Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Biochem. J. (2008) 410, 245–253 (Printed in Great Britain) 245 doi:10.1042/BJ20071136 Activators of the farnesoid X receptor negatively regulate androgen glucuronidation in human prostate cancer LNCAP cells Jenny KAEDING*†, Emmanuel BOUCHAERT‡, Julie BÉLANGER*†, Patrick CARON*†, Sarah CHOUINARD*§, Mélanie VERREAULT*†, Olivier LAROUCHE*§, Georges PELLETIER*§, Bart STAELS‡, Alain BÉLANGER*§ and Olivier BARBIER*†1 *Molecular Endocrinology and Oncology Research Center, CHUL Research Center, Quebec, QC, Canada G1V 4G2, †Faculty of Pharmacy, Laval University, QC, Canada G1K 7P4, ‡Unité 545 INSERM, Lille, Département d’athérosclérose, Institut Pasteur de Lille and Faculté de Pharmacie, Université Lille II, Lille F-59019, France, and §Faculty of Medicine, Laval University, QC, Canada G1K 7P4 Androgens are major regulators of prostate cell growth and physiology. In the human prostate, androgens are inactivated in the form of hydrophilic glucuronide conjugates. These metabolites are formed by the two human UGT2B15 [UGT (UDP-glucuronosyltransferase) 2B15] and UGT2B17 enzymes. The FXR (farnesoid X receptor) is a bile acid sensor controlling hepatic and/or intestinal cholesterol, lipid and glucose metabolism. In the present study, we report the expression of FXR in normal and cancer prostate epithelial cells, and we demonstrate that its activation by chenodeoxycholic acid or GW4064 negatively interferes with the levels of UGT2B15 and UGT2B17 mRNA and protein in prostate cancer LNCaP cells. FXR activation also causes a drastic reduction of androgen glucuronidation in these cells. These results point out activators of FXR as negative regulators of androgen-conjugating UGT expression in the prostate. Finally, the androgen metabolite androsterone, which is also an activator of FXR, dose-dependently reduces the glucuronidation of androgens catalysed by UGT2B15 and UGT2B17 in an FXR-dependent manner in LNCaP cells. In conclusion, the present study identifies for the first time the activators of FXR as important regulators of androgen metabolism in human prostate cancer cells. INTRODUCTION ive reduced metabolites ADT and 3α-DIOL; however, since these metabolic reactions can be reversed, complete androgen inactivation is only ensured through glucuronidation [2]. The importance of glucuronidation for androgen metabolism in the human prostate has further been emphasized by the observation that polymorphisms within androgen-glucuronidating genes are associated with an increased risk for prostate cancer [8–11]. Among the 18 functional UGT (UDP-glucuronosyltransferase) enzymes identified in humans, UGT2B7, UGT2B15 and UGT2B17 have a remarkable capacity to conjugate androgens [2,12]. However, only UGT2B15 and UGT2B17 are expressed at detectable levels in the human prostate [5,13]. The sequence homology between these two UGTs is very high (95 %), but they possess a distinct substrate specificity towards 5α-reduced androgens. UGT2B17 glucuronidates the 3-hydroxy position of ADT and the 17-hydroxy group of 3α-DIOL or DHT and has been identified as the major glucuronidating enzyme for ADT in the prostate. The conjugating activity of UGT2B15 is limited to the 17-hydroxy position of 3α-DIOL and DHT [13,14]. The FXR (farnesoid X receptor; NR1H4) is a bile acid sensor that belongs to the family of ligand-activated transcription factors [15]. The regulatory activity of FXR is primarily associated with the control of genes involved in hepatic bile acid homoeostasis [16]. However, recent data suggest that FXR plays a broader role in controlling numerous genes involved in glucose, lipid, cholesterol and steroid metabolism [17,18]. Various observations suggest that FXR may also be involved in androgen activation and activity: (i) FXR is a receptor for certain androgen metabolites such as ADT and etiocholanolone [19]; (ii) FXR The glucuronidation reaction, which corresponds to the transfer of the polar glucuronosyl group from UDP-GA (UDP-glucuronic acid) to an acceptor substrate, is an essential metabolic pathway for a wide variety of endogenous and exogenous molecules [1]. Among glucuronidated endobiotics, androgens occupy an important place. Indeed, there is much clinical and in vivo evidence that glucuronidation constitutes the major elimination route for the active androgen DHT (dihydrotestosterone) and its 5α-reduced metabolites 3α-DIOL (androstane-3α,17β-diol) and ADT (androsterone) [2]. Likewise, ADT-3G (androsterone-3glucuronide) and 3α-DIOL-17G (3α-DIOL-17-glucuronide) are the predominant androgen metabolites found in circulation in men [2]. Whereas glucuronidation is generally considered as a hepatic detoxification mechanism, extrahepatic glucuronidation is now established as an efficient way to locally inactivate endogenous bioactive molecules [2,3]. This is particularly true for androgens, which are efficiently glucuronidated within their target tissues such as the human prostate [2,4,5]. The prostate’s growth and function are tightly regulated by androgens, and changes in the synthesis and inactivation of the active hormone can lead to excessive androgen influence and thereby initiate or support the progression of prostate cancer [6]. Androgenic precursors secreted from adrenals or testis are converted in the prostate into DHT [2], the active androgen that activates the AR (androgen receptor; NR3C4) and thereby initiates the transcription of genes involved in prostate cell proliferation and differentiation [7]. DHT is locally converted into the inact- Key words: androgen metabolism, farnesoid X receptor (FXR), gene expression, glucuronidation, prostate, UDP-glucuronosyltransferase (UGT) 2B15 (UGT2B15) and UGT2B17. Abbreviations used: ADT, androsterone; ADT-3G, androsterone-3-glucuronide; AP-1, activator protein-1; AR, androgen receptor; CDCA, chendeoxycholic acid; DHT, dihydrotestosterone; 3α-DIOL, androstane-3α,17β-diol; 3α-DIOL-17G, 3α-DIOL-17-glucuronide; DTT, dithiothreitol; FBS, fetal bovine serum; FXR, farnesoid X receptor; HSD, hydroxysteroid dehydrogenase; LC, liquid chromatography; LC–MS/MS, LC–tandem MS; PrEC, prostate epithelial cells; UDP-GA, UDP-glucuronic acid; UGT, UDP-glucuronosyltransferase. 