Departamento de Bioquímica y Biología Molecular, Facultad de Medicina, Universidad de Alcalá, Alcalá de Henares, 28871 Madrid, Spain. E-mail: email@example.com
The response of anandamide is terminated by a carrier-mediated transport followed by degradation catalyzed by the cloned enzyme fatty acid amidohydrolase (FAAH). In this study, we provide biochemical data showing an anandamide uptake process and the expression of FAAH in human prostate.
Anandamide was accumulated in PC-3 cells by a saturable and temperature-dependent process. Kinetic studies of anandamide uptake, determined in the presence of cannabinoid and vanilloid antagonists, revealed apparent parameters of KM=4.7±0.2 μM and Vmax=3.3±0.3 pmol min−1 (106 cells)−1.
The accumulation of anandamide was moderately inhibited by previously characterized anandamide transporter inhibitors (AM404, UCM707 and VDM11) but was unaffected by inhibitors of other lipid transport systems (phloretin or verapamil) and moderately affected by the FAAH inhibitor methyl arachidonyl fluorophosphonate.
The presence of FAAH in human prostate epithelial PC-3 cells was confirmed by analyzing its expression by Western blot and measuring FAAH activity.
To further study the structural requirements of the putative carrier, we synthesized a series of structurally different compounds 1–8 and evaluated their capacity as uptake inhibitors. They showed different inhibitory capacity in PC-3 cells, with (9Z,12Z)-N-(fur-3-ylmethyl)octadeca-9,12-dienamide (4, UCM119) being the most efficacious, with maximal inhibition and IC50 values of 49% and 11.3±0.5 μM, respectively.
In conclusion, PC-3 cells possess a complete inactivation system for anandamide formed by an uptake process and the enzyme FAAH. These results suggest a possible physiological function of anandamide in the prostate, reinforcing the role of endocannabinoid system as a neuroendocrine modulator.
Although research on the molecular bases of the physiological effects of marijuana use was slowed for many years due to the lack of specific technology, much progress has been achieved in cannabinoid research in the last decade. Many of the physiological and pharmacological actions of plant-derived cannabinoids are mimicked by endogenous substances named endocannabinoids. The first characterized endocannabinoid was N-arachidonoylethanolamine (anandamide, ANA) (Devane et al., 1992), which has been found in several mammalian tissues and reproductive fluids (Schuel et al., 2002). Endocannabinoids are now presented as new neuromodulators acting through specific cannabinoid receptors (Di Marzo et al., 1998; Wilson & Nicoll, 2002). To date, two cannabinoid receptors that belong to the seven-transmembrane G protein-coupled receptors superfamily, CB1 and CB2 (cannabinoid receptor of type 1 and 2, respectively), have been cloned in humans (Sugiura & Waku, 2002). CB1 receptors are expressed in several brain regions (reviewed in Ameri, 1999) and also in some peripheral tissues and cell lines, whereas CB2 receptors are mostly confined to immune tissues (Klein et al., 2001). Additionally, anandamide can also activate the binding site for capsaicin, the pungent principle in hot pepper, named vanilloid VR1 (vanilloid receptor of type 1) receptor (Smart et al., 2000). The termination of the action of anandamide is achieved through its uptake by cells and subsequent enzymatic degradation, catalyzed by the specific enzyme fatty acid amidohydrolase (FAAH) (Giuffrida et al., 2001; Deutsch et al., 2002; Cravatt & Lichtman, 2003). The uptake of anandamide by intact neurons is mediated by a high-affinity, rapid, saturable, selective and temperature-dependent carrier that is not affected by fatty acids, arachidonate metabolites, neurotransmitters or biogenic amines (firstly described by Di Marzo et al. (1994) and reviewed in Giuffrida et al. (2001), Hillard & Jarrahian (2000) and Fowler & Jacobsson (2002)). Mechanisms of endocannabinoid inactivation may also exist in peripheral tissues, and they have been detected in macrophages and rat basophilic leukemia cells (RBL-2H3) (Bisogno et al., 1997), platelets (Maccarrone et al., 2001), human