Wine polyphenols exert antineoplasic effect on androgen resistant PC-3 cell line through the inhibition of the transcriptional activity of COX-2 promoter mediated by NF-κβ

Abstract

Objective

Mediterranean diet may play a role in the prevention of prostate cancer (PCa) development and progression. Cyclooxygenase-2 (COX-2) expression is associated with increased cellular proliferation, prevents apoptosis and favors tumor invasion. We intend to clarify whether resveratrol and other polyphenols effectively inhibit COX-2 activity and induce apoptosis in hormone-resistant PC-3 cell line.

Material and method

PC-3 cells were cultured and treated with different concentrations of gallic acid, tannic acid, quercetin, and resveratrol in presence of phorbol myristate acetate (PMA; 50 μg/ml) that induces COX-2 expression. Total RNA was extracted and COX-2 expression was analyzed by relative quantification real-time PCR (ΔΔCt method). COX-2 activity was determined by PGE-2 detection using ELISA. Caspase 3/7 luminescence assay was used to disclose apoptosis. Transitory transfection with short human COX-2 (phPES2 −327/+59) and p5xNF-κβ-Luc plasmids determined COX-2 promoter activity and specifically that dependant of NF-κβ.

Results

COX-2 expression was not modified in media devoid of PMA. However, under PMA induction tannic acid (2.08 ± .21), gallic acid (2.46 ± .16), quercetin (1.78 ± .14) and resveratrol (1.15 ± .16) significantly inhibited COX-2 mRNA with respect to control (3.14 ± .07), what means a 34%, 23%, 46% and 61% reduction, respectively. The inhibition in the levels of PGE-2 followed a similar pattern. All compounds studied induced apoptosis at 48 h, although at a different rate. PMA caused a rise in activity 7.4 ± .23 times phPES2 −327/+59 and 2.0 ± .1 times p5 × NF-κβ-Luc at 6 h compared to basal. Resveratrol suppressed these effects 17.1 ± .21 and 32.4 ± .18 times, respectively. Similarly, but to a lesser extent, the rest of evaluated polyphenols diminished PMA inductor effect on the activity of both promoters.

Conclusions

Polyphenols inhibit transcriptional activity of COX-2 promoter mediated by NF-κβ. This effect could explain, at least in part, the induction of apoptosis in vitro by these substances in castration resistant PCa.

Resumen

Objetivo

La dieta mediterránea puede tener un papel en la prevención del desarrollo y progresión del cáncer de próstata (CaP). La expresión de ciclooxigenasa-2 (COX-2) se asocia con la proliferación celular aumentada, previene la apoptosis y favorece la invasión tumoral. Se intenta clarificar si el resveratrol y otros polifenoles del vino inhiben de forma efectiva la actividad de COX-2 e inducen apoptosis en la línea celular PC-3 de cáncer hormonorrefractario.

Material y método

Células PC-3 fueron cultivadas y tratadas con diferentes concentraciones de ácido gálico, ácido tánico, quercetina y resveratrol en presencia del éster de forbol PMA (50 μg/ml) que induce la expresión de COX-2. Se extrajo ARN total y se analizó la expresión de COX-2 mediante cuantificación relativa por PCR a tiempo real (método ΔΔCt). La actividad COX-2 se determinó mediante detección de PGE-2 por la técnica ELISA. Se empleó ensayo de luminiscencia caspasa-3/7 para evaluar la apoptosis. La transfección transitoria con plásmidos short human COX-2 (phPES2 −327/+59) y p5xNF-κβ-Luc determinó la actividad del promotor de COX-2, y de forma particular, la dependiente de NF-κβ.