1 To whom correspondence should be addressed, at Laboratory of Transcriptomics and Molecular Pharmacology, CHUL Research Center (CHUQ), 2705, Blvd Laurier, Quebec, QC, Canada G1V 4G2 (email [email protected]). c The Authors Journal compilation c 2008 Biochemical Society 246 J. Kaeding and others activation results in the modulation of genes encoding androgen precursor-synthesizing enzymes, namely dehydroepiandrosterone sulfotransferase (SULT2A1), 5α-reductase and 3β-HSD (3βhydroxysteroid dehydrogenase) in the liver [20,21]; (iii) FXR was recently identified as a negative modulator of the androgen– oestrogen-converting aromatase enzyme in human breast carcinoma MDA-MB-468 cells [18]; and (iv) FXR is expressed at very high levels in the adrenal gland, where androgen precursor synthesis occurs, and at detectable levels in androgen-sensitive tissues, such as the prostate [22]. On the other hand, several UGT enzymes have already been shown to be regulated by FXR. In rat hepatocytes treated with the specific FXR agonist GW4064, a significant increase in UGT2B1 and UGT1A7 mRNA levels has been reported [20]. Similarly, in human hepatocytes, FXR activation resulted in increased UGT2B4 expression [23], whereas Lu et al. [24] reported that ligand-activated FXR inhibits the UGT2B7 expression in intestinal Caco-2 cells. Based on these observations, we hypothesized that FXR may also control the expression and activity of the androgenconjugating UGT2B15 and UGT2B17 enzymes in the prostate. To address this hypothesis, we have investigated the presence of FXR transcript and proteins in human prostate tissue, epithelial cells and/or in the androgen-sensitive prostate cancer LNCaP cell line. Furthermore, the effect of endogenous as well as synthetic FXR activators on UGT2B15 and UGT2B17 expression and activity was analysed in LNCaP cells. MATERIALS AND METHODS Materials Prostate tissue slides were purchased from US Biomax (Toronto, ON, Canada) [25]. UDP-GA, poly(L-lysine) and steroid substrates (3α-DIOL, ADT and DHT) were obtained from Sigma (St. Louis, MO, U.S.A.). CDCA (chendeoxycholic acid), 3α-DIOL-17G, ADT-3G and DHT-17G were purchased from Steraloids (Newport, RI, U.S.A.). Protein assay reagents were obtained from Bio-Rad Laboratories (Marnes-la-Coquette, France). Cell culture materials and Lipofectin® transfection reagent were obtained from Invitrogen (Burlington, ON, Canada). SYBR Green PCR Mastermix was purchased from Applied Biosystems (Foster City, CA, U.S.A.). Casodex (bicalutamide) was provided by Endorecherche (Quebec, QC, Canada). R1881 was from PerkinElmer (Woodbridge, ON, Canada) and GW4064 was as described previously [23]. The anti-human UGT2B15 and UGT2B17 antibodies have been described previously [5,12,26]; the rabbit polyclonal anti-human FXR antibodies were from Abcam (ab28676; Cambridge, MA, U.S.A.) and Santa Cruz Biotechnology (sc-13063; Santa Cruz, CA, U.S.A.) respectively, the rabbit polyclonal anti-human actin antibody was purchased from Sigma (A5060) and the secondary anti-rabbit IgG (RPN4301) was obtained from Amersham (Oakville, ON, Canada). Immunohistochemistry Immunohistochemical staining was conducted using standard techniques as previously described [25,27]. Briefly, immunostaining was performed by using anti-human FXR antibodies diluted in Tris saline (1:20 for sc-13063 and 1:200 for ab28676) (pH 7.6) for 1 h at room temperature (22 ◦C). Sections were subsequently washed with PBS and incubated with a biotin-labelled goat anti-rabbit γ -globulin for 10 min. After incubation, the sections were treated with streptavidin coupled with peroxidase, and diaminobenzidine was used as the chromogen to visualize the c The Authors Journal compilation c 2008 Biochemical Society Table 1 Primers and conditions used in PCR analyses FWD, forward; REV, reverse. Gene Primers Final concn. (nM) 28S FWD: 5 -AAACTCTGGTGGAGGTCCGT-3 REV: 5 -CTTACCAAAGTGGCCCACTA-3 200 UGT2B15 FWD: 5 -GTGTTGGGAATATTATGACTACAGTAAC-3 REV: 5 -GGGTATGTTAAATAGTTCAGCCAGT-3 125 UGT2B17 FWD: 5 -TGACTTTTGGTTTCAAGC-3 REV: 5 -TTCCATTTCCTTAGGCAA-3 300 biotin–streptavidin peroxidase complex (Vectastain ABC kit; Vector Laboratories). The sections were then counterstained with haematoxylin. For negative control, the antibodies were replaced by a rabbit pre-immune serum. Cell culture Human PrEC (prostate epithelial cells; Cambrex Walkersville, MD, U.S.A.) and human prostate cancer LNCaP cells (A.T.C.C., Rockville, MD, U.S.A.) were cultured as described in [25]. For all experiments, LNCaP cells were used between passages 21 and 33 and in culture plates pretreated with poly(L-lysine) overnight prior to cell seeding. CDCA, GW4064, DHT, Casodex and androgens were dissolved in ethanol and the final concentration of the solvent never exceeded 0.1 % (v/v) in cell culture experiments. For RNA, protein and cell viability analyses, 3 × 105 LNCaP cells per well were seeded on to 6-well plates and cultured in RPMI 1640 medium supplemented with 0.2 % FBS (fetal bovine serum) for 48 h to allow cell attachment. Cells were then treated with vehicle [0.1 % (v/v) ethanol], CDCA, GW4064, DHT or ADT alone and/or in combination with Casodex at the indicated concentrations in RPMI 1640 medium supplemented with 0.2 % FBS for different time periods. For glucuronidation assays, 2 × 106 cells were seeded on to 10 cm culture dishes and cultured in RPMI 1640 medium, supplemented with 0.2 % FBS, for 48 h. Cells were then treated with vehicle (ethanol), CDCA (50 µM), GW4064 (5 µM) or ADT (10 µM) for 24 or 72 h in the same medium, which was changed after 48 h during the 72 h treatments. Finally, cells were harvested and resuspended in PBS (pH 7.4; 50 mM) with DTT (dithiothreitol; 0.5 mM) and homogenized by pipetting. RNA analysis Total RNA from human hepatocytes was obtained as previously described [28]. Extraction of total RNA from PrEC or LNCaP cells and reverse transcription of 1 µg of RNA were performed using respectively the TRI Reagent acid phenol protocol (Molecular Research Center, Cincinnati, OH, U.S.A.) and 200 units of SuperScript II reverse transcriptase (Invitrogen) as previously reported [23,25]. UGT2B15 and UGT2B17 mRNA levels were determined by real-time PCR with SYBR Green by using the ABI Prism 7500 detection system from Applied Biosystems. For each reaction, the final volume of 20 µl comprised 10 µl of SYBR Green PCR mix, 2 µl of each primer (Table 1) and 6 µl of a reverse transcription product (dilution 1:500). The cycle parameters for real-time PCR were as follows: initial 95 ◦C for 10 min, followed by 40 cycles of 95 ◦C for 15 s and 56 ◦C for 1 min. The specific amplification of each UGT mRNA was ensured by testing PCR strategies with all human UGT2B cDNAs (1.0 ng) and by direct sequencing of PCR products. Threshold cycle (Ct ) values were analysed using the comparative Ct (Ct ) method as described by the manufacturer Farnesoid X receptor reduces androgen glucuronidation in LNCaP cells (Applied Biosystems). The amount of target gene (2−Ct ) was obtained by normalizing to the endogenous reference (28S) and was expressed relatively to vehicle-treated control as the baseline. For each gene, the amplification efficiency and the accuracies of Ct of target genes compared with 28S were tested using 2–5 log of concentrations of cDNA produced from LNCaP-cell purified mRNA. The absolute quantification of FXR cDNA in reversetranscribed mRNA (1 µg) from human hepatocytes, PrEC and LNCaP cells was performed using the absolute standard curve method as recommended by the manufacturer’s instructions (Applied Biosystems). Briefly, FXR cDNA was quantified in both reverse-transcribed mRNA and in aqueous solutions containing increasing amounts (from 1 fg to 1 ng) of FXR cDNA by using the TaqMan® gene expression assay Hs00231968_m1 (Applied Biosystems). Threshold cycle (Ct ) values from the standard curve were interpolated to calculate the FXR mRNA concentration in each reverse-transcribed total RNA. In these experiments, the cycle parameters were: initially 95 ◦C for 10 min, followed by 40 cycles of 95 ◦C for 15 s and 60 ◦C for 1 min. Cell viability After treatment, LNCaP cells were trypsinized, centrifuged and resuspended in PBS containing propidium iodide (20 µg/ml). After 15 min of incubation on ice, the reaction was stopped by adding 25 % ethanol in PBS. Cells were then stained with Hoechst 33 342 (112 µg/ml) for 10 min. Analysis by flow cytometry was performed on an Epics Elite ESP instrument (Beckman Coulter) by using an argon laser (488 nm) and a helium–cadmium laser (325 nm). For each condition, 100 000 cells were evaluated. Protein analyses Total proteins from LNCaP cells were purified according to the TRI Reagent acid phenol protocol. Human liver tissue samples as previously described [28] were homogenized in PBS (pH 7.4; 50 mM) with DTT (0.5 mM). Nuclear extracts of confluent LNCaP cells (approx. 