endothelial cells (Maccarrone et al., 2000a) and mouse Sertoli cells (Maccarrone et al., 2003). Because of this rapid deactivation process, it could be deemed likely that cannabinoid receptors and the putative cannabinoid transporter are close together in the same or neighboring cells. It has been previously described that human prostate epithelial cells (Ruiz et al., 1999; Melck et al., 2000; Velasco et al., 2001) and human prostate gland (Ruiz-Llorente et al., 2003) express cannabinoid receptors suggesting a functional role of cannabinoids in the prostate gland. In fact, morphological and functional alterations of human prostate gland have been observed in chronic exposure of laboratory animals to (−)-Δ9-tetrahydrocannabinol (THC) (Harclerode, 1984) and in male rats exposed in utero to THC (Ahluwalia et al., 1985). It has also been described that cannabinol and THC inhibit specific binding of dihydrotestosterone to the androgen receptor in the prostate gland (Purohit et al., 1980). Moreover, there is evidence that suggests that the endocannabinoid system has an important role in the regulation of endocrine functions, and recent results support the consideration of endocannabinoids as neuroendocrine modulators (Wenger et al., 2001; Maccarrone et al., 2002, 2003; Wenger & Moldrich, 2002). Accordingly, it is of importance to determine the mechanisms of degradation of endocannabinoids, in particular anandamide as the most representative one, in target glands of the hypothalamic-pituitary-gonadal axis. It has been recently shown that cannabinoids regulate the proliferation of prostate cells (Melck et al., 2000; Mimeault et al., 2003; Sánchez et al., 2003), suggesting that cannabinoids not only modulate hormone levels but have a direct action on prostate development. In this study we focus on the degradation system of anandamide in prostate, an issue that has remained unapproached to date. We show that anandamide is accumulated in prostate carcinome cells (PC-3) cells by a saturable and temperature-dependent process. The accumulation of anandamide is inhibited by previously characterized anandamide transporter inhibitors such as N-(4-hydroxyphenyl)arachidonamide (AM404) and N-(fur-3-ylmethyl)arachidonamide (UCM707), which block transport with moderate efficacy, and N-(4-hydroxy-2-methylphenyl) arachidonamide (VDM11), which caused a 49% of inhibition. New synthesized inhibitors (1–8) were also tested for their efficacy to inhibit anandamide transporter. Among them, (9Z,12Z)-N-(fur-3-ylmethyl)octadeca-9,12-dienamide (compound 4, UCM119) deserves special attention as it was as potent as VDM11 in PC-3 cells (UCM119 maximal inhibition=49%; IC50=11.3 μM). Additionally, we provide evidence about the expression of FAAH by human prostate gland and prostate epithelial cell lines. All these data suggest the existence of a complete anandamide inactivation system in PC-3 cells constituted by an uptake process together with the expression of the FAAH. This study reinforces the notion that the endocannabinoid system plays a physiological function in the prostate.
Arachidonyl-[5,6,8,9,11,12,14,15-3H]-ethanolamide ([3H]-anandamide, [3H]-ANA, 200 Ci mmol−1) was obtained from American Radiolabeled Chemicals, Inc. (St Louis, MO, U.S.A.). Unlabeled anandamide was from Sigma (St Louis, MO, U.S.A.). AM404, VDM11, methyl arachidonyl fluorophosphonate (MAFP) and capsazepine were purchased from Tocris (Avonmouth, Bristol, U.K.). Phloretin and (±)-verapamil hydrochloride were from Sigma (St Louis, MO, U.S.A.). Cannabinoid receptor antagonists SR141716A (N-(piperidin-1-yl)-5-(4-chlorophenyl)-1-(2,4-dichlorophenyl)-4-methyl-1H-pyrazole-3-carboxamide hydrochloride) and SR144528 (N-[(1S)-endo-1,3,3-trimethyl-bicyclo[2.2.1]heptan-2-yl]-5-(4-chloro-3-methylphenyl)-1-(4-methylbenzyl)-pyrazole-3-carboxamide) were a kind gift from Sanofi Recherche (Montpellier, France). Other agents were from Sigma (St Louis, MO, U.S.A.). FAAH antiserum was generously donated by C.J. Hillard (Medical College of Wisconsin, U.S.A.). Peroxidase-conjugated goat anti-rabbit immunoglobulin (Ig) G was from Santa Cruz Biotechnology (Santa Cruz, CA, U.S.A.).
