Cancer Cell Biology
Expression and function of fatty acid amide hydrolase in prostate cancer
Article first published online: 19 JUN 2008
Copyright © 2008 Wiley-Liss, Inc.
International Journal of Cancer
Volume 123, Issue 6, pages 1318–1326, 15 September 2008
How to Cite
Endsley, M. P., Thill, R., Choudhry, I., Williams, C. L., Kajdacsy-Balla, A., Campbell, W. B. and Nithipatikom, K. (2008), Expression and function of fatty acid amide hydrolase in prostate cancer. Int. J. Cancer, 123: 1318–1326. doi: 10.1002/ijc.23674
- Issue published online: 16 JUL 2008
- Article first published online: 19 JUN 2008
- Manuscript Accepted: 2 APR 2008
- Manuscript Received: 9 NOV 2007
- Wisconsin Breast Cancer Showhouse for a Cure
- Ridin' for Research Funds
- The Cancer Center of the Medical College of Wisconsin
- The National Institute on Drug Abuse. Grant Number: DA-09155
- fatty acid amide hydrolase;
- cell invasion;
- prostate cancer;
The hydrolysis of endocannabinoids has profound effects on the function of the endocannabinoid signaling system in the regulation of prostate carcinoma cells. Prostate carcinoma cells exhibit a wide range of hydrolysis activity for 2-arachidonoylglycerol (2-AG), the major endocannabinoid. However, enzyme(s) responsible for 2-AG hydrolysis and their functions in prostate cancer have not been characterized. In this study, we demonstrated that fatty acid amide hydrolase (FAAH) was differentially expressed in normal and prostate carcinoma cells. In PC-3 cells, overexpression of FAAH resulted in increased FAAH protein, 2-AG hydrolysis, cell invasion and cell migration. Conversely, small-interfering RNA (siRNA) knockdown of FAAH in LNCaP cells decreased FAAH protein, 2-AG hydrolysis and cell invasion. Furthermore, CAY10401, a FAAH inhibitor, decreased cell invasion and it enhanced the reduction of invasion in FAAH siRNA-transfected LNCaP cells. Immunohistochemistry staining of commercial tissue microarrays (TMAs) demonstrated FAAH staining in 109 of 157 cores of prostate adenocarcinomas but weak staining in 1 of 8 cores of normal prostate tissues. These results suggest that FAAH regulates 2-AG hydrolysis and invasion of prostate carcinoma cells and is potentially involved in prostate tumorigenesis. © 2008 Wiley-Liss, Inc.
2-Arachidonoylglycerol (2-AG) is an endocannabinoid, an endogenous ligand for the cannabinoid (CB) receptors.1, 2 In prostate carcinoma cells, 2-AG is present at high concentrations3 whereas anandamide (AEA), the other well-characterized endocannabinoid, is at very low concentrations. 2-AG acts as an endogenous anti-invasive factor in prostate carcinoma cells; however, this effect is profoundly influenced by its hydrolysis.3–5 The detrimental effects of 2-AG hydrolysis are; (i) it reduces 2-AG concentrations as a ligand for the CB receptors and (ii) it generates free arachidonic acid (AA) that can be further metabolized to a variety of biologically-active eicosanoids. In fact, exogenously added 2-AG produced an increase rather than a decrease in invasion of PC-3 cells.5 This effect is mediated by rapid hydrolysis of 2-AG to AA followed by 12-lipoxygenase metabolism of AA to 12(S)-hydroxyeicostetraenoic acid (12(S)-HETE).5 Both free AA and 12(S)-HETE are effective promoters of invasion in prostate carcinoma cells.5–8 These results demonstrate the importance of 2-AG hydrolysis in regulating invasion of prostate carcinoma cells.
There are 2 well-characterized enzymes that hydrolyze 2-AG toAA and glycerol, fatty acid amide hydrolase (FAAH) and monoacylglycerol lipase (MGL).9–11 MGL only hydrolyzes monoacylglycerols like 2-AG while FAAH hydrolyzes both 2-AG and AEA.10, 12 MGL predominantly localizes to the cytosol; however, MGL activity has also been reported in cellular membranes.10, 13–15 On the other hand, FAAH is a membrane-associated protein found in microsomal and mitochondrial fractions.16–20 FAAH has been previously shown to be present in PC-3 and LNCaP cells and normal human prostate tissue.21 However, there has not been a comprehensive study to determine FAAH expression, its function in hydrolysis of endocannabinoids and its regulation of prostate carcinoma cells.
