Cancer Research Audrey Hepburn Genetics and Biology of Brain Cancer
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 Email this article to a friend
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 Davis, J. N.
Right arrow Articles by Day, M. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Davis, J. N.
Right arrow Articles by Day, M. L.
[Cancer Research 65, 3633-3642, May 1, 2005]
© 2005 American Association for Cancer Research

Cell and Tumor Biology

Disruption of Rb/E2F Pathway Results in Increased Cyclooxygenase-2 Expression and Activity in Prostate Epithelial Cells

Joanne N. Davis1, Michael T. McCabe1,2, Simon W. Hayward3, John M. Park1 and Mark L. Day1,2

1 Department of Urology and 2 Program in Cellular and Molecular Biology, Michigan Urology Center, University of Michigan, Ann Arbor, Michigan; and 3 Department of Urology, Vanderbilt University Medical Center, Nashville, Tennessee

Requests for reprints: Mark L. Day, Department of Urology, University of Michigan, 6131 CCGC, 1500 East Medical Center Drive, Ann Arbor, MI 48109-0944. Phone: 734-647-2528; Fax: 734-947-9271; E-mail: mday{at}umich.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The loss of the retinoblastoma tumor suppressor gene (RB) is common in many human cancers, including prostate. We previously reported that engineered deletion of RB in prostate epithelial cells results in sustained cell growth in serum-free media, a predisposition to develop hyperplasia and dysplasia in prostate tissue recombinant grafts, and sensitization to hormonal carcinogenesis. Examining the molecular consequence of RB loss in this system, we show that cyclooxygenase-2 (COX-2) is significantly up-regulated following RB deletion in prostate tissue recombinants. To study the effect of RB deletion on COX-2 regulation, we generated wild-type (PrE) and Rb–/– (Rb–/–PrE) prostate epithelial cell lines rescued by tissue recombination. We show elevated COX-2 mRNA and protein expression in Rb–/–PrE cell lines with increased prostaglandin synthesis. We also find that loss of Rb leads to deregulated E2F activity, with increased expression of E2F target genes, and that exogenous expression of E2F1 results in elevated COX-2 mRNA and protein levels. COX-2 promoter studies reveal that E2F1 transcriptionally activates COX-2, which is dependent on the transactivation and DNA-binding domains of E2F1. Further analysis revealed that the E2F1 target gene, c-myb, is elevated in Rb–/–PrE cells and E2F1-overexpressing cells, whereas ectopic overexpression of c-myb activates the COX-2 promoter in prostate epithelial cells. Additionally, cotransfection with E2F1 and a dominant-negative c-myb inhibited E2F1 activation of the COX-2 promoter. Taken together, these results suggest activation of a transcriptional cascade by which E2F1 regulates COX-2 expression through the c-myb oncogene. This study reports a novel finding describing that deregulation of the Rb/E2F complex results in increased COX-2 expression and activity.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Cyclooxygenase (COX) or prostaglandin endoperoxide synthase is the rate-limiting enzyme that catalyzes the formation of prostaglandins from arachidonic acid. Two COX enzymes have been identified with similar enzymatic features; however, significant differences exist between them (reviewed in refs. 1, 2). COX-1 has been shown to be constitutively expressed in a majority of cells and performs housekeeping functions requiring immediate prostaglandin synthesis for vascular homeostasis, water resorption, and platelet aggregation. COX-2 has been described as an immediate early response gene and expressed predominantly in low levels. COX-2 has shown to be proinflammatory and can be activated by mitogens, tumor promoters, cytokines, and stress-inducing agents (1, 2).

There is accumulating evidence suggesting that COX-2 plays a role in human cancer. Elevated COX-2 expression has been observed in human colon cancer (3), lung cancer (4), cervical cancer (5), and breast cancer (6). Studies in animal models have shown that overexpression of COX-2 induced tumorigenesis and promoted cancer progression (7). In contrast, inhibition of COX-2 decreased tumor incidence and progression (8, 9). Although there is strong evidence implicating COX-2 in the development of certain human cancers, the role of COX-2 in prostate cancer remains unclear. In some studies, increased COX-2 expression has been detected in human prostate cancer tissues with a positive correlation between increased COX-2 with advancing prostate tumor stage (10–12). In contrast, other studies showed positive COX-2 staining in normal tissues with loss of expression in prostate cancer (13, 14). Other studies show that COX-2 is involved in early stages of prostate cancer with elevated COX-2 levels in benign proliferative hyperplastic lesions, high-grade prostatic intraepithelial neoplasia (HGPIN), and areas of the prostate with postinflammatory hyperproliferation, a condition thought to be a precursor to prostate cancer (13, 15). To help define a role of COX-2 in prostate cancer, COX-2 has been examined in transgenic mouse models of prostate cancer TRAMP (16) and LPB-Tag (17), which utilize the viral oncogene SV40 large T antigen to disrupt Rb and p53 tumor suppressor pathways to promote tumorigenesis. In both studies, COX-2 expression was elevated in HGPIN and early prostatic lesions, but was absent in late-stage tumors and sites of metastasis. These results support the role of COX-2 in the early development of prostate cancer and also suggest that disruption of the p53 and Rb pathways may be involved in COX-2 regulation. Because of the contrasting observations of COX-2 in prostate cancer, one endorsing elevated COX-2 and the other showing COX-2–independent mechanisms for prostate cancer progression, additional studies are required to examine the function and regulation of COX-2 in diverse populations of prostate epithelial and cancer cells.

Disruption of the Rb/E2F1 pathway may be a key event in the development of human cancers. The loss of the RB tumor suppressor gene has been reported in many human tumors including prostate (18), breast (19), colon (20), and bladder (21). The RB tumor suppressor gene encodes a 110 kDa phosphoprotein (Rb) that regulates the transition between G1 and S phases of the cell cycle (reviewed in ref. 22). Hypophosphorylated Rb binds to and sequesters the transcription factor E2F, thereby inhibiting transactivation of E2F target genes and ultimately inducing G1 cell cycle arrest. Inactivation of Rb results in increased E2F activity, enhanced transcriptional activation of E2F target genes, and forward progression of the cell cycle (23). Elevated E2F1 activity can induce hyperproliferation, hyperplasia, and p53-dependent apoptosis in E2F1 transgenic mice and transgenic mice that overexpress E2F1 under the keratin 5 promoter develop spontaneous tumors in keratin 5–expressing tissues (24, 25). E2F1 can also cooperate with Ras to induce transformation of rat embryonic fibroblasts, induce tumors in mice, and can increase the invasive potential of cancer cells (26, 27). Elevated expression of E2F1 has been observed in human breast cancers (28), non–small cell lung cancer (29), and salivary gland tumors (30). The oncogenic capacity of E2F is thought to be related to its regulatory role in the expression of genes involved in cell cycle and cell proliferation; however, mRNA profiling of E2F1-overexpressing cells indicates that E2F1 regulates hundreds of genes, not only those involved in cell cycle but also oncogenes, genes involved in apoptosis, signal transduction, and transcriptional regulation, suggesting that E2F1 may have multiple roles in tumorigenesis (31). Therefore, activation of E2F1 by disruption of the Rb tumor suppressor pathway may play a role in the development of cancer independent of cell cycle disruption.