20 million) were prepared by the protocol of Liu et al. [30]. For Western-blot experiments, 1 µg of human liver microsomes (BD Biosciences Discovery Labware, Woburn, MA, U.S.A.) and 10 µg of total proteins were size-separated by SDS/10 %-(w/v)-PAGE and immunoblotted with anti-UGT2B15 or -UGT2B17 antibodies as previously reported [5,12]. The same amount of protein was also immunoblotted with an antiactin antibody as a loading control. Relative protein levels were quantified by using BioImage Visage 110s (Genomic Solution, Ann Arbor, MI, U.S.A.). Glucuronidation assays and steroid quantification Glucuronidation assays were performed for 1 h at 37 ◦C by incubating 3 µg of total protein from homogenized LNCaP cells in the presence of 1 mM UDP-GA, 200 µM ADT, DHT or 3α-DIOL in a glucuronidation assay buffer as previously described [13,25]. Assays were ended by adding 100 µl of methanol with 0.02 % BHT (butylated hydroxytoluene) followed by centrifugation at 14 000 g for 10 min. The formation of glucuronide conjugates was resolved by LC (liquid chromatography) coupled with MS [LC–MS/MS (LC–tandem MS)] by using a previously established analytical method [12,25]. The LC–MS/MS system was composed of an Alliance 2690 apparatus (Waters, Milford, MA, U.S.A.) coupled with an API-3200 mass spectrometer (Applied Biosystems, Concord, ON, Canada). Non-conjugated DHT was determined by GC–MS (gas chromatography coupled with MS) on an HP5973 quadrupole mass spectrometer equipped with a chemical ionization source as previously reported [31]. 247 Statistical analyses All results are presented as means + − S.D. Comparisons between two groups were performed using a two-tailed Student’s t test. Comparisons between three or more groups were made using ANOVA followed by the Dunnett’s procedure for multiple comparisons as the post hoc analysis (SAS V9.1.3 software; SAS Institute, Cary, NC, U.S.A.). In all cases, P 0.05 was considered statistically significant. RESULTS FXR is expressed in LNCaP cells Previous studies reported the presence of FXR transcripts in the human prostate [22]; however, the receptor expression in normal or cancer prostate cells has never been investigated. To address this question, we quantified FXR mRNA levels in total RNA from PrEC in primary culture, prostate cancer LNCaP cells and human hepatocytes (positive control) (Figure 1A). PrEC and LNCaP cells contained similar concentrations of FXR mRNA (54 and 42 fg/µg of total RNA respectively). As expected, FXR mRNA was approx. 30-fold more abundant in human hepatocytes when compared with prostate cells (Figure 1A). It is of interest that such a variation is close to the 56-fold difference in FXR mRNA levels reported between human liver and prostate tissues [22]. Since the presence of the FXR protein has never been demonstrated in the human prostate tissue and in prostate cancer cell lines, Western-blot and immunohistochemistry analyses were performed with nuclear extract from LNCaP cells and slices from the human prostate respectively (Figures 1B and 1C). Westernblot analyses performed with the anti-FXR antibody (ab28676; Abcam) revealed the presence of two bands in both LNCaP cell nuclear extracts and human liver homogenate (positive control) (Figure 1B). These 54 and 56 kDa bands present similar migration properties as the previously reported functional isoforms of FXRα identified in humans: 1–2 and 3–4 [17,32,33]. Interestingly, LNCaP nuclear extracts are enriched in the FXRα3/4 isoforms, which contain an extended N-terminus and are fully active to regulate gene transcription [32,33]. The expression of the FXR proteins in normal prostate tissue has also been investigated through immunohistochemistry, by using two different sources of anti-FXR antibodies (sc-13063; Santa Cruz Biotechnology; and Ab28676; Abcam). Both antibodies yielded an intense staining of nuclei from both androgens-responsive luminal and androgensproducing basal cells of the prostate epithelium (Figure 1C). The nuclear localization of the FXR protein is consistent with labelling reported with anti-FXR antibodies in cell nuclei from other human tissues [34,35]. Overall, these observations demonstrate that FXR mRNA and protein are expressed at significant levels in androgen-sensitive prostate cells. CDCA represses the expression of androgen-glucuronidating UGT enzymes in LNCaP cells To investigate whether the FXR activator CDCA affects UGT2B15 and UGT2B17 gene expressions, LNCaP cells were treated with 50 µM CDCA for the indicated time periods (Figures 2A and 2B). UGT2B15 and UGT2B17 mRNA levels were time-dependently reduced, with maximal inhibition after 72 h of treatment (Figures 2A and 2B). Likewise, treatment of LNCaP cells with increasing concentrations of CDCA (5–50 µM) for 48 h resulted in a dose-dependent decrease in both UGT2B15 and UGT2B17 transcript levels in the range of 15–50 µM (Figures 2C and 2D). A 5 µM portion of CDCA slightly induced UGT2B15 c The Authors Journal compilation c 2008 Biochemical Society 248 J. Kaeding and others not significantly altered after 24 h of CDCA treatment (50 µM; Figure 3A), longer exposure (72 h) resulted in 69 and 78 % reduction in UGT2B15 and UGT2B17 respectively (Figure 3B). Subsequently, enzymatic assays were performed using homogenized LNCaP cells incubated in the presence of ethanol (vehicle) or 50 µM CDCA for 24 or 72 h (Figures 3C and 3D). Apart from DHT, its two major metabolites 3α-DIOL and ADT were used as substrates. After 24 h of exposure to CDCA, no significant changes in the formation of 3α-DIOL-17G, DHT-17G and ADT-3G were observed (Figure 3C). However, after 72 h of treatment, the formation of 3α-DIOL-17G, DHT-17G and ADT3G was 55, 52 and 54 % reduced respectively when compared with vehicle-treated cells (Figure 3D). Bile acids are natural detergents that exert pro-apoptotic and pro-necrotic effects at high concentrations [36]. Thus it was possible that the reduction of UGT2B only reflected non-specific toxicity of CDCA. To exclude this possibility, we performed a flow cytometric cell viability/toxicity