In this study, we characterized the differential expression of FAAH in normal prostate epithelial and carcinoma cells, as well as the functional role of FAAH in regulating the motility of prostate carcinoma cells. Furthermore, we demonstrated FAAH expression in human normal prostate and tumor tissues.
Material and methods
Prostate carcinoma PC-3, DU-145 and LNCaP cells and human fibroblasts were obtained from the American Type Culture Collection (Rockville, MD). Normal prostate epithelial cells (PrEC) and PrEC medium were obtained from Cambrex (Walkersville, MD). Polyclonal FAAH primary antibody, CAY10401, AEA, [2H8]AEA, 2-AG and [2H8]2-AG were obtained from Cayman Chemical Co (Ann Arbor, MI). Goat anti-rabbit and rabbit anti-mouse IgG-HRP secondary antibodies were purchased from Zymed Laboratories Inc. (South San Francisco, CA). [3H]-AEA and [3H]-2-OG were obtained from American Radiolabeled Chemical (St. Louis, MO). pCMV6-XL5 vector and pCMV6-XL5 containing FAAH cDNA were purchased from Origene Technologies (Rockford, MD). RNase-free DNA-Free was obtained from Ambion (Austin, TX). RT2 Real-Time SYBR Green/Fluorescein PCR Master Mix, FAAH and GAPDH primers were obtained from SuperArray (Frederick, MD). FAAH siRNA, a functional non-targeting control (siControl) and DharmaFECT2 transfection reagent were purchased from Dharmacon, Inc. (Lafayette, CO). Prostate cancer tissue microarrays (TMAs) were obtained from US Biomax (Rockville, MD). 3,3′-Diaminobenzidine tetrahydrochloride (DAB) and β-actin primary antibody were obtained from Sigma-Aldrich (St. Louis, MO). VECTASTATIN Elite ABC and biotinylated goat anti-rabbit secondary antibody were purchased from Vector Laboratories (Burlingame, CA).
Cells and cell culture
PC-3, DU-145 and LNCaP cells were maintained in RPMI 1640 supplemented with 10% fetal bovine serum, L-glutamine (2 mM), streptomycin (100 μg/mL) and penicillin (100 U/mL). PrEC cells were maintained in PrEC growth media. Cells were grown in 75-cm2 polystyrene tissue culture flasks at 37°C in 5% CO2 to about 65–75% confluency before use. The media for human fibroblasts were changed to serum-free RPMI and cultured overnight. Then, the conditioned media were used as the chemoattractant in the invasion assay.
Quantitative reverse-transcriptase polymerase chain reaction of FAAH and GAPDH
Total RNA was extracted from PC-3, DU-145, LNCaP and PrEC cells using Trizol followed by treatment with RNase-free DNA-Free. One microgram of total RNA was reverse-transcribed using oligo-dT primers with SuperScript III First-Strand Synthesis Kit (Invitrogen). cDNA samples were quantified using RT2 Real-Time SYBR Green/Fluorescein PCR Master Mix with human FAAH and GAPDH primers (SuperArray; proprietary primers, sequence not disclosed). The real-time PCR cycling conditions were as follows: 95°C for 10 min, followed by 45 cycles at 95°C for 30 sec, then 55°C for 60 sec and finally at 72°C for 30 sec followed by fluorescence measurement on a BioRad iCycler. The relative expression of FAAH mRNA was normalized to GAPDH for each cell line. FAAH expression was presented as arbitrary units of the ratio between expressions of FAAH in prostate carcinoma cells compared with PrEC FAAH mRNA expression.
Western blot analysis
Cells were grown, harvested and separated by centrifugation as previously described.3, 4 Isolated crude membrane fractions were resuspended in lysis buffer (50 mM Tris-HCl, 150 mM NaCl, 1.5 mM MgCl2, 1 mM EGTA, 10% (v/v) glycerol, 1% (v/v) Triton X-100, pH 7.5) supplemented with protease inhibitors (Roche, Indianapolis, Indiana). Membrane proteins were separated on SDS-PAGE BioRad Ready Gels and transferred to a nitrocellulose membrane. Blots were incubated with FAAH primary antibody (1:1,000) followed by goat anti-rabbit IgG-HRP. Protein loading and β-actin were used as loading controls. Detection was made by using ECL Western Blotting Substrate (Pierce, Rockford, IL) and captured by Fuji film X-ray (Tokyo, Japan). Band densities were analyzed using Image J software from the NIH.