Homozygous deletion of RB is lethal in embryonic mice, making it difficult to study the effects of RB loss in epithelial cells. To solve this problem, we developed a prostate tissue recombination model with homozygous deletion of RB, where RB–/– prostatic tissues are combined with rat urogenital mesenchyme (UGM) to generate UGM+Rb–/– prostate tissues and cell lines (32, 33). As a consequence of RB deletion, RB–/– cells are able to proliferate in serum-free conditions, have increased development of hyperplasia, and are susceptible to hormonal carcinogenesis (32, 33). In the current study, we describe one of the consequences of RB loss in prostate epithelial cells, up-regulation of COX-2. We utilized wild-type (PrE) and RB–/– (Rb–/–PrE) prostate epithelial cells lines rescued by tissue recombination to study the effect of RB inactivation on COX-2. Elevated COX-2 levels were present in Rb–/–PrE cell lines and UGM+Rb–/– recombinant tissue grafts in vivo. The increase in COX-2 expression correlated with elevated levels of prostaglandins and increased sensitivity to the COX-2 inhibitor NS-398. To further understand the mechanism(s) by which homozygous deletion of RB resulted in increased COX-2, we assessed whether the transcription factor, E2F1, plays a role in regulating COX-2. We show that Rb–/–PrE cells have increased E2F activity and exogenous expression of E2F1 transcriptionally regulates COX-2. Furthermore, we show that E2F1 activation of COX-2 is dependent on the c-myb oncogene. This is the first study to describe a role of Rb/E2F1 in regulating COX-2 expression.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Tissue culture reagents and chemicals. The generation of PrE and Rb–/–PrE cell lines was previously described (32). Cells were maintained in RPMI 1640 culture medium supplemented with 5% fetal bovine serum (FBS), 100 units/mL penicillin, 100 µg/mL streptomycin, and 2 mmol/L L-glutamine (Life Technologies, Inc., Gaithersburg, MD) and kept at 37°C in a humidified atmosphere of 5% CO2. The COX-2 inhibitor NS-398 and COX-1 inhibitor valeroyl salicylate acid (VSA) were purchased from Cayman Chemical, Co. (Ann Arbor, MI). For treatment with NS-398 or VSA, 5 x 105 cells were plated in 100 mm2 dishes for 24 hours. NS-398 or VSA was added to cells at the indicated concentrations and incubated for 4 days. Cell viability was monitored by trypan blue exclusion assay (Life Technologies).

Immunohistochemistry. The generation of UGM+Rb+/+ and UGM+Rb–/– tissues have been previously described (33). Tissue sections from UGM+Rb+/+ and UGM+Rb–/– were deparaffinized in xylene and rehydrated in graduated alcoholic solutions and PBS. Endogenous peroxidase activity was blocked with 0.5% hydrogen peroxide in methanol for 5 minutes and washed in PBS before staining. Immunocytochemical staining for COX-2 (Cayman Chemical) was done at a 1:200 dilution at room temperature overnight. Goat anti-rabbit secondary antibody was applied for 1 hour at room temperature. Antigen retrieval was accomplished with a sodium citrate solution for 10 minutes in a microwave.

Reverse transcription-PCR. RNA was prepared from cells using Qiagen RNAEasy kit per manufacturer's protocol (Qiagen, Valencia, CA). RNA was quantitated spectrophotometrically and 2 µg RNA were used to generate cDNA using Thermoscript RT-PCR Reaction System (Invitrogen, Carlsbad, CA) per manufacturer's protocol. PCR reaction was done using the following primers: E2F1, 5'-CATCCAGCTCATTGCCAAGAAG-3', 5'-GTCCGGTCCTCCCCAGAGGATG-'3'; E2F2, 5'-AAGAAGTTCATTTACCTCCTGA-3', 5'-AATCACTGTCTGCTCCTTGAA-3'; E2F4, 5'-CAAAGAGCTGTCAGAAATCTTC-3', 5'-TTGTAGATGTAATCGTGGTCTCC-3'; E2F5, 5'-GAATTGAAGGAAAGAGAACTTG-3', 5'-AAACCACTGGCTTAGATGAA-3'; DP-1, 5'-CGCCTCCTCCCAACTCTGTCATC-3', CAGCATCCCATCTGCACCATTGCT-3'; DP-2, 5'-TGCAGCATCTCCAGTGAC-3', 5'-CAGAGAGCATTTGCCTGAC-3'; and hypoxanthine phosphoribosyltransferase (HPRT), 5'-CAGTACAGCCCCAAAATGGT-3', 5'-TTACTAGGCAGATGGCCACA-3'. PCR was done using a Thermocycler with an annealing temperature of 59°C. HPRT was used as an internal control. Ten-microliter aliquots were removed at cycle number 25, 28, 31, and 34. Reaction products were electrophoresed in a 0.8% agarose gel in 0.5x TAE buffer (20 mmol/L Tris, 0.5 mmol/L EDTA, adjusted to pH 7.5) for 2 hours at 100 V. DNA products were visualized with ethidium bromide staining under UV light.

Immunofluorescence. Cells grown in four-well chamber slides were washed twice with PBS before fixation in 4% formaldehyde for 5 minutes followed by fixation in 100% ice-cold ethanol for 5 minutes. Fixed cells were washed twice in PBS and then incubated with 0.1% FBS/1% milk/TBST [10 mmol/L Tris (pH 7.6), 250 mmol/L NaCl, 0.25% Tween 20] for 30 minutes to block nonspecific binding. Cells were incubated with antibodies to E2F1 (C-20, Santa Cruz Biotechnology, Santa Cruz, CA), DP-1 (K-20, Santa Cruz Biotechnology), and Rb (PharMingen) in 0.1% FBS/1% milk/TBST for 1 hour at room temperature followed by incubation with Rhodamine-conjugated goat anti-rabbit or goat anti-mouse secondary antibody (Santa Cruz Biotechnology). Nuclear staining was detected using 1 µg/mL 4',6-diamidino-2-phenylindole (DAPI, Molecular Probes, Eugene, OR). Cells were washed with TBST and mounted with Aqua Poly/Mount (Polysciences, Inc., Warrington, PA). To image the signal, conventional fluorescence microscopy was done with a photomicroscope (Olympus, Melville, NY) equipped with an epifluorescence attachment.

Western blot analysis. Whole cell protein lysates were prepared from PrE and Rb–/–PrE cells as previously described (32). Supernatants were quantitated using a Bradford microtiter assay and denatured with a reducing 2x sample loading buffer for 5 minutes at 100°C. All proteins were separated on Tris/glycine precast NOVEX gels and analyzed utilizing the ECL detection system (Amersham Biosciences, Piscataway, NJ) as previously described (32). Primary antibodies were obtained as follows: COX-1 (Cayman Chemical), COX-2 (Cayman Chemical), Rb (14001A, PharMingen), E2F1 (KH-20, PharMingen), p19ARF (Ab-80, Abcam, Cambridge, United Kingdom), Cyclin A (H-432, Santa Cruz Biotechnology), Cyclin D (A-12, Santa Cruz Biotechnology), Cyclin E (M-20, Santa Cruz Biotechnology), c-myb (C-2, Santa Cruz Biotechnology), and ß-actin (Santa Cruz Biotechnology). Appropriate secondary antibodies goat anti-rabbit and goat anti-mouse were obtained from Bio-Rad (Hercules, CA). Signals were detected by enhanced chemiluminescence signal (Bio-Rad) by X-ray film (Kodak, Rochester, NY).