analysis using combined staining with propidium iodide and Hoechst 33342 (Figure 4). Treatment with 50 µM CDCA caused significant reduction of cell viability only after 72 h (31 % reduction; P < 0.05; Figure 4), whereas viability of cells treated for 24 and 48 h was not affected by CDCA. In contrast, exposure to CDCA for 96 h resulted in 50 % induced cell mortality, which was not statistically significant due to high variability. We also analysed the effect of the selective FXR agonist GW4064 (5 µM) on cell viability as a negative control (Figure 4). This synthetic activator of FXR did not show any significant effect on cell viability, even after treatment for 96 h. Taken together, these observations demonstrate that CDCA specifically reduces the expression and androgen-conjugating activity of the UGT2B15 and UGT2B17 enzymes in prostate cancer LNCaP cells. FXR mediates the CDCA-dependent inhibition of androgen glucuronidation in LNCaP cells Figure 1 The FXR receptor is expressed in human prostate epithelial and LNCaP cells (A) FXR mRNA levels were quantified in PrEC in primary culture, in LNCaP cells as well as human hepatocytes (HH) by using the absolute standard curve method of real-time PCR as described in the Materials and methods section. (B) LNCaP nuclear extracts (100 µg) were immunoblotted with an anti-human FXR antibody (Abcam; 1:500) using human liver homogenate (100 µg) as a positive control. (C) Normal prostate tissues (n = 4) were subjected to immunohistochemical analyses using two different anti-human FXR antibodies: from Abcam (Ab28676; 1:200; 1, 3, 5) and Santa Cruz Biotechnology (sc-13063; 1:20; 7). A rabbit pre-immune serum was used as the negative control (2, 4, 6, 8). Magnifications are × 400 (1–6) and × 200 (7, 8). Arrows indicate luminal cells and arrowheads indicate basal cells. expression; however, this effect was not statistically significant and lower doses were inefficient in affecting UGT expression in LNCaP cells (results not shown). CDCA decreases UGT2B15 and UGT2B17 protein levels and androgen-glucuronidation activity in LNCaP cells without affecting cell viability Analysis of UGT2B15 and UGT2B17 protein levels in CDCAtreated LNCaP cells further confirmed the time-dependent reduction of UGT2B15 and UGT2B17 enzymes (Figures 3A and 3B). Whereas UGT2B15 and UGT2B17 protein levels were c The Authors Journal compilation c 2008 Biochemical Society Because CDCA may also exert FXR-independent effects, the influence of GW4064 on UGT2B15 and UGT2B17 gene expressions was also investigated (Figure 5). We observed that GW4064 (5 µM) reduced UGT2B15 and UGT2B17 mRNA (Figure 5A) and protein (Figure 5B) levels to a similar order of magnitude as CDCA (Figures 2 and 3). However, in contrast with CDCA, GW4064 already yielded a clear reduction in UGT protein levels as early as 24 h when compared with vehicle-treated cells (Figure 5B). Consistent with the reduced concentration of UGT2B15 and UGT2B17 proteins, we also observed that treatment with GW4064 for 72 h drastically reduced the glucuronide conjugation of DHT, 3α-DIOL and ADT (78, 77 and 73 % reduction respectively) (Figure 5C). Overall, these results confirm that FXR activation reduces UGT2B gene expression and activity in prostate cancer LNCaP cells. ADT also affects androgen glucuronidation in an FXR-dependent manner The 5α-reduced androgen metabolite ADT was recently identified as an activator of FXR [37]. Since ADT is a substrate for UGT2B17 in the prostate, it was of interest to determine whether this androgen derivative could also reduce its own glucuronidation by activating FXR. To verify this hypothesis, LNCaP cells were treated with increasing concentrations of ADT, and levels of UGT2B15 and UGT2B17 mRNA, protein and activity were determined (Figure 6). No significant reduction in UGT2B15 and UGT2B17 mRNA levels was observed in the presence of 10 nM ADT (Figure 6A). However, a significant dose-dependent Farnesoid X receptor reduces androgen glucuronidation in LNCaP cells Figure 2 249 CDCA represses androgen-glucuronidating UGT2B15 and UGT2B17 mRNA expression in human prostate cancer cells (A, B) LNCaP cells were treated with ethanol (vehicle) or CDCA at 50 µM in RPMI 1640 medium supplemented with 0.2 % FBS for the indicated time periods. (C, D) LNCaP cells were treated with ethanol (vehicle) or increasing concentrations of CDCA in RPMI 1640 medium supplemented with 0.2 % FBS for 48 h. UGT2B15 (A, C) and UGT2B17 (B, D) mRNA levels were measured by real-time PCR and normalized to 28S expression. Values are expressed as means + − S.D. (n = 3) and statistically significant differences between control- and CDCA-treated samples are indicated by asterisks [(A, B) Student’s t test; (C, D) one-way ANOVA; n.s., not significant; ∗ P < 0.05; ∗∗ P < 0.005; ∗∗∗ P < 0.001]. reduction in UGT mRNA levels was observed in the presence of 100 nM to 10 µM ADT (Figure 6A). Furthermore, UGT2B protein levels also decreased significantly after 24 h of treatment with ADT (10 µM) and were further reduced after exposure for 72 h (Figure 6B). Likewise, ADT (10 µM) treatment for 72 h resulted in a significant reduction in 3α-DIOL-17G, DHT-17G and ADT-3G formation (Figure 6C). DHT is also a potent reducer of UGT2B15 and UGT2B17 expression in LNCaP cells [25]. Because 3α-HSD enzymes, which can re-convert ADT into DHT, are expressed in LNCaP cells [38], it remained plausible that the reduction in UGT expression observed in the presence of ADT only reflected an effect of the activated AR after re-conversion of ADT into DHT. To test this hypothesis, we determined the formation of DHT in treated LNCaP cells (Table 2) and compared the response to ADT and DHT treatment in the presence or absence of Casodex, a specific AR antagonist (Figures 6D and 6E). Casodex treatment alone did not result in DHT formation at a detectable level in LNCaP cells (Table 2). However, the addition of ADT (10 µM) led to the synthesis of approx. 3 nM DHT in the presence of Casodex (Table 2). When LNCaP cells were treated with the corresponding concentration of DHT (3 nM), UGT2B15 and UGT2B17 levels were reduced by 50 and 35 % respectively (Figures 6D and 6E). This effect was significantly weaker than treatment with 10 µM ADT, which resulted in 85 and 75 % reduction of UGT2B15 and UGT2B17 mRNA expression respectively (Figures 6D and 6E). In addition, co-treatment with Casodex resulted in a complete loss of the DHT-dependent inhibition of the expression of both UGT enzymes, but failed to significantly alter the effect of ADT (Figures 6D and 6E). These results demonstrate that ADT inhibits androgen glucuronidation in LNCaP cells in an AR-independent manner. DISCUSSION The present study demonstrates the presence of FXR in epithelial cells of the human prostate and establishes its activators as important regulators of androgen metabolism in prostate cancer LNCaP cells. FXR is classically considered as a bile acid sensor that regulates hepatic and intestinal bile acid synthesis, metabolism and transport [16]. However, this receptor