Determination of AEA in prostate cancer cells by liquid chromatography-electrospray ionization-mass spectrometry
AEA in prostate carcinoma cells were analyzed by LC-ESI-MS using previously described method for 2-AG.3 [2H8]AEA and [2H8]2-AG were used as internal standards. Samples were extracted by solid phase extraction as previously described,22 redissolved in 20 μL acetonitrile and analyzed by using LC-ESI-MS (Agilent 1100 LC-MSD, SL model) as previously described.3 The detection was made in the positive ion mode and the m/z 348, 356, 379 and 387 were used for AEA, [2H8]AEA, 2-AG and [2H8]2-AG, respectively. The concentrations of AEA and 2-AG were calculated by comparing their ratios of peak areas to the standard curves and normalized to protein concentrations.
To determine the relative hydrolysis activity of FAAH, membrane proteins were incubated with FAAH substrates ([3H]-AEA or [3H]-2-OG). [3H]-2-OG was used as a 2-AG analog. Hydrolysis was determined as previously described.3, 4 Briefly, cells were washed with PBS buffer, homogenized and separated by centrifugation (25,000g). The 0.5 mL assay solution consisting of PBS buffer (pH 7.4), 10 μg membrane proteins and [3H]-2-OG or [3H]-AEA (50 nM) were incubated at 37°C for 10 min. The reaction was stopped by the addition cyclohexane:ethyl acetate (1:1 v/v), vortexed and placed on ice for 5 min with intermittent vortexing. The separated organic and aqueous layers were counted for radioactivity by liquid scintillation. The radioactivity in the organic phase represents the unmetabolized AEA or 2-OG and the radioactivity in the aqueous phase represents the hydrolyzed ethanolamide or glycerol. Hydrolysis activity was calculated as the percentage of substrate conversions compared to control; however, specific hydrolysis activity was calculated as the amount of moles hydrolyzed per mg protein per minute of incubation.
Since LNCaP cells express high levels of FAAH mRNA, protein levels and 2-AG hydrolysis, we used this cell line for FAAH specific siRNA experiments. Maximal and specific suppression of FAAH expression was optimized by evaluating different siRNA concentrations and transfection times of 4 separate siRNA sequences. In addition, a functional nontargeting siRNA that was bioinformatically designed by Dharmacon Inc. to have ≥4 mismatches with known human and mouse genes was included as a control (siControl). LNCaP cells were transfected for 24 hr at 37°C in antibiotic-free medium with DharmaFECT2 alone, siControl (40 nM), or FAAH siRNA (40 nM). Transfection medium was replaced with 10% serum feed medium for 12 hr and cells were harvested for Western blot, enzyme activity and invasion analysis.
Overexpression of FAAH
Since PC-3 cells express very low levels of FAAH protein, we used this cell line to overexpress a cDNA encoding FAAH. PC-3 cells were transfected using Lipofectamine with 5 μg of purified pCMV6-XL5 without (vector control) or with FAAH cDNA for 5 hr. After 5 hr incubation, cell medium was replaced with 10% serum feed medium and harvested at 24 hr post-transfection for Western immunoblot, enzyme activity, invasion and migration analysis.
Cell invasion assay
Cell invasion assays were performed as previously described.3 Briefly, cells were incubated with thymidine [methyl-3H] (1 μCi/mL) in 10% serum feed medium, rinsed with fresh 10% serum feed medium to remove unbound thymidine and detached. LNCaP cells were incubated with vehicle, a FAAH-specific inhibitor (CAY10401; 0.1, 1 and 10 μM) or transfected with siControl (40 nM) or FAAH-siRNA (40 nM). LNCaP cells (100,000 cells) were added to the upper compartment of Transwells containing a Matrigel-coated 8.0-μm membrane (Corning Inc., Corning, NY).
Similarly, PC-3 cells (50,000 cells) transfected with vector-control or vector-containing FAAH cDNA were added to the upper compartment. Fibroblast conditioned media (400 μL) was added in the lower compartment of the well as a chemoattractant. Cells were incubated at 37°C for 5 hr. Cells that migrated into the lower compartment were detached using 0.75% trypsin and counted for radioactivity using a liquid scintillation counter. Each treatment was performed on 6 wells. Two or three separate experiments were performed. The invasion was reported as the percentage of the invasion of the control cells.