Northern blot analysis. Total RNA was extracted from cells using Qiagen RNA Easy kit per manufacturer's protocol. Northern blot analysis was done using 20 µg total RNA run on a 4% formaldehyde denaturing agarose gel in 1x N-morpholinopropanesulfonic acid (MOPS) buffer [0.2 mol/L MOPS, 10 mmol/L EDTA, 10 mmol/L sodium acetate (pH 7.0)] and transferred to a nylon membrane (Stratagene, La Jolla, CA) in 20x SSC [3 mol/L NaCl, 0.3 mol/L sodium citrate (pH 7.0)] overnight using capillary action. The RNA was cross-linked to the membrane using a Stratagene UV cross-linker and then incubated for 15 minutes at 65°C in hybridization buffer (250 mmol/L Na2HPO4 and 7% SDS). Hybridization was done at 65°C overnight in hybridization buffer containing cDNA probes for COX-2, E2F-1, or ß-actin labeled with {alpha}-32P ATP (2 x 106 cpm/mL) using Random Primer Labeling kit (Stratagene) per manufacturer's protocol. The COX-2 cDNA probe was a 581 bp PCR product made using the following primers: sense 5'-ACTCACTCAGTTTGTTGAGTCATTC-3', antisense 5'-TTTGATTAGTACTGTAGGGTTAATG-3'. The E2F1 cDNA probe is a 1.1 kb fragment generated from mouse E2F1 cDNA cut with BamHI and Xho1 restriction endonucleases. The ß-actin probe is a 900 bp PCR product generated from the following primers: sense 5'-TACAGCTTCACCACCACAGC-3', antisense 5'-ACAGAAGCAATGCTGTCACC-3'. Following the overnight incubation with the respective 32P-labeled probes, the membrane was washed twice with 20 mmol/L Na2HPO4 and 5% SDS for 45 minutes each at 65°C, followed by two washes with 20 mmol/L Na2HPO4 and 1% SDS for 45 minutes each at 65°C. The membrane was exposed to film (Kodak) overnight.

Prostaglandin E2 analysis. Cells (1 x 105) were plated per well in a six-well plate. Culture media was collected at 24, 48, 72, and 96 hours. Fifty microliters of centrifuged supernatant from each sample were analyzed for prostaglandin E2 (PGE2) production using PGE2 enzyme immunoassay system per manufacturer's protocol (Cayman Chemical). PGE2 from each supernatant was measured in triplicate using Dynatech Laboratories Microplate Reader (Dynatech Laboratories, Chantilly, VA).

Generation of E2F1-overexpressing cells. PrE cells were trypsinized, plated in 100 mm2 dishes, and allowed to reach 70% confluence before transfection. Ten micrograms E2F1 cDNA or pCDNA (empty vector control) plasmids containing a neomycin resistance gene were transfected into PrE cells using TFx50 transfection reagent (Promega, Madison, WI) following manufacturer's protocol. Cells were incubated for 72 hours, and then the media was changed to RPMI 1640 containing 5% FBS, 0.1% penicillin/streptomycin, and 200 µg G418 (Sigma Chemical, Co., St. Louis, MO). Cells were replenished with fresh media containing 200 µg G418 every 4 days and selected for stable expressing clones.

Transient transfection and plasmid constructs. PrE and Rb–/–PrE cells were plated at a density of 2 x 105 cells per well in a six-well dish and incubated for 24 hours at 37°C before transfection. Cells were cotransfected with 1 µg COX2-Luc, dihydrofolate reductase (DHFR)-Luc, or E2F-Luc with or without 0.5 µg pCDNA (empty vector control), wild-type E2F1, Eco 132, or E2F11-284. ß-galactosidase (0.1 µg) was cotransfected for normalization purposes. The COX2-Luc construct was purchased from Oxford Biomedical Research Laboratories (Rochester, MI); E2F1 cDNA construct was kindly provided by Tony Kouzarides (Wellcome/Cancer Research UK Institute, Cambridge, United Kingdom); DHFR-Luc, E2F-Luc, Eco132, and E2F11-284 were kindly provided by W. Douglas Cress (Moffitt Cancer Center, Tampa, FL); the full-length c-myb and dominant-negative c-myb constructs were kindly provided by Mansoor Husain (University of Toronto, Toronto, Canada); and the ß-galactosidase cDNA construct was purchased from Promega. Cells were transfected with the appropriate constructs using Tfx50 transfection reagent (Promega) following the manufacturer's protocol for 72 hours. To prepare whole cell protein lysates, cells were washed once with PBS and cells were harvested by cell scraping. Cells were lysed in Tropix Lysis Buffer (Applied Biosystems, Foster City, CA), freeze thawed, and centrifuged. Supernatants were collected, luciferase activity was monitored using luciferin substrate (Promega), and relative light units were measured using a Monolight 2010 luminometer. To assay ß-galactosidase activity, cell lysates were incubated for 1 hour with ß-galactosidase substrate (Applied Biosystems) and luminescence was monitored using Monolight 2010 luminometer. Each sample was done in triplicate and the luciferase activity was normalized to ß-galactosidase activity to determine results. The results are expressed as the average of three independent experiments with the indicated SD.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Increased COX-2 expression in UGM+Rb–/– prostate tissues and Rb–/–PrE prostate epithelial cell lines. To identify novel properties that distinguish wild-type prostate epithelial cells from prostate epithelial cells with RB deletion, we compared gene expression profiles of established wild-type PrE and Rb–/–PrE prostate epithelial cell lines using microarray analysis. Among the genes that were differentially expressed, we noted that COX-2 was up-regulated 8-fold in Rb–/–PrE prostate epithelial cells compared with wild-type PrE cells (data available at http://www.ncbi.nlm.nih.gov/geo/, series number GSE934). This gene was of immediate interest because of multiple studies implicating COX-2 as an important modulator of tumorigenesis (6, 7, 10, 34). Supporting these findings, we independently observed that COX-2 staining was substantially elevated in UGM+Rb–/– prostate tissue recombinants. In wild-type UGM+Rb+/+ tissues (Fig. 1A-a and c), the majority of the glands were lined with single columnar luminal epithelial cells with low COX-2 staining; however, there were some sporadic areas of epithelial hyperplasia, which correlated with increased COX-2 stain intensity. In contrast, UGM+Rb–/– tissues predominantly exhibited regions of hyperplasia with strong COX-2 staining (Fig. 1A-b and d). Overall, UGM+Rb–/– tissue recombinants have increased COX-2 staining in individual cells and an increase in the number of COX-2–positive cells in areas of hyperplasia, suggesting that loss of Rb leads to increased COX-2 expression in prostate epithelial cells.



View larger version (64K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Increased COX-2 expression in prostate epithelial cells with homozygous deletion of Rb. A, color photomicrographs of representative histologic sections from UGM+Rb+/+ (a, x20 magnification; c, x60 magnification) and UGM+Rb–/– prostate recombinant tissue grafts (b, x20 magnification; d, x60 magnification) stained for COX-2. B, Western blot analysis of COX-1, COX-2, and Rb in two wild-type PrE and four Rb–/– cell lines. ß-actin is shown as a loading control. C, Northern blot analysis of COX-2 mRNA levels in PrE and Rb–/– cells. ß-actin is shown as a loading control. D, PrE and Rb–/–PrE cells were transiently transfected with 1 µg COX2-Luc and 0.1 µg ß-galactosidase for normalization purposes. Forty-eight hours after transfection, cells were lysed and assayed for luciferase and ß-gal expression. The results are expressed as COX2-Luc activity divided by ß-gal. Columns, mean value of three independent experiments; bars, SD.

 
We next examined COX-2 protein expression in two wild-type PrE and four Rb–/–PrE cell lines derived from UGM+Rb+/+ and UGM+Rb–/– recombinant tissues. Figure 1B shows increased COX-2 expression in Rb–/–PrE cell lines compared with PrE control cell lines. COX-2 levels were relatively low in PrE cell lines and high in all Rb–/–PrE cell lines. As a control, we show positive Rb protein expression in PrE cells and absence of Rb in Rb–/–PrE cell lines. We also examined COX-1 levels in PrE and Rb–/–PrE cells and observed relatively low levels of COX-1 in Rb–/–PrE cells and slightly elevated levels of COX-1 in PrE cells (Fig. 1B). Increased COX-2 mRNA was confirmed by Northern blot analysis from total RNA isolated from PrE and Rb–/–PrE cell lines (Fig. 1C). These results confirm that COX-2 expression is up-regulated in Rb–/–PrE cell lines and UGM+Rb–/– tissues.

