1

ANANDAMIDE BIOSYNTHESIS CATALYZED BY THE PHOSPHODIESTERASE GDE1 ANDDETECTION OF GLYCEROPHOSPHO-N-ACYL ETHANOLAMINE PRECURSORS IN

MOUSE BRAINGabriel M. Simon and Benjamin F. Cravatt

From the Skaggs Institute for Chemical Biology and The Department of Chemical Physiology, TheScripps Research Institute, 10550 N. Torrey Pines Rd., La Jolla, CA 92037

Running title: Anandamide biosynthesis catalyzed by GDE1 from GP-NArE precursorAddress correspondence to: Benjamin F. Cravatt, The Scripps Research Institute, 10550 N. Torrey PinesRd. La Jolla, CA 92037. Tel. 858-784-8633; Fax. 858-784-8023; Email: cravatt@scripps.edu

http://www.jbc.org/cgi/doi/10.1074/jbc.M707807200The latest version is at JBC Papers in Press. Published on January 27, 2008 as Manuscript M707807200

Copyright 2008 by The American Society for Biochemistry and Molecular Biology, Inc.

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

2

Anandamide (AEA) is an endogenousligand of cannabinoid receptors and a well-characterized mediator of many physiologicalprocesses including inflammation, pain, andappetite. The biosynthetic pathway(s) foranandamide and its N -acyl ethanolamine(NAE) congeners remain enigmatic.Previously, we proposed an enzymatic routefor producing NAEs that involves the double-O - d e a c y l a t i o n o f N -acylphosphatidylethanolamines (NAPEs) byalpha-beta hydrolase 4 (ABDH4 or Abh4) toform glycerophospho (GP)-NAEs, followedby conversion of these intermediates to NAEsby an unidentified phosphodiesterase. Here,we report the detection and measurement ofGP-NAEs, including the anandamidep r e c u r s o r g l y c e r o p h o p s p h o - N -arachidonoylethanolamine (GP-NArE), asendogenous constituents of mouse braintissue. Inhibition of the phosphodiesterase-mediated degradation of GP-NAEs ex vivoresulted in a striking accumulation of theselipids in brain extracts, suggesting a rapidendogenous flux through this pathway.F u r t h e r m o r e , w e i d e n t i f yglycerophosphodiesterase 1 (GDE1), alsoknown as MIR16, as a broadly expressedmembrane enzyme with robust GP-NAEphosphodiesterase activity. Together, thesedata provide evidence for a multi-steppathway for the production of anandamide inthe nervous system by the sequential actionsof Abh4 and GDE1.

N-acylated ethanolamines (NAEs) werefirst isolated from preparations of egg yolk,peanut oils, and soybeans in the 1950s due totheir anti-inflammatory and anti-allergenicproperties (1). In the ensuing decades, NAEshave been found in nearly all mammalian tissuesand ascribed a number of bioactivities (2).Studies by Schmid and colleagues first identifiedN-acylated phosphatidylethanolamines (NAPEs)as markers of ischemic shock in brain and hearttissues and postulated that these lipids were themetabolic precursors of NAEs (3). In 1992,anandamide (N-arachidonoyl ethanolamine,NArE, AEA), was identified as an endogenousligand for the brain (type I) cannabinoid receptor(4). The discovery of AEA as an

endocannabinoid sparked renewed interest in NAEbiochemistry and a search for pathways responsiblefor NAE biosynthesis and degradation. The integralmembrane protein fatty acid amide hydrolase(FAAH) has been identified as the primary enzymeresponsible for NAE degradation in most tissues,including brain (5). Pharmacological (6) or genetic(7) blockade of FAAH leads to elevated NAE levelsand anti-inflammatory and analgesic effectsmediated through cannabinoid receptors. Despitethese advances, a detailed understanding of thebiosynthetic pathway(s) for NAEs has proven moreelusive.

Substantial evidence implicates NAPEs asthe key metabolic precursor for NAEs. NAPEs areformed by the calcium-dependent transfer of an acylchain from the sn-1 position of phosphatidylcholineto the amine of phosphatidylethanolamine by an as-of-yet unidentified enzyme (8). N-ArachidonoylNAPE, which serves as the AEA precursor, is foundat very low levels in brain tissue, consistent with thelow abundance of arachidonate in the sn-1 positionof phosphatidylcholine (9). Recently, a calcium-independent transacylase capable of generatingNAPE from PE and PC was cloned, however thisprotein displays markedly different properties fromthe brain enzyme that generates NAPEs (10).