Wound healing assay
In vitro wound healing assay was performed to assess cell migration. PC-3 cells were plated at 1 × 106 cells into 35-mm dishes. Cells were transfected with vectors-containing FAAH cDNA as stated above. After transfection, confluent cells were scraped to form a wound (with a small screwdriver). Cells were washed with serum-free media and photographs were taken at 0 and 24-hr incubation at 37°C. Each transfection was performed in 3 dishes and repeated in 2 separate experiments. Migration was determined by the difference between the initial wound widths (0 hr) from the final wound widths (24 hr). The migration was reported as the percentage of the migration of the vector transfected cells.
Tissue microarray analysis
Human prostate cancer tissue microarrays (TMA) were deparaffinized with xylene and antigen retrieval was completed by incubating slides at 95°C in citrate buffer (pH 6.0). Endogenous peroxidase activity was blocked by incubation with a 3% H2O2 solution for 30 min at 25°C. Tissue microarray slides were washed in Tris-buffered saline supplemented with 0.1% Tween-20 (TBST) followed by blocking in Protein Block serum-free (DakoCytomation) for 20 min at 25°C. After the blocking solution was removed, slides were incubated with the FAAH primary antibody (1:10 dilution) for 1 hr at 25°C. Slides were washed with TBST and incubated with biotinylated secondary antibody for 1 hr at 25°C. Slides were washed in TBST and incubated with VECTASTATIN Elite ABC for 2 hr at 25°C and color development was achieved through addition of the chromophore, 3,3′-diaminobenzidine tetrahydrochloride (DAB). Slides were dehydrated with increasing ethanol concentrations and counterstained using hematoxylin. FAAH immunostaining was assessed semiquantitatively by 2 pathologists (I.C. and A. K-B.) and assigned scores as 0 (no staining), 1 (weak staining), 2 (moderate staining) and 3 (strong/bright staining). The scores were grouped according to the Gleason scores of the tissue arrays; for normal, well-differentiated (Gleason scores 2–4), moderately-differentiated (Gleason scores 5–6) and poorly-differentiated/undifferentiated (Gleason scores 7–10) tissues.
The means of the measured values of each treatment group were compared using Student's t-test. Means were considered statistically different from one another if p < 0.05.
Differential expression of FAAH mRNA and protein in prostate carcinoma cells
To determine the relative amounts of FAAH mRNA, quantitative RT-PCR was performed on various cell lines to determine the cycle thresholds (bold line, Fig. 1a). Cycle thresholds (Ct) of the prostate carcinoma cell lines were normalized to the housekeeping gene, GADPH (data not shown). Normalized FAAH mRNA expression levels in prostate carcinoma cells were compared to the expression in normal prostate epithelial (PrEC) cells (Fig. 1b). In comparison to PrEC cells, FAAH mRNA expression was 20-fold higher in LNCaP cells, 1.52-fold higher in DU-145 cells and 40-fold lower in PC-3 cells (Fig. 1b).
FAAH protein expression in membrane fractions of PrEC, LNCaP, DU-145 and PC-3 cells was determined by Western immunoblot analysis with a polyclonal anti-FAAH antibody (Fig. 1c). PrEC cells expressed relatively low level of FAAH protein (band intensity ratio to β-actin = 0.06). In prostate carcinoma cells, LNCaP cells expressed high level of FAAH protein (ratio = 2.42) while DU-145 and PC-3 cells expressed low level of FAAH protein (ratio = 0.08 and 0.01, respectively). FAAH protein expression correlated well with FAAH mRNA in these cells. FAAH protein was not detected in cytosolic fractions of these cells indicating that FAAH is present only in the membrane fractions. These results indicated differential FAAH expression in normal prostate epithelial cells and the 3 prostate carcinoma cell lines.
Specific hydrolysis activities in prostate carcinoma cell membranes
Since AEA is known to be hydrolyzed by FAAH only, it was used to demonstrate the enzymatic activity of FAAH. The specific hydrolysis activities for AEA were 32.19, 7.73, 30.70 and 104.88 pmol/min/mg protein by PrEC, PC-3, DU-145 and LNCaP cells, respectively (Fig. 1d). These specific FAAH hydrolysis activities correlated with the FAAH protein expression levels (Fig. 1c). The specific hydrolysis activities for 2-OG were 96.52, 86.53, 159.20 and 166.02 pmol/min/mg protein by PrEC, PC-3, DU-145 and LNCaP cells, respectively (Fig. 1d).