Starting materials used were high-grade commercial products from Aldrich, Acros, Fluka, Merck or Panreac. Arachidonic (AA) (90% of purity), palmitic, oleic, linoleic and linolenic acids were purchased from Sigma. All reagents were used as supplied except methylene chloride, which was distilled over CaH2 before use. Fur-3-ylmethylamine, (5Z,8Z,11Z,14Z)-N-(fur-3-ylmethyl)icosa-5,8,11,14-tetraenamide (1, UCM707), S-(fur-2-ylmethyl) (5Z,8Z,11Z,14Z)-icosa-5,8,11,14-tetraenethioate (6), (±) (5Z,8Z,11Z,14Z)-N-[(1-methyl-2-thien-3-yl)ethyl]icosa-5,8,11,14-tetraenamide (7) and 2-(1H-indol-3-yl)ethyl (5Z,8Z,11Z,14Z)-icosa-5,8,11,14-tetraenoate (8) were synthesized as previously described (López-Rodríguez et al., 2001a, 2001b; 2003; Fowler et al., 2003).
Infrared (IR) spectra were determined on a Perkin-Elmer 781 or Shimadzu-8300 infrared spectrophotometer. 1H- and 13C-NMR spectra were recorded on a Varian VXR-300S, Bruker Avance 300-AM or Bruker 200-AC instrument at room temperature (RT). Chemical shifts (δ) are expressed in parts per million relative to internal tetramethylsilane; coupling constants (J) are in hertz. Satisfactory elemental analyses were obtained for all the newly synthesized analogs and are within ±0.4% of the theoretical values. Thin-layer chromatography (TLC) was run on Merck silica gel 60 F-254 plates. For normal pressure chromatography, Merck silica gel type 60 (size 70-230) was used.
General procedure for the synthesis of derivatives 2–5
A solution of dicyclohexylcarbodiimide (DCC, 1 equivalent) and 4-(dimethylamino)pyridine (DMAP, 0.068 equivalents) in dry methylene chloride (3 ml mmol−1 DCC) was added dropwise to a stirred solution of one equivalent (0.33 mmol) of the corresponding fatty acid in dry methylene chloride (1 ml mmol−1) and fur-3-ylmethylamine (1.5 equivalents) in dry methylene chloride (1 ml mmol−1) at −20°C in a salt-ice-bath and under argon. The mixture was stirred for 5 min at this temperature and then the cooling bath was removed and stirred at RT (3–6 h) until no further evolution was observed by TLC (chloroform : methanol, 95 : 5). The dicyclohexylurea was filtered off, the filtrate evaporated under reduced pressure and the residue obtained was taken up in methylene chloride (20 ml mmol−1 of fatty acid). This resulting organic phase was washed with a cooled 0.5 M hydrochloric acid solution and brine, and the organic extracts dried over Na2SO4 or MgSO4. Then, the solvent was evaporated under reduced pressure and the product purified by column chromatography on silica gel using the appropriate eluent.
Human prostate PC-3 cells were purchased from American Type Culture Collection (Rockville, MD, U.S.A.). They were routinely grown in RPMI 1640 medium supplemented with 10% fetal calf serum. For experiments, cells were seeded at 22,000 cells cm−2 in six-well dishes and grown for 2 days. At 24 h before the experiment, the serum-containing medium was removed and cells were transferred to a chemically defined medium consisting of RPMI 1640 supplemented with 5 μg ml−1 insulin, 5 μg ml−1 transferrin and 5 ng ml−1 sodium selenite.
The number of cells was determined in parallel using a Neubauer chamber.
Determination of cellular [3H]-anandamide uptake
Incorporation of ANA into PC-3 cells was determined by incubating cells with a mixture of [3H]-ANA (see reagents for [3H]ANA specification) and ANA in a 1/500 relationship to different final concentrations and during different intervals according to the experiments. The cellular uptake was terminated by placing the plates on ice, rapid aspiration of the medium, washing four times with 0.5 ml of ice-cold phosphate-buffered saline (PBS), pH 7.4, containing 1% fatty acid free bovine serum albumin (BSA) and subsequent protein precipitation with 10% trichloroacetic acid (TCA). The plates were kept on ice for additional 15 min and 0.5 ml of a 1 M sodium hydroxide solution was added. Then, the plates were incubated for 1 h at RT and assayed for tritium content by liquid scintillation.
For inhibition assays, cells were preincubated at 37°C for 15 min in the presence or absence of several concentrations of the tested carrier-mediated anandamide transport inhibitors and incubated with 100 nM ANA for 5 min.