Resultados

La expresión de COX-2 no se modificó en ausencia de PMA. No obstante, bajo la inducción con PMA, ácido tánico (2,08 ± 0,21), ácido gálico (2,46 ± 0,16), quercetina (1,78 ± 0,14) y, especialmente resveratrol (1,15 ± 0,16) inhibieron significativamente la expresión de COX-2 respecto al control (3,14 ± 0,07), causando una reducción del 34, 23, 46 y 61%, respectivamente. La inhibición en los niveles de PGE-2 siguió un patrón similar. Todos los compuestos estudiados indujeron apoptosis a las 48 h, aunque en diferente proporción. El PMA causó un aumento de la actividad de 7,4 ± 0,23 veces phPES2 −327/+59 y 2,0 ± 0,1 veces p5 × NF-κβ-Luc a 6 h sobre los niveles basales. El resveratrol suprimió estos efectos 17,1 ± 0,21 y 32,4 ± 0,18 veces, respectivamente. De forma similar, aunque en menor medida, el resto de polifenoles evaluados disminuyeron el efecto inductor de PMA sobre la actividad de ambos promotores.

Conclusiones

Los polifenoles inhiben la actividad transcripcional del promotor de COX-2 mediada por NF-κβ. Este efecto podría explicar, al menos en parte, la inducción in vitro de apoptosis por estas sustancias en el CaP resistente a la castración.

Introduction

Epidemiological studies provide rather consistent evidence that a diet which has a high content of fruits and vegetables reduces the risk for several types of cancer.¹ However, the components of these nutrients that exert this protective effect and the precise mechanism by which they exert these effects are not known with certainty. Prostate cancer (PCa) remains a considerable health problem for men around the world. There is a large body of literature devoted to describe the potential chemopreventive or chemotherapeutic effect of certain phytochemicals in red wine polyphenols, including enhanced apoptosis, growth arrest, modification of the cell cycle, inhibition of DNA synthesis, and modulation expression of enzymes, such as cyclooxygenases or prostaglandin H2 synthase (COXs).^2–5

Polyphenolic compounds constitute one of the largest and most ubiquitous group of phytochemicals. These compounds, especially flavonoids and phenolic acids, are an import part of human diet.^1,3 The biochemical properties, intake, and bioavailability of different polyphenolic compounds that have been reviewed for this topic are of particular interest in the issue of prostate cancer chemoprevention and/or modulation of therapies.^6–8

There are two isoforms of COX that catalyze the formation of prostaglandins (PG) from arachidonic acid. Whereas COX-1 is a housekeeping gene that is expressed constitutively and is generally responsible for the production of PGs under physiological conditions, COX-2 is a highly inducible gene by different stimuli. Furthermore, COX-2 is thought to be the isoform responsible for the production of pro-inflammatory PGs and it is very important for tumorigenesis.⁹

Cyclooxygenase-2 (COX-2) is considered a potential target for cancer therapy, because COX-2 levels are elevated in the majority of human tumors compared to corresponding normal tissues. There are different possible mechanisms that could account for the relation between COX-2 and cancer. Increased levels of PGs and COX-2 activity have been detected in multiple epithelial cancers (e.g., lung, colon, prostate),^10,11 and especially PGE₂ can affect and increases cell proliferation and invasiveness, promoting angiogenesis and inhibiting apoptosis in malignant cells.

Phorbol esters such as phorbol 12-myristate 13-acetate (PMA) function as tumor promoters and have been reported to modulate diverse cellular responses including cell growth, programmed cell death, differentiation, and gene transcription. It has been established that besides PMA, cytokines and lipopolysaccharide also upregulate COX-2 expression.^12–14 We have demonstrated that regulation of androgen receptor (AR) transcription mediates the antiproliferative effect of resveratrol and other wine polyphenols in androgen-sensitive LNCaP cell line.^15,16 Now we intend to study in vitro the effects of these compounds in castration-resistant PCa and the mechanisms involved.