The high specific hydrolysis activities for both AEA and 2-OG in LNCaP cells indicates that FAAH in most likely responsible for the hydrolysis, which was supported by the high levels of FAAH gene and protein expression in LNCaP cells. High specific hydrolysis activity for 2-OG suggests other 2-OG (2-AG) hydrolyzing enzyme(s) are present, particularly in DU-145 cells.
Effects of FAAH siRNA on endocannabinoid hydrolysis and invasion of LNCaP cells
Since LNCaP cells have very high levels of FAAH mRNA and protein expression (Figs. 1b and 1c), the effects of FAAH siRNA on endocannabinoid hydrolysis and cell invasion were investigated. LNCaP cells were treated with 2 separated siRNA sequences; siRNA no. 1-GUGCAGAAGUUACACAGUAUU or siRNA no. 2-GAAGGGCUGUGUCUAUGGAUU, which both effectively inhibited FAAH protein expression. However, siRNA no. 1 (40 nM) is presented due to the lower effective concentration needed compared to siRNA no. 2 (80 nM). LNCaP cells transfected with a FAAH-specific siRNA exhibited an 80% reduction in FAAH protein expression (band intensity ratio to β-actin = 0.29) as compared to nontransfected control or siControl transfected cells (ratio = 1.49 and 1.38, respectively) (Fig. 2a). There was approximately a 70% reduction in AEA hydrolysis in FAAH siRNA-transfected LNCaP cell membranes as compared to control and siControl-transfected membranes (Fig. 2b). A 30% reduction in 2-OG hydrolysis was observed in these same membrane fractions (Fig. 2b), suggesting the presence of other enzyme(s) that contribute to 2-OG hydrolysis. These results demonstrated that FAAH contributes to the hydrolysis of AEA and 2-OG (2-AG) in LNCaP cell membranes.
FAAH siRNA transfection significantly reduced invasion of LNCaP cells (74.9% ± 4.3%) as compared to control (100.0% ± 6.0%) and siControl (94.0% ± 7.7%) (Figs. 2c and 2d). Exogenously added 2-AG (10 μM) to FAAH siRNA-transfected LNCaP cells further reduced cell invasion to 53.0% ± 2.3% (Fig. 2c). This reduction in cell invasion was blocked by SR141716A, a selective CB1 receptor antagonist (500 nM) (Fig. 2c).
Interestingly, exogenous AEA (1 μM) reduced invasion of FAAH siRNA-transfected LNCaP cells to 50.8% ± 1.6% (Fig. 2d). Again, SR141716A (500 nM) blocked the reduction of cell invasion by AEA (Fig. 2d). These results demonstrated that blocking endocannabinoid hydrolysis by FAAH siRNA enhanced the inhibition of cell invasion by AEA and 2-AG and the inhibition was abolished by the selective CB1 receptor antagonist.
Effects of CAY10401, a specific FAAH inhibitor and FAAH siRNA on AEA hydrolysis and invasion of LNCaP cells
To further demonstrate the role of FAAH in cell invasion, FAAH activity in LNCaP cells was inhibited with both pharmacological inhibitor and FAAH siRNA transfection. LNCaP cells treated with CAY10401, a specific FAAH inhibitor, at 0.1, 1.0 and 10.0 μM reduced cell invasion to 68.4% ± 2.9%, 56.6% ± 1.5% and 59.6% ± 3.0%, respectively (Fig. 3a). CAY10401 (100 nM) inhibited specific AEA hydrolysis activity in LNCaP membranes (71% reduction) compared to vector transfected membranes (Fig. 3b). Furthermore, CAY10401 (100 nM) augmented the inhibition of specific AEA hydrolysis activity in FAAH siRNA membranes (90% reduction) of LNCaP cells (Fig. 3b).
FAAH siRNA transfection reduced invasion of LNCaP cells (74.9% ± 4.3%) as compared with siControl transfected cells (siControl). CAY10401 (100 nM) reduced invasion of siControl transfected cells to 68.37% ± 1.71%. CAY10401 (100 nM) further reduced invasion of FAAH siRNA transfected LNCaP cells to 50.7% ± 2.86% (Fig. 3c). These studies demonstrated that a combined inhibition of FAAH by a specific FAAH inhibitor and FAAH siRNA further reduced endocannabinoid hydrolysis and cell invasion.
