In Prostate Cancer Cells

ANTICANCER RESEARCH 27: 2393-2402 (2007)

Anticancer Effects of Licofelone (ML-3000) in Prostate Cancer Cells

NARAYANAN K. NARAYANAN1, DOMINIC NARGI1, MUKUNDAN ATTUR2, STEVEN B. ABRAMSON2,3 and BHAGAVATHI A. NARAYANAN1

1Department of Environmental Medicine, 2Hospital for Joint Diseases, 3Department of Medicine (Rheumatology) and Pathology, New York University School of Medicine, New York, NY, U.S.A.

Abstract. Background: Licofelone, a potent anti- prostate cancer among men who are regularly taking aspirin inflammatory agent has been reported to interfere with the or other nonsteroidal anti-inflammatory drugs (NSAIDs) (4- cyclooxygenase-2 (COX-2) and 5-lipoxygenase (5-LOX) 7). Despite the potential benefits, side-effects due to the use signaling pathways with few side-effects. However, the of higher doses of cyclooxygenase-2 (COX-2) inhibitors are underlying mechanism of licofelone against human cancer is a major concern (8-9). A growing number of studies have not understood. Materials and Methods: Human and mouse indicated that the metabolism of arachidonic acid (AA) prostate cancer cells were exposed to licofelone in a time- and either by the COX or the lipoxygenase (LOX) pathways dose-dependent manner. Cell growth/cell viability, apoptosis, generates eicosanoids involved in the tumor promotion, and expression of COX-2 and 5-LOX at the gene and protein progression and metastasis of prostate cancer (10, 11). It levels were investigated. Results: For the first time, it was has been pointed out that inhibiting only one of these demonstrated that licofelone inhibited prostate cancer cell COX/LOX biosynthetic pathways cannot disrupt the growth and significantly down-regulated COX-2 and 5-LOX metabolism of AA involved in human carcinogenesis (12). expression. A weak inhibitory effect on COX-1 protein was also Increasing interest in the simultaneous blocking of the observed. Conclusion: Licofelone inhibited COX-2 and 5-LOX COX/LOX pathways by interfering with the production of simultaneously and prevented overall cancer cell growth by both prostaglandins and the biosynthesis of leukotrienes enhancing apoptosis in both androgen-dependent and (LTs) is emerging as a promising approach for cancer androgen-independent prostate cancer cells. Validating the chemoprevention and treatment. dual role of licofelone in animal models of prostate cancer is Licofelone ([2,2-dimethyl-6-(4-chlorophenyl)-7-phenyl- critical for promoting its use as a potential chemopreventive or 2,3-dihydro-1H-pyrrolizine-5-yl]-acetic acid), previously therapeutic agent. named as ML3000, (Figure 1) has been demonstrated to inhibit COX-2 and 5-LOX. This dual COX/LOX inhibitor Prostate cancer is one of the most common malignancies has been shown to be an effective anti-inflammatory agent among men, and is the second leading cause of cancer death against carrageenan-arachidonic acid- and bradykinin- in the United States (1, 2). African-American men have far induced paws edema and bradykinin-induced mechanical higher rates of prostate cancer incidence in the U.S. hyperalgesia (13-17). Licofelone is currently under compared to their white counterparts (3). Most importantly, evaluation as a treatment for osteoarthritis (OA), the most epidemiological reports have shown a decreased risk of common form of arthritis, and has been shown to be effective when administered orally and well tolerated on single or repeated administration (15). The pharmacological Abbreviations: COX-2, cyclooxygenase-2; 5-LOX, 5-lipoxygenase; profile of licofelone against arthritis has been well VEGF, vascular endothelial growth factor; TRAMP, transgenic characterized based on the COX-1, COX-2 and 5-LOX adenocarcinoma of the mouse prostate. enzyme activities (13, 18). Licofelone exhibits anti- inflammatory and analgesic activity comparable to that of Correspondence to: Narayanan K. Narayanan, Ph.D., Assistant conventional NSAIDs and has also shown an improved Professor, Department of Environmental Medicine, New York gastrointestinal safety profile (15, 19, 20). Based on these University School of Medicine, Tuxedo, NY, 10987, U.S.A. Tel: +1 845 earlier findings, it was decided to test its effect against 731 3625, Fax: +1 845 351 4510, e-mail: [email protected] prostate cancer involving AA metabolism. To determine the Key Words: Licofelone, COX/LOX inhibitor, NSAIDs, TRAMP, chemopreventive efficacy of licofelone, cell viability was prostate cancer. determined after the application of various concentrations