Materials and methods

Materials

PMA and polyphenolic compounds (tannic acid, gallic acid, quercetin and resveratrol) were purchased from Sigma–Aldrich (St. Louis, MO, USA) and dissolved in ethanol at stock concentration and stored at −20 °C. Final concentrations used for different experiments were prepared by diluting the stock with RPMI before use. Culture plastic was obtained from Corning-Costar (NY, USA). RPMI-1640 medium with 100 U/ml penicillin G, 0.1 mg/ml streptomycin, 50 μg/ml gentamicin, and 2 mM l-glutamine were from Gibco BRL (Paisley, UK). Fetal bovine serum (FBS), Phosphate-buffered saline (PBS) and trypsin-ethylenediamine tetraacetate (EDTA) 0.025% were from Biological Industries (Beit-Hamek, Israel), Amresco (Solon, OH, USA), and Gibco BRM (Paisley, UK), respectively. MMLV reverse transcriptase was from Sigma–Aldrich (St. Louis, MO, USA).

Cell culture and treatments

The androgen-insensitive PC-3 cell line was obtained from the American type cultura collection (ATCC, Rockville, MD). Cells were maintained in RPMI 1640 (Gibco BRL, Spain) supplemented with 5% heat-inactivated fetal bovine serum (FCS, LINUS, Spain). Cells were cultured in a humidified atmosphere of 95% air and 5% CO₂ at 37 °C and they were grown at 60–70% confluence, trypsinized with 0.05% trypsin-2 mM EDTA (Gibco BRL, Spain), and plated for experimental use. For experiments, the cells were treated with or without PMA (50 μg/ml) and 10 M of tannic acid, gallic acid, quercetin, and resveratrol or vehicle (ethanol 0.01%) for 24 h for mRNA expression and transfection assay, and for 48 h to study the effect of polyphenols on PGE₂ secretion and apoptosis in PC-3 cells.

RNA extraction and real-time PCR for COX-2 expressions

Total RNA from cell samples was obtained using RNeasy^® Mini kit, from Qiagen (Valencia, CA, USA) following the manufacturer's instructions. Yield of RNA was determined spectrophotometrically at 260/280 nm and density (OD₂₆₀/OD₂₈₀) ratios were measured to ensure the quality of the isolated RNA using ND-100 spectrophotometer (Nanodrop Technology). The average ratio was 1.8 to 2.1 for all samples. For RT, an aliquot containing 1–2 μg of total RNA from each sample was used for first-strand cDNA in a final volume of 20–40 l using 0.75 μl (0.5 mg) of random primers (Promega, Madison, WI) and 0.5 μl (12.5 mmol/L) dNTPs mixture (Ecogen, Spain), incubated at 70 °C for 10 min and immediately cooled in ice to avoid re-naturalization. The following RT mixture was prepared in 10–20 μl for each sample: 4–8 μl MMLV RT buffer (5×) (Sigma), 0.5 μl (20 U) of RNase inhibitor (RNasin, Promega) and 1 μl (100 U) of Moloney murine leukemia virus (MMLV) reverse transcriptase and incubated at 25 °C for 5 min and 37 °C for 50 min. The reaction was stopped at 95 °C for 5 min to inactivate the RT. Real-time PCR was performed using a standard TaqMan^® PCR kit protocol on a fluorescence temperature cycler, iCycler (Bio-Rad Laboratories, Hercules, CA, USA). The 20 l PCR included 0.1 ng of cDNA, 2·l TaqMan^® Universal PCR Master Mix (Applied Biosystems), 1·l of a mix of unlabelled PCR primers (COX-2 or 18S rRNA), and Taqman MGB probe (TaqMan^® Assays, Applied Biosystems, Madrid, Spain). The reactions were incubated in a 96-well plate at 95 °C for 10 min, followed by 40 cycles of 95 °C for 15 and 60 °C for 1 min. All reactions were run in triplicate. The threshold cycle (CT) is defined as the fractional cycle number at which the fluorescence passes the fixed threshold. To determine the relative expression of AR in each sample, the results were normalized to 18S (used as an internal standard) and the comparative CT method was used (User Bulletin #2 (2001), Applied Biosystem, Madrid, Spain).