Release of NAEs from NAPEs wasoriginally postulated to occur via a calcium-stimulated D-type phospholipase (PLD). In 2004,Ueda and colleagues isolated and identified anNAPE-selective PLD (NAPE-PLD) (11). Mice witha disrupted NAPE-PLD gene [NAPE-PLD(-/-) mice]were found to exhibit dramatically reduced brainlevels of long chain, saturated NAEs. In contrast,polyunsaturated NAEs, including AEA, wereunaltered in these animals (12). These resultsindicate that enzymes distinct from NAPE-PLD areinvolved in the biosynthesis of AEA in vivo.Consistent with this premise, RNA interference-mediated knockdown of NAPE-PLD in RAW264.7macrophages failed to alter AEA levels in these cells(13).

That NAPE-PLD does not appear tocontribute to AEA biosynthesis, at least in certaincell and tissue types, necessitates the existence ofalternative biosynthetic enzymes. Consistent withthis premise, substantial rates of conversion ofNAPE to NAE were observed in tissue extracts fromNAPE-PLD(-/-) mice (12). Interestingly, thisresidual activity was not due to another D-type

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

3

phospholipase, but rather to a multi-enzymecascade that involves the initial A/B-typephospholipase-catalyzed double O-deacylationof NAPEs to generate glycerophospho-(GP)-NAEs, followed by phosphodiesterase-catalyzedconversion of these intermediates to NAEs(Figure 1). This pathway was further found tobe inhibited by millimolar concentrations ofcalcium which may explain why it hadpreviously gone undetected in assays for NAPE-PLD activity that included high (2-10 mM)levels of exogenous calcium. Conventionalprotein chromatography coupled with activity-based proteomic analysis identified the hithertounannotated enzyme alpha/beta-hydrolase 4(Abh4 or ABHD4) as a B-type NAPE lipasecapable of removing both O-acyl chains fromNAPE to yield GP-NAE (14). This findings u g g e s t e d t h a t g l y c e r o p h o s p h o -N-arachidonoylethanolamine (GP-NArE) may be aphysiologically relevant precursor of AEA andpointed to the GP-NAE phosphodiesterase as acritical final step leading to AEA release.However, to our knowledge, GP-NAEs have notpreviously been identified as natural brainconstituents, nor has a phosphodiesterasecapable of converting these lipids to NAEs beencharacterized. Herein, we describe the detectionand measurement of GP-NAEs, including theAEA precursor GP-NArE, as endogenous brainconstituents. We observe the enzymaticformation of endogenous GP-NAEs in ex vivobrain preparations and, moreover, we identifyt h e i n t e g r a l m e m b r a n e e n z y m eglycerophosphodiesterase 1 (GDE1) as a GP-NAE phosphodiesterase enriched in brain that iscapable of generating NAEs, including AEA, invitro.

EXPERIMENTAL PROCEDURES

Materials- 1-Oleoyl-2-hydroxy-sn-glycero-3-phosphoethanolamine was purchased fromAvanti Polar Lipids (Alabaster, AL).Pentadecenoyl chloride, palmitoyl chloride,oleoyl chloride, arachidonoyl chloride,eicosanoyl chloride, and docosahexaenoylchloride were purchased from NuCheck Prep(Elysian, MN). [1’-14C]Palmitic acid waspurchased from Moravek Biochemicals (Brea,CA). Methoxy arachidonoylfluorophosphonate

(MAFP) was purchased from Cayman Chemicals(Ann Arbor, MI).