To determine differences in basal transcriptional activity of the COX-2 promoter, we assayed expression levels of a promoter construct containing the 5'-flanking sequences corresponding to bp –3,195 to +39 relative to the transcription start site of the mouse COX-2 promoter, cloned in front of a luciferase reporter gene (COX2-Luc). To normalize for transfection efficiency, cells were cotransfected with a cytomegalovirus promoter–driven ß-galactosidase reporter gene construct (ß-gal). Figure 1D shows that Rb–/–PrE cells have 4-fold increase in basal COX-2 promoter activity compared with PrE cells. As a negative control, cells were also transfected with a promoterless expression vector (Basic-Luc). This result confirms with the microarray analysis and Northern blot analysis and shows that COX-2 is transcriptionally up-regulated in prostate epithelial cells with homozygous deletion of RB. Together, these results suggest that the loss of RB in prostate epithelial cells results in increased COX-2 expression, as shown in vitro and in vivo.

High levels of prostaglandins are produced by Rb–/–PrE cells. The rate-limiting step in the conversion of free arachidonic acid to prostaglandins is catalyzed by COX (reviewed in ref. 2). Given the observation that COX-2 levels are elevated in Rb–/–PrE cells, we wanted to determine if prostaglandin synthesis was elevated Rb–/–PrE prostate epithelial cells in vitro. Rb–/–PrE cells produced 10-fold greater PGE2 levels compared with PrE cells after 24 hours (2600 pg/mL for Rb–/– cells versus 226 pg/mL for PrE cells) and continued to have elevated PGE2 levels over 4 days (Fig. 2A). To show that COX-2 is a major contributor to PGE2 synthesis, we treated cells with increasing concentrations of the specific COX-2 inhibitor, NS-398. As shown in Fig. 2B, NS-398 significantly inhibited PGE2 synthesis in PrE and Rb–/–PrE cell lines. After 24 hours, the lowest concentration, 10 µmol/L NS-398, decreased PGE2 levels 90% in PrE cells (316-3.2 pg/mL) and 99% in Rb–/–PrE cells (3,312-2.0 pg/mL). These results show that COX-2 is a major contributor to PGE2 synthesis in prostate epithelial cells. As a control, we also examined whether COX-1 contributes to PGE2 synthesis in PrE and Rb–/–PrE cells. Cells were treated with increasing concentrations of the specific COX-1 inhibitor; VSA and supernatants were assayed for PGE2 synthesis. As shown in Fig. 2B, VSA did not significantly inhibit PGE2 levels in PrE or Rb–/–PrE cells. This suggests that COX-2 is the major contributor to PGE2 production in PrE and Rb–/–PrE cells.



View larger version (29K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Rb–/–PrE cells have increased PGE2 synthesis and are sensitive to the growth inhibitory effects of NS-398. A, increased PGE2 synthesis in Rb–/–PrE cells. An equal number of PrE and Rb–/–PrE cells were plated and incubated for 4 days. Supernatants were removed at the indicated time and assayed for PGE2 production by ELISA assay. Each sample was measured in triplicate. Columns, mean of three independent experiments; bars, SD. B, PrE and Rb–/– cells were treated with 0, 10, 25, 50, and 100 µmol/L NS-398 or VSA for 24 hours. Supernatants were removed at the indicated time and assayed for PGE2 production by ELISA assay. Each sample was measured in triplicate. Columns, mean of three independent experiments; bars, SD. C, NS-398 inhibits the growth of Rb–/–PrE cells. PrE and Rb–/–PrE cells were treated with either 0, 25, 50, or 100 µmol/L NS-398 for 4 days. Cell growth was determined by trypan blue exclusion assay. Each sample was measured in triplicate. Points, mean; bars, SD.

 
Rb–/–PrE cells are sensitive to the cyclooxygenase-2 inhibitor NS-398. Prostaglandins have been shown to increase cell growth of prostate epithelial cells (35) and COX-2 inhibitors have been shown to inhibit cell growth of LNCaP, PC3, and DU-145 prostate cancer cells (36). We wanted to assess whether NS-398 had the ability to regulate the growth of PrE and Rb–/–PrE cells. Cells were treated with increasing concentrations of NS-398 for 4 days and cell growth was determined. Figure 2C indicates that NS-398 exhibited differential effects on the proliferation of PrE and Rb–/–PrE cells. PrE cells were less susceptible to the antiproliferative effect of NS-398 compared with Rb–/–PrE cells. No appreciable effect was noted in PrE cells with NS-398 until the concentration reached 100 µmol/L. In contrast, NS-398 inhibited the growth of Rb–/–PrE cells dose dependently with an IC50 value of 25 µmol/L. After 4 days of treatment, 50 and 100 µmol/L NS-398 completely inhibited Rb–/–PrE cell growth. We also examined growth rates of Rb–/–PrE and PrE cells to the COX-1 inhibitor VSA (data not shown). VSA did not have an effect on cell growth of PrE or Rb–/–PrE cells. These results show that Rb–/– cells are much more sensitive to the COX-2 inhibitor, NS-398, in contrast to normal prostate epithelial PrE cells.

Increased E2F1 activity in Rb–/–PrE cells. We wanted to explore the mechanism by which loss of RB in prostate epithelial cells results in increased COX-2 expression. Studies assessing RB deletion have reported increased E2F activity following loss of RB (37, 38). Considering that E2F is a major Rb-regulated protein and a transcriptional regulator of many genes, we hypothesized that the loss of RB in prostate epithelial cells leads to increased activation of E2F, which then increases COX-2 transcription. To test this hypothesis, we first assessed expression levels of E2F family members in PrE and Rb–/–PrE cells by semiquantitative reverse transcription-PCR (RT-PCR) analysis, using specific primers to E2F 1 to 5. After confirmation of linear amplification of the PCR products, as shown in Fig. 3A, the gene expression levels were compared. The mRNA levels of the transcriptional activator, E2F1, were slightly increased in Rb–/–PrE cells compared with PrE cells. E2F1 has E2F binding sites in its promoter and is transcriptionally regulated by E2F (39). Furthermore, increased E2F1 expression has been reported in cells with germ line deletion of RB (40), which is consistent with our results. Similar gene expression levels were detected for the transcriptional activator E2F2 and the transcriptional repressors, E2F4 and E2F5. We could not detect E2F3 expression in either PrE or Rb–/–PrE cells using multiple primers to E2F3; however, we were able to detect E2F3 in NIH 3T3 positive control cells (data not shown), suggesting that E2F3 levels are below detection by RT-PCR. These results show that E2F1, E2F2, E2F4, and E2F5 are present in PrE and Rb–/–PrE cells and that E2F1 levels are elevated in Rb–/–PrE cells, which is consistent with previous studies describing the effects of RB deletion (40).



View larger version (44K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Increased E2F1 activity in Rb–/– cells. A, semiquantitative RT-PCR of E2F family members 1 to 5 and DP family members in PrE and Rb–/–PrE cells. The linear amplified curve of the PCR product of each sample was examined at three-cycle intervals. PCR products were prepared under appropriate conditions and aliquots were taken at the end of the 25th, 28th, 31st, and 34th cycles. B, subcellular localization of E2F1, DP-1, DP-2, and pRb determined by immunofluorescence. PrE and Rb–/–PrE cells were grown on coverslips, fixed in –20°C ethanol, and immunostained with antibodies against E2F1, Rb, DP-1, and DP-2 followed by Rhodamine-conjugated secondary antibody. Nuclei were visualized with DAPI staining and corresponding panels are represented below appropriate immunostained panel. C, Western blot analysis of PrE and Rb–/– cells for the detection of E2F1 target genes, Cyclin A, Cyclin D, Cyclin E, and p19ARF. ß-actin is shown as a loading control. D, PrE and Rb–/– cells were transiently transfected with 1 µg of either DHFR-Luc or E2F-Luc for 72 hours. Cells were cotransfected with ß-gal for normalization purposes. Results are expressed as fold activation. Columns, mean value of three independent experiments; bars, SD.