Determination of caspase 3/7 activity

The cells were grown in 6-well plates at a density of 1 × 10⁵ cells per well and treated with the highest doses of each polyphenol used in the proliferation study. At 24 and 48 h of treatment, the media was aspirated and the cells were washed once with PBS. Then we added M-PER mammalian protein extraction reagent (Pierce, Rockford, IL, USA) with fresh protease inhibitor cocktail (Boehringer Mannheim, Germany). The cells were gently stirred for 5 min and the lysate was collected in a microfuge tube. The lysate was cleared by centrifugation at 13,000 rpm for 10 min and the supernatant was either used or immediately stored at −80 °C. The protein concentration was determined by BCA Protein Assay using the manufacturer's protocol (Pierce, Rockford, IL, USA). For apoptosis quantification we used a luminescent assay that measures caspase-3 and -7 activities (Caspase-Glo 3/7 Assay, Promega, Madison, WI, USA). Briefly, the assay provides a proluminescent substrate for caspase 3/7 in a reagent optimized for caspase activity, luciferase activity, and cell lysis. The addition of reagent results in cell lysis, caspase cleavage of the substrate, and generation of a “glow-type” luminescent signal produced by luciferase. Luminescence is proportional to the amount of caspase activity present. For this assay we used 1 μg protein for each determination.

Measurements of PGE₂

The cells were treated as described in the determination of caspase 3/7 activity. Levels of PGE₂ released by the cells were measured by enzyme immmunoassay according to the manufacter’ instructions. Rates of produccion of PGE₂ were normalized to protein concentrations (R&D Systems, Minneapolis, USA).

Reporter plasmids and transient transfection experiments

The reporter plasmids used were 5 × NF-κB-luc (Stratagene, La Jolla, CA, USA), and luciferase-based reporter plasmid corresponding to the 5′ flanking regulatory region of human-sort COX-2 (phPESs−327/+59). These two expression vectors were transfected in PC-3 cells using Megafectin-20 (Qbiogene, Carlsbad, CA). Briefly, PC-3 cells were grown in 24-well plates at 60–80% confluence. The culture medium was then replaced by vehicle medium, that is, serum-free medium supplemented with 0.1% BSA, for 18–20 h. Then, PC-3 cells were lipotransfected with the mixture consisting of 3–4 μg of the above-mentioned plasmids incubated at room temperature for 15 min with 0.1 μl of enhancer, 3 μl of Hepes, and 2 μl of Megafectin-20 in vehicle medium. After incubation, the transfection mixture was added to cell cultures for 4 h at 37 °C in RPMI-1460 containing 5% FCS but without antibiotic liposomes. Afterwards, the transfection mixture was replaced by fresh vehicle medium containing the different compounds to be tested. At 24 h, PC-3 cells were extracted with 1_Reporter Lysis Buffer (Promega, Madison, WI, USA) for analysis of reporter gene expression. Luciferase activity was expressed as relative luciferase units/mg of protein.

Statistical analysis

Results are expressed as mean ± s.e.m. One-way ANOVA with Dunnett t-test post hoc analysis was used to compare ARM expression analysis between treated and control groups. Comparisons with regard to caspase 3/7 activity were carried out with Student’ unpaired t-test. Significance level was reached when p < 0.05.

