oliver BB, Ringold GM, Kliewer SA. Peroxisome proliferator-activated receptors {alpha} and {gamma} are activated by indomethacin and other non-steroidal anti-inflammatory drugs. J Biol Chem 1997;272:3406–10.[Abstract/Free Full Text]
  16. Issemann I, Green S. Activation of a member of the steroid hormone receptor superfamily by peroxisome proliferators. Nature 1990;347:645–50.[CrossRef][Medline]
  17. Eibl G, Wente MN, Reber HA, Hines OJ. Peroxisome proliferator-activated receptor {gamma} induces pancreatic cancer cell apoptosis. Biochem Biophys Res Commun 2001;287:522–9.[CrossRef][Medline]
  18. Elstner E, Muller C, Koshizuka K, et al. Ligands for peroxisome proliferator-activated receptorgamma and retinoic acid receptor inhibit growth and induce apoptosis of human breast cancer cells in vitro and in BNX mice. Proc Natl Acad Sci U S A 1998;95:8806–11.[Abstract/Free Full Text]
  19. Kitamura S, Miyazaki Y, Hiraoka S, et al. PPAR{gamma} inhibits the expression of c-MET in human gastric cancer cells through the suppression of Ets. Biochem Biophys Res Commun 1999;265:453–6.[CrossRef][Medline]
  20. Mueller E, Smith M, Sarraf P, et al. Effects of ligand activation of peroxisome proliferator-activated receptor {gamma} in human prostate cancer. Proc Natl Acad Sci U S A 2000;97:10990–5.[Abstract/Free Full Text]
  21. Tsubouchi Y, Sano H, Kawahito Y, et al. Inhibition of human lung cancer cell growth by the peroxisome proliferator-activated receptor-{gamma} agonists through induction of apoptosis. Biochem Biophys Res Commun 2000;270:400–5.[CrossRef][Medline]
  22. Lefebvre AM, Chen I, Desreumaux P, et al. Activation of the peroxisome proliferator-activated receptor {gamma} promotes the development of colon tumors in C57BL/6J-APCMin/+ mice. Nat Med 1998;4:1053–7.[CrossRef][Medline]
  23. Saez E, Tontonoz P, Nelson MC, et al. Activators of the nuclear receptor PPAR{gamma} enhance colon polyp formation. Nat Med 1998;4:1058–61.[CrossRef][Medline]
  24. Saez E, Rosenfeld J, Livolsi A, et al. PPAR {gamma} signaling exacerbates mammary gland tumor development. Genes Dev 2004;18:528–40.[Abstract/Free Full Text]
  25. Panigrahy D, Singer S, Shen LQ, et al. PPAR{gamma} ligands inhibit primary tumor growth and metastasis by inhibiting angiogenesis. J Clin Invest 2002;110:923–32.[CrossRef][Medline]
  26. Yamakawa K, Hosoi M, Koyama H, et al. Peroxisome proliferator-activated receptor-{gamma} agonists increase vascular endothelial growth factor expression in human vascular smooth muscle cells. Biochem Biophys Res Commun 2000;271:571–4.[CrossRef][Medline]
  27. Kokawa A, Kondo H, Gotoda T, et al. Increased expression of cyclooxygenase-2 in human pancreatic neoplasms and potential for chemoprevention by cyclooxygenase inhibitors. Cancer 2001;91:333–8.[CrossRef][Medline]
  28. Koshiba T, Hosotani R, Miyamoto Y, et al. Immunohistochemical analysis of cyclooxygenase-2 expression in pancreatic tumors. Int J Pancreatol 1999;26:69–76.[CrossRef][Medline]
  29. Merati K, Said SM, Andea A, et al. Expression of inflammatory modulator COX-2 in pancreatic ductal adenocarcinoma and its relationship to pathologic and clinical parameters. Am J Clin Oncol 2001;24:447–52.[CrossRef][Medline]
  30. Molina MA, Sitja-Arnau M, Lemoine MG, Frazier ML, Sinicrope FA. Increased cyclooxygenase-2 expression in human pancreatic carcinomas and cell lines: growth inhibition by nonsteroidal anti-inflammatory drugs. Cancer Res 1999;59:4356–62.[Abstract/Free Full Text]
  31. Niijima M, Yamaguchi T, Ishihara T, et al. Immunohistochemical analysis and in situ hybridization of cyclooxygenase-2 expression in intraductal papillary-mucinous tumors of the pancreas. Cancer 2002;94:1565–73.[CrossRef][Medline]
  32. Tucker ON, Dannenberg AJ, Yang EK, et al. Cyclooxygenase-2 expression is up-regulated in human pancreatic cancer. Cancer Res 1999;59:987–90.[Abstract/Free Full Text]
  33. Yip-Schneider MT, Barnard DS, Billings SD, et al. Cyclooxygenase-2 expression in human pancreatic adenocarcinomas. Carcinogenesis 2000;21:139–46.[Abstract/Free Full Text]
  34. Elbrecht A, Chen Y, Cullinan CA, et al. Molecular cloning, expression and characterization of human peroxisome proliferator activated receptors {gamma} 1 and {gamma} 2. Biochem Biophys Res Commun 1996;224:431–7.[CrossRef][Medline]
  35. Eibl G, Bruemmer D, Okada Y, et al. PGE(2) is generated by specific COX-2 activity and increases VEGF production in COX-2-expressing human pancreatic cancer cells. Biochem Biophys Res Commun 2003;306:887–97.[CrossRef][Medline]