0250-7005/2007 $2.00+.40 2393 ANTICANCER RESEARCH 27: 2393-2402 (2007)

Cell viability determination. Cell viability after licofelone treatment was determined by the trypan blue (0.4%) exclusion assay (22). After 6, 12, 24, and 48 h exposure to 2.5, 5, 10, 20, 30, 40 or 50 ÌM licofelone, adherent and floating cells were harvested by trypsinization and recovered by centrifugation. Trypan blue staining of the cells enabled easy identification of dead cells because they took up the dye and appeared blue with uneven cell membranes. By contrast, the living cells repelled the dye and appeared retractile and colorless. The viable cells from three parallel sets of experiments were counted to determine the % of cell survival, and to compare the rate of cell growth inhibition between the control (DMSO) and licofelone treatment.

Figure 1. Structure of licofelone (ML-3000). Detection of apoptosis by DAPI and annexinV staining. The rate of apoptosis induced by licofelone in prostate cancer cells was assessed first with DAPI staining of the nuclear material. Briefly, cells grown in 35-mm dishes were treated with licofelone 10 ÌM (IC50) of licofelone on human prostate cancer cells and cells (determined by a dose-response study) for 48 h, and then the derived from a transgenic adenocarcinoma of the mouse floating and adherent cells were fixed in 10% formalin for 15 min. prostate (TRAMP). Its effect on COX-2, 5-LOX and VEGF After washing with phosphate-buffered saline (PBS), the cells were treated with 0.1% Triton X-100, 4 M HCl, and sodium tetraborate. (vascular endothelial growth factor) expression, both at the Each treatment was performed for 15 min and was followed by a mRNA and protein level, and on other proinflammatory PBS wash. The cells were then stained with DAPI in 80% methanol genes at the transcription level was examined. for 30 min and again washed with PBS, and were viewed under a fluorescence microscope as described previously by Narayanan et al. Materials and Methods (22). In order to detect early apoptotic cells induced by licofelone, parallel experiments with annexin V staining of the membrane for Cell culture and treatments. The cancer cell types used in this study PS (phosphatidylserine externalization) were performed in the included human prostate cancer cells (PC-3) obtained from the control and licofelone-treated cells. The cells showing positive American Type Culture Collection (Manassas, VA, USA) and staining for annexin V with the characteristic morphological human benign prostatic hyperplasia (BPH-1) cells, a gift from changes of apoptosis were detected using high power microscopy Hayward et al. (21). These cells were grown in RPMI (Gibco, CA, and quantified using 100 cells per field for 10 fields. USA) with 5% fetal bovine serum (FBS) in standard cell culture conditions (22). The TRAMP-derived mouse prostate cancer cell Immunohistochemical detection of COX-2 and 5-LOX in TRAMP line TR-75 used in the present study was established at Dr. tissues. Paraffin-embedded, 5-Ìm thickness dorsolateral (DL) Bhagavathi Narayanan's laboratory for ongoing studies (23). These prostate tissue sections of TRAMP mice were used for all the cells were grown in DMEM (F12) high glucose W/L-glutamine immunohistochemical analyses. Following rehydration, the antigens media supplemented with 5% Nu-Serum IV, a growth medium were retrieved by a process which involved microwaving with supplement from BD Biosciences (San Jose, CA, USA) and 5% antigen-unmasking fluid (Vector Laboratories CA, USA) twice for FBS, 5 Ìg/ml insulin, 25 units/ml of penicillin-streptomycin and 10–8 5 min with a 3-min interval. After 15 min at room temperature the M dihydrotestosterone (23, 24). A uniform number of cells plated in sections were washed and blocked with 10% normal horse serum. T-25 and/or 35-mm cell-culture dishes were placed in a humidified The sections were then incubated for 1 h with primary mouse anti- incubator at 37ÆC with 5% CO2 before treatment. All the cells that COX-2 (Cayman, Ann Arbor, MI, USA), and anti-5-LOX reached 75% confluence were used for various treatments. (Biomole, Philadelphia, PA, USA) antibodies at room Dr. Mukundan Attur and Dr. Steven B. Abramson at the Hospital temperature. The overall expression levels of COX-2 and 5-LOX for Joint Diseases, New York University School of Medicine, were detected using a universal labeling kit (Hrp/DAB) from provided the compound licofelone (ML-3000) for this study. The Ventana Medical Systems (Tucson, AZ, USA). The Image Pro range of licofelone concentrations selected for this study was based software program (Media Cybernetics, Silver Spring, MD, USA) on various previously reported doses used in several cell culture was used to quantify the total number of positively stained cells in studies, namely human umbilical vein endothelial cells (14), rat a minimum of using 100 cells per field for 10 fields at x40 basophilic leukemia cells (25), human osteoarthritic chondrocytes magnification (28). (26) and gastric parietal cells (27). The licofelone was dissolved in DMSO and added to the cell culture media to arrive at the final Immunoflourescence detection of COX-2 and 5-LOX in TR-75 cells concentrations and incubated with the cells for various lengths of of TRAMP. In order to determine the effect of licofelone on time. Profiling the pharmacologically effective dose of licofelone on COX-2 and 5-LOX expression, immunofluorescence detection COX-2 and 5-LOX enzyme activity was not an objective of this study, procedures were used with TR-75 cells treated with 10 ÌM since the main focus was to demonstrate the anticancer effect of licofelone. Briefly, cells grown in 35 mm dishes treated with licofelone and its ability to modulate genes and proteins associated licofelone for 48 h were washed in PBS, fixed in 10% formalin, with prostate cancer cell growth. Controls received 1% DMSO only. and pretreated with 0.1% Triton X-100 for 15 min each. After All the experiments were repeated three times for confirmation. blocking with 1% bovine serum albumin (BSA) for an hour,