Results

Polyphenols suppress PMA-mediated increase in the production of PGE₂

In these experiments, four polyphenols were tested with regard to their effect on the production on PGE₂ to determine the anti-inflammatory effects of these polyphenols: the stilbene resveratrol, the flavonol quercetin, the tannin tannic acid, and the benzoic acid and gallic acid (Fig. 1). Phorbol esters are potent inducers of COX-2, so we determined the effect of tannic, gallic acid, quercetin, and resveratrol on the synthesis of PGs by PC-3 cells in which COX-2 was induced by PMA. As shown in Fig. 2A, PMA caused about a 4-fold increase in the synthesis of PGE₂. This effect was markedly supressed by treatment with resveratrol (82%), tannic acid (64%), and quercetin (41%); and it was lightness inhibited by gallic acid (17%). To evaluate whether the inhibition of PGE₂ synthesis was because of inhibition of COX-2 or COX-1, we compared the effects of NS-398, a selective inhibitor of COX-2. Pretreatment of cells with NS-398 (0.1–10 M) inhibited the synthesis of PGs to less than 10% of the PMA-stimulated control level (Fig. 2B). This result means that more than 90% of the PMA-stimulated COX activity in PC-3 cells was because of the COX-2 isoform. We did not find any change in the levels of PGE₂ synthesis in absence of PMA (data not shown).

Figure 1.

Biochemical structure of the different polyphenols studied (resveratrol, quercetin, tannic acid and gallic acid) and chemical groups they correspond to.

Figure 2.

Polyphenols suppress phorbol ester-mediated elevated in the production of PGE₂. PC-3 cells were treated with or without PMA (50 μg/ml) and 10 μM of gallic and tannic acid, quercetin, resveratrol and vehicle (0.01% ethanol) (A) or NS398 (0–10 μM) (B) for 48 h. The medium was collected to determine the rate of synthesis of PGE₂. Production of PGE₂ was determined by enzyme immunoassay. ^*p < 0.05; ^**p < 0.01 vs. PMA treatment.

Polyphenols inhibit PMA-mediated increase in COX-2 mRNA

To determine whether the above effects on production of PGE₂ could be related to differences in levels of mRNA COX-2, real-time RT-PCR of total RNA was carried out. Fig. 3 shows that treatment with PMA resulted in a marked increase (3-fold) in levels of COX-2 mRNA expression versus no treated cells (p ≤ 0.01). This effect was partially suppressed by polyphenols. Resveratrol and quercetin inhibited COX-2 expression 63% and 47%, respectively. However, tannic and gallic acids inhibited COX-2 mRNA expression to a lesser extent (20–25%). These data indicated that these polyphenols inhibited COX-2 expression at the transcription levels, resveratrol and quercetin being the most potent inhibitors. We did not find any change in the levels of COX-2 mRNA in absence of PMA (data not shown).

Figure 3.

Effect of polyphenols on PMA-induced COX-2 mRNA levels in PC-3 cells treated with or without PMA (50 μg/ml) and 10 μM of gallic and tannic acid, quercetin, resveratrol or vehicle (0.01% ethanol) for 24 h. Total cellular RNA was isolated and analyzed by real-time RT-PCR for COX-2 expression. The 18S rRNA gene was amplified as an internal control. A significant decrease in COX-2 mRNA expression was observed for all polyphenols. Resveratrol was the most potent. ^*p < 0.05; ^**p < 0.01 vs. PMA treatment.

Effect of polyphenols on NF-κB and Cox-2 activity

To further investigate the importance of PMA and polyphenols in modulating the expression of COX-2, transient transfections were performed using a reporter plasmid bearing the short (−327/+59pb) human COX-2 promoter. As in mRNA, the treatment with PMA for 24 h significantly increased activity of the luciferase-based promoter construct about >7-fold. All polyphenols caused inhibition of PMA-mediated induction of COX-2 promoter activity (Fig. 4).

Figure 4.

Effect of polyphenols on PMA-induced COX-2 promoter activity. PC-3 cells were transiently transfected with 3–4 μg of plasmid phCOX-2 (−327/+59)-luc. After transfection, cells were treated with vehicle (white column), PMA (50 μg/ml, black column), or PMA and polyphenols (10 μM, rest of columns). Reporter activity was measured in cellular extracts 24 h later. Luciferase activity represents data that have been normalized with protein concentration. ^*p < 0.05; ^**p < 0.01 versus PMA treatment.