 
The transcriptional regulatory activity of E2F depends on the ability to heterodimerize to DP-1 and DP-2; therefore, we examined expression levels of DP-1 and DP-2 by semiquantitative RT-PCR over a linear range of amplification. We observed a slight increase in DP-1 mRNA expression Rb–/–PrE cells compared with PrE cells (Fig. 3A), whereas DP-2 levels were similar between the cell lines. Elevated DP-1 mRNA levels have also been observed in mouse embryonic fibroblasts with RB deletion and in cells with increased E2F activity, indicating that Rb/E2F complex may transcriptionally regulate DP-1 (41, 42).

Subcellular localization of E2F is critical for its transcriptional activity. E2F1 can be differentially located in cells, but must be in the nucleus to be transcriptionally active. We assessed the subcellular localization of E2F1, the most well-characterized E2F family member, in PrE and Rb–/–PrE cells by immunohistochemistry. We observed differential subcellular localization of E2F1 in PrE cells compared with Rb–/–PrE cells. Figure 3B shows that E2F1 is predominantly cytoplasmic in PrE cells with minimal staining in the nucleus. In contrast, there is greater E2F1 nuclear staining in Rb–/–PrE cells. The subcellular localization of DP-1 and DP-2 were similar in PrE and Rb–/–PrE cells (Fig. 3B). The increase in E2F1 nuclear localization in Rb–/–PrE cells is suggestive of increased E2F activity; however, to directly assess E2F activation in PrE and Rb–/–PrE cells, we examined protein expression levels of known E2F target genes. E2F target genes known to be activated by E2F1 (Cyclin A, Cyclin D, Cyclin E, and p19ARF) were selected for Western blot analysis. Figure 3C shows increased protein expression of Cyclin A, Cyclin D, Cyclin E, and p19ARF in all four Rb–/–PrE cell lines compared with PrE cells. This is consistent with the model that E2F activity is increased in Rb–/–PrE cells.

To further show specific activation of E2F target promoters in Rb–/–PrE cells, we investigated basal activation of two E2F target promoters, one containing four adjacent E2F consensus binding sites linked to a luciferase reporter gene (E2F-Luc) and a second construct containing the well-characterized E2F target gene promoter, DHFR, linked to a luciferase reporter gene (DHFR-Luc). Cells were cotransfected with ß-gal to normalize for transfection efficiency. Figure 3D shows that E2F-Luc activity is 4-fold higher in Rb–/–PrE cells compared with PrE cells, and DHFR-Luc activity is 7-fold higher in Rb–/–PrE cells compared with wild-type control. Taken together, increased E2F1 nuclear expression in Rb–/–PrE cells, with elevated levels of E2F target genes (E2F1, Cyclin A, Cyclin D, Cyclin E, and p19ARF) and increased activation of E2F specific promoter-luciferase constructs, strongly supports the model of increased E2F activity in Rb–/–PrE cells. Our results are consistent with other reports demonstrating increased E2F target gene expression and E2F activity in cells with loss of RB (37, 38, 40).

E2F1-overexpressing prostate epithelial cells have increased cyclooxygenase-2 mRNA and protein levels and increased prostaglandin E2 synthesis. Given the increased transcriptional activity of E2F and increased COX-2 expression levels in Rb–/–PrE cells, we investigated whether E2F1 regulates COX-2 expression in prostate epithelial cells. We transfected PrE cells with an E2F1 cDNA construct and generated two stable E2F1-overexpressing cell lines, named PrE2F1-1 and PrE2F1-2, as well as a stable expressing empty vector control (PrE-pCDNA). Using Northern blot analysis, we show increased expression of E2F1 mRNA levels in the stable expressing cells compared with the empty vector control (Fig. 4A). To show that E2F1-overexpressing clones have increased E2F activity, we show increased protein expression of known E2F target genes, Cyclin A, Cyclin D, Cyclin E, and p19ARF (Fig. 4B), and increased activation of DHFR-Luc in PrE2F1-1 and PrE2F1-2 cells (Fig. 4C). Exogenous expression of E2F1 resulted in increased COX-2 mRNA and protein levels increased COX-2 promoter activity and elevated PGE2 levels (Fig. 4D). These results show that exogenous expression of E2F1 positively regulates COX-2 expression and activity in prostate epithelial cells.



View larger version (29K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Overexpression of E2F1 increases COX-2 levels in prostate epithelial cells. A, Northern blot analysis of stably transfected PrE-pCDNA (empty vector control) and two E2F1-overexpressing cell lines, PrE2F1-1 and PrE2F1-2. Membranes were incubated with probes for either E2F1 or COX-2. ß-actin is shown as a loading control. B, Western blot analysis to detect COX-2 and E2F target genes Cyclin A, Cyclin D, Cyclin E, and p19ARF in stably transfected PrE-pcDNA, PrE2F1-1, and PrE2F1-2 cells. C, control and E2F1-overexpessing cells were transiently transfected with 1 µg of either Basic-Luc, COX2-Luc, or DHFR-Luc for 72 hours. Cells were cotransfected with ß-gal for normalization purposes. Results are expressed as fold activation. Columns, mean value of three independent experiments; bars, SD. D, PrE-E2F1 overexpressing cells have increased PGE2 synthesis. Supernatants from cells stably transfected with either pCDNA (empty vector) or E2F1 were analyzed for PGE2 production using an ELISA assay. Each sample was measured in triplicate. Points, mean; bars, SD.

 
Transcriptional activation of the cyclooxygenase 2 promoter by E2F1. To determine whether E2F1 transcriptionally regulates COX-2, COX-2 promoter/luciferase constructs were utilized. PrE and Rb–/–PrE cells were cotransfected with COX2-Luc with increasing amounts of wild-type E2F1 expression vector. As shown in Fig. 5A, 1 µg E2F1 activated COX-2 promoter activity 7-fold compared with empty vector control (pCDNA3) in PrE cells and 8-fold in Rb–/–PrE cells. As a positive control, we show that E2F1 activates the well-characterized E2F-inducible promoter DHFR-Luc 10-fold (Fig. 5B). This activation was dependent on an intact E2F1 DNA-binding domain and an intact transactivation domain because the DNA-binding mutant (Eco132) and deletion of the transactivation domain of E2F1 (E2F11-284) were incapable of activating COX-2 reporter activity. As expected, the E2F1 mutants did not activate the DHFR-Luc, which is known to require both the transactivation and DNA-binding domains. These results indicate that both the DNA-binding domain and transactivation domain of E2F1 are required to activate the COX-2 promoter.



View larger version (16K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5. E2F1 transcriptionally activates COX-2 promoter. PrE and Rb–/– cells were cotransfected with either 1 µg COX2-Luc (A) or DHFR-Luc (B), and either 0, 0.5, 1 µg of E2F1, or 1 µg of the E2F1 mutants Eco132 or E2F11-284. Forty-eight hours after transfection, cells were lysed and assayed for luciferase and ß-gal activity. To normalize for transfection efficiency, 0.1 µg ß-gal was included in transfection reaction and results are shown as luciferase activity divided by ß-gal activity. Each experiment was done in triplicate. Columns, mean; bars, SD.

 
Using computational analysis, we examined the COX-2 promoter for potential E2F consensus binding sites (TTTGCGG/CG/CAAA); however, we were unable to identify any strong E2F1 consensus binding sites. The lack of E2F binding sites in the COX-2 promoter suggests that other mechanisms downstream from E2F1 may be regulating COX-2 transcriptional activity; therefore, we examined our microarray analysis, comparing Rb–/–PrE and PrE cells, for potential transcription factors downstream of E2F1. We found that the transcription factor c-myb was up-regulated 4-fold in Rb–/–PrE cells compared with PrE cells (data available at http://www.ncbi.nlm.nih.gov/geo/, series number GSE934); furthermore, the c-myb transcription factor has been shown to directly activate the COX-2 promoter in human colon cancer cells (43). We, therefore, investigated whether E2F1-activation of COX-2 is mediated by c-myb.