Effects of overexpression of FAAH on endocannabinoid hydrolysis and invasion of PC-3 cells
PC-3 cells have extremely low levels of FAAH protein expression; therefore, they were transiently transfected with pCMV6-XL5 vector containing FAAH cDNA to overexpress FAAH. Overexpression of FAAH markedly increased FAAH protein in PC-3 cells (Fig. 4a). AEA hydrolysis in the membrane fractions of PC-3 cells overexpressing FAAH increased 5-fold compared to control cells (Fig. 4b); however, the hydrolysis of 2-OG increased by about 1.8-fold compared to control PC-3 cell membranes (Fig. 4b).
Overexpression of FAAH in PC-3 cells led to an increase of cell invasion (118.5% ± 4.0%) compared to the vector control (100.0% ± 3.2%) (Fig. 4c). The addition of exogenous 2-AG (10 μM) further enhanced cell invasion of PC-3 cells in both vector control (122.1% ± 1.8%) and FAAH overexpressing cells (131.3% ± 3.54%). A similar increase in cell invasion was observed when PC-3 cells overexpressing FAAH were treated with AEA (1 μM). However, AEA (1 μM) reduced invasion of vector transfected PC-3 cells (Fig. 4d). The inhibition of cell invasion by AEA is due to the low FAAH expression and AEA hydrolysis activity (Fig. 1) in PC-3 cells.
Endogenous AEA concentrations were 166 ± 13, 164 ± 14 and 160 ± 18 (pg/mg protein) in PC-3, DU-145 and LNCaP cells, respectively. These concentrations were relatively low as compared with 2-AG concentrations. Since LNCaP cells have high FAAH expression and AEA hydrolysis, these results may suggest that LNCaP cells have relatively higher AEA synthesis than other cells.
FAAH expression in human prostate tissues
FAAH protein expression was determined in commerically available prostate cancer TMAs with different Gleason scores. In the normal stroma and prostate glands, a weak FAAH staining (level of staining of 1) was detected in 1 of 8 cores. Within 4 of 12 cores of benign prostatic hyperplasia (BPH), cells of the basal layer showed moderate granular cytoplasmic positivity while the luminal cells had no staining (Fig. 5a). Well-differentiated prostate cancer tissues (Gleason scores 2–4), FAAH protein staining was detected in 18 of 19 cores with the averaged staining level of 1.94 ± 0.17. In moderately differentiated prostate cancer tissues (Gleason score 5–6), FAAH protein staining was detected in 6 of 8 cores with averaged staining level of 1.50 ± 0.22 (Fig. 5b). For poorly differentiated or undifferentiated prostate cancer tissues (Gleason score = 7–10), FAAH protein staining was detected in 85 of 130 cores with an average staining of 1.94 ± 0.09 (Fig. 5c). These TMA staining patterns of a limited number of tissue samples suggested that FAAH is expressed in 69% (109 of 157) of prostate cancer tissues while it has a low incidence (12% or 1 of 8 cores) in normal tissues (Fig. 5d).
Each member of the ‘endocannabinoid signaling system’ significantly contribute to the effectiveness of the endocannabinoids. Enzymes that synthesize and metabolize the endocannabinoids limit the concentrations and actions of these endogenous CB ligands. 2-AG, acting through the CB1 receptor, inhibits invasion of prostate carcinoma cells; however, the function of 2-AG is profoundly mediated by its hydrolysis.3–5 In this study, we characterized FAAH mRNA and protein expression in normal prostate epithelial cells (PrEC) and prostate carcinoma cell lines (LNCaP, DU-145 and PC-3). Furthermore, we demonstrated the role of FAAH in the regulation of endocannabinoid hydrolysis, invasion and migration of prostate carcinoma cells as well as FAAH expression in human prostate tissues.
FAAH mRNA and protein expression markedly varied between prostate carcinoma cells. LNCaP cells exhibited high FAAH expression compared to PrEC, DU-145 and PC-3 cells. The high hydrolysis of endocannabinoids in LNCaP cells may contribute to the insensitivity of cell invasion by pharmacological agents as compared with other cell lines.3 Interestingly, a previous study demonstrated that PC-3 and DU-145 cells treated with AEA underwent apoptosis but, LNCaP cells were more resistant to AEA-induced apoptosis.23 The resistance to AEA-induced apoptosis in LNCaP cells probably due to the high AEA hydrolysis by FAAH as demonstrated in this study.