Synthesis of GP-NAE standards andsubstrates was performed essentially as described(14). Briefly, acid chlorides were reacted with anexcess of 1-oleoyl-2-hydroxy-sn-glycero-3-phosphoethanolamine and allowed to react for 1 h inCH2CL2 with a catalytic amount of triethylamine.N-acylated lipids were then purified by silica-gelflash column chromatography. The resultinglysoNAPE was stirred vigorously in 2:1:1CHCl3:MeOH:2N LiOH for 4 hours beforeacidification and purification via silica-gel flashcolumn chromatography.Brain-lipid preparation and tandem MS analysis ofGP-NAE- 8-week old male C57BL/6 mice wereanesthetized with 50% CO2 and killed bydecapitation. Brains were removed within 30seconds of decapitation and immediately frozen inliquid N2. Frozen brains were placed in a dounce-homogenizer with 8 mL 2:1:1 CHCl3:MeOH:50 mMTris pH 8.0 and 200 pmols of 1,2-dihydroxy-sn-glycero-3-phospho(N-pentadecenoyl)ethanolaminewas added as an internal standard. Mild basificationof the aqueous layer was required to prevent theartifactual accumulation of GP-NAEs due to thenon-enzymatic (acid-catalyzed) breakdown of sn-1alkenyl lyso-NAPEs. Brains were immediatelyhomogenized with 8-10 strokes of the pestle andcentrifuged for 10 minutes at 1,400 x g to separatephases. The organic phase was extracted and anadditional 4 mL of CHCl3 added to the remaininghomogenate, which was vortexed, re-centrifuged,and extracted again. The pooled organic phaseswere evaporated to dryness under N2 yieldingapproximately 20-25 mg lipid per brain. For the exv i v o analysis of GP-NAEs, brains werehomogenized in a dounce homogenizer in 2mL of 50mM Tris pH 8.0 or 50 mM Tris pH 8.0 with 10 mMEDTA. 1.4 µL of 28 mM MAFP in DMSO (for 20µM final concentration) or DMSO alone was addedto homogenates immediately after homogenization.Samples were either extracted immediately orincubated at room temperature for 4 hours withrotation prior to extraction and GP-NAE quantitationas described above.

Lipid extracts were resuspended in 120 µL2:1 CHCl3:MeOH and 30 µL of the solution wasinjected into an Orbitrap mass spectrometer (ThermoFisher Scientific, Waltham, MA) with an Agilent1100 autosampler. Samples were chromatographed

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

4

on a Gemini 5µm C18 column (50 x 4.6mm,Phenomenex, Torrance, CA) with a 50 minutegradient from 0 to 100% buffer B followed by 7minutes of isocratic flow with 100% buffer B.Buffer A consisted 5:95 MeOH:H2O with 0.1%of 28% NH4OH and buffer B consisted of5:35:60 H2O:MeOH:iPrOH with 0.1% of 28%NH4OH. Analysis was performed with an ESIsource in the negative ionization mode with asource-voltage of -3.5 kV, flow-rate of 0.4mL/min and capillary temperature of 325 °C.Care was taken to avoid carryover by running‘blank’ samples prior to all endogenousmeasurements. The method included a high-resolution full-scan in the Orbitrap (operated at30,000 resolution) followed by sixfragmentation events in which fragment-ionswere analyzed in the ion-trap and/or sent to theOrbitrap for high-resolution mass analysisdepending on the experiment. Fragmentationevents were chosen to detect palmitoyl-, oleoyl-,arachidonoyl-, eicosanoyl-, docosahexaenoyl-,and pentadecenoyl-GP-NAEs by targeting m/zvalues of 452, 478, 500, 508, 524, and 436, forMS/MS analysis, respectively. For quantitationextracted ion chromatograms of the phospho-NAE peaks ([M-H-74]-)were integrated andcompared to the internal N -C15:1 GP-NAEstandard. Extractions using synthetic standardsrevealed extraction-efficiencies of 3%, 25%,45%, 8%, 18%, and 77% for N-C15:1, C20:4,C22:6, C16:0, C18:1, and C20:0 GP-NAEs,respectively, and measurements of endogenousGP-NAEs were corrected accordingly.Subcloning and recombinant expression ofGDE-domain containing enzymes- Thefollowing ESTs were purchased fromOpenBioSystems (Huntsville, AL) andsubcloned into pcDNA3.1-myc/His usingconventional techniques: GDE1/MIR16 –IMAGE:3601420, GDE2 – IMAGE: 4224574,GDE3 – IMAGE: 4457335, GDE4 –IMAGE:3602449, GDE7: 3486726. COS-7cells were grown at 37°C with 5% CO2 to ~70%confluence in Dulbecco’s modified Eagle’smedium (DMEM) containing 10% fetal calfserum in 10-cm dishes and transfected with 6 µgof plasmid DNA using the FuGENE 6 (RocheApplied Science) transfection reagent accordingto the manufacturer’s protocol. After 48 h, thecells were washed twice with phosphate-