2394 Narayanan et al: Anticancer Effects of Licofelone

Figure 2. TR-75 cells showing morphological changes associated with cell growth inhibition and reduced cell survival rate with different doses of licofelone treatment for 48 h. A, control; licofelone treatment: B, 2.5 ÌM; C, 5 ÌM; D, 10 ÌM; E, 20 ÌM; F, 30 ÌM. At higher concentrations of 40 and 50 ÌM, fewer adherents and more floating cells were observed (Figures not shown).

COX-1, COX-2 and 5-LOX expression was detected after Western blot analysis. Prostate cancer cells TR-75 and PC-3 incubating the cells with the specific antibody conjugated to FITC treated with 10 ÌM licofelone for 48 h were harvested by (Cayman, Ann Arbor MI, USA) for 1 h at room temperature. trypsinization. The total protein was isolated with protein Green fluorescence signaling of COX-1, COX-2 and 5-LOX extraction buffer containing 150 mM NaCl, 10 mM Tris (pH 7.2), expression with reference to DNA staining with DAPI was viewed 5 mM EDTA, 0.1% Triton X-100, 5% glycerol and 2% sodium under an Olympus AX-70 epifluorescence microscope (Olympus dodecyl sulphate (SDS), in addition to a mixture of protease America Inc., Center Valley, PA, USA) as described previously inhibitors (Boehringer Mannheim, GmbH, Mannheim, Germany). (29). Quantification of the green fluorescence was used to Equal amounts of protein (50 Ìg/lane) were fractionated on 10% measure the level of expression. SDS-polyacrylamide gel electrophoresis (PAGE) gels and transferred to PVDF (polyvinylidene difluoride) membranes. The RNA isolation and quantitative real-time PCR analysis. A two-step Western blot procedure was carried out as described previously real-time reverse transcription-polymerase chain reaction (RT- (29). The 5-LOX and VEGF antibodies were purchased from PCR) was carried out using the total RNA extracted from the TR- Biomole (Philadelphia, PA, USA) and Santa Cruz Biotechnology, 75 cells as described previously (23). Briefly, 5 Ìg of the total RNA (Santa Cruz, CA, USA), respectively. The antibodies for COX-1 extracted from cells treated with 10 ÌM licofelone for 24 h were and COX-2 were purchased from Cayman Chemicals (USA). The subjected to real-time RT-PCR analysis using gene-specific primer reactive protein bands developed were detected using sequences for COX-1, COX-2, prostaglandin E receptor 2 chemiluminescence reagents (ECL from GE Healthcare (PTGER2), 5-LOX, 12-LOX, VEGF, NF-kBp65, TNF·, with Amersham Bio-Sciences, Piscataway, NJ, USA). Densitometric amplification of GAPDH (glyceraldehyde-3-phosphate analysis for quantification of the protein bands was performed dehydrogenase) used as the internal control. All of the templates with the software Gel-Pro Analyzer (Media Cybernetics). were initially denatured for 2 min at 94ÆC, and the amplification was extended at a final temperature of 72ÆC for 7 min. Real-time Statistical analysis. The significant differences in the cellular effects RT-PCR analysis using a PCR reaction mix containing Syber green induced by licofelone including cell growth inhibition, apoptosis, intercalating dye (Bio-Rad Laboratories, Hercules, CA, USA) was and expression of COX-2 and 5-LOX between control and carried out to evaluate the mRNA expression. PCR amplification licofelone-treated cells were compared using one-way analysis of was performed using Cepheid Smart Cycler II (Cepheid, variance (ANOVA) followed by Tukey's multiple comparisons Sunnyvale, CA, USA). procedure (30).

2395 ANTICANCER RESEARCH 27: 2393-2402 (2007)

Figure 4. Licofelone induced apoptosis in prostate cancer cells. Annexin V staining was used to detect the rate of apoptosis as described in Materials and Methods. Control cells were treated with DMSO. The results presented are from three sets of similar experiments.

Results

Licofelone inhibits prostate cancer cell growth. Although cell viability assays were performed to determine the effect of licofelone against prostate cancer cell growth for various time-points, the results are presented only for 48 h. Phase- contrast microscopic observations revealed a dose- dependent effect on growth inhibition in the TR-75 cells with an IC50 of 10 ÌM (Figures 2 and 3). Significant growth inhibition was observed at 5, 10 and 20 ÌM concentrations (p<0.01). However, at higher concentrations of 30, 40 and 50 ÌM, fewer adherent, but more floating cells were observed. In addition, PC-3 and BPH-1 cells also revealed a similar trend in cell growth inhibition with licofelone treatment (Figure 3).

Licofelone induces apoptotic cell death. In order to confirm whether licofelone-induced cell growth inhibition was mediated through apoptosis, the rate of apoptosis was measured by annexin V staining, which determines early apoptotic cells characterized by phosphatidylserine externalization (PS). All the three prostate cancer cells (TR- 75, PC-3 and BPH cells) treated with 10 ÌM licofelone for 48 h showed a higher number of early apoptotic cells compared to that of the control. A semi-quantification of apoptotic cells revealed a significant increase in the rate of apoptosis in all three cell types (p<0.001) and the results Figure 3. Dose-dependent effect of licofelone ranging from 2.5 to 50 ÌM are presented in Figure 4. concentrations for 48 h. The results presented indicate cell growth inhibition in terms of cell survival in human prostate cancer cells (PC-3), TRAMP tissues express COX-2 and 5-LOX. COX-2 and cells from benign prostatic hyperplasia (BPH) and TR-75 cells. For each cell type, three parallel sets of experiments were conducted. The viable cells 5-LOX expression in TRAMP tissues was confirmed by (%) were determined by a trypan blue exclusion assay as described in the immunohistochemical detection using specific antibodies. Materials and Methods section. As shown in Figure 5, a higher expression of both COX-2

2396 Narayanan et al: Anticancer Effects of Licofelone

Figure 5. COX-2 and 5-LOX expression in TRAMP tissues. A, H&E staining of paraffin-embedded prostate tissue sections indicating the cellular changes, associated with PIN and adenocarcinoma stages of TRAMP (A1, A2). The immunohistochemical detection shows the COX-2 (A3, A4) and 5-LOX (A5, A6) expression in TRAMP tissues. B, The chart illustrates the quantification of the mean number of positively stained cells per field for COX-2 and 5-LOX expression in the PIN and adenocarinoma of the TRAMP at x40 magnification.