A recent study demonstrated that AEA antagonizes the transient receptor potential melastatin family member 8 (TRPM8) channels.24 TRPM8 channel is strongly expressed in the prostate and prostate carcinoma cells.25 Thus, TRPM8 channel may have an important role in mediating the function of endocannabinoids, particularly AEA in prostate cancer and it needs further investigation.
Androgen-dependent LNCaP cells, exhibited very high levels of FAAH mRNA and protein expression. However, the androgen-independent prostate carcinoma DU-145 and PC-3 cells, exhibited low levels of FAAH expression. The molecular mechanism(s) responsible for increased FAAH expression in LNCaP cells is currently unknown. FAAH expression is higher in an adult prostate gland than a pubertal prostate gland. An androgen signaling crosstalk has been suggested to regulate FAAH expression.26 Collectively, high FAAH expression in LNCaP cells may be associated in part by androgen-dependent cell phenotype, which needs further investigation.
Interestingly, DU-145 and PC-3 cells express low levels of FAAH and hydrolysis activity of AEA (FAAH substrate) but have high hydrolysis activity of 2-OG (substrate for FAAH, MGL and possibly other enzymes). The results from this study suggest the presence of another enzyme(s) in prostate carcinoma cells that may be responsible for 2-OG (2-AG) hydrolysis. In agreement with this observation, several recent studies suggested an unidentified enzyme(s) responsible for 2-AG hydrolysis in rat cerebellar membranes and microglial cells.27–29 The identity of 2-AG hydrolyzing enzyme(s) in prostate carcinoma cells is currently under investigation.
Inhibition of FAAH activity in LNCaP cells by either a specific FAAH inhibitor or FAAH siRNA transfection reduced endocannabinoid hydrolysis and cell invasion. The inhibition was further enhanced with the addition of 2-AG or AEA and it was blocked by the selective CB1 receptor antagonist, SR-141716A. These results are consistent with our previous finding that the activation of the CB1 receptor inhibits prostate carcinoma cell invasion.3 The results indicate that CAY10401 and FAAH siRNA inhibit cell invasion through the inhibition of FAAH and the FAAH expression regulates endocannabinoid hydrolysis and invasion in LNCaP cells.
Overexpression of FAAH protein was sufficient to increase cell invasion and migration of PC-3 cells. Exogenous AEA and 2-AG further enhanced PC-3 cell invasion compared to vehicle treated cells, consistent with our previous finding that hydrolysis of 2-AG enhances PC-3 cell invasion.5 Interestingly, exogenous AEA reduced invasion of vector transfected PC-3 cells due to the very low FAAH expression, AEA is not hydrolyzed and activates CB1 receptor to inhibit cell invasion. This finding is in agreement with previous studies that the stable analog of AEA, R(+)-methanandamide inhibits invasion of prostate, breast and cervical cancer cells through a CB1-dependent mechanism.3, 30, 31
Prostate tissue microarray analyses indicated an increase of FAAH protein expression in adenocarcinomas compared to normal prostate tissue samples. Definite conclusion of whether FAAH expression is correlated with tumor Gleason score can be made with larger number of samples for each Gleason score. Elevated FAAH expression in prostate cancer tissues may result in an increase of free arachidonic acid and eicosanoids that are known to increase prostate carcinoma cell proliferation, progression and metastasis.6, 8, 32–34 In a mouse colon carcinogenic model, inhibition of FAAH increases endocannabinoid concentrations and reduced aberrant crypt formations (earliest identifiable neoplastic lesions) through increased apoptosis,35 indicating that FAAH expression and activity contribute to tumorigenesis.
This is the first study to demonstrate FAAH in the regulation of prostate carcinoma cell motility and potential involvement in tumorigenesis. Inhibition of FAAH activity by specific inhibitors may be one therapeutic target for the treatment of prostate cancer.
The authors would like to thank Ms. Ana-Doris Gomez for her excellent technical assistance, Ms. Lynn Gruman for her suggestions on the immunohistochemistry technique and Mr. Jon Hansen for the rat whole brain lysates. M.P.E. was supported by Predoctoral Fellowship Award from Emory T. Clark Foundation.