also exerts important functions in the control of lipid and glucose homoeostasis [17]. The results presented here further reinforce the idea that FXR plays different physiological roles in various extrahepatic tissues. FXR activation caused a drastic reduction of UGT2B15 and UGT2B17 gene expression, resulting in a decreased glucuronide conjugation of the active androgen DHT and its reduced metabolites ADT and 3α-DIOL. The observation that FXR affects UGT2B15 and UGT2B17 gene expressions in an identical manner is consistent with the very high homology of their proximal promoter regions (more than 90 % nucleotide homology within the proximal 1600 bp sequences of the UGT2B15 and UGT2B17 genes) [39,40] and supports the idea that both genes share an identical FXR response element within this region. In human c The Authors Journal compilation c 2008 Biochemical Society 250 Figure 3 J. Kaeding and others CDCA decreases UGT2B15 and UGT2B17 protein levels and glucuronidation of androgen metabolites in LNCaP cells (A, B) LNCaP cells were treated with ethanol (vehicle) or CDCA (50 µM) in RPMI 1640 medium supplemented with 0.2 % FBS for the indicated time periods. Then, 10 µg of total protein of each sample was immunoblotted with the anti-UGT2B15 antibody (1:1500; top panels) or the anti-UGT2B17 antibody (1:2000; middle panels). Human liver microsomes (2 µg) were used as a positive control. The equal loading of each lane was ensured by hybridizing with an anti-human actin antibody (1:1000; bottom panels). Immunoreactions with anti-UGT antibodies were quantified by densitometry and are expressed as relative quantity of proteins compared with vehicle-treated cells (set as 1). (C, D) LNCaP cells were treated with ethanol (vehicle) or CDCA (50 µM) in RPMI 1640 medium supplemented with 0.2 % FBS, for 24 h (C) or 72 h (D), and glucuronidation assays were performed with 3α-DIOL, ADT and DHT as substrates (200 µM) using 3 µg of cell homogenates. Glucuronide formation was quantified by LC–MS/MS analyses and corrected according to the amount of protein. Values (means + − S.D.; n = 4) are expressed as µg of glucuronide formation/µg of protein per hour, and statistically significant differences between control- and CDCA-treated samples are indicated by asterisks (Student’s t test; n.s., not significant; ∗ P < 0.05; ∗∗∗ P < 0.001). Figure 4 Treatment with FXR agonists does not significantly interfere with LNCaP cell viability LNCaP cells were treated with ethanol (vehicle) or CDCA at 50 µM in RPMI 1640 medium supplemented with 0.2 % FBS for the indicated time periods. Cell viability was determined by flow cytometry after staining with propidium iodide and Hoechst 33342 as described in the Materials and methods section. Significant differences between control- and CDCA-/GW4064-treated samples are indicated by asterisks (Student’s t test; ∗ P < 0.05). prostate epithelium, the UGT2B15 protein is found in luminal cells, whereas UGT2B17 is expressed in the basal layer [5,25]. Thus the presence of FXR protein in the two cell types is also consistent with its regulatory role in the control of the respective UGT enzyme expression. Nevertheless, the exact mechanism by which FXR negatively regulates UGT2B15 and UGT2B17 genes remains to be determined. Negative FXR response elements have been reported in bile acid-down-regulated genes [24,41], including the human UGT2B7 gene [24]. However, bioinformatics analysis of the UGT2B15 and UGT2B17 promoter se c The Authors Journal compilation c 2008 Biochemical Society quences failed to identify such a responsive sequence (results not shown). While we cannot exclude the presence of other negative FXR elements within UGT promoters, this last observation does not support a direct down-regulation of the UGT2B15 and UGT2B17 genes by FXR. On the other hand, FXR also negatively regulates gene transcription by interfering with other regulatory pathways. Indeed, FXR was suggested as a mediator of the bile acid-dependent activation of the AP-1 (activator protein-1) transcription pathway in colon cancer cells [2]. Considering the importance of AP-1 factors in the pathogenesis of prostate cancer [8,42], it would be possible that the FXR-dependent regulatory pathway of UGT2B genes in LNCaP cells involves AP-1 factors. Such a hypothesis is supported by the presence of at least four conserved AP-1-binding sites within the highly homologous 1.6 kb proximal promoter sequences of both UGT2B15 and UGT2B17 genes [39,40]. The FXR-dependent inhibition of androgen-glucuronidating UGT2B enzymes in LNCaP cells was unexpected, since previous studies reported that CDCA fails to modulate UGT2B15 mRNA levels in human hepatocytes [23]. However, the present observation suggests that FXR regulates UGT2B15 expression in a tissue-specific manner. Similar observations were reported for the UGT2B7 gene, which was inhibited by bile acid-activated FXR in colon carcinoma Caco-2 cells, while remaining unaffected in hepatocytes [23,24]. It is generally accepted that such a differential regulation may be associated with the tissue-specific balance between nuclear receptor co-activators and co-repressors [18]. We also observed that Ugt2b expression seems to increase in the prostate from FXR-null mice, compared with wild-type Farnesoid X receptor reduces androgen glucuronidation in LNCaP cells 251 Figure 5 GW4064 also reduces UGT2B15 and UGT2B17 expression and androgen glucuronidation activity in LNCaP cells (A) LNCaP cells were treated with ethanol (vehicle) or GW4064 (5 µM) in RPMI 1640 medium supplemented with 0.2 % FBS for 24 h. UGT2B15 and UGT2B17 mRNA levels were measured by real-time PCR and normalized to 28S expression. Values are expressed as means + − S.D. (n = 3) and statistically significant differences between control- and GW4064-treated samples are indicated by asterisks (Student’s t test: ∗∗∗ P < 0.001). (B) LNCaP cells were treated with ethanol (vehicle) or GW4064 (5 µM) in RPMI 1640 medium supplemented with 0.2 % FBS for 24 h. Then, 10 µg of total protein of each sample was immunoblotted with the anti-UGT2B15 antibody (1:1500; top panel) or the anti-UGT2B17 antibody (1:2000; middle panel). Human liver microsomes (HL) (2 µg) were used as a positive control. The equal loading of each lane was ensured by hybridizing with an anti-human actin antibody (1:1000; lower panel). Immunoreactions with anti-UGT antibodies were quantified by densitometry and are expressed as relative quantity of protein content compared with vehicle-treated cells (set as 1). (C) LNCaP cells were treated with ethanol (vehicle) or GW4064 (5 µM) in RPMI 1640 medium supplemented with 0.2 % FBS, for 72 h, and glucuronidation assays were performed with 3α-DIOL, ADT and DHT (200 µM) by using 3 µg of cell homogenate as described in the Materials and methods section. Glucuronide formation was quantified by LC–MS/MS analysis and corrected according to the amount of protein. Statistically significant differences between control- and GW4064-treated samples are indicated by asterisks (Student’s t