buffered saline, scraped, resuspended in 50 mM TrispH 8.0, and sonicated to lyse. The lysates were thenspun at 145,000 x g for 45 min to isolate theinsoluble (membrane) fraction, which was sonicatedto resuspend in 50mM Tris pH 8.0. Cell-membranesthus prepared were stored as single-use aliquots at0.01 mg/mL for enzyme assays or 2 mg/mL forwestern blots at -80°C. For western blots the cell-membranes were treated with (or without) PNGase F(New England Biolabs, Ipswich, MA) according tothe manufacturer’s protocol prior to SDS-PAGEanalysis.GP-NAE phosphodiesterase activity assay- Enzymeassays were performed in 4 mL glass vials in 50 mMTris pH 8.0 with 2 mM MgCl2 except for thefollowing experiments. pH-rate profiles wereperformed in a buffers containing 50 mM Tris, 50mM HEPES, and 50 mM glycine that was adjustedwith HCl to the desired pH. The cation-sensitivityassays were performed in 50 mM Tris pH 8.0 andthe indicated cation (or EDTA) at 2 mM finalconcentration. For all assays, 10 µL protein (100 µgfor cell-membranes or 500 µg for brain samples)was added to 85 µL of the specified buffer and 5 µLGP-NAE substrate (2 mM in DMSO or EtOH, 100µM final concentration). Vials were incubated at 37°C for 30 min and then quenched with 1.5 mL 2:1CHCl3:MeOH and 400µL 1% formic acid. For LC-MS assays, d4-N-oleoylethanolamine (10 µL of 100µM stock in EtOH; 1 nmol total) was added as aninternal standard. Phases were separated bycentrifugation at 1,400 x g for 3 min and the organicphase transferred to a new vial and evaporated todryness under a stream of N2. For LC-MS assaysthe dried reactions were resuspended in 500 µLMeOH and 50 µL injected into an Agilent 1100-series LC-MS for analysis using an 8-minutegradient from 30% to 100% buffer B. Buffer Aconsisted 5:95 MeOH:H2O with 0.1% formic acidand buffer B consis ted of 5:35:60H2O:MeOH:iPrOH with 0.1% formic acid. NAEformation was quantified by comparison with theinternal d4-N-oleoylethanolamine standard. Forradiochromatographic assays, the dried reactionswere dissolved in 20 µL of 2:1 CHCl3:MeOH,spotted on thin layer silica plates, and developed in40:10:1 CHCl3:MeOH:28% NH4OH. Distribution ofradioactivity on the plate was quantified by aphosphorimaging device (Packard), and productswere identified by comparison with 14C-radiolabledsynthetic standards. NAE formation was calculated

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

5

from percentage values of the total radioactivityon the TLC plates. All of the assays wereconducted with n=2 or 3, and error barsrepresent the standard error of the mean.Generation of polyclonal antibodies againstmGDE1 and western blotting- Polyclonalantibodies that recognize GDE1 were preparedby expressing an N-terminal GST-fusion of theC-terminal 163 amino acids of GDE1 preparedusing traditional subcloning techniques. Thisfusion protein was expressed in BL21-DE3 cellsby growing at 37 °C to OD600 = 0.4 and inducingexpression with 1 mM isopropyl β -D-1-thiogalactopyranoside (IPTG). After 4 h, cellswere harvested and sonicated. Lysate wascentrifuged for 30 min at 12,000 x g and thepellet was resuspended in phosphate bufferedsaline and visualized by SDS-PAGE and proteinstaining. The fusion protein was excised fromthe gel, homogenized, mixed 1:1 with RIBIadjuvant, and injected three times into rabbits attwo-week intervals. One week after the finalinjection, blood was taken and antibodiesaffinity purified with the original antigenimmobilized on PVDF membrane. Thesepolyclonal antibodies were used at a 1:200dilution in TBST (20 mM Tris-HCl, 150 mMNaCl, 0.2% Tween-20) overnight at 4 °Cfollowed by incubation with secondary HRP-conjugated goat anti-rabbit (Bio-RadLaboratories, Hercules, CA) at 1:5000 in TBSTprior to developing with chemiluminescent ECLsubstrate (Pierce, Rockford, IL) and exposing toX-ray film. Specificity of antibodies was testedvia immunoblots of mock- and GDE1-transfected COS-7 cel l membranes(Supplemental Figure 1).Reverse Transcription-PCR- cDNA wasprepared using the Superscript III reversetranscriptase kit (Invitrogen) according to themanufacturer’s protocol. PCR was performedusing 30 cycles with cycling times of 30 s ofmelting at 95° C, 30 s of annealing at 58° C, and30 s of extension at 72° C with Taq DNApolymerase using the following primers: GDE1o l i g o n u c l e o t i d e s , 5 ’ -CCGACTATGTGATCTGACATTTGAACAAG T T A G G - 3 ’ a n d 5 ’ -CCATACTGTTATTGTACAGCTGTGGAAATTCCG-3’; and glyceraldehyde-3-phosphatedehydrogenase oligonucleotides, 5’-

