Portal de Periódicos da CAPES

Formation of prostamides from anandamide in FAAH knockout mice analyzed by HPLC with tandem mass spectrometry

2004; Elsevier BV; Volume: 45; Issue: 4 Linguagem: Inglês

10.1194/jlr.m300475-jlr200

1539-7262

Allan Weber, Jinsong Ni, Kah-Hiing John Ling, Andrew Acheampong, Diane D‐S. Tang‐Liu, Robert M. Burk, Benjamin F. Cravatt, David F. Woodward,

Forensic Toxicology and Drug Analysis

We investigated the formation of PGF2α 1-ethanolamide, PGE2 1-ethanolamide, and PGD2 1-ethanolamide (prostamides F2α, E2, and D2, respectively) in liver, lung, kidney, and small intestine after a single intravenous bolus administration of 50 mg/kg of anandamide to normal and fatty acid amide hydrolase knockout (FAAH −/−) male mice. One group of three normal mice was not dosed (naïve) while another group of three normal mice received a bolus intravenous injection of 50 mg/kg of anandamide. Three FAAH −/− mice also received an intravenous injection of 50 mg/kg of anandamide. After 30 min, the lung, liver, kidney, and small intestine were harvested and processed by liquid-liquid extraction. The concentrations of prostamide F2α, prostamide E2, prostamide D2, and anandamide were determined by HPLC-tandem mass spectrometry. Prostamide F2α was detected in tissues in FAAH −/− mice after administration of anandamide. Concentrations of anandamide, prostamide E2, and prostamide D2 in liver, kidney, lung, and small intestine were much higher in the anandamide-treated FAAH −/− mice than those of the anandamide-treated control mice.This report demonstrates that prostamides, including prostamide F2α, were formed in vivo from anandamide, potentially by the cyclooxygenase-2 pathway when the competing FAAH pathway is lacking. We investigated the formation of PGF2α 1-ethanolamide, PGE2 1-ethanolamide, and PGD2 1-ethanolamide (prostamides F2α, E2, and D2, respectively) in liver, lung, kidney, and small intestine after a single intravenous bolus administration of 50 mg/kg of anandamide to normal and fatty acid amide hydrolase knockout (FAAH −/−) male mice. One group of three normal mice was not dosed (naïve) while another group of three normal mice received a bolus intravenous injection of 50 mg/kg of anandamide. Three FAAH −/− mice also received an intravenous injection of 50 mg/kg of anandamide. After 30 min, the lung, liver, kidney, and small intestine were harvested and processed by liquid-liquid extraction. The concentrations of prostamide F2α, prostamide E2, prostamide D2, and anandamide were determined by HPLC-tandem mass spectrometry. Prostamide F2α was detected in tissues in FAAH −/− mice after administration of anandamide. Concentrations of anandamide, prostamide E2, and prostamide D2 in liver, kidney, lung, and small intestine were much higher in the anandamide-treated FAAH −/− mice than those of the anandamide-treated control mice. This report demonstrates that prostamides, including prostamide F2α, were formed in vivo from anandamide, potentially by the cyclooxygenase-2 pathway when the competing FAAH pathway is lacking. Anandamide (arachidonyl ethanolamide) is a potent endogenous ligand of central and peripheral cannabinoid receptors (1Bisogno T. Maurelli S. Melck D. De Petrocellis L. Di Marzo V. Biosynthesis, uptake and degradation of anandamide and palmitoylethanolamide in leukocytes.J. Biol. Chem. 1997; 272: 3315-3323Abstract Full Text Full Text PDF PubMed Scopus (349) Google Scholar). It possesses various cannabimimetic activities in vitro and in vivo, including antinociception, hypotension, hypothermia, hypomotility, and catalepsy (2Sugiura T. Konko S. Sukagawa A. Tonegawa T. Nakane S. Yamashita A. Waku K. Enzymatic synthesis of anandamide, an endogenous cannabinoid receptor ligand, through N-acylphosphatidylethanolamine pathway in testis: involvement of Ca2+-dependent transacylase and phosphodiesterase activities.Biochem. Biophys. Res. Commun. 