2397 ANTICANCER RESEARCH 27: 2393-2402 (2007)

Figure 6. Cellular localization of COX-2 and 5-LOX in TR-75 cells. The bar graph represents the percentage of immunofluorescent cells for COX-1, COX-2 and 5-LOX in TR-75 cells exposed to licofelone (10 ÌM) for 48 h. DAPI was used for the DNA staining as control. FITC conjugated COX-2 and 5-LOX mouse monoclonal antibodies were used to detect the levels of protein expression in the cytoplasm and/or in the nucleus.

and 5-LOX was evident in the TRAMP tissue than in the PTGER2, 5-LOX, 12-LOX, NF-κBp65 and VEGF, while PIN, suggesting that the cells derived from TRAMP tumor indicating a small change in the expression levels of were suitable for testing the effect of licofelone against GAPDH and COX-1 compared to the level in the untreated prostate cancer. Haematoxylin and eosin-stained prostate control. These preliminary findings suggest a definite anti- tissue sections indicated the cellular changes associated with inflammatory effect of licofelone against prostate cancer. the PIN and adenocarcinoma stages of the mouse prostate However, the lack of impact on TNF· may be indicative of (Figure 5). yet another mode of action.

Licofelone inhibits COX-2 and 5-LOX in TRAMP cells. Licofelone inhibits COX-2 and 5-LOX protein expression. Immunofluorescence detection indicated that cells exposed Western blot analysis showed a marked inhibition of COX-2 to 10 ÌM licofelone for 48 h significantly (p<001) reduced and 5-LOX protein in both the TR-75 and PC-3 cells treated COX-2 and 5-LOX expression (Figure 6) compared to that with 10 Ìm licofelone compared to the controls (Figure 8). It in the control cells. Our results indicated that there is no is also important to mention here that no significant effect on significant effect on COX-1 expression in the TR-75 cells COX-1 protein was seen in the licofelone-treated cells. with licofelone treatment. Discussion Licofelone alters the transcription levels of proinflammatory genes. In order to determine the impact of licofelone on key Chemopreventive agents of both natural and pro-inflammatory gene targets at the transcription level, a pharmaceutical sources are known to play a critical role two-step real-time RT-PCR was carried out with total RNA in prostate cancer prevention (31-36). In this study extracted from the TR-75 cells. RT-PCR was performed as licofelone clearly demonstrated anti-inflammatory activity described earlier (23). As shown in Figure 7, licofelone- by blocking the 5-LOX and COX-2 expression and treated TR-75 cells showed down-regulation of COX-2, inhibited human and mouse prostate cancer cell growth in

2398 Narayanan et al: Anticancer Effects of Licofelone

Figure 7. RT-PCR. A two-step real-time RT-PCR was carried out with 5 Ìg of total RNA extracted from TR-75 cells treated with 10 ÌM of licofelone for 24 h, as described in the Materials and Methods section. Real-time RT-PCR amplification of COX-1, COX-2, prostaglandin E receptor 2 (PTGER2), 5- LOX, 12-LOX, VEGF, TNF· and NF-kBp65 were determined using gene-specific primers. Amplification of GAPDH was used as the internal control. Real-time PCR amplification of the gene products were indicated by the Syber green signal intensity, as described by Narayanan et al. (21). Note: The number of PCR cycles is inversely related to the amount of mRNA.

a dose-dependent manner. By using both androgen- licofelone could be mediated by enhancing the apoptosis dependent TR-75 and androgen-independent PC-3 mechanism and abrogating the antiapoptotic influence of prostate cancer cells, overall cancer cell growth inhibition COX-2 and 5-LOX, but sparing the COX-1 expression. induced by licofelone has been demonstrated for the first Further analysis of down-regulated pro-inflammatory time. These findings indicate its potential role in genes including COX-2 and 5-LOX along with other preventing early and metastatic prostate cancer, whether mediators of inflammation in TR-75 cells, clearly showed it is driven by hormone dependent or independent factors. the effect of licofelone as an anti-inflammatory agent Most importantly, our findings suggest that the effect of against prostate cancer.