Nonspecific ANA incorporation was determined by parallel incubations carried out at 4°C. These values were subtracted from the total amount of ANA incorporated. To determinate the non-specific binding to plastic plates, six well plates without cells at 37°C were also run in parallel for some of the experiments and the results obtained turned out to be essentially comparable to the values obtained at 4°C in the presence of cells.
To study whether ANA bound to CB1, CB2 or VR1 could influence total ANA accumulation, incubations were carried out as previously described preincubating cells for 15 min with 50 μM SR141716A, 50 μM SR144528 and 50 μM capsazepine.
To study the influence of FAAH or other lipid transporters in ANA uptake, another set of studies were carried out in the presence of 100 μM of the FAAH inhibitors MAFP and phenylmethanesulfonyl fluoride (PMSF) or other lipid uptake inhibitors such as phloretin or verapamil.
Kinetic studies of the ANA incorporation were performed following the same procedure at different incubation times, and using [3H]-ANA diluted with cold ANA to yield total concentrations ranging from 10 nM to 25 μM ANA.
The Michaelis constant (KM) and maximum accumulation rate (Vmax) values of the uptake kinetics were calculated after subtraction of the uptake at 4°C from that at 37°C and were determined by fitting the data to the Michaelis–Menten equation using nonlinear regression (GraphPad Prism, Graphpad Software, San Diego, CA, U.S.A.).
IC50 values were determined by nonlinear least square fitting of the data (one site competition), by the GraphPad Prism software.
Each graph point represents triplicate determinations, and results are based on at least three independent experiments.
Expression of FAAH
PC-3 cells were scrapped into 0.5 ml of ice-cold lysis buffer (50 mM Tris/HCl, pH 7.4; 5 mM ethylenediaminetetraacetic acid (EDTA); 1 mM ethylene glycol-bis (β-amino ethyl ether) tetraacetic acid (EGTA); 10 mM 2-mercaptoethanol containing 1 μg ml−1 leupeptin, 1 μg ml−1 aprotinin, 10 μg ml−1 soybean trypsin inhibitor and 1 mM PMSF). Cells were frozen, then thawed and subsequently disrupted by sonication after which they were centrifuged at 100,000 × g for 1 h. The pellet was resuspended in lysis buffer. Human prostate tissue specimens were obtained from 55- to 75-year-old patients undergoing open prostatectomy. The study protocol conformed to the Declaration of Helsinki, and was approved by the corresponding ethic committee. Prostate tissue was homogenized with a Polytron PT-20 tissue homogenizer (Kinematica, AG, Lucerne, Switzerland) at maximal speed in lysis buffer at 4°C. The homogenized tissue was filtered through a nylon mesh and centrifuged at 3500 × g for 5 min. The resulting supernatant was centrifuged at 30,000 × g for 20 min at 4°C. The pellet was washed in lysis buffer and centrifuged again at 30,000 × g for 20 min. The resulting pellet was homogenized and considered as membrane fraction. A similar protocol was followed for rat cerebellum tissue.
The amount of protein in each sample was determined using the protein assay kit (Bio-Rad, Richmond, CA, U.S.A.) and equivalent amounts (20 μg of protein) were analyzed by Western blot as previously described (Díaz-Laviada et al., 1990) using a primary polyclonal antibody against a peptide corresponding to residues 561–579 of the C-terminus of FAAH and previously characterized (Tsou et al., 1998). Blots were developed with the SuperSignal® chemiluminescent substrate from Pierce (Rockford, IL, U.S.A.).