test; n.s., not significant; ∗∗ P < 0.005; ∗∗∗ P < 0.001). animals (see Supplementary Figure 1 at http://www.BiochemJ. org/bj/410/bj4100245add.htm). While being consistent with the idea that FXR down-regulates UGT2B enzymes in the prostate, the physiological implications of this observation are unclear since androgen glucuronidation is almost absent in the rodent prostate [2,43,44]. While being structurally different, the bile acid CDCA and the 5α-reduced androgen metabolite ADT are two ultimate chol- Figure 6 Elevated concentrations of ADT also affect androgen-conjugating UGT expression and activity in an FXR-dependent manner (A) LNCaP cells were treated with ethanol (vehicle) or ADT at indicated concentrations in RPMI 1640 medium supplemented with 0.2 % FBS for 48 h. UGT2B15 and UGT2B17 mRNA levels were measured by real-time PCR and normalized to 28S. Values are expressed as means + − S.D. (n = 3) and statistically significant differences between control- and ADT-treated samples are indicated by asterisks (one-way ANOVA; n.s., not significant; ∗ P < 0.05; ∗∗∗ P < 0.001). (B) LNCaP cells were treated with ethanol (vehicle) or ADT (10 µM) in RPMI 1640 medium supplemented with 0.2 % FBS for 24 or 72 h. For each sample, 10 µg of cell homogenate was immunoblotted with the anti-UGT2B15 antibody (1:1500; top panel) or the anti-UGT2B17 antibody (1:2000; middle panel). The equal loading of each lane was ensured by hybridizing with an anti-human actin antibody (1:1000; bottom panel). Immunoreactions with anti-UGT antibodies were quantified by densitometry and are expressed as relative quantity of protein content compared with vehicle-treated cells (set as 1). (C) LNCaP cells were treated with ethanol (vehicle) or ADT (10 µM) in RPMI 1640 medium supplemented with 0.2 % FBS, for 72 h, and glucuronidation assays were performed with 200 µM of 3α-DIOL, ADT or DHT by using 3 µg of cell homogenate. Glucuronide formation was quantified by LC–MS/MS analysis and corrected for the amount of protein. Values (means + − S.D.; n = 4) are expressed as µg of glucuronide formation/µg of protein per hour, and statistically significant differences between control- and GW4064-treated samples are indicated by asterisks (Student’s t test; ∗∗ P < 0.005; ∗∗∗ P < 0.001). (D) LNCaP cells were treated with ethanol (vehicle), ADT (10 µM) and DHT (3 nM) in the presence or absence of Casodex (10 µM) in RPMI 1640 medium supplemented with 0.2 % FBS for 24 h. UGT2B15 and UGT2B17 mRNA levels were measured by real-time PCR and normalized to 28S. Values are expressed as means + − S.D. (n = 3). Statistically significant differences between control and ADT/DHT treatments, as well as their effects in the presence or absence of casodex, were determined by two-way ANOVA and are indicated by asterisks (n.s., not significant; ∗ P < 0.05; ∗∗ P < 0.005; ∗∗∗ P < 0.001). c The Authors Journal compilation c 2008 Biochemical Society 252 Table 2 J. Kaeding and others DHT formation in LNCaP cells in the presence of Casodex BLQ, below limit of quantification. Treatment [DHT] (nM) Casodex (10 µM) ADT (10 µM) + Casodex (10 µM) BLQ 3.04 + − 0.45 esterol catabolites, which have been identified as FXR activators [37,45]. Since the exact mechanism by which FXR negatively regulates UGT2B15 and UGT2B17 genes remains to be determined in the prostate, it was of interest to investigate whether ADT can affect its own glucuronidation. We observed that ADT modulates UGT2B15 and UGT2B17 expression and activity at concentrations above 100 nM. While this concentration is drastically elevated when compared with levels of unconjugated ADT found in human circulation (i.e. approx. 1–3 nM) [46,47], it is more similar to the concentration range of circulating ADT3G (i.e. 35–190 nM) [2,47–50]. Because ADT is efficiently glucuronidated and constitutes the most abundant androgen metabolite found in humans, this observation suggests that intratissular concentrations of ADT may be much higher than expected from plasma quantification. Thus it is likely that ADTdependent activation of FXR is of physiological importance and is able to result in decreased UGT in the human prostate tissue. Furthermore, ADT activates FXR at a concentration close to the K m value of its glucuronidation by UGT2B17: 400–600 nM [14,51]. From this point of view, ADT glucuronidation may be considered as a relevant inactivating pathway, which can compete with the activation of FXR by ADT, as was previously demonstrated for CDCA [28]. The reduction of UGT2B17 expression in the presence of ADT may therefore be interpreted as a feed-forward mechanism by which this androgen metabolite is able to sustain its FXR-mediated biological activity through reduction of its inactivation. Therefore our results provide the first evidence for a likely physiological role of ADT as an FXR ligand in the human prostate. However, it is also possible that FXR is activated by a yet unidentified endogenous metabolite present at physiologically relevant levels in the human prostate. The present study constitutes the first evidence that FXR plays an important role in the control of androgen metabolism in prostate cells. However, activation of the FXR regulatory pathway also negatively interferes with the PMA-dependent induction of aromatase gene expression in human breast cancer cells [18]. Aromatase catalyses the conversion of androgens into oestrogens. Together with the present study, this observation suggests that in hormone-sensitive tissues, such as the mammary and prostate glands, FXR activation may result in accumulation of androgens. In prostate cancer cells, FXR activation and subsequent reduction of androgen glucuronidation may correspond to a procarcinogenic mechanism. Indeed, it has already been reported that stimuli that reduce androgen glucuronidation also stimulate prostate cancer cell proliferation [2,26]. However, Choi et al. [52] reported that CDCA was ineffective in modulating prostate carcinoma PC-3 cell growth, which does not support the abovestated hypothesis. However, and in contrast with LNCaP, PC-3 cells do not express the androgen-conjugating UGT2B enzymes and grow in an androgen-independent manner [53]. Consequently, the effects we report here would have not been observed in PC-3 cells. Our results clearly show that FXR is expressed in PrEC, where its activation reduces the glucuronide conjugation of androgens. Interestingly, the development of cholestasis has been reported in c The Authors Journal compilation c 2008 Biochemical Society various patients with prostate cancer [54–57]. During cholestasis, plasma levels of bile acids are drastically increased [16], and based on the present study, it is tempting to speculate that, in such patients, glucuronidation of androgen may be reduced in the prostate tissue. On the other hand, GW4064 is considered as a promising treatment for cholestasis [16], and based on the present study, it would be of importance to monitor circulating markers of prostate cancer in GW4064-treated patients. In conclusion, the present study shows for