The c-myb oncogene mediates E2F1-dependent activation of cyclooxygenase-2 promoter. The c-myb oncogene is an E2F1 target gene that is up-regulated in E2F overexpressing cells (44). We confirmed that c-myb protein expression was elevated in Rb–/–PrE cells and PrE2F1-overexpressing cells by Western blot analysis. Figure 6A shows increased c-myb protein expression in Rb–/–PrE cells compared with PrE cells and increased c-myb expression in prostate epithelial cells stably transfected with E2F1. To further characterize a mechanism responsible for increased COX-2 expression in Rb–/–PrE and E2F1 overexpressing cells, we investigated whether E2F1 activation of COX-2 is mediated by c-myb. To test this, we first assessed whether c-myb could activate the COX-2 promoter in prostate epithelial cells. PrE cells were cotransfected with COX2-Luc and a full-length c-myb cDNA construct. Figure 6B shows that c-myb activates COX2-Luc 7-fold, similar to results reported in human colon cancer cells by Ramsay et al. (43). Cotransfection with a dominant-negative c-myb construct, which lacks the DNA-binding domain (DN c-myb), completely inhibited activation of COX2-Luc by c-myb (Fig. 6B). As shown in Fig. 6C, cotransfection of dominant-negative c-myb with E2F1 completely abrogated E2F1 activation of COX2-Luc, indicating that activation of the COX-2 promoter by E2F1 is dependent on the activity of c-myb.



View larger version (29K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6. Effect of dominant-negative c-myb on E2F1-induced COX-2 promoter activation. A, the oncogene c-myb is up-regulated in Rb–/–PrE and E2F1-overexpressing cells. Whole cell lysates from PrE, Rb–/–PrE, PrE-pCDNA (empty vector control), PrE2F1-1, and PrE2F1-2 cells were subjected to Western blot analysis for the detection of c-myb. ß-actin is shown as a loading control. B, oncogenic c-myb activates COX-2 promoter activity in prostate epithelial cells. PrE cells were transiently transfected with COX2-Luc and either pCDNA (empty vector), E2F1, c-myb, or dominant-negative c-myb (DN c-myb). Forty-eight hours after transfection, cells were lysed and assayed for luciferase and ß-gal. Each experiment was done in triplicate. Columns, mean; bars, SD. C, dominant-negative c-myb abrogates E2F1-induced COX-2 promoter activation. PrE cells were cotransfected with COX2-Luc and a combination of pCDNA, E2F1, c-myb and DN c-myb. Forty-eight hours after transfection, cells were lysed and assayed for luciferase and ß-gal. Each experiment was done in triplicate. Columns, mean; bars, SD.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
There is strong evidence to implicate COX-2 in human colon cancer, breast cancer, and lung cancer; however, a common consensus for COX-2 in prostate tumorigenesis has not been established (3–6). Some reports document elevated COX-2 levels in prostate cancer tissue and HGPIN lesions as well as areas of proliferative inflammatory atrophy, a precursor to prostate cancer (10–12, 17). In contrast, other studies have shown increased COX-2 expression in normal prostate tissue and minimal COX-2 expression in prostate cancer and prostate cancer cell lines (13, 14). Animal models of prostate cancer have shown increased COX-2 expression in hyperplasia, in prostate cancer, and in metastatic lesions (16, 17). Overexpression of COX-2 in LNCaP cells promotes tumors in severe combined immunodeficient mice and inhibition of COX-2 can prevent tumor formation in metastasis in animal models (7). In our study, we observed increased COX-2 expression in UGM+Rb–/– prostate recombinant tissue grafts, which was confirmed in Rb–/–PrE prostate epithelial cell lines. Increased COX-2 expression correlated with areas of hyperplasia in vivo. Interestingly, the UGM+Rb–/– tissues are susceptible to hormonal carcinogenesis in vivo (33), suggesting a role of COX-2 in precursor or early prostatic lesions.

Elevated COX-2 expression has been observed in two transgenic mouse models of prostate cancer with nonfunctional Rb (16, 17). In both studies, the probasin promoter was used to target SV40 large T antigen to the luminal prostatic epithelial cells. SV40 large T disrupts both p53 and Rb to promote tumorigenesis. COX-2 expression was high in HGPIN lesions as well as early stages of prostate cancer, but absent in late-stage tumors and in metastasis (16, 17). Furthermore, the COX-2 inhibitor Celecoxib inhibited prostate tumor progression and reduced metastasis in the TRAMP model. Although the use of viral oncogenes, such as large T antigen, are useful to promote oncogenesis, they also target additional cellular molecules, thereby limiting the ability to study a specific molecular mechanism and making interpretation of experimental results difficult. In our study, we specifically disrupted RB while maintaining wild-type p53 in the Rb–/– cells (32) and show that either loss of RB or exogenous overexpression of E2F1 results in increased COX-2 expression with increased prostaglandin synthesis. The increase in COX-2 was associated with areas of hyperplasia in vivo and increased E2F1 activity in vitro, supporting the role that disruption of Rb or activation of E2F1 is important in COX-2 regulation and may be important in early prostate tumorigenesis. In our study, we report that the COX-2 inhibitor NS-398 was selective in inhibiting Rb–/–PrE cell growth and did not inhibit growth of normal prostate epithelial cells. Because the Rb–/– prostate recombinant tissues develop prostate cancer following testosterone and estrogen treatment, it would be interesting to determine if COX-2 inhibitors could prevent hormonal carcinogenesis in Rb–/– prostate tissues.

The exact function of COX-2 in promoting tumor development is not known; however, there are several molecular mechanisms described for COX-2 in promoting cell growth. COX-2 induces prostaglandin synthesis and prostaglandins have growth-promoting effects in prostate cancer cells (35). PGE2 stimulated primary prostate cancer cell growth through activation of interleukin-6 and signal transducers and activators of transcription 3 signaling molecules (36). The COX-2 inhibitor Celecoxib can inhibit Akt activation and induce apoptosis in human cancer cells (45). It remains to be determined if NS-398 induced apoptosis in Rb–/–PrE cells. Other mechanisms have been described for COX-2, including the production of reactive oxygen radicals during the conversion of arachidonic acid to prostaglandin, which may contribute to DNA oxidative damage and induction of genetic mutations (46). COX-2 may play a role in angiogenesis by up-regulating vascular endothelial growth factor (VEGF) levels (47). Interestingly, we detected increased VEGF mRNA levels in Rb–/–PrE cells by microarray analysis (http://www.ncbi.nlm.nih.gov/geo/, series number GSE934), suggesting another consequence of Rb/E2F deregulation is up-regulation of VEGF.

Deregulation of E2F1 through impairment of Rb occurs in tumors. This deregulation can occur through mutation of the RB gene, overexpression of Cyclin D, or inactivation of 16INK4A cyclin-dependent kinase (23). The growth inhibitory protein p16INK4A inhibits Cyclin D–dependent kinases, which, in turn, phosphorylate Rb. In a recent study by Crawford et al. (48), silencing of p16INK4A gene expression through promoter hypermethylation resulted in increased COX-2 levels and elevated PGE2 synthesis in mammary epithelial cells. The increased COX-2 expression contributed to increased angiogenic activity and invasiveness; furthermore, the COX-2 inhibitor NS-398 was able to reduce cell survival and increase apoptosis in mammary epithelial cells. These studies support our findings that disruption of the Rb pathway plays a regulatory role in COX-2.