Enzyme assay for FAAH and HPLC analysis
The assay of FAAH (EC 126.96.36.199) was performed by measuring the release of [3H]AA from [3H]ANA (see reagents for [3H]ANA specification), using HPLC in cell extracts. PC-3 cells were washed in PBS, scraped in 50 mM Tris, pH 7.4, 1 mM EDTA and sonicated on ice with a microtip at maximum power three times for 10 s each in a cell disruptor Branson sonifier (American Laboratory Trading LLC). The homogenate was then centrifuged at 1000 × g for 10 min and the resulting pellet was diluted to a protein concentration of 1 mg ml−1. For FAAH activity assay, several concentrations of anandamide in the presence or absence of 20 μM MAFP were added to a total volume of 200 μl hydrolase assay buffer (50 mM Tris-HCl, pH 9.0). The reaction was initiated by the addition of 20 μg of cellular extract and after incubation at 37°C for 20 min the reaction was stopped by the addition of 300 μl KOH (16%, v v−1). The samples were treated sequentially with chloroform/methanol (2 : 1, v v−1) and distilled water to obtain the lipidic and aqueous-soluble fractions as reported previously (Folch et al., 1957). The lipid extract was resuspended in methanol and used for anandamide and AA separation by HPLC. Separation was accomplished by reverse-phase HPLC with a Luna 5 μm-pore-size C18 column (250 nm × 4.60 mm; Phenomenex, Torrance, CA, U.S.A.). Anandamide and AA were eluted with acetonitrile/water (80 : 20, v v−1) at a flow rate of 1.2 ml min−1. The column effluent was monitored simultaneously by UV absorption spectroscopy (Beckman 168 variable-wavelength detector; Beckman Instruments, Palo Alto, CA, U.S.A.) and online radioactivity detection (using an LB-506 C-1 radioactivity detector; Berthold, Bad Widbad, Germany). Anandamide and AA were identified by comparison of the retention time of pure standards. FAAH activity was expressed as pmoles of arachidonate released per minute per milligram of protein.
Assay of affinity at cannabinoid receptors
Assay of affinity of UCM119 at cannabinoid receptors was carried out as previously described (López-Rodríguez et al., 2001b). Tissue and ligands, respectively, were for CB1 receptor, rat cerebellar membranes and 0.5 nM [3H]-WIN552122 and for CB2 receptor, HEK293EBNA CB2-transfected cell membranes and 0.3 nM [3H]-CP55940 (for details, see López-Rodríguez et al., 2001b).
Characterization of anandamide uptake by PC-3 cells
PC-3 cells were able to accumulate [3H]-ANA, in a process which was temperature- (Figure 1a, only total and nonspecific ANA accumulation for one representative concentration is shown for the sake of clarity), concentration- (data not shown) and time-dependent and was saturable at 37°C (Figure 1b). Kinetic analyses revealed that [3H]-ANA uptake can be represented by two components of different affinities (Figure 1c). The higher affinity component (KM=15.8 nM, Vmax=0.09 pmol ANA min−1 (106 cells)−1) may reflect very specific binding sites such as receptors, whereas the lower affinity component (KM=1.2 μM, Vmax=1.04 pmol ANA min−1 (106 cells)−1) is suggestive of a carrier-mediated transport as the apparent KM value is similar to those of previously described anandamide transporters in other cells (Hillard & Jarrahian, 2000).
Influence of CB1, CB2 and VR1 antagonists in the anandamide uptake
Due to the demonstrated presence of CB1 and CB2 cannabinoid receptors (Ruiz et al., 1999; Ruiz-Llorente et al., 2003) and to the potential existence of vanilloid receptor type 1 (VR1) in PC-3 cells (unpublished results), and to displace the high-affinity component, we carried out an additional series of experiments aimed at determining the contribution of receptor binding in the quantification of anandamide incorporated into cells. The VR1 antagonist capsazepine was able to displace anandamide uptake in a dose-dependent manner, with 50 μM being the most effective concentration inferred from its dose–inhibition curve (data not shown). Anandamide uptake was then measured in the presence of the vanilloid antagonist capsazepine (50 μM), cannabinoid CB1 antagonist SR141716A (50 μM) and cannabinoid CB2 antagonist SR144528 (50 μM). The three antagonists, when incubated together, were able to displace about 40% of the total specific bound [3H]-ANA (Figure 2); so in order to avoid this receptor-bound contribution, all the experiments were performed in the presence of 50 μM of the antagonists. Anandamide accumulation, measured under these conditions, was time-, concentration- and temperature-dependent (Figure 3a–c). Kinetic studies of anandamide uptake, determined under these conditions (Figure 3d), fitted to one straight line, which revealed apparent parameters of KM=4.9±0.2 μM and Vmax=3.3±0.3 pmol min−1 (106 cells)−1.