1996; 218: 113-117Crossref PubMed Scopus (177) Google Scholar, 3Smith P.B. Compton D.R. Welch S.P. Razdan R.K. Mechoulam R. Martin B.R. The pharmacological activity of anandamide, a putative endogenous cannabinoid in mice.J. Pharmacol. Exp. Therap. 1994; 270: 219-227PubMed Google Scholar, 4Rodriguez de Fonseca F. Del Arco I. Martin-Calderon J.L. Gorriti M.A. Navarro M. Role of the endogenous cannabinoid system in the regulation of motor activity.Neurobiol. Dis. 1998; 5: 483-501Crossref PubMed Scopus (139) Google Scholar, 5Chakrabarti A. Ekuta J.E. Onaivi E.S. Neurobehavioral effects of anandamide and cannabinoid receptor gene expression in mice.Brain Res. Bull. 1998; 45: 67-74Crossref PubMed Scopus (37) Google Scholar, 6Watanabe K. Matsunaga T. Nakamura S. Kimura T. Ho I.K. Yoshimura H. Yamamoto I. Pharmacological effects in mice of anandamide and its related fatty acid acid ethanolamides, and enhancement of cataleptogenic effect of anandamide by phenylmethylsulfonyl fluoride.Bio. Pharm. Bull. 1999; 22: 366-370Crossref PubMed Scopus (23) Google Scholar). One major enzymatic pathway responsible for the regulation of anandamide is hydrolysis of anandamide to arachidonic acid and ethanolamine via fatty acid amide hydrolase (FAAH) (7Cravatt B.F. Giang D.K. Mayfield S.P. Boger D.L. Lerner R.A. Gilula N.B. Molecular characterization of an enzyme that degrades neuromodulatory fatty-acid amides.Nature. 1996; 384: 83-87Crossref PubMed Scopus (1806) Google Scholar, 8Giang D.K. Cravat B.F. Molecular characterization of human and mouse fatty acid amide hydrolases.Proc. Natl. Acad. Sci. USA. 1997; 94: 2238-2242Crossref PubMed Scopus (395) Google Scholar, 9Willoughby K.A. Moore S.F. Martin B.R. Ellis E.F. The biodisposition and metabolism of anandamide in mice.J. Pharmacol. Exp. Ther. 1997; 282: 243-247PubMed Google Scholar). Studies have indicated that anandamide could be converted to PGE2 1-ethanolamide (prostamide E2) (10Yu M. Ives D. Ramesha C.S. Synthesis of prostaglandin E2 ethanolamide from anandamide by cyclooxygenase-2.J. Biol. Chem. 1997; 272: 21181-21186Abstract Full Text Full Text PDF PubMed Scopus (352) Google Scholar, 11Burstein S.H. Rossetti R.G. Yagen B. Zurier R.B. Oxidative metabolism of anandamide.Prostaglandins Other Lipid Mediat. 2000; 61: 29-41Crossref PubMed Scopus (132) Google Scholar) in the presence of cyclooxygenase-2 (COX-2) in vitro. COX-2 has been known to convert arachidonic acid to various prostaglandins (12Marnett L.J. Rowlinson S.W. Goodwin D.C. Kalgutkar A.S. Lanzo C.A. Arachidonic acid oxygenation by COX-1 and COX-2.J. Biol. Chem. 1999; 274: 22903-22906Abstract Full Text Full Text PDF PubMed Scopus (465) Google Scholar) that possess potent biological activity. COX-2 may be a potential key enzyme in anandamide conversion to prostamides (prostanoid ethanolamides) in vivo. These prostamides possess biological activity (13Woodward D.F. Krauss A. H-P. Chen J. Gil D.W. Kedzie K.M. Protzman C.E. Shi L. Chen R. Krauss H.A. Bogardus A. Dinh H.T.T. Wheeler L.A. Andrews S.W. Burk R.M. Gac T. Roof M.B. Garst M.E. Kaplan L.J. Sachs G. Pierce K.L. Regan J.W. Ross R.A. Chan M.F. Replacement of the carboxylic group of prostaglandin F2α with a hydroxyl or methoxy substituent provides biologically unique compounds.Br. J. Pharmacol. 2000; 130: 1933-1943Crossref PubMed Scopus (32) Google Scholar, 14Woodward D.F. Krauss A. H-P. Chen J. Lai R.K. Spada C.S. Burk R.M. Andrews S.W. Shi L. Liang Y. Kedzie K.M. Chen R. Gil D.W. Kharlamb A. Acheampong A. Ling J. Madhu C. Ni J. Rix P. Usansky J. Usansky H. Weber A. Welty D. Yang W. Tang-Liu D. D-S. Garst M.E. Brar B. Wheeler L.A. Kaplan L.J. Pharmacology of bimatoprost (Lumigan).Surv. Ophthalmol. 2001; 45: S337-S345Abstract Full Text Full Text PDF PubMed Scopus (199) Google Scholar), including contraction of feline cat iris and the reduction of intraocular pressure in primates. Furthermore, these prostamides have longer plasma elimination half-life when compared with prostaglandins (15Kozak K.R. Crews B.C. Ray J.L. Tai H.H. Morrow J.D. Marnett L.J. Metabolism of prostaglandin glycerol esters and prostaglandin ethanolamides in vitro and in vivo.J. Biol. Chem. 2001; 276: 36993-36998Abstract Full Text Full Text PDF PubMed Scopus (127) Google Scholar). The longer plasma half-life may allow them to exert biological effects remote from the site of synthesis. If the FAAH pathway in rodents was absent or disrupted by knocking out the FAAH gene, then substantial concentrations of anandamide after exogenous administration of anandamide may be available to the COX-2 pathway for the potential production of prostamides in vivo. This may more closely model higher species where anandamide and 2-arachidonylglycerol exhibit more resistance to ester or amide enzymatic hydrolysis (15Kozak K.R. Crews B.C. Ray J.L. Tai H.H. Morrow J.D. Marnett L.J. Metabolism of prostaglandin glycerol esters and prostaglandin ethanolamides in vitro and in vivo.J. Biol. Chem. 2001; 276: 36993-36998Abstract Full Text Full Text PDF PubMed Scopus (127) Google Scholar). In this study, we investigated levels of biosynthesized prostamides by a sensitive and specific method involving HPLC tandem mass spectrometry (HPLC-MS/MS). The small intestine and kidney, which constitutively expressed COX-2 (16Zimmerman K.C. Sarbia M. Schror K. Weber A.A. Constitutive cyclooxygenase expression in healthy humans and rabbit gastric mucosa.Mol. Pharmacol. 1998; 54: 536-540Crossref PubMed Scopus (116) Google Scholar, 17Nantel F. Meadows E. Denis D. Connolly B. Metters K.M. Giaid A. Immunolocalization of cyclooxygenase-2 in macula densa of human elderly.FEBS Lett. 1999; 457: 475-477Crossref PubMed Scopus (133) Google Scholar) and other key tissues (i.e., liver and lung), were examined for anandamide remaining and for prostamide formation after intravenous bolus administration of 50 mg/kg anandamide to normal mice and FAAH knockout (FAAH −/−) mice. Anandamide was purchased from Cayman Chemicals (Ann Arbor, MI) with 98% purity as determined by HPLC. An IV formulation of anandamide (20 mg/ml) in 1:1:18 (ethanol:Incrocas 30:0.9% saline; v/v/v) was used in the study. Incrocas-30 (PEG-30:castor oil) was a gift from Croda, Inc. (Parsippany, NY). Reference standards PGF2α 1-ethanolamide (prostamide F2α), PGE2 1-ethanolamide (prostamide E2), and PGD2 1-ethanolamide (prostamide D2), as well as the internal standard d8-anandamide, were purchased from Cayman Chemicals. The internal standard, d4-prostamide F2α, was synthesized by Allergan. All other chemicals used in the study were of reagent grade or better. Nine male Swiss Webster mice, 5–6 months old and weighing 20–30 g, were used in the study with six normal mice purchased from Charles River Laboratories (Portage, MI), and three FAAH −/− mice supplied from the Scripps Institute (San Diego, CA). The animal procedures that were used have been approved by the Allergan's Animal Care and Use Committee (AACUC). The FAAH −/− mice have been previously characterized, demonstrating the absence of the FAAH protein and related activity (18Cravatt B.F. Demarest K. Patricelli M.P. Bracey M.H. Giang D.K. Martin B.R. Lichtman A.H. Supersenstivity to anandamide and enhanced endogenous cannabinoid signaling in mice lacking fatty acid amide hydrolase.Proc. Natl. Acad. Sci. USA. 2001; 98: 9371-9376Crossref PubMed Scopus (1117) Google Scholar). Three groups of mice (two groups of normal mice and one group of FAAH −/− mice; n = 3/group) were cannulated with 0.3 mm silastic tubing (Dow Chemical, Midland, MI) in the jugular vein under isoflurane anesthesia the day before anandamide administration. One group of normal mice did not receive anandamide and another group of normal mice received intravenous administration of anandamide at 50 mg/kg. This dose of anandamide is similar to the dose that produced behavioral effects in rats and mice (3Smith P.B. Compton D.R. Welch S.P. Razdan R.K. Mechoulam R. Martin B.R. The pharmacological activity of anandamide, a putative endogenous cannabinoid in mice.J. Pharmacol. Exp. Therap. 1994; 270: 219-227PubMed Google Scholar, 9Willoughby K.A. Moore S.F. Martin B.R. Ellis E.F. The biodisposition and metabolism of anandamide in mice.J. Pharmacol. Exp. Ther. 1997; 282: 243-247PubMed Google Scholar). The FAAH −/− group also received intravenous administration of anandamide at 50 mg/kg. Thirty minutes after the bolus intravenous administration of anandamide, the treated control and the treated FAAH −/− mice were euthanized by CO2 inhalation. Liver, lung, kidney, and small intestine were surgically removed from all the animals and were kept on ice until processing. On the day tissues were harvested, samples were minced into 1 mm2 sections and extracted with 5 ml of acetonitrile overnight at 4°C. The mixture was centrifuged at 2,500 g for 10 min at 4°C. The supernatant was evaporated to dryness under nitrogen. The resulting dry residue was stored at −70°C until analysis. For liquid chromatography-MS/MS (LC-MS/MS) analysis, dried residues were spiked with the 10 ng of deuterated internal standards, d4-prostamide F2α, and d8-anandamide. The internal standards were added post extraction to compensate for the variability of the mass spectrometric response. The samples were evaporated to dryness at 40°C and reconstituted with 150 μl of 1:1 (v/v) mixture of 0.5% formic acid in water and acetonitrile. Standards of prostamide F2α, prostamide E2, prostamide D2, and anandamide were prepared in mobile phase at concentrations from 0.05 ng/ml to 100 ng/ml and were processed in the same fashion as the samples. Fifty microliters of extract were injected into the LC-MS/MS for analysis. The Shimadzu HPLC system (Shimadzu Scientific Instruments, Columbia, MD) consisted of SCL-10A vp system controller, LC-10AD VP liquid chromatogram, and SIL-10AD VP autoinjector. A Luna C8 3 μm (100 × 2.0 mm) column maintained at ambient temperature was used in the analysis. The mobile phase A consisted of 0.5% formic acid in water and the mobile phase B consisted of 0.5% of formic acid in acetonitrile. The flow rate was set at 0.2 ml/min with a gradient depicted in Table 1.TABLE 1HPLC gradient used for analysisTimeAaA = 0.5% formic acid in acetonitrile; B = 0.5% formic acid in water.BaA = 0.5% formic acid in acetonitrile; B = 0.5% formic acid in water.Gradientmin%%09010Initial conditions0–0.190 → 7010 → 30Step0.1–5.070 → 4030 → 60Linear5.0–5.140 → 060 → 100Step5.1–9.10100Isocratic9.1–9.20 → 90100 → 10Step9.2–13.09010IsocraticProstamide D2, PGD2 1-ethanolamide; prostamide E2, PGE2 1-ethanolamide; prostamide F2α, PGF2α 1-ethanolamide.a A = 0.5% formic acid in acetonitrile; B = 0.5% formic acid in water. Open table in a new tab Prostamide D2, PGD2 1-ethanolamide; prostamide E2, PGE2 1-ethanolamide; prostamide F2α, PGF2α 1-ethanolamide. LC-MS/MS analysis was performed using Sciex API 3000 triple quadrupole mass spectrometer (Applied Biosystems, Foster City, CA). The mass spectrometry system was operated under positive ion turbo-ionspray ionization mode. The entire LC eluent was analyzed using a turbo ionspray ionization source and a tandem mass spectrometer with 5,000 V applied to the spray needle. The turbo ionspray temperature was set at 350°C, with the declustering potential set at 60 V. Previous studies (15Kozak K.R. Crews B.C. Ray J.L. Tai H.H. Morrow J.D. Marnett L.J. Metabolism of prostaglandin glycerol esters and prostaglandin ethanolamides in vitro and in vivo.J. Biol. Chem. 2001; 276: 36993-36998Abstract Full Text Full Text PDF PubMed Scopus (127) Google Scholar, 19Giuffrida A. Rodriguez de Fonseca F. Nava F. Loubet-Lescoulie P. Piomelli D. Elevated circulating levels of anandamide after administration of the transport inhibitor, AM404.Eur. J. Pharmacol. 2000; 408: 161-168Crossref PubMed Scopus (116) Google Scholar) have used LC-MS with single-ion monitoring for analysis of anandamide. For our study, samples were analyzed by using a product ion scan for characterization of metabolites or enzymatic products and by using multiple-reaction monitoring (MRM) for quantitation of anandamide and prostamides. Quantitation was performed using two sets of MRM transitions: one MRM set involved monitoring parent ions (MH+) → water loss (MH − H2O)+ daughter ions; and the second set of MRM monitored parent ions (MH+) → ethanolamide selective m/62 daughter ions. The first and second experiments utilized MRM ion pair transitions for quantitation (Table 2).TABLE 2MRM transitions for LC-MS/MS quantitationCompoundRetention TimeMRM Ion Pair TransitionFirst Setminm/zAnandamide7.8 348.1–62.2Prostamide F2α4.0 398.2–380.0Prostamides D2+E2aProstamide D2 and prostamide E2 coeluted.4.4 396.2–378.0Second SetAnandamide7.8 348.1–62.2Prostamide F2α4.0 398.2–62.2Prostamides D2+E2aProstamide D2 and prostamide E2 coeluted.4.4 396.2–62.2LC-MS/MS, liquid chromatography tandem mass spectrometry; MRM, multiple-reaction monitoring.a Prostamide D2 and prostamide E2 coeluted. Open table in a new tab LC-MS/MS, liquid chromatography tandem mass spectrometry; MRM, multiple-reaction monitoring. The peak areas, peak area ratios, linear regression, assayed concentrations, and other quantitative analysis calculations were generated using the Sciex Analyst 1.1 quantitation software (Applied Biosystems, Foster City, CA). The Analyst software was used to construct weighted linear regression curves relating to peak area ratios of the analyte/internal standard to the concentration of analyte in the calibration standard. The assayed concentrations from extracted samples were determined from the calibration curve using the Analyst software. For each compound, the assayed concentration was multiplied by the volume of the extract then divided by the tissue weight. Thus, the results were expressed as ng/g of tissue. Chromatograms of reference standards containing anandamide, prostamide F2α, prostamide D2, and prostamide E2, with the internal standards depicted in Fig. 1. Typical MRM chromatograms of extracted lung from an FAAH −/− mouse dosed with anandamide with the internal standards d4-prostamide F2α and d8-anandamide are displayed in Fig. 2. Prostamide E2 and prostamide D2, which have identical molecular weights, were not separated chromatographically in samples; therefore, the concentrations of these two prostamides were reported as the sum, prostamides E2 and D2. The linearity of the calibration curves for anandamide, prostamide F2α, and prostamides D2+E2 was assessed by analyzing eight calibration standards of each compound over the assay range: 0.05, 0.125, 0.25, 0.5, 2.5, 5, 25, and 50 ng/ml. The calibration curve was determined by least-square linear regression analysis of peak area ratios of the analyte/internal standard. The d4-prostamide F2α was used as an internal standard for quantifying prostamide E2+D2 and prostamide F2α. Likewise, the d8-anandamide was used as an internal standard to quantify anandamide. The calibration range for anandamide was from 0.05 ng/ml to 50 ng/ml, and correlation coefficient was at 0.9973. The calibration range for prostamide F2α was from 0.125 ng/ml to 50 ng/ml, and correlation coefficient was at 0.9927. The calibration range for prostamide E2+D2 was from 0.05 ng/ml to 50 ng/ml, and correlation coefficient was at 0.9894. The accuracy of the back-calculated concentrations of standards was within 80–120%. The precision (CV%) of the assays was ±20%.Fig. 2LC-MS/MS chromatograms of anandamide, prostamide F2α, prostamide E2, and prostamide D2 with d4-prostamide F2α and d8-anandamide as internal standards in the lung of fatty acid amide hydrolase knockout (FAAH −/−) mouse dosed with anandamide.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Mice were fully recovered from the cannulation prior to anandamide administration. FAAH −/− mice receiving anandamide became unconscious for 10–15 min but gradually recovered prior to euthanasia. The sample preparation for the LC-MS/MS quantitation was effective as demonstrated by the ≥80% recovery of radioactivity using liquid-liquid extraction with acetonitrile from mouse plasma, human plasma, and buffered solutions spiked with [3H]prostamide F2α (data not shown). Concentrations of prostamides F2α, E2, and D2 and anandamide in the liver, kidney, lung, and small intestine are depicted in Table 3.TABLE 3The concentrations of prostamide F2α, prostamide E2+D2, and anandamide in the liver, kidney, lung, and small intestine of naïve control mice, treated control mice, and treated FAAH −/− mice (n = 3)a→ Naïve—Naïve normal mice. Treated controls—normal mice receiving 50 mg/kg of intravenous anandamide. Treated FAAH knockouts—FAAH knockout mice receiving 50 mg/kg of intravenous anandamide.Anandamide (ng/g)TissuesNaïveTreated ControlsTreated FAAH −/−MeanSDMeanSDMeanSDLiver0.140.0454.942.7239169Kidney0.630.2888.025.8347314Lung0.320.1985.631.2286184Small intestine3.225.475.226.7185.477.8Prostamide F2α (ng/g)TissuesNaïveTreated ControlsTreated FAAH −/−Liver BLQb→ BLQ is "below the limit of quantitation" defined as less than 50 pg/ml for each compound.—BLQ—14.12.9Kidney BLQ—BLQ—6.154.88Lung BLQ—BLQ—12.514.7Small intestine BLQ—BLQ—0.450.51Prostamide E2+D2 (ng/g)c→ Prostamide E2 and prostamide D2 were not always separated chromatographically; therefore, the sum of prostamide E2 and D2 was reported as prostamide E2+D2.Tissues NaïveTreated ControlsTreated FAAH −/−Liver BLQ—BLQ—0.120.21Kidney BLQ—0.360.083.726.44Lung BLQ—3.171.9513.412.2Small intestine BLQ—BLQ—0.450.51FAAH, fatty acid amide hydrolase; FAAH −/−, FAAH knockout.a → Naïve—Naïve normal mice. Treated controls—normal mice receiving 50 mg/kg of intravenous anandamide. Treated FAAH knockouts—FAAH knockout mice receiving 50 mg/kg of intravenous anandamide.b → BLQ is "below the limit of quantitation" defined as less than 50 pg/ml for each compound.c → Prostamide E2 and prostamide D2 were not always separated chromatographically; therefore, the sum of prostamide E2 and D2 was reported as prostamide E2+D2. Open table in a new tab FAAH, fatty acid amide hydrolase; FAAH −/−, FAAH knockout. There were strong correlations between the tissue concentrations obtained from parent ion → (MH-H2O)+ MRM (reported in Table 1) and those obtained from parent ion → m/z 62. Although the former MRM method was more sensitive than the latter one, the latter was more selective as shown by less interfering peaks (Figs. 2, 3). The limited sample size did not allow for meaningful statistical analysis. The variability for tissue concentration of the endogenous compounds in this study had approximate CV values approaching 100%; however, the observed concentrations in tissues between the different groups (n = 3) increased by several orders of magnitude. Anandamide was found in all tissues collected from all mice. This is consistent with published results that anandamide was found in rodent plasma and brain (19Giuffrida A. Rodriguez de Fonseca F. Nava F. Loubet-Lescoulie P. Piomelli D. Elevated circulating levels of anandamide after administration of the transport inhibitor, AM404.Eur. J. Pharmacol. 2000; 408: 161-168Crossref PubMed Scopus (116) Google Scholar, 20Arai Y. Fukushima T. Shirao M. Yang X. Imai K. Sensitive determination of anandamide in rat brain utilizing a coupled-column HPLC with fluorimetric detection.Biomed. Chromatogr. 2000; 14: 118-124Crossref PubMed Scopus (29) Google Scholar), porcine ocular tissues (21Matsuda S. Kanemitsu N. Nakamura A. Mimura Y. Ueda N. Kurahashi Y. Yamamoto S. Metabolism of anandamide, an endogenous cannabinoid receptor ligand in porcine ocular tissues.Exp. Eye Res. 1997; 64: 707-711Crossref PubMed Scopus (83) Google Scholar), and our finding that anandamide was found in ocular tissues (22Woodward, D. F., J. Chen, A. H-P. Krauss, W. Yang, R. M. Burk, S. W. Andrews, M. E. Garst, and L. A. Wheeler. 2001. Prostamide F2α pharmacological characterization of a novel, naturally occurring substance (Abstract at Association for Research in Vision and Ophthalmology (ARVO) meeting. Ft. Lauderdale, FL, April 29–May 4, 2001).Google Scholar), indicating that endogenous anandamide may be ubiquitous throughout the body. Concentrations of anandamide increased over 100× in liver, kidney, and lung in normal mice receiving exogenous anandamide when compared with that in naïve animals. The small intestine showed only a modest increase in anandamide concentration after exogenous administration. Additional increases of 3–4× in concentration were found in the treated FAAH −/− mice when compared with normal mice receiving exogenous anandamide in liver, lung, and kidney. In the small intestine, the increase with treated FAAH −/− mice was more than 15× over treated normal mice. Prostamide F2α was not detected in the tissues from control mice but was found in all tissues from treated FAAH −/− mice, with the highest amount found in the liver (14.1 ± 2.9 ng/g) and the lowest amount in the small intestine (0.45 ± 0.51 ng/g). Prostamides E2+D2 were found only in animals receiving the exogenous anandamide but were highest in the FAAH −/− mice. The lung had the highest concentrations of prostamides E2+D2: 3.17 ± 1.95 ng/g in the normal mice receiving exogenous anandamide and 13.4 ± 12.2 ng/g in the treated FAAH −/− mice. Multiple peaks eluted during the chromatography, which have the same transition periods as prostamide F2α and prostamide E2+D2, suggest that unknown anandamide metabolites may exist with the same molecular weight as prostamide F2α and prostamide E2+D2 but with different retention times. Figure 3 is a representative chromatogram with a m/z 398/380 transition pair for both liver samples in a normal mouse receiving anandamide and in an FAAH −/− mouse receiving anandamide. The peaks in the chromatogram from the FAAH −/− mouse were 10-fold higher than those in the chromatogram of the normal mouse. Analysis of tissues, using the MRM ion pair of MH+ > ion of m/z 62 and retention times, confirmed that these were prostamides containing the ethanolamide moiety, HO-(CH2)2-NH2, and the peaks quantitated were indeed prostamides. The product ion spectra of anandamide and its metabolite prostamide F2α detected in FAAH −/− mice were compared with those of anandamide and prostamide F2α reference standards. The product ion spectra for standards of anandamide, prostamide F2α, prostamide D2, and prostamide E2 are depicted in Fig. 4. Anandamide observed in the liver of FAAH −/− mice dosed with anandamide showed the characteristic protonated ethanolamine ion at m/z 62 in the product ion spectrum of protonated anandamide at m/z 348 (Fig. 5). The retention time of the peak and the fragmentation pattern were consistent with those of the anandamide reference standard. Similarly, prostamide F2α observed in the liver of FAAH −/− mice dosed with anandamide (Fig. 5) also shows the characteristic protonated ethanolamine ion at m/z 62, as well as the peaks corresponding to the sequential water loss from the parent drug in the product ion spectrum of m/z 398 (Figs. 4, 5). The retention times of the peak and fragmentation pattern were consistent with those of the prostamide F2α reference standard.Fig. 5Product ion spectrum of proposed anandamide (A) and prostamide F2α (B) moieties observed in the liver of an FAAH −/− mouse dosed with anandamide.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Both arachidonic acid and anandamide have been shown to be substrates of COX-2, an inducible COX isoform forming prostaglandins or prostamides, respectively. Prostaglandins are short-lived and yet can have substantial biological effects. In contrast, prostamides and glyceryl esters are not efficient substrates for 15-hydroxyprostaglandin dehydrogenase; therefore, endocannabinoid-derived COX-2 metabolites may be sufficiently stable to exert systemic activity (15Kozak K.R. Crews B.C. Ray J.L. Tai H.H. Morrow J.D. Marnett L.J. Metabolism of prostaglandin glycerol esters and prostaglandin ethanolamides in vitro and in vivo.J. Biol. Chem. 2001; 276: 36993-36998Abstract Full Text Full Text PDF PubMed Scopus (127) Google Scholar). Furthermore, prostamides are not inhibitors of FAAH hydrolysis of anandamide (23Matias, I., J. Chen, L. DePetrocellis, T. Bisongno, A. Ligresti, F. Fezza, A. H-P. Krauss, L. Shi, C. E. Protzman, C. Li, Y. Liang, A. I. Nieves, K. M. Kedzie, R. M. Burke, V. Di Marzo, and D. Woodward. Prostamides: pharmacology and metabolism in vitro. J. Pharm. Exp. Ther. In press.Google Scholar), therefore, prostamides are not substrates of FAAH. Pharmacological investigation of prostamide activity is in its infancy, but studies to date indicate potent ocular hypotensive activity and a pharmacological activity profile that is distinct from known prostaglandin receptors (14Woodward D.F. Krauss A. H-P. Chen J. Lai R.K. Spada C.S. Burk R.M. Andrews S.W. Shi L. Liang Y. Kedzie K.M. Chen R. Gil D.W. Kharlamb A. Acheampong A. Ling J. Madhu C. Ni J. Rix P. Usansky J. Usansky H. Weber A. Welty D. Yang W. Tang-Liu D. D-S. Garst M.E. Brar B. Wheeler L.A. Kaplan L.J. Pharmacology of bimatoprost (Lumigan).Surv. Ophthalmol. 2001; 45: S337-S345Abstract Full Text Full Text PDF PubMed Scopus (199) Google Scholar, 24Ross R.A. Craib S.J. Stevenson L.A. Pertwee R.G. Henderson A. Toole J. Ellington H.C.J. Pharmacological characterization of the anandamide cyclooxygenase metabolite: prostaglandin E2 ethanolamide.J. Pharmacol. Exp. Ther. 2002; 301: 900-907Crossref PubMed Scopus (94) Google Scholar). The endogenous pool of anandamide is substantially less than that of arachidonic acid. Consequently, the detection of the prostamides is more difficult. Nevertheless, by using FAAH −/− mice we were able to measure prostamides and thereby demonstrate that prostamides can be formed in vivo by the proposed biosynthetic pathway from anandamide as depicted in Fig. 6. The major pathway of anandamide metabolism is the hydrolysis by FAAH; however, other pathways involving COX-2 can eventually lead to the formation of prostamides. Prostamides were detected in tissues of FAAH −/− mice after administration of anandamide using LC-MS/MS analysis. Concentrations of anandamide, prostamide F2α, and prostamide E2+D2 in liver, kidney, lung, and small intestine were much higher in the anandamide-treated FAAH −/− mice than those in anandamide-treated normal mice. This is the first report demonstrating that prostamides, including prostamide F2α, were formed from anandamide in vivo, possibly by the COX-2 pathway.