Because activation of NF-κB is critical for induction of COX-2 by phorbol esters, we also studied whether one of the main transcription factors involved in the inflammatory response, NF-κB, is inhibited in the signal transducction pathway leading to COX-2 expression caused by PMA. For this purpose, we determined NF-κB-dependent transcription basal activity in PC-3 transiently transfected with p5 × NF-κB-luc. Stimulation of cells with PMA for 24 h results in a 2-fold increased NF-κB luciferase activity. This stimulation was partially suppressed by gallic (24%), tannic acid (33%), and quercetin (37%) (Fig. 5). Resveratrol was the most potent, and it was capable of inhibiting almost totally (>90%) the activity of NF-κB promoter even with respect to basal conditions, in the absence of PMA (>90% inhibition).

Figure 5.

Effect of polyphenols on PMA-induced NF-κB promoter activity. PC-3 cells were transiently transfected with 3–4 μg of plasmid phCOX-2 (−327/+59)-luc. After transfection, cells were treated with vehicle (white column), PMA (50 μg/ml, black column), or PMA and polyphenols (10 μM, rest of columns). Reporter activity was measured in cellular extracts 24 h later. Luciferase activity represents data that have been normalized with protein concentration. ^*p < 0.05; ^**p < 0.01 versus PMA treatment.

Induction of apoptosis by polyphenols on PC-3 cells

Caspase 3/7 activity in control and polyphenols treated cells was determined to assess whether the inhibition of COX-2 metabolism (PGE2 expression and secretion) may cause induction of apoptosis. PC-3 cells were treated for 24 and 48 h in presence of different polyphenols (10 M), lysed and measured for the presence of active caspase 3/7 with Caspase-Glo 3/7 Assay. Caspase 3/7 activity was significantlly increased by all polyphenols tested with respect to control at 48 h, although at a different rate: quercetin 1.6, gallic acid 1.7, tannic acid 3.5, and resveratrol 42.5 times compared to control (Fig. 6A). Minor differences can be better perceived in detail (Fig. 6B).

Figure 6.

Induction of apoptosis by polyphenols on PC-3 cells determined by caspase 3/7 activity in control and treated cells. All compounds induced apoptosis at 48 h, although at a different rate: tannic acid 3.5, gallic acid 1.7, quercetin 1.6 and resveratrol 42.5 times compared to control (A). Minor differences can be appreciated in detail (B).

Discussion

Resveratrol is a phytoalexin found in high concentrations in grapes and has been shown to inhibit chemical carcinogens in several models.¹⁷ This natural compound has been proposed of therapeutic potential in preclinical, animal models, and human studies in several malignancies.¹⁸ Specific interest has emerged in prostate cancer, since resveratrol has proved AR expression transcriptional regression, accelerated AR degradation, and modulation of AR co-activators.^16,19–21 Resveratrol in the diet spontaneously suppresses developing prostate tumors in the Transgenic Adenocarcinoma Mouse Prostate (TRAMP) model through down-regulation of phospho-ERK, a downstream effector of sex steroid receptor and growth factor signaling.²² High concentrations of resveratrol have also resulted in dose-dependent significant inhibition of PKB/Akt phosphorylation, both in PC-3 and LNCaP cell lines.^23,24 Resveratrol-mediated PI3Kinhibition may be mediated in part by inhibition of FOXO family tumor suppression phosphorylation, thus allowing for proapototic targets like TRAIL or p27.²⁵