Expression and activity of FAAH in prostate PC-3 cells
Considering that diverse studies (Deutsch et al., 2001; Patricelli & Cravatt, 2001; Hillard & Jarrahian, 2003) suggest the existence of crossregulation between the FAAH and the anandamide transporter, we analyzed the expression of FAAH by prostate cells. Both epithelial prostate cells and human prostate tissue express FAAH protein, as detected by Western blot (Figure 4). FAAH polyclonal antibodies, previously validated by antigen competition experiments (Tsou et al., 1998), recognized two bands in human prostate tissue. The major band corresponds to a molecular mass of 69 kDa, which is a little higher than the predicted molecular mass for the human FAAH (Giang & Cravatt, 1997) but is similar to the FAAH band observed in mouse uterus (Maccarrone et al., 2000b) and to the band detected in rat cerebellum, used as positive control (Figure 4). The minor band of 73.2 kDa was also detected in the prostate cell lines PC-3 and LNCaP (Figure 4) and could correspond to different isoforms.
To further investigate the activity of the FAAH expressed in prostate PC-3 cells, we measured the released [3H]AA from [3H]ANA in cell extracts by HPLC. Sharp and well-resolved ANA and AA peaks were obtained (Figure 5a). FAAH activity, expressed as pmoles of AA released per minute per milligram of protein, was concentration-dependent and saturable (Figure 5b). FAAH activity detected in PC-3 cells was almost completely inhibited by 20 μM MAFP (Figure 5b), a potent and selective inhibitor of FAAH (Deutsch et al., 1997), reinforcing the notion that FAAH is expressed in prostate PC-3 cells.
Evaluation of the influence of FAAH or other lipid transporters in anandamide uptake
Since PC-3 cells express FAAH, we next tested the influence of the FAAH inhibitors MAFP and PMSF on anandamide transport, in order to assess whether anandamide hydrolysis was affecting the uptake process. In these experiments, PMSF, tested at a concentration of 100 μM, was not able to induce significant inhibition (less than 9%) in the anandamide uptake process whereas MAFP inhibited anandamide incorporation into cells by 30% (Figure 6).
Additionally, and in order to rule out the involvement of other lipid transporters previously described, we examined the effect of several nonselective transport inhibitors on the accumulation of anandamide such as verapamil or phloretin at concentrations previously used to characterize other anandamide uptake processes (Beltramo et al., 1997; Hillard et al., 1997). Neither verapamil (100 μM), which behaves as a phospholipid uptake inhibitor (Beltramo et al., 1997), nor phloretin (100 μM), which has been characterized as an inhibitor of diverse carrier-mediated uptake processes including the accumulation of fatty acids by adipocytes and glucose and anion transport into erythrocytes (Hillard et al., 1997), was able to exert any appreciable capacity to inhibit anandamide transport (Figure 6), with inhibition values not significantly different from control, of about 0 and 20%, respectively.
Synthesis of new anandamide uptake inhibitors
As the following step in the search of more potent anandamide transport inhibitors in order to get deeper insights into the structural requirements involved in the recognition of compounds for this carrier, we synthesized a series of structurally different analogs (1–8) and evaluated their capacity as transporter inhibitors. Compounds 1–8 (Figure 7) were synthesized following the methods detailed in the Methods section.
Inhibition of the anandamide uptake
The newly synthesized compounds (tested at 100 μM as the maximal concentration) showed different efficacy as uptake inhibitors, ranging from negligible capacity of compound 3 to 49% of inhibition produced by compound 4, which was comparable to the inhibition produced by the transporter selective inhibitor VDM11 (Figure 8). [3H]ANA uptake was moderately inhibited in a dose-dependent manner by the endocannabinoid transporter inhibitor AM404 (Figure 9). However, the maximal concentration used, 100 μM, which was previously shown to cause 70% inhibition in neurons and astrocytes (Beltramo et al., 1997), only reduced the transport to about a 25% of the untreated control (Figure 9). Compound 4 (UCM119) exhibits an IC50 value of 11.3 μM as shown from its dose–response curve (Figure 9) and was as potent as VDM11 (IC50 value of 10.89 μM) in PC-3 prostate cells (Figure 9).
Additionally, and to study the selectivity degree of this compound, we carried out binding assays to determine the affinity for both cannabinoid receptors. Compound 4 shows little affinity for both cannabinoid receptors (Ki (CB1)>5000 nM; Ki (CB2)>1000 nM) especially if we compare these values with those obtained for other cannabinoid ligands such as WIN552122 or CP55940 with Ki values in the low nanomolar range (López-Rodríguez et al., 2001b).