2399 ANTICANCER RESEARCH 27: 2393-2402 (2007)

Figure 8. Western blot analysis. COX-2 and 5-LOX protein expressions were analyzed using specific antibody as described in the Materials and Methods section. A, TR-75 and B, PC-3 cells showing reduced COX-2 and 5-LOX protein expression and bar graph representing the quantification of the respective protein expression levels.

The debate on whether selective inhibition of COX-2 before choosing dual inhibitors for clinical use. Although with or without concomitant inhibition of COX-1 or significant progress has been made in understanding the whether selective inhibition of the lipoxygenase pathway role of bioactive lipids in cancer prevention, many could affect the reproductive tissues is still inconclusive. questions remain unanswered. Ensuring an appropriate The role of COX/LOX inhibitors, such as licofelone, risk-to-benefit ratio remains a significant challenge for the present major advantages when compared to the selective development of viable chemopreventives targeting COX-2 inhibitors in that they act on two major pathways, inflammatory pathways. As the uncertainty over moreover associated with AA metabolism; the dual cardiovascular complications with selective COX-2 inhibitors are reported to have no gastric toxicity (37, 38). inhibitors indicates the need for alternative agents, Most importantly, however, all in vivo data must be taken validating the dual role of licofelone in animal models of into account to understand the comparative efficacy of prostate cancer is very critical for promoting its use as a individual agents and safety as indicated by Clark (39) potential chemopreventive or therapeutic agent for