In mammary adenocarcinoma, resveratrol has been proved to inhibit the enzyme COX-2, inducible isoform of the rate-limiting enzymes that convert arachidonic acid to proinflammatory prostaglandins, which is increased following stimulation of mitogens, such as phorbol ester.^18,26 This inhibition was manifested at different levels, which include direct enzymatic activity, mRNA and protein synthesis, cyclic AMP response element, and AP-1 mediated gene expression.^26,27 Resveratrol has also been confirmed to suppress COX-2 promoter activity in DLD-1 human colonic cancer cells.²⁸ Regarding the prostate, COX-2 is up-regulated in proliferative inflammatory atrophy (PIA) of the prostate, but not in PCa,²⁹ although it has been advocated to promote prostate cancer progression.³⁰ Catechins present in green tea inhibit COX-2 without affecting COX-1 expression at mRNA and protein levels, both in androgen-responsive LNCaP and androgen-refractory PC-3. This action has been proven both in vitro and in vivo using the TRAMP mouse model.^31,32 Also the effects of green tea polyphenols on PCa are synergistic with COX-2 inhibitors.³³

The COX-2 inflammatory pathway is considered a potential target in epithelial neoplasm, and also specifically in PCa. The activation of NF-κB is considered critical for induction of COX-2 by phorbol esters.³⁴ What is more, overexpressed COX-2 is secondary to overexpression of NF-κβ.³⁵ Hussain et al. demonstrated that the constituent of green tea epigallocatechin-3-gallate selectively inhibits COX-2 in human prostate carcinoma cells, both LNCaP and PC-3.³⁶ This action is the same we demonstrate is exerted by resveratrol and other polyphenols in red wine, all abundant in the Mediterranean diet.³⁷

NF-κβ is a transcription factor often overexpressed that regulates different cellular activities including inflammation, immunity, and cell death. Downregulation of NF-κβ is known to trigger apoptotic and antiproliferative effects and is considered a likely mechanism to render health benefits of green tea polyphenols intake.³⁸ However, reality is very complex because green tea polyphenols target many other different mechanisms, including cell-cycle arrest, detoxification enzymes, IGF, TRAIL-induced apoptosis, MAP kinases, Hh (Hedgehog signaling) PI3K-Akt (rapamycin signaling), and PKC.^39,40 According to our results, polyphenols in red wine also target the inflammatory pathways and could be responsible for the in vitro antiproliferative effect of resveratrol, quercetin, gallic acid, and tannic acid on PC-3 cell line. This signaling appears totally different from that of the AR observed in LNCaP cell line.¹⁶ Inhibition of NF-κβ signaling in TRAMP mice activates apoptosis trough increased Bax/Bcl-2 ratio.⁴¹ This effect has also been demonstrated for pomegranate juice of the Punica granatum fruit⁴²; delphinidin, a major anthocyanid present in many pigmented fruits and vegetables,⁴³ and also recently α-tomatine, the major saponin present in tomato (Lycopersicon esculentum).^44,45 It is very understandable that the mechanism we describe for resveratrol and other red wine polyphenols on androgen-refractory PC-3 cell line of prostate cancer follows a very similar rationale.

In this study, we show that different polyphenols present in red wine inhibit PMA stimulation of COX-2 expression, promoter activity for COX-2, and PGE₂ release in androgen-insensitive PC-3 human prostate carcinoma leading to increased apoptosis. Despite the large accumulated experience in vitro very few clinical trials using precise concentrations of likely chemopreventive compounds have been conducted. In this sense, a randomized phase II study has recently shown that pomegranate extract is associated with more than 6-month increase in PSA doubling time in men with biochemical recurrence after local therapy; however, no control arm was used.⁴⁶

Definitely there is no evidence to consider resveratrol a therapeutic antineoplasic agent. However, it has been considered to enhance radiation sensitivity in several malignancies, including prostate.^47,48 Possibly in the future some beneficial role might be found as adjunct in the management of castrate-resistant disease, or at least be beneficial to attenuate PCa progression. At present, we must admit that, from the clinical perspective, the effect of dietary antioxidants on prostate cancer remains undefined.⁴⁹

Conflict of interest

The authors declare that they have no conflict of interest.

Funding

This work has been funded by the grant Foundation for Research in Urology (FIU) of the Spanish Association of Urology.