The objective of these studies was to demonstrate the existence of an inactivation system for anandamide, including a carrier-mediated process responsible for anandamide uptake as well as the expression of FAAH, and to characterize this putative transporter in epithelial PC-3 cells.
The accumulation of exogenous [3H]-ANA by these cells fulfills the main features of a carrier-mediated transport. It is temperature-, concentration- and time-dependent, showing a t1/2 value of about 5 min. The KM and Vmax values obtained from the kinetic studies revealed similar maximal accumulation rates and affinity values of this transporter compared with other studies previously reported examining anandamide uptake in both peripheral and central nervous system-derived cells (for a review, see López-Rodríguez et al., 2002).
Kinetic assays carried out in the presence of CB1, CB2 and VR1 receptor antagonists revealed KM and Vmax values of 4.9 μM and 3.3 pmol min−1 (106 cells)−1. As expected, the kinetic analysis fitted a straight line, which reflects exclusively the interaction with the transporter, once the receptor influence is eliminated. This KM value, in the low micromolar range, is very similar to those obtained for other neurotransmitter uptake systems (Beltramo et al., 1997).
In order to consider only the transporter component, we carried out the rest of the experiments coincubating with these three antagonists. However, and even under these conditions, neither AM404 nor UCM707 was able to induce any remarkable inhibitory effects. However, this is not a surprising finding since important differences among different cells have been reported regarding different sensitivities toward chemicals such as NO or alkylating agents, the proper AM404 or even the endocannabinoids (for a review, see López-Rodríguez et al., 2002). Our results are similar to those obtained in platelets in which an insensitivity to AM404 inhibition was also observed (Maccarrone et al., 2001).
Prostate PC-3 cells as well as human prostate tissue express FAAH enzyme corroborated by Western blot and activity assay. FAAH antibodies recognized two bands of 69 and 73.2 kDa, whereas mainly the upper band was detected in prostate cells. Data from our laboratory reveal that there is differential expression of this two bands among different patients (unpublished results). As FAAH expression is steroid-regulated (Maccarrone et al., 2000b; Waleh et al., 2002), the expression of the upper band in prostate cell lines could be due to the particular hormone situation of these cells. FAAH expressed in PC-3 cells was active and was inhibited by the potent FAAH inhibitor MAFP. The results demonstrate the existence of an anandamide degradation system in PC-3 cells.
There exists a recent controversy about the existence of a specific anandamide transporter suggesting that intracellular anandamide hydrolysis by FAAH may be responsible for the observed saturations of anandamide incorporation into cells (Glaser et al., 2003; Hillard & Jarrahian, 2003). In a first attempt to examine the influence of FAAH as well as other lipid transporters, we determined the ability of PMSF and MAFP to inhibit the anandamide uptake. Only MAFP produced a 30% inhibition of the incorporation process, suggesting that anandamide hydrolysis may participate in the anandamide uptake process but it is not responsible for all the accumulated anandamide. The FAAH inhibitor MAFP has been used in previous studies to examine the role of FAAH activity in the transport of anandamide (Day et al., 2001; Deutsch et al., 2001; Glaser et al., 2003). However, in these studies, a 50–90% of transport inhibition was obtained. Moreover, in cells lacking FAAH, MAFP but not PMSF reduced anandamide transport (Day et al., 2001) suggesting that another mechanism different from FAAH exists and it may be inhibited by MAFP (Day et al., 2001). On the other hand, PMSF only inhibited anandamide transport in FAAH expressing cells, suggesting that the inhibition detected with PMSF is due to the FAAH component. In our studies, PMSF did not cause transport inhibition, supporting the notion that other components besides FAAH participate in the process of anandamide accumulation. Another recent study shows that in rabbit platelets, which do not possess a carrier-mediated mechanism for anandamide but express an active FAAH, MAFP did not affect anandamide uptake although it inhibited anandamide degradation (Fasia et al., 2003).
We tested nonselective inhibitors of other lipid transporter systems like verapamil and phloretin, and neither of these agents appreciably inhibited anandamide transport, a fact that indicates the existence in PC-3 cells of an anandamide uptake system different from other lipid transporter proteins.
To search for more potent anandamide transport inhibitors in PC-3 cells, we synthesized and tested a series of structural analogs of anandamide (compounds 1–8) and compared them with previously characterized inhibitors like AM404, VDM11 and UCM707, which have allowed us to set up a structure–activity relationship (SAR) study focused on determining the main structural features involved in the interaction of compounds with the putative anandamide transporter characterized in PC-3 cells.