the first time a regulatory function of the nuclear receptor FXR in androgen metabolism in prostate cancer LNCaP cells. In concert with other recent reports, this suggests that FXR may be an important, yet not well studied, regulator of androgen homoeostasis. This work was supported by the CIHR (Canadian Institute of Health Research) (MOP118446 to O. B. and MOP64273 to A. B.), FRSQ (Fonds de la Recherche en Santé du Québec) and CFI (Canada Foundation for Innovation) (to O. B.) and the Faculty of Pharmacy, Laval University. S. C. is supported by ‘Fonds pour la Recherche et l’Enseignement’ (to R. V. and S. K.) from the CIHR and O. B. and J. K. by the Health Research Foundation of Rx&D-CIHR. We thank Endorecherche for providing Casodex and Professor F. Gonzalez (National Institutes of Health, Bethesda, MD, U.S.A.) for providing the FXR-null animals. We are grateful to Valérie Jomphe from the Department of Statistics, Laval University, for her assistance with statistical analysis. We also thank Dr Virginie Bocher for a critical reading of this paper and Dr Chantal Guillemette for helpful discussions. REFERENCES 1 Mackenzie, P., Walter Bock, K., Burchell, B., Guillemette, C., Ikushiro, S., Iyanagi, T., Miners, J., Owens, I. S. and Nebert, D. W. (2005) Nomenclature update for the mammalian UDP glycosyltransferase (UGT) gene superfamily. Pharmacogenet. Genomics 15, 677–685 2 Bélanger, A., Pelletier, G., Labrie, F., Barbier, O. and Chouinard, S. (2003) Inactivation of androgens by UDP-glucuronosyltransferase enzymes in humans. Trends Endocrinol. Metab. 14, 473–479 3 Guillemette, C., Bélanger, A. and Lépine, J. (2004) Metabolic inactivation of estrogens in breast tissue by UDP-glucuronosyltransferase enzymes: an overview. Breast Cancer Res. 6, 246–254 4 Barbier, O. and Bélanger, A. (2003) The cynomolgus monkey (Macaca fascicularis ) is the best animal model for the study of steroid glucuronidation. J. Steroid Biochem. Mol. Biol. 85, 235–245 5 Barbier, O., Lapointe, H., El Alfy, M., Hum, D. W. and Bélanger, A. (2000) Cellular localization of uridine diphosphoglucuronosyltransferase 2B enzymes in the human prostate by in situ hybridization and immunohistochemistry. J. Clin. Endocrinol. Metab. 85, 4819–4826 6 Soronen, P., Laiti, M., Torn, S., Harkonen, P., Patrikainen, L., Li, Y., Pulkka, A., Kurkela, R., Herrala, A. and Kaija, H. (2004) Sex steroid hormone metabolism and prostate cancer. J. Steroid Biochem. Mol. Biol. 92, 281–286 7 Platz, E. A. and Giovannucci, E. (2004) The epidemiology of sex steroid hormones and their signaling and metabolic pathways in the etiology of prostate cancer. J. Steroid Biochem. Mol. Biol. 92, 237–253 8 Park, J., Chen, L., Ratnashinge, L., Sellers, T. A., Tanner, J.-P., Lee, J.-H., Dossett, N., Lang, N., Kadlubar, F. F., Ambrosone, C. B. et al. (2006) Deletion polymorphism of UDP-glucuronosyltransferase 2B17 and risk of prostate cancer in African American and Caucasian men. Cancer Epidemiol. Biomarkers Prev. 15, 1473–1478 9 Park, J., Chen, L., Shade, K., Lazarus, P., Seigne, J., Patterson, S., Helal, M. and Pow-Sang, J. (2004) Asp85tyr polymorphism in the UDP-glucuronosyltransferase (UGT) 2B15 gene and the risk of prostate cancer. J. Urol. 171, 2484–2488 10 Hajdinjak, T. and Zagradisnik, B. (2004) Prostate cancer and polymorphism D85Y in gene for dihydrotestosterone degrading enzyme UGT2B15: frequency of DD homozygotes increases with Gleason score. Prostate 59, 436–439 11 MacLeod, S. L., Nowell, S., Plaxco, J. and Lang, N. P. (2000) An allele-specific polymerase chain reaction method for the determination of the D85Y polymorphism in the human UDP-glucuronosyltransferase 2B15 gene in a case-control study of prostate cancer. Ann. Surg. Oncol. 7, 777–782 12 Chouinard, S., Pelletier, G., Bélanger, A. and Barbier, O. (2004) Cellular specific expression of the androgen-conjugating enzymes UGT2B15 and UGT2B17 in the human prostate epithelium. Endocr. Res. 30, 717–725 13 Turgeon, D., Carrier, J.-S., Lévesque, E., Hum, D. W. and Bélanger, A. (2001) Relative enzymatic activity, protein stability, and tissue distribution of human steroid-metabolizing UGT2B subfamily members. Endocrinology 142, 778–787 Farnesoid X receptor reduces androgen glucuronidation in LNCaP cells 14 Dubois, S., Beaulieu, M., Lévesque, E., Hum, D. W. and Bélanger, A. (1999) Alteration of human UDP-glucuronosyltransferase UGT2B17 regio-specificity by a single amino acid substitution. J. Mol. Biol. 289, 29–39 15 Laffitte, B. A., Kast, H. R., Nguyen, C. M., Zavacki, A. M., Moore, D. D. and Edwards, P. A. (2000) Identification of the DNA binding specificity and potential target genes for the farnesoid X-activated receptor. J. Biol. Chem. 275, 10638–10647 16 Trottier, J., Milkiewicz, P., Kaeding, J., Verreault, M. and Barbier, O. (2006) Coordinate regulation of hepatic bile acid oxidation and conjugation by nuclear receptors. Mol. Pharm. 3, 212–222 17 Lee, F. Y., Lee, H., Hubbert, M. L., Edwards, P. A. and Zhang, Y. (2006) FXR, a multipurpose nuclear receptor. Trends Biochem. Sci. 31, 572–580 18 Swales, K. E., Korbonits, M., Carpenter, R., Walsh, D. T., Warner, T. D. and Bishop-Bailey, D. (2006) The farnesoid X receptor is expressed in breast cancer and regulates apoptosis and aromatase expression. Cancer Res. 66, 10120–10126 19 Howard, W. R., Pospisil, J. A., Njolito, E. and Noonan, D. J. (2000) Catabolites of cholesterol synthesis pathways and forskolin as activators of the farnesoid X-activated nuclear receptor. Toxicol. Appl. Pharmacol. 163, 195–202 20 Pircher, P. C., Kitto, J. L., Petrowski, M. L., Tangirala, R. K., Bischoff, E. D., Schulman, I. G. and Westin, S. K. (2003) Farnesoid X receptor regulates bile acid–amino acid conjugation. J. Biol. Chem. 278, 27703–27711 21 Miyata, M., Matsuda, Y., Tsuchiya, H., Kitada, H., Akase, T., Shimada, M., Nagata, K., Gonzalez, F. J. and Yamazoe, Y. (2006) Chenodeoxycholic acid-mediated activation of the farnesoid X receptor negatively regulates hydroxysteroid sulfotransferase. Drug Metab. Pharmacokinet. 21, 315–323 22 Nishimura, M., Naito, S. and Yoko, T. (2004) Tissue-specific mRNA expression profiles of human nuclear receptor subfamilies. Drug Metab. Pharmacokinet. 19, 135–149 23 Barbier, O., Torra, I. P., Sirvent, A., Claudel, T., Blanquart, C., Duran-Sandoval, D., Kuipers, F., Kosykh, V., Fruchart, J. C. and Staels, B. (2003) FXR induces the UGT2B4 enzyme in hepatocytes: a potential mechanism of negative feedback control of FXR activity. Gastroenterology 124, 1926–1940 24 Lu, Y., Heydel, J.-M., Li, X., Bratton, S., Lindblom, T. and Radominska-Pandya, A. (2005) Lithocholic acid decreases expression of UGT2B7 in Caco-2 cells: a potential role for a negative farnesoid X receptor response element. Drug Metab. Dispos. 