TGTCTTCACCACCATGGAGAAGGC-3’ and 5’-TGGCAGTGATGGCATGGAACTGTGG-3’. Theproducts were visualized on a 2% agarose gel.

RESULTS

Measurement of endogenous GP-NAE-Initial attempts to detect endogenous GP-NAE inmouse brain were frustrated by the presence ofabundant and isobaric lyso-phosphatidylethanolamine lipids that displayed overlappingchromatographic behavior. To accurately detect andmeasure GP-NAEs, we employed LC-MS/MS toidentify characteristic fragment ion peaks (Figure 2)(15). First, synthetic GP-NAEs were prepared andtheir chromatographic behavior and fragmentationstudied. Fragmentation of GP-NAEs in the negativemode results in three dominant ions: a phospho-NAE peak at [M-H-74]- and two peakscorresponding to glycerol phosphate and itsdehydrate at 171 and 153, respectively (Figure 2Aand D). The phospho-NAE peak was selected foridentification and quantification by multiple reactionmonitoring, as this peak also verified the N-acylchain of the parent GP-NAE species. Thus, byscanning for the [M-H]- > [M-H-74]- transition,several endogenous GP-NAEs, including C16:0,C18:1, and C20:4 GP-NAEs, could be detected thatdisplayed identical retention and fragmentation tosynthetic standards (Table 1 and Figure 2). Parentand fragment mass signals were strongest for themore abundant GP-NAEs (C16:0, C18:1) (Figure2), but could be detected for all GP-NAEs listed inTable 1. Furthermore, we employed a hybrid ion-trap/Orbitrap instrument, which enabled high-resolution mass measurements of parent andfragment-ions. This instrument yieldedmeasurements of endogenous GP-NAEs and keyfragments that were within ~2 ppm of predictedmasses thereby confirming the identity of these NAEprecursors (Supplemental Table 1). Quantificationof endogenous GP-NAEs was performed bycomparison to an internal unnatural (N-C15:1) GP-NAE standard. Endogenous GP-NAEs werepresent at levels somewhat lower than endogenousNAEs (Table 1).

Due to the low levels of endogenous GP-NAEs, we sought additional evidence that theselipids represented potential intermediates in theenzymatic biosynthesis of NAE. To measureenzymatic turnover of GP-NAEs we employed an ex

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

6

vivo assay. Previous studies had determined thatbrain GP-NAE phosphodiesterase activity wasinhibited by EDTA (14). Here, we found thattreatment of brain extracts with EDTA resultedin dramatic accumulation of endogenous GP-NAEs in brain homogenates over time (Figure3A). Methoxy arachidonoylfluorophosphonate(MAFP) is a general inhibitor of serine-hydrolases including the NAPE-lipase Abh4(14). The accumulation of GP-NAEs wascompletely blocked by addition of MAFPthereby confirming that these lipids originatedfrom the enzymatic sn-1/sn-2-deacylation ofNAPE precursors. Notably, long chain saturatedGP-NAEs (e.g., C20:0) did not accumulate,suggesting that this pathway may only be activeon GP-NAEs bearing select N-acyl chains (seeDiscussion). The rapid enzymatic turnover ofGP-NAEs in brain extracts, combined with thelow endogenous levels of these precursors,suggests that they represent transientintermediates in the biosynthesis of NAEs.

The significant accumulation of theAEA precursor C20:4 GP-NAE in EDTA-treated brain extracts permitted acquisition ofmuch cleaner MS profiles for this lipid, whichmatched precisely those observed with asynthetic standard (Figure 3C and D).