Previous reports have documented that loss of RB results in increased expression of E2F target genes and increased E2F activity (37, 38, 49). We have observed increased expression of E2F target genes, Cyclin A, Cyclin D, Cyclin E, p19ARF, E2F1, and c-myb, as well as increased activation of E2F-specific promoter/luciferase constructs in the Rb–/–PrE cells, supporting the model of increased activation of E2F in Rb–/–PrE cells. We argue that the increase in COX-2 expression observed in Rb–/–PrE cells is mediated in part by E2F. This hypothesis is supported by the fact that exogenous E2F1 transcriptionally activated COX-2 expression in prostate epithelial cells, which required both E2F1 DNA-binding and transactivation domains. We have also shown that c-myb, an E2F1 target gene (44, 50), is up-regulated in Rb–/–PrE cells and E2F1-overexpressing cells and exogenous expression of c-myb activated the COX-2 promoter in mouse prostate epithelial cells. These results agree with the study by Ramsay et al. (43), who identified c-myb binding sites in the COX-2 promoter and showed transcriptional regulation of the COX-2 promoter by c-myb in human colon cancer cells. We show that a dominant-negative c-myb construct, which lacks the DNA-binding domain, abrogates E2F1-mediated activation of the COX-2 promoter. We suggest that c-myb mediates E2F1-directed transcription of COX-2. Because c-myb is an oncogenic transcription factor and is overexpressed in human leukemias (51), lung cancer (52), and colon cancer (53), the expression and activity of c-myb should be examined in prostate tumorigenesis.

In summary, we have in vitro and in vivo evidence to show that deletion of Rb and increased expression of E2F1 result in increased COX-2 expression and function in prostate epithelial cells. COX-2 expression correlates with increased PGE2 synthesis. Our data indicate that NS-398 can directly target epithelial cells with loss of Rb and increased COX-2 expression, thereby supporting the role for COX-2 inhibitors in prostate chemoprevention. Finally, the identification of E2F1 as a transcriptional regulator of COX-2 implicates that E2F1 has multiple targets in addition to cell cycle regulatory proteins that can contribute to cancer development and progression, possibly through up-regulation of the oncogene, c-myb.


   Acknowledgments
 
Grant support: American Cancer Society-Clyde Dixon Trust Postdoctoral Fellowship, PF-03-252-01-TBE (J.N. Davis), U01 CA96403 (S.W. Hayward), and NIH grants DK 061488 and DK 066610 (M.L. Day).

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 8/30/04. Revised 12/17/04. Accepted 1/ 5/05.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Smith WL, DeWitt DL, Garavito RM. Cyclooxygenases: structural, cellular, and molecular biology. Annu Rev Biochem 2000;69:145–82.[CrossRef][Medline]
  2. Vane JR, Bakhle YS, Botting RM. Cyclooxygenases 1 and 2. Annu Rev Pharmacol Toxicol 1998;38:97–120.[CrossRef][Medline]
  3. Sano H, Kawahito Y, Wilder RL, et al. Expression of cyclooxygenase-1 and -2 in human colorectal cancer. Cancer Res 1995;55:3785–9.[Abstract/Free Full Text]
  4. Wolff H, Saukkonen K, Anttila S, Karjalainen A, Vainio H, Ristimaki A. Expression of cyclooxygenase-2 in human lung carcinoma. Cancer Res 1998;58:4997–5001.[Abstract/Free Full Text]
  5. Nagai N, Tian X, Mukai K, et al. Overexpression of cyclooxygenase-2 protein and its relationship to apoptosis in cervical carcinoma treated with neoadjuvant chemotherapy. Int J Mol Med 2003;12:709–14.[Medline]
  6. Hwang D, Scollard D, Byrne J, Levine E. Expression of cyclooxygenase-1 and cyclooxygenase-2 in human breast cancer. J Natl Cancer Inst 1998;90:455–60.[Abstract/Free Full Text]
  7. Fujita H, Koshida K, Keller ET, et al. Cyclooxygenase-2 promotes prostate cancer progression. Prostate 2002;53:232–40.[CrossRef][Medline]
  8. Oshima M, Dinchuk JE, Kargman SL, et al. Suppression of intestinal polyposis in Apc {delta}716 knockout mice by inhibition of cyclooxygenase 2 (COX-2). Cell 1996;87:803–9.[CrossRef][Medline]
  9. Chulada PC, Thompson MB, Mahler JF, et al. Genetic disruption of Ptgs-1, as well as Ptgs-2, reduces intestinal tumorigenesis in Min mice. Cancer Res 2000;60:4705–8.[Abstract/Free Full Text]
  10. Gupta S, Srivastava M, Ahmad N, Bostwick DG, Mukhtar H. Over-expression of cyclooxygenase-2 in human prostate adenocarcinoma. Prostate 2000;42:73–8.[CrossRef][Medline]
  11. Yoshimura R, Sano H, Masuda C, et al. Expression of cyclooxygenase-2 in prostate carcinoma. Cancer 2000;89:589–96.[CrossRef][Medline]
  12. Madaan S, Abel PD, Chaudhary KS, et al. Cytoplasmic induction and over-expression of cyclooxygenase-2 in human prostate cancer: implications for prevention and treatment. BJU Int 2000;86:736–41.[CrossRef][Medline]
  13. Zha S, Gage WR, Sauvageot J, et al. Cyclooxygenase-2 is up-regulated in proliferative inflammatory atrophy of the prostate, but not in prostate carcinoma. Cancer Res 2001;61:8617–23.[Abstract/Free Full Text]
  14. Subbarayan V, Sabichi AL, Llansa N, Lippman SM, Menter DG. Differential expression of cyclooxygenase-2 and its regulation by tumor necrosis factor-{alpha} in normal and malignant prostate cells. Cancer Res 2001;61:2720–6.[Abstract/Free Full Text]
  15. Wang W, Bergh A, Damber JE. Chronic inflammation in benign prostate hyperplasia is associated with focal upregulation of cyclooxygenase-2, Bcl-2, and cell proliferation in the glandular epithelium. Prostate 2004;61:60–72.[CrossRef][Medline]
  16. Gupta S, Adhami VM, Subbarayan M, et al. Suppression of prostate carcinogenesis by dietary supplementation of celecoxib in transgenic adenocarcinoma of the mouse prostate model. Cancer Res 2004;64:3334–43.[Abstract/Free Full Text]
  17. Shappell SB, Olson SJ, Hannah SE, et al. Elevated expression of 12/15-lipoxygenase and cyclooxygenase-2 in a transgenic mouse model of prostate carcinoma. Cancer Res 2003;63:2256–67.[Abstract/Free Full Text]
  18. Phillips SM, Morton DG, Lee SJ, Wallace DM, Neoptolemos JP. Loss of heterozygosity of the retinoblastoma and adenomatous polyposis susceptibility gene loci and in chromosomes 10p, 10q and 16q in human prostate cancer. Br J Urol 1994;73:390–5.[Medline]
  19. Pietilainen T, Lipponen P, Aaltomaa S, Eskelinen M, Kosma VM, Syrjanen K. Expression of retinoblastoma gene protein (Rb) in breast cancer as related to established prognostic factors and survival. Eur J Cancer 1995;31A:329–33.
  20. Gope R, Gope ML. Abundance and state of phosphorylation of the retinoblastoma susceptibility gene product in human colon cancer. Mol Cell Biochem 1992;110:123–33.[CrossRef][Medline]
  21. Primdahl H, von der Maase H, Christensen M, Wolf H, Orntoft TF. Allelic deletions of cell growth regulators during progression of bladder cancer. Cancer Res 2000;60:6623–9.[Abstract/Free Full Text]
  22. Nevins JR. Toward an understanding of the functional complexity of the E2F and retinoblastoma families. Cell Growth Differ 1998;9:585–93.[Medline]
  23. Sherr CJ. Cancer cell cycles. Science 1996;274:1672–7.[Abstract/Free Full Text]
  24. Pierce AM, Gimenez-Conti IB, Schneider-Broussard R, Martinez LA, Conti CJ, Johnson DG. Increased E2F1 activity induces skin tumors in mice heterozygous and nullizygous for p53. Proc Natl Acad Sci U S A 1998;95:8858–63.[Abstract/Free Full Text]
  25. Pierce AM, Schneider-Broussard R, Gimenez-Conti IB, Russell JL, Conti CJ, Johnson DG. E2F1 has both oncogenic and tumor-suppressive properties in a transgenic model. Mol Cell Biol 1999;19:6408–14.[Abstract/Free Full Text]
  26. Pierce AM, Fisher SM, Conti CJ, Johnson DG. Deregulated expression of E2F1 induces hyperplasia and cooperates with ras in skin tumor development. Oncogene 1998;16:1267–76.[CrossRef][Medline]
  27. Zhang SY, Liu SC, Johnson DG, Klein-Szanto AJ. E2F-1 gene transfer enhances invasiveness of human head and neck carcinoma cell lines. Cancer Res 2000;60:5972–6.[Abstract/Free Full Text]
  28. Zhang SY, Liu SC, Al-Saleem LF, et al. E2F-1: a proliferative marker of breast neoplasia. Cancer Epidemiol Biomarkers Prev 2000;9:395–401.[Abstract/Free Full Text]
  29. Gorgoulis VG, Zacharatos P, Mariatos G, et al. Transcription factor E2F-1 acts as a growth-promoting factor and is associated with adverse prognosis in non-small cell lung carcinomas. J Pathol 2002;198:142–56.[CrossRef][Medline]
  30. Etges A, Nunes FD, Ribeiro KC, Araujo VC. Immunohistochemical expression of retinoblastoma pathway proteins in normal salivary glands and in salivary gland tumours. Oral Oncol 2004;40:326–31.[CrossRef][Medline]
  31. Ma Y, Croxton R, Moorer RL Jr, Cress WD. Identification of novel E2F1-regulated genes by microarray. Arch Biochem Biophys 2002;399:212–24.[CrossRef][Medline]
  32. Day KC, McCabe MT, Zhao X, et al. Rescue of embryonic epithelium reveals that the homozygous deletion of the retinoblastoma gene confers growth factor independence and immortality but does not influence epithelial differentiation or tissue morphogenesis. J Biol Chem 2002;277:44475–84.[Abstract/Free Full Text]
  33. Wang Y, Hayward SW, Donjacour AA, et al. Sex hormone-induced carcinogenesis in Rb-deficient prostate tissue. Cancer Res 2000;60:6008–17.[Abstract/Free Full Text]
  34. Liu CH, Chang SH, Narko K, et al. Overexpression of cyclooxygenase-2 is sufficient to induce tumorigenesis in transgenic mice. J Biol Chem 2001;276:18563–9.[Abstract/Free Full Text]
  35. Tjandrawinata RR, Dahiya R, Hughes-Fulford M. Induction of cyclo-oxygenase-2 mRNA by prostaglandin E2 in human prostatic carcinoma cells. Br J Cancer 1997;75:1111–8.[Medline]
  36. Liu XH, Kirschenbaum A, Lu M, et al. Prostaglandin E(2) stimulates prostatic intraepithelial neoplasia cell growth through activation of the interleukin-6/GP130/STAT-3 signaling pathway. Biochem Biophys Res Commun 2002;290:249–55.[CrossRef][Medline]
  37. Macleod KF, Hu Y, Jacks T. Loss of Rb activates both p53-dependent and independent cell death pathways in the developing mouse nervous system. EMBO J 1996;15:6178–88.[Medline]
  38. Ruiz S, Santos M, Segrelles C, et al. Unique and overlapping functions of pRb and p107 in the control of proliferation and differentiation in epidermis. Development 2004;131:2737–48.[Abstract/Free Full Text]
  39. Neuman E, Flemington EK, Sellers WR, Kaelin WG Jr. Transcription of the E2F-1 gene is rendered cell cycle dependent by E2F DNA-binding sites within its promoter. Mol Cell Biol 1994;14:6607–15.[Abstract/Free Full Text]
  40. Sage J, Miller AL, Perez-Mancera PA, Wysocki JM, Jacks T. Acute mutation of retinoblastoma gene function is sufficient for cell cycle re-entry. Nature 2003;424:223–8.[CrossRef][Medline]
  41. Black EP, Huang E, Dressman H, et al. Distinct gene expression phenotypes of cells lacking Rb and Rb family members. Cancer Res 2003;63:3716–23.[Abstract/Free Full Text]
  42. Ishida S, Huang E, Zuzan H, et al. Role for E2F in control of both DNA replication and mitotic functions as revealed from DNA microarray analysis. Mol Cell Biol 2001;21:4684–99.[Abstract/Free Full Text]
  43. Ramsay RG, Friend A, Vizantios Y, et al. Cyclooxygenase-2, a colorectal cancer nonsteroidal anti-inflammatory drug target, is regulated by c-MYB. Cancer Res 2000;60:1805–9.[Abstract/Free Full Text]
  44. Sala A, Nicolaides NC, Engelhard A, et al. Correlation between E2F-1 requirement in the S phase and E2F-1 transactivation of cell cycle-related genes in human cells. Cancer Res 1994;54:1402–6.[Abstract/Free Full Text]
  45. Hsu AL, Ching TT, Wang DS, Song X, Rangnekar VM, Chen CS. The cyclooxygenase-2 inhibitor celecoxib induces apoptosis by blocking Akt activation in human prostate cancer cells independently of Bcl-2. J Biol Chem 2000;275:11397–403.[Abstract/Free Full Text]
  46. Nikolic D, van Breemen RB. DNA oxidation induced by cyclooxygenase-2. Chem Res Toxicol 2001;14:351–4.[CrossRef][Medline]
  47. Leung WK, To KF, Go MY, et al. Cyclooxygenase-2 upregulates vascular endothelial growth factor expression and angiogenesis in human gastric carcinoma. Int J Oncol 2003;23:1317–22.[Medline]
  48. Crawford YG, Gauthier ML, Joubel A, et al. Histologically normal human mammary epithelia with silenced p16(INK4a) overexpress COX-2, promoting a premalignant program. Cancer Cell 2004;5:263–73.[CrossRef][Medline]
  49. Hurford RK Jr, Cobrinik D, Lee MH, Dyson N. pRB and p107/p130 are required for the regulated expression of different sets of E2F responsive genes. Genes Dev 1997;11:1447–63.[Abstract/Free Full Text]
  50. Campanero MR, Armstrong M, Flemington E. Distinct cellular factors regulate the c-myb promoter through its E2F element. Mol Cell Biol 1999;19:8442–50.[Abstract/Free Full Text]
  51. Slamon DJ, Boone TC, Murdock DC, et al. Studies of the human c-myb gene and its product in human acute leukemias. Science 1986;233:347–51.[Abstract/Free Full Text]
  52. Griffin CA, Baylin SB. Expression of the c-myb oncogene in human small cell lung carcinoma. Cancer Res 1985;45:272–5.[Abstract/Free Full Text]
  53. Alitalo K, Winqvist R, Lin CC, de la Chapelle A, Schwab M, Bishop JM. Aberrant expression of an amplified c-myb oncogene in two cell lines from a colon carcinoma. Proc Natl Acad Sci U S A 1984;81:4534–8.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Cancer Res.Home page
J. N. Davis, K. J. Wojno, S. Daignault, M. D. Hofer, R. Kuefer, M. A. Rubin, and M. L. Day
Elevated E2F1 Inhibits Transcription of the Androgen Receptor in Metastatic Hormone-Resistant Prostate Cancer
Cancer Res., December 15, 2006; 66(24): 11897 - 11906.
[Abstract] [Full Text] [PDF]

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 Email this article to a friend
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 Davis, J. N.
Right arrow Articles by Day, M. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Davis, J. N.
Right arrow Articles by Day, M. L.

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