33, 937–946 25 Chouinard, S., Pelletier, G., Bélanger, A. and Barbier, O. (2006) Isoform-specific regulation of UDP-glucuronosyltransferase (UGT)2B enzymes in the human prostate: differential consequences for androgen and bioactive lipid inactivation. Endocrinology 147, 5431–5442 26 Guillemette, C., Lévesque, E., Beaulieu, M., Turgeon, D., Hum, D. W. and Bélanger, A. (1997) Differential regulation of two uridine diphospho-glucuronosyltransferases, UGT2B15 and UGT2B17, in human prostate LNCaP cells. Endocrinology 138, 2998–3005 27 Barbier, O., Lapointe, H., El Alfy, M., Hum, D. W. and Bélanger, A. (2000) Cellular localization of uridine diphosphoglucuronosyltransferase 2B enzymes in the human prostate by in situ hybridization and immunohistochemistry. J. Clin. Endocrinol. Metab. 85, 4819–4826 28 Trottier, J., Verreault, M., Grepper, S., Monté, D., Bélanger, J., Kaeding, J., Caron, P., Inaba, T. T. and Barbier, O. (2006) Human UDP-glucuronosyltransferase (UGT)1A3 enzyme conjugates chenodeoxycholic acid in the liver. Hepatology 44, 1158–1170 29 Reference deleted 30 Liu, Y.-Y., Nguyen, C. and Peleg, S. (2000) Regulation of ligand-induced heterodimerization and coactivator interaction by the activation function-2 domain of the vitamin D receptor. Mol. Endocrinol. 14, 1776–1787 31 Labrie, F., Bélanger, A., Bélanger, P., Bérubé, R., Martel, C., Cusan, L., Gomez, J., Candas, B., Chaussade, V., Castiel, I. et al. (2007) Metabolism of DHEA in postmenopausal women following percutaneous administration. J. Steroid Biochem. Mol. Biol. 103, 178–188 32 Zhang, Y., Kast-Woelbern, H. R. and Edwards, P. A. (2003) Natural structural variants of the nuclear receptor farnesoid X receptor affect transcriptional activation. J. Biol. Chem. 278, 104–110 33 Huber, R. M., Murphy, K., Miao, B., Link, J. R., Cunningham, M. R., Rupar, M. J., Gunyuzlu, P. L., Haws, T. F., Kassam, A., Powell, F. et al. (2002) Generation of multiple farnesoid-X-receptor isoforms through the use of alternative promoters. Gene 290, 35–43 34 De Gottardi, A., Dumonceau, J. M., Bruttin, F., Vonlaufen, A., Morard, I., Spahr, L., Rubbia-Brandt, L., Frossard, J. L., Dinjens, W. N., Rabinovitch, P. S. et al. (2006) Expression of the bile acid receptor FXR in Barrett’s esophagus and enhancement of apoptosis by guggulsterone in vitro . Mol. Cancer 5, 48 253 35 Bishop-Bailey, D., Walsh, D. T. and Warner, T. D. (2004) Expression and activation of the farnesoid X receptor in the vasculature. Proc. Natl. Acad. Sci. U.S.A. 101, 3668–3673 36 Trauner, M. and Boyer, J. L. (2003) Bile salt transporters: molecular characterization, function, and regulation. Physiol. Rev. 83, 633–671 37 Wang, S., Lai, K., Moy, F. J., Bhat, A., Hartman, H. B. and Evans, M. J. (2006) The nuclear hormone receptor farnesoid X receptor (FXR) is activated by androsterone. Endocrinology 147, 4025–4033 38 Simard, J., Ricketts, M. L., Gingras, S., Soucy, P., Feltus, F. A. and Melner, M. H. (2005) Molecular biology of the 3beta-hydroxysteroid dehydrogenase/delta5-delta4 isomerase gene family. Endocr. Rev. 26, 525–582 39 Beaulieu, M., Lévesque, É., Tchernof, A., Beatty, B. G., Bélanger, A. and Hum, D. W. (1997) Chromosomal localization, structure, and regulation of the UGT2B17 gene, encoding a C19 steroid metabolizing enzyme. DNA Cell Biol. 16, 1143–1154 40 Turgeon, D., Carrier, J. S., Lévesque, É., Beatty, B. G., Bélanger, A. and Hum, D. W. (2000) Isolation and characterization of the human UGT2B15 gene, localized within a cluster of UGT2B genes and pseudogenes on chromosome 4. J. Mol. Biol. 295, 489–504 41 Claudel, T., Sturm, E., Duez, H., Torra, I. P., Sirvent, A., Kosykh, V., Fruchart, J. C., Dallongeville, J., Hum, D. W., Kuipers, F. et al. (2002) Bile acid-activated nuclear receptor FXR suppresses apolipoprotein A-I transcription via a negative FXR response element. J. Clin. Invest. 109, 961–971 42 Jochum, W., Passegue, E. and Wagner, E. F. (2001) AP-1 in mouse development and tumorigenesis. Oncogene 20, 2401–2412 43 Barbier, O. and Bélanger, A. (2003) The cynomolgus monkey (Macaca fascicularis ) is the best animal model for the study of steroid glucuronidation. J. Steroid Biochem. Mol. Biol. 85, 235–245 44 Guillemette, C., Hum, D. W. and Bélanger, A. (1996) Levels of plasma C19 steroids and 5 alpha-reduced C19 steroid glucuronides in primates, rodents, and domestic animals. Am. J. Physiol. Endocrinol. Metab. 271, E348–E353 45 Parks, D. J., Blanchard, S. G., Bledsoe, R. K., Chandra, G., Consler, T. G., Kliewer, S. A., Stimmel, J. B., Willson, T. M., Zavacki, A. M., Moore, D. D. et al. (1999) Bile acids: natural ligands for an orphan nuclear receptor. Science 284, 1365–1368 46 Williams, R. H., Larsen, P. R., Kronenberg, H. M., Melmed, S., Polonsky, K. S., Wilson, J. D. and Foster, D. W. (2002) Williams Text Book of Endocrinology, Saunders, Philadelphia, PA, U.S.A. 47 Bélanger, A., Candas, B., Dupont, A., Cusan, L., Diamond, P., Gomez, J. and Labrie, F. (1994) Changes in serum concentrations of conjugated and unconjugated steroids in 40–80-year-old men. J. Clin. Endocrinol. Metab. 79, 1086–1090 48 Hong, Y., Gagnon, J., Rice, T., Perusse, L., Leon, A., Skinner, J., Wilmore, J., Bouchard, C. and Rao, D. (2001) Familial resemblance for free androgens and androgen glucuronides in sedentary black and white individuals: the HERITAGE family study. Health, risk factors, exercise training and genetics. J. Endocrinol. 170, 485–492 49 Guillemette, C., Hum, D. W. and Bélanger, A. (1996) Levels of plasma C19 steroids and 5 alpha-reduced C19 steroid glucuronides in primates, rodents, and domestic animals. Am. J. Physiol. Endocrinol. Metab. 271, E348–E353 50 Wu, A. H., Whittemore, A. S., Kolonel, L. N., Stanczyk, F. Z., John, E. M., Gallagher, R. P. and West, D. W. (2001) Lifestyle determinants of 5{alpha}-reductase metabolites in older African-American, white, and Asian-American men. Cancer Epidemiol. Biomarkers Prev. 10, 533–538 51 Beaulieu, M., Lévesque, E., Hum, D. W. and Bélanger, A. (1996) Isolation and characterization of a novel cDNA encoding a human UDP-glucuronosyltransferase active on C19 steroids. J. Biol. Chem. 271, 22855-22862 52 Choi, Y. H., Im, E. O., Suh, H., Jin, Y., Yoo, Y. H. and Kim, N. D. (2003) Apoptosis and modulation of cell cycle control by synthetic derivatives of ursodeoxycholic acid and chenodeoxycholic acid in human prostate cancer cells. Cancer Lett. 199, 157–167 53 Koh, E., Kanaya, J. and Namiki, M. (2001) Adrenal steroids in human prostatic cancer cell lines. Arch. Androl. 46, 117–125 54 Ben-Ishay, D., Slavin, S., Levij, I. S. and Eliakim, M. (1975) Obstructive jaundice associated with carcinoma of the prostate. Isr. J. Med. Sci. 11, 838–844 55 Bloch, W. E. and Block, N. L. (1992) Metastatic prostate cancer presenting as obstructive jaundice. Urology 40, 456–457 56 Koruk, M., Buyukberber, M., Savas, C. and Kadayifci, A. (2004) Paraneoplastic cholestasis associated with prostate carcinoma. Turk. J. Gastroenterol. 15, 53–55 57 Karakolios, A., Kasapis, C., Kallinikidis, T., Kalpidis, P. and Grigoriadis, N. (2003) Cholestatic jaundice as a paraneoplastic manifestation of prostate adenocarcinoma. Clin. Gastroenterol. Hepatol. 1, 480–483 Received 17 August 2007/26 October 2007; accepted 7 November 2007 Published as BJ Immediate Publication 7 November 2007, doi:10.1042/BJ20071136 c The Authors Journal compilation c 2008 Biochemical Society