Identification of GDE1 as the GP-NAEphosphodiesterase- Having confirmed that GP-NAEs are endogenous constituents of mousebrain, we next attempted to identify theenzyme(s) that catalyze hydrolysis of thephosphodiester bond of these lipids to releaseNAEs. One candidate enzyme class is theglycerophosphodiester-phosphodiesterases(GDEs), which hydrolyze the phosphodiesterbond of glycerophosphodiesters to yield glycerolphosphate and an alcohol (16). Enzymescontaining this GDE domain have been shown toplay roles in glycerophospholipid metabolism inyeast and bacteria (17). Searches of mammaliangenomes revealed 7 genes encoding proteinscontaining GDE domains. The GP-NAEphosphodiesterase activity in brain tissue wasfound exclusively in the membrane fraction(Figure 5B) and treatment with NaCl or Na2CO3

did not solubilize the activity (not shown),indicating that it derives from an integralmembrane protein. Of the seven genes encodingmammalian GDE-domain containing proteins,

six are predicted to have transmembrane domainsand of these, only five are expressed in brain, asdetermined by EST counts from the UniGenedatabase and reports from the literature (18). Thesefive predicted GDEs were heterologously expressedin COS-7 cells as epitope-tagged proteins. Followingconfirmation of expression by western blotting,membrane fractions from these cells were tested forGP-NAE phosphodiesterase activity using an LC-MS-based substrate assay with synthetic N-C16:0GP-NAE. One enzyme, GDE1, displayed robustGP-NAE phosphodiesterase activity, while the otherfour GDEs were indistinguishable from a mock-transfected control (Figure 4A and B).

Brain GP-NAE phosphodiesterase andrecombinant GDE1 exhibit similar biochemicalproperties- We next compared the activity of GDE1to the GP-NAE phosphodiesterase activity detectedin mouse brain. First, we confirmed by westernblotting that GDE1 (Figure 5A), like the GP-NAEphosphodiesterase activity, resides in brainmembrane fractions (Figure 5B). Treatment with theglycosidase PNGaseF reduced the apparentmolecular mass of GDE1 by ~5 kDa, indicating that,as previously reported, the enzyme is an N-linkedglycoprotein (19). These data are consistent withhydropathy plot analyses that predict the enzyme isan integral membrane protein with the majority ofthe protein sequence, including the catalytic domain,facing the luminal/extracellular compartments of thecell. We next compared the sensitivity of GDE1 andbrain GP-NAE phosphodiesterase to various divalentcations and found that both activities were inhibitedby EDTA, calcium chloride, and zinc chloride, andmarkedly enhanced by magnesium chloride (Figure6B ). Recombinant GDE1 and brain GP-NAEphosphodiesterase activity also displayed equivalentpH-rate profiles, showing a slightly alkaline pHoptimum (Figure 6A). Finally, GDE1 and GP-NAEphosphodiesterase activity were found to exhibitnearly identical tissue distributions, with highestlevels being observed in brain and spinal cord,followed by kidney, liver, and testis (Figure 7). Incontrast, little or no GP-NAE phosphodiesteraseactivity or GDE1 expression was detected in heart orspleen. Interestingly, this tissue distributionresembles that of other enzymes that participate inNAE metabolism, including FAAH (5) and Abh4(14). These data collectively suggest that GDE1 is aprincipal GP-NAE phosphodiesterase in mammaliantissues.

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

7

Finally, we measured the ability ofGDE1 to hydrolyze GP-NAEs with various N-acyl chains. GDE1 was able to generate NAEsfrom synthetic GP-NAEs bearing multiple typesof N-acyl chains, showing the highest activitywith C16:0, C18:1, and C20:4 substrates, andlower activity with C20:0 and C22:6 substrates(Figure 8).