2400 Narayanan et al: Anticancer Effects of Licofelone prostate cancer. Most of the present data on the effect of 13 Singh VP, Patil CS and Kulkarni SK: Effect of licofelone licofelone on several of the molecular targets has been against NSAIDs-induced gastrointestinal ulceration and investigated using TR-75 cells because of our future inflammation. Indian J Exp Biol 43: 247-253, 2005. 14 Ulbrich H, Soehnlein O, Xie X, Eriksson EE, Lindbom L, interest in testing the efficacy of licofelone in the TRAMP Albrecht W, Laufer S and Dannhardt G: Licofelone, a novel 5- model assay for prostate cancer. LOX/COX-inhibitor, attenuates leukocyte rolling and adhesion on endothelium under flow. Biochem Pharmacol 70: 30-36, 2005. Acknowledgements 15 Laufer S, Augustin J, Dannhardt G and Kiefer W: (6,7- Diaryldihydropyrrolizin-5-yl) acetic acids, a novel class of potent We acknowledge the support in part by USPHS grant CA106296 dual inhibitors of both cyclooxygenase and 5-lipoxygenase. J from the National Cancer Institute to Dr. B. Narayanan for this Med Chem 37: 1894-1897, 1994. study. We thank Helen Duss at NYU School of Medicine for 16 Laufer S, Tries SJ, Augustin J and Dannhardt G: Pharmacological editing the manuscript. profile of a new pyrrolizine derivative inhibiting the enzymes cyclo-oxygenase and 5-lipoxygenase, Arzneim.-Forsch./Drug Res 44: 629-636, 1994. References 17 Cicero AFG, Derosa G and Gaddi A: Combined lipoxygenase/ cyclo-oxygenase inhibition in the elderly: the example of 1 Jemal A, Siegel R, Ward E, Murray T, Xu J, Smigal C and Thun licofelone. Drugs Aging 22: 393-403, 2005. MJ: Cancer statistics 2006. CA Cancer J Clin 56: 106-130, 2006. 18 Hampton T: Arthritis clinical trial results revealed. JAMA 297: 2 Hsing AW and Chokkalingam AP: Prostate cancer 28-29, 2007. epidemiology. Front Biosci 11: 1388-1413, 2006. 19 Shureiqi I, Chen D, Lotan R, Yang P, Newman RA, Fischer 3 Powell IJ: Epidemiology and pathophysiology of prostate cancer SM and Lippman SM: 15-lipoxygenase-1 mediates nonsteroidal in African-American men. J Urol 77: 444-449, 2007. anti-inflammatory drug-induced apoptosis independently of 4 Kelloff GJ, Lieberman R, Steele VE, Boone CW, Lubet RA, cyclooxygenase-2 in colon cancer cells. Cancer Res 60: 6846- Kopelovich L, Malone WA, Crowell JA, Higley HR and Sigman 6850, 2000. CC: Agents, biomarkers, and cohorts for chemopreventive 20 Bias P, Buchner A, Klesser B and Laufer S: The gastrointestinal agent development in prostate cancer. Urology 57: 46-51, 2001. tolerability of the LOX/COX inhibitor, licofelone, is similar to 5 Norrish AE, Jackson RT and McRae CU: Non-steroidal anti- placebo and superior to naproxen therapy in healthy volunteers: inflammatory drugs and prostate cancer progression. Int J results from a randomized, controlled trial. Am J Gastroenterol Cancer 77: 511-515, 1998. 99: 611-618, 2004. 6 Nelson JE and Harris RE: Inverse association of prostate 21 Hayward SW, Dahiya R, Cunha GR, Bartek J, Deshpande N cancer and non-steroidal anti-inflammatory drugs (NSAIDs): and Narayan P: Establishment and characterization of an results of a case-control study. Oncol Rep 7: 169-170, 2000. immortalized but non-transformed human prostate epithelial 7 Waskewich C, Blumenthal RD, Li H, Stein R, Goldenberg DM cell line: BPH-1. In Vitro Cell Dev Biol Anim 31: 14-24, 1995. and Burton J: Celecoxib exhibits the greatest potency amongst 22 Narayanan NK, Narayanan BA and Reddy BS: A combination cyclooxygenase (COX) inhibitors for growth inhibition of COX- of docosahexaenoic acid and celecoxib prevents prostate cancer 2-negative hematopoietic and epithelial cell lines. Cancer Res cell growth in vitro and is associated with modulation of nuclear 62: 2029-2033, 2002. factor-kappa B, and steroid hormone receptors. Int J Oncol 26: 8 Solomon SD, McMurray JJ, Pfeffer MA, Wittes J, Fowler R, 785-792, 2005. Finn P, Anderson WF, Zauber A, Hawk E and Bertagnolli M: 23 Narayanan BA, Narayanan NK, Davis L and Nargi D: RNA Cardiovascular risk associated with celecoxib in a clinical trial interference-mediated cyclooxygenase-2 inhibition prevents for colorectal adenoma prevention. N Eng J Med 353: 1071- prostate cancer cell growth and induces differentiation: 1080, 2005. modulation of neuronal protein synaptophysin, cyclin D1, and 9 Solomon SD: Cyclooxygenase-2 inhibitors and cardiovascular androgen receptor. Mol Cancer Ther 5: 1117-1125, 2006. risk. Curr Opin Cardiol 21: 613-617, 2006. 24 Foster BA, Kaplan PJ and Greenberg NM: Peptide growth 10 Shappell SB, Manning S, Boeglin WE, Guan YF, Roberts RL, factors and prostate cancer: new models, new opportunities. Davis L, Olson SJ, Jack GS, Coffey CS, Wheeler TM, Breyer Cancer Metastasis Rev 17: 317-324, 1998. MD and Brash AR: Alterations in lipoxygenase and 25 Tries S, Neupert W and Laufer S: The mechanism of action of cyclooxygenase-2 catalytic activity and mRNA expression in the new antiinflammatory compound ML3000: inhibition of prostate carcinoma. Neoplasia 3: 287-303, 2001. 5-LOX and COX-1/2. Inflamm Res 51: 135-143, 2002. 11 Jack GS, Brash AR, Olson SJ, Manning S, Coffey CS, Smith JA 26 Boileau C, Pelletier JP, Tardif G, Fahmi H, Laufer S, Lavigne M Jr and Shappell SB: Reduced 15-lipoxygenase-2 immuno- and Martel-Pelletier J: The regulation of human MMP-13 by staining in prostate adenocarcinoma: correlation with grade and licofelone, an inhibitor of cyclo-oxygenases and 5-lipoxygenase, in expression in high-grade prostatic intraepithelial neoplasia. human osteoarthritic chondrocytes is mediated by the inhibition Hum Pathol 31: 1146-1154, 2000. of the p38 MAP kinase signalling pathway. Ann Rheumatic 12 Marcouiller P, Pelletier JP, Guevremont M, Martel-Pelletier J, Diseases 64: 891-898, 2005. Ranger P, Laufer S and Reboul P: Leukotriene and 27 Smolka AJ, Goldenring JR, Gupta S and Hammond CE: prostaglandin synthesis pathways in osteoarthritic synovial Inhibition of gastric H,K-ATPase activity and gastric epithelial membranes: regulating factors for interleukin 1beta synthesis. cell IL-8 secretion by the pyrrolizine derivative ML 3000. BMC J Rheumatol 32: 704-712, 2005. Gastroenterol 4: 1-11, 2004.

2401 ANTICANCER RESEARCH 27: 2393-2402 (2007)

28 Narayanan BA, Narayanan NK, Pittman P and Reddy BS: 36 Narayanan BA: Chemopreventive agents alters global gene Adenocarcina of the mouse prostate growth inhibition by expression pattern: Predicting their mode of action and targets. celecoxib: downregulation of transcription factors involved in Curr Cancer Drug Targets 6: 711-728, 2006. COX-2 inhibition. Prostate 66: 257-265, 2006. 37 Gaytan M, Bellido C, Morales C, Sanchez-Criado JE and 29 Narayanan NK, Narayanan BA, Bosland MC, Condon MS and Gaytan F: Effects of selective inhibition of cyclooxygenase and Nargi D: Docosahexaenoic acid in combination with celecoxib lipooxygenase pathways in follicle rupture and ovulation in the modulates HSP70 and p53 proteins in prostate cancer cells. Intl rat. Reproduction 132: 571-577, 2006. J Cancer 119: 1586-1598, 2006. 38 Julemont F, Dogne JM, Pirotte B and de Leval X: Recent 30 Miller RG: Simultaneous Statistical Inference. 2nd edition. New development in the field of dual COX / 5-LOX inhibitors. Mini York: Springer-Verlag, p. 37, 1981. Rev Med Chem 4: 633-638, 2004. 31 Singh RP and Agarwal R: Mechanisms of action of novel agents 39 Clark TP: The clinical pharmacology of cyclooxygenase-2- for prostate cancer chemoprevention. Endocr Relat Cancer 13: selective and dual inhibitors. Vet Clin North Am Small Anim 751-778, 2006. Pract 36: 1061-1085, 2006. 32 Basler JW and Piazza GA: Nonsteroidal anti-inflammatory drugs and cyclooxygenase-2 selective inhibitors for prostate cancer chemoprevention. J Urol 171: S59-62, 2004. 33 Kelloff GJ, Lieberman R, Steele VE, Boone CW, Lubet RA, Kopelovich L, Malone WA, Crowell JA, Higley HR and Sigman CC: Agents, biomarkers, and cohorts for chemopreventive agent development in prostate cancer. Urology 57: 46-51, 2001. 34 Sabichi AL and Lippman SM: COX-2 inhibitors and other NSAIDs in bladder and prostate cancer. Prog Exp Tumor Res 37: 163-178, 2003. 35 Steele VE, Moon RC, Lubet RA, Grubbs CJ, Reddy BS, Wargovich M, McCormick DL, Pereira MA, Crowell JA, Bagheri D, Sigman CC, Boone CW and Kelloff GJ: Preclinical efficacy evaluation of potential chemopreventive agents in animal carcinogenesis models: methods and results from the Received January 25, 2007 NCI Chemoprevention Drug Development Program. J Cell Revised April 4, 2007 Biochem Suppl 20: 32-54, 1994. Accepted May 8, 2007

2402