In these derivatives, we have carried out structural modifications in different parts of anandamide including changes in the fatty acid chain as well as the substitution of its ethanolamine moiety for a heterocycle (Figure 7). Thus, as a first approach, we have modified the heterocyclic moiety, the length and branching between the AA chain and the heterocycle, and we have also replaced the amide moiety for ester or thioester groups (compounds 1, 6–8). Among these AA derivatives, the fur-3-ylmethylamine moiety together with an amide group exhibits the major inhibitory potency. As representative examples, Figure 8 shows the inhibitory values obtained for compounds 6, 7 and 8 vs 1 (16.8, 9 and 22.4 vs 25%, respectively).
In order to evaluate the influence of the fatty acid chain in the interaction with the transporter, we replaced the AA moiety with different fatty acid chains bearing distinct chain lengths (16 and 18 carbon atoms) and diverse number of insaturations (compounds 1–5).
Regarding the number of insaturations, the presence of two double bonds enhances the affinity for the transporter, as deduced from the inhibition value obtained for compound 4, the most potent derivative of this series of compounds, which was able to inhibit nearly half of anandamide transport (49%).
Increasing or decreasing the number of double bonds implied in all cases important reductions in the inhibitory ability. For instance, compound 2, with one double bond, and compound 5, with three double bonds, exhibit comparable inhibition values (15.8 and 18.1%, respectively). These results point to the existence of an optimum folded conformation that depends on the number of insaturations.
Additionally, compound 3 allowed us to evaluate the effect of reducing the hydrophobic length chain. The fact that this compound is not able to inhibit the transporter reflects the ability of the carrier to distinguish between structural analogs.
Although previous SAR studies have identified a series of more potent anandamide uptake inhibitors in other cell types such as astrocytoma, neurons or lymphoma cells (Piomelli et al., 1999; Jarrahian et al., 2000; López-Rodríguez et al., 2003), the series of analogs reported here is the first one that has been analyzed in PC-3 prostate cells. Moreover, the partial efficacy that all of these inhibitors (1–8) and previously characterized inhibitors (AM404 and VDM11) display to inhibit the anandamide uptake in PC-3 cells even at the highest doses used reveals the lack of an optimal molecular interaction between the putative carrier and the synthesized analogs. Further structural optimization of compounds is still required in order to develop inhibitors that are able to block completely the anandamide uptake process in PC-3 cells.
In this context, these new synthesized inhibitors, especially UCM119, suppose a starting point in the development of new tools for the study of other anandamide transport systems in which AM404 or other classical transporter inhibitors are not effective.
Taken together, the above data show the existence of recognition selectivity among analogs with similar chemical structures, one of the most important features exhibited by protein-mediated processes, thus giving further support to the existence of a facilitated uptake phenomenon responsible for anandamide transport in PC-3 cells. Undoubtedly, more detailed studies will be necessary to further characterize the transporter and to establish the involvement and importance of this process both physiologically and therapeutically, especially with regard to their potential utility for the management of tumoral processes in which this cell line is implicated (Ruiz et al., 1999).
In conclusion, the results reported here demonstrate that prostate cells incorporate anandamide in a process kinetically comparable to that described in other cells. The transport process is different from other lipid transporters, since it is not affected by inhibitors of other lipid carriers. Moreover, the minor effect observed after incubation in the presence of PMSF and MAFP seems to indicate that anandamide hydrolysis is not the only driving force for anandamide transport into PC-3 cells, at least within the time frame of our experiments.
The identification of anandamide carrier-mediated transport together with the expression of FAAH in PC-3 cells represents the first characterization of an anandamide degradation system in prostate cells and reinforces the role of endocannabinoids as neuroendocrine regulators.
This investigation was supported by Ministerio de Ciencia y Tecnología (BQU 2001-1459 and SAF 2002-01572) and by Agencia Antidroga de la Comunidad de Madrid. L. Ruiz-Llorente and S. Ortega-Gutiérrez are fellows from the Ministerio de Educación, Cultura y Deporte and M.G. Sánchez is fellow from Universidad de Alcalá. We thank Unidad de Cultivos from Universidad de Alcalá for technical assistance and Sanofi Recherche for SR141716A and SR144528 gift.