  36. Gurnell M, Wentworth JM, Agostini M, et al. A dominant-negative peroxisome proliferator-activated receptor {gamma} (PPAR{gamma}) mutant is a constitutive repressor and inhibits PPAR{gamma}-mediated adipogenesis. J Biol Chem 2000;275:5754–9.[Abstract/Free Full Text]
  37. Takata Y, Kitami Y, Okura T, Hiwada K. Peroxisome proliferator-activated receptor-{gamma} activation inhibits interleukin-1ß -mediated platelet-derived growth factor-{alpha} receptor gene expression via CCAAT/enhancer-binding protein-{delta} in vascular smooth muscle cells. J Biol Chem 2001;276:12893–7.[Abstract/Free Full Text]
  38. Hotz HG, Reber HA, Hotz B, et al. An orthotopic nude mouse model for evaluating pathophysiology and therapy of pancreatic cancer. Pancreas 2003;26:e89–98.[CrossRef][Medline]
  39. Nithipatikom K, Laabs ND, Isbell MA, Campbell WB. Liquid chromatographic-mass spectrometric determination of cyclooxygenase metabolites of arachidonic acid in cultured cells. J Chromatogr B Analyt Technol Biomed Life Sci 2003;785:135–45.[Medline]
  40. Tontonoz P, Singer S, Forman BM, et al. Terminal differentiation of human liposarcoma cells induced by ligands for peroxisome proliferator-activated receptor {gamma} and the retinoid X receptor. Proc Natl Acad Sci U S A 1997;94:237–41.[Abstract/Free Full Text]
  41. Collin M, Patel NS, Dugo L, Thiemermann C. Role of peroxisome proliferator-activated receptor-{gamma} in the protection afforded by 15-deoxy{delta}12,14 prostaglandin J2 against the multiple organ failure caused by endotoxin. Crit Care Med 2004;32:826–31.[CrossRef][Medline]
  42. Chu J, Lloyd FL, Trifan OC, Knapp B, Rizzo MT. Potential involvement of the cyclooxygenase-2 pathway in the regulation of tumor-associated angiogenesis and growth in pancreatic cancer. Mol Cancer Ther 2003;2:1–7.[Abstract/Free Full Text]
  43. Clay CE, Namen AM, Fonteh AN, Atsumi G, High KP, Chilton FH. 15-deoxy-{Delta}(12,14)PGJ(2) induces diverse biological responses via PPAR{gamma} activation in cancer cells. Prostaglandins Other Lipid Mediat 2000;62:23–32.[CrossRef][Medline]
  44. Clay CE, Namen AM, Atsumi G, et al. Magnitude of peroxisome proliferator-activated receptor-{gamma} activation is associated with important and seemingly opposite biological responses in breast cancer cells. J Investig Med 2001;49:413–20.[Medline]
  45. Girnun GD, Smith WM, Drori S, et al. APC-dependent suppression of colon carcinogenesis by PPAR{gamma}. Proc Natl Acad Sci U S A 2002;99:13771–6.[Abstract/Free Full Text]
  46. Krey G, Braissant O, L'Horset F, et al. Fatty acids, eicosanoids, and hypolipidemic agents identified as ligands of peroxisome proliferator-activated receptors by coactivator-dependent receptor ligand assay. Mol Endocrinol 1997;11:779–91.[Abstract/Free Full Text]
  47. Singla AK, Chawla M, Singh A. Nimesulide: some pharmaceutical and pharmacological aspects—an update. J Pharm Pharmacol 2000;52:467–86.[CrossRef][Medline]
  48. Inoue H, Takamori M, Shimoyama Y, Ishibashi H, Yamamoto S, Koshihara Y. Regulation by PGE2 of the production of interleukin-6, macrophage colony stimulating factor, and vascular endothelial growth factor in human synovial fibroblasts. Br J Pharmacol 2002;136:287–95.[CrossRef][Medline]
  49. Mukutmoni M, Hubbard NE, Erickson KL. Prostaglandin E(2) modulation of vascular endothelial growth factor production in murine macrophages. Prostaglandins Leukot Essent Fatty Acids 2001;65:123–31.[CrossRef][Medline]
  50. Pai R, Szabo IL, Soreghan BA, Atay S, Kawanaka H, Tarnawski AS. PGE(2) stimulates VEGF expression in endothelial cells via ERK2/JNK1 signaling pathways. Biochem Biophys Res Commun 2001;286:923–8.[CrossRef][Medline]
  51. Amano H, Hayashi I, Endo H, et al. Host prostaglandin E(2)-EP3 signaling regulates tumor-associated angiogenesis and tumor growth. J Exp Med 2003;197:221–32.[Abstract/Free Full Text]
  52. Fukuda R, Kelly B, Semenza GL. Vascular endothelial growth factor gene expression in colon cancer cells exposed to prostaglandin E2 is mediated by hypoxia-inducible factor 1. Cancer Res 2003;63:2330–4.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Cancer Res.Home page
H. Funahashi, M. Satake, D. Dawson, N.-A. Huynh, H. A. Reber, O. J. Hines, and G. Eibl
Delayed Progression of Pancreatic Intraepithelial Neoplasia in a Conditional KrasG12D Mouse Model by a Selective Cyclooxygenase-2 Inhibitor
Cancer Res., August 1, 2007; 67(15): 7068 - 7071.
[Abstract] [Full Text] [PDF]

Home page
 Molecular Cancer TherapeuticsHome page
M. T. Yip-Schneider, H. Wu, M. Ralstin, C. Yiannoutsos, P. A. Crooks, S. Neelakantan, S. Noble, H. Nakshatri, C. J. Sweeney, and C. M. Schmidt
Suppression of pancreatic tumor growth by combination chemotherapy with sulindac and LC-1 is associated with cyclin D1 inhibition in vivo
Mol. Cancer Ther., June 1, 2007; 6(6): 1736 - 1744.
[Abstract] [Full Text] [PDF]

Home page
 DiabetesHome page
B. Askari, J. E. Kanter, A. M. Sherrid, D. L. Golej, A. T. Bender, J. Liu, W. A. Hsueh, J. A. Beavo, R. A. Coleman, and K. E. Bornfeldt
Rosiglitazone Inhibits Acyl-CoA Synthetase Activity and Fatty Acid Partitioning to Diacylglycerol and Triacylglycerol via a Peroxisome Proliferator-Activated Receptor-{gamma}-Independent Mechanism in Human Arterial Smooth Muscle Cells and Macrophages
Diabetes, April 1, 2007; 56(4): 1143 - 1152.
[Abstract] [Full Text] [PDF]

Home page
 J. Clin. Pathol.Home page
G Perrone, D Santini, A Verzi, B Vincenzi, D Borzomati, F Vecchio, R Coppola, A Antinori, P Magistrelli, G Tonini, et al.
COX-2 expression in ampullary carcinoma: correlation with angiogenesis process and clinicopathological variables
J. Clin. Pathol., May 1, 2006; 59(5): 492 - 496.
[Abstract] [Full Text] [PDF]

Home page
 Mol Cancer ResHome page
H. Sawai, J. Liu, H. A. Reber, O. J. Hines, and G. Eibl
Activation of Peroxisome Proliferator-Activated Receptor-{gamma} Decreases Pancreatic Cancer Cell Invasion through Modulation of the Plasminogen Activator System
Mol. Cancer Res., March 1, 2006; 4(3): 159 - 167.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
H. Sawai, Y. Okada, K. Kazanjian, J. Kim, S. Hasan, O. J. Hines, H. A. Reber, D. S.B. Hoon, and G. Eibl
The G691S RET Polymorphism Increases Glial Cell Line-Derived Neurotrophic Factor-Induced Pancreatic Cancer Cell Invasion by Amplifying Mitogen-Activated Protein Kinase Signaling
Cancer Res., December 15, 2005; 65(24): 11536 - 11544.
[Abstract] [Full Text] [PDF]

Home page
 J. Pharmacol. Exp. Ther.Home page