DISCUSSION

Despite over 20 years of study, thebiosynthesis of AEA and other NAEs remainspoorly understood (20). It is generally acceptedthat N-acyl phosphatidylethanolamines (NAPEs)are the precursors for NAEs, but the preciseenzymatic steps leading to release of NAEs fromNAPEs are unclear. Four different enzymaticroutes have been proposed: 1) Direct hydrolyticcleavage of the phosphodiester bond of NAPE torelease NAE, a reaction that is catalyzed by theNAPE-PLD enzyme (3). Although NAPE-PLD(-/-) mice possess reduced brain levels of long-chain saturated NAEs, these animals do notdisplay significant alterations in polyunsaturatedNAE in the nervous system (including AEA).These findings invoke the existence ofadditional enzymatic routes for AEA formationin the brain. 2) Phospholipase-C-type cleavageof the phosphodiester bond releasing phospho-NAE, which is rapidly converted to NAE by alipid phosphatase (11). An enzyme displayingNAPE-phospholipase C-type activity remains tobe identified, but a candidate phospho-NAEphosphatase, PTPN22, was cloned fromRAW293 immune cells and is also expressed atlow levels in the brain (13). However,PTPN22(-/-) mice possess essentially wild-typelevels of AEA in the brain (21). 3) De-acylation of NAPE to generate lysoNAPE whichis then converted to NAE by alysophospholipase D. Evidence for this pathwayincludes the identification of a secretedphospholipase, sPLA2 IB, that singly deacylatesNAPE (22). LysoNAPE generated by thisenzyme was converted by brain extracts to NAEin a reaction presumed to be catalyzed by a D-type lysophospholipase. More recent datasuggest that this lysoNAPE-to-NAEtransformation actually reflects a two-stepconversion (see enzymatic route 4 ) (14). 4 )

Double-deacylation of NAPE to yieldglycerophospho-(GP)-NAE, which is then convertedto NAE by a metal-dependent phosphodiesterase.This pathway is supported by two principalobservations. First, conversion of NAPE or(lyso)NAPE to NAE in NAPE-PLD(-/-) brains isblocked by MAFP, an inhibitor of serine hydrolases,including many A/B-type phospholipases.Similarly, an NAPE analog with non-hydrolyzableether bonds in place of natural esters was notconverted to NAE in NAPE-PLD(-/-) brain extracts,demonstrating the absolute requirement for NAPEdeacylation (14). Collectively, these observationssuggest that double-O -deacylation, as would beperformed by an MAFP-sensitive hydrolase, isrequired for the PLD-independent conversion ofNAPE to NAE. This model was bolstered by theidentification and molecular characterization ofAbh4 as a B-type phospholipase selective forNAPEs over other phospholipid substrates. T h eresulting GP-NAE products are then rapidlyconverted by brain homogenates to NAEs by ametal-dependent phosphodiesterase. If this fourthenzymatic pathway is responsible for convertingNAPEs to NAEs in brain, then this tissue shouldcontain both the GP-NAE intermediates and a GP-NAE phosphodiesterase.

Here, we have identified the integralmembrane enzyme GDE1 as a GP-NAEphosphodiesterase. GDE1 belongs to larger familyof enzymes that hydrolyze the phosphodiester bondof glycerol-phosphodiesters releasing glycerol-phosphate and a free alcohol. GDE domain-containing enzymes are found in bacteria, yeast,plants, and mammals and are ascribed functions thatinclude membrane remodeling, phospholipidturnover, and phosphate scavenging. To date, GDEshave been primarily characterized in the context ofg l y c e r o p h o s p h o e t h a n o l a m i n e ( G P - E ) ,g l y c e r o p h o s p h o c h o l i n e ( G P - C ) , o rglycerophosphoinositol (GP-I) metabolism (23-26).

To determine if any of the mammalianGDE-domain containing enzymes possess GP-NAEphosphodiesterase activity, we expressed theseenzymes in COS-7 cells. Among the five GDEstested, one enzyme, GDE1, exhibited robust activitywith GP-NAE. GDE1 was originally identified in ayeast two-hybrid screen for proteins that interactwith RGS16, a regulator of G-protein signaling (19).Based on its interactions with RGS16, GDE1 is alsoreferred to as MIR16 (for Membrane Interacting

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

8

protein of RGS16). Subsequently GDE1 hasbeen shown to cleave GP-inositol substrates invitro, although the endogenous substrate(s) forthis enzyme remain unknown (27). The Farquhargroup went on to demonstrate that GPCRagonists could modulate the activity of GDE1suggesting a provocative bidirectionalinteraction between GPCR signaling and GDE1activity.

Several additional lines of biochemicalevidence support GDE1 as the principal GP-NAE phosphodiesterase in brain tissue. First,GDE1 displayed a pH optimum and cationsensitivity nearly identical to those observed forbrain GP-NAE phosphodiesterase activity. Bothactivities displayed slightly alkaline pH optima,inhibition by EDTA, calcium, and zinc, andactivation by magnesium. Second, GDE1 is anintegral membrane protein, which matches thesubcellular distribution of the brain GP-NAEphosphodiesterase activity. Finally, the tissuedistribution of GDE1 matched closely that ofGP-NAE phosphodiesterase activity, as well asother enzymes involved in NAE metabolism,being highly expressed in brain and spinal cord,intermediately expressed in testes, kidney, andliver, and only weakly expressed in heart andspleen.

We have also reported herein thedetection and quantification of GP-NAEs asendogenous brain constituents. To ourknowledge, these lipids have not previouslybeen reported as natural products in mammaliansystems. Using liquid chromatography coupledwith high-resolution tandem mass spectrometrywe detected multiple GP-NAE species,including the proposed AEA precursor GP-NArE, in normal mouse brain. Theseendogenous metabolites were detected at levelsapproximately 10-fold lower than NAEs

suggesting a rapid enzymatic flux through thispathway. Consistent with this premise, substantialaccumulation of endogenous GP-NAEs wasobserved upon incubation of brain lysates with theGP-NAE phosphodiesterase inhibitor EDTA. Thisaccumulation was also blocked by MAFP, whichpresumably prevents the serine hydrolase (e.g.,Abh4)-mediated sn-1/sn-2 deacylation of NAPEprecursors. It is interesting to note that, the EDTA-dependent accumulation of GP-NAEs was acyl-chain dependent. While significant increases wereobserved in long chain polyunsaturated (e.g., C20:4,C22:6), as well as shorter chain saturated and mono-unsaturated (e.g., C16:0, C18:1) GP-NAEs, noaccumulation of the long chain saturated C20:0 GP-NAE was detected. We also detected a number ofother long-chain saturated GP-NAEs (e.g., C22:0,C24:0, C24:1) in brain extracts that did notaccumulate upon incubation with EDTA (data notshown). These findings presents a striking foil tothe levels of NAEs in NAPE-PLD(–/–) mice, wheresignificant decreases were observed exclusively inlong chain (> C20) saturated/monounsaturatedmembers of this lipid class (12). Together, these datasuggest that NAEs may be biosynthesized in vivo bya bifurcated pathway dependent on the identity oftheir acyl chains.

In summary, this work, in conjunction withprevious studies (12,14), supports a model for AEAbiosynthesis that involves double O-deacylation byAbh4 to yield GP-NArE followed by GDE1-mediated conversion of this lipid intermediated toAEA. Validation of this route for AEA biosynthesisthrough, for example, the creation of mice lackingAbh4 or GDE1, would designate these enzymes asprincipal regulators of endocannabinoid signaling invivo and potential therapeutic targets for a range ofhuman disorders, including obesity, smokingcessation, and cancer (28).

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

9

References

1. Kuehl, F. A., Jacob, T. A., Ganley, O. H., Ormond, R. E., and Meisinger, M. A. P. (1957)J. Am. Chem. Soc. 79, 5577-5578

2. Bachur, N. R., Masek, K., Melmon, K. L., and Udenfriend, S. (1965) J Biol Chem 240,1019-1024

3. Natarajan, V., Reddy, P. V., Schmid, P. C., and Schmid, H. H. (1982) Biochim BiophysActa 712, 342-355

4. Devane, W. A., Hanus, L., Breuer, A., Pertwee, R. G., Stevenson, L. A., Griffin, G.,Gibson, D., Mandelbaum, A., Etinger, A., and Mechoulam, R. (1992) Science 258, 1946-1949

5. Cravatt, B. F., Giang, D. K., Mayfield, S. P., Boger, D. L., Lerner, R. A., and Gilula, N.B. (1996) Nature 384, 83-87

6. Kathuria, S., Gaetani, S., Fegley, D., Valino, F., Duranti, A., Tontini, A., Mor, M.,Tarzia, G., La Rana, G., Calignano, A., Giustino, A., Tattoli, M., Palmery, M., Cuomo,V., and Piomelli, D. (2003) Nat Med 9, 76-81

7. Cravatt, B. F., Demarest, K., Patricelli, M. P., Bracey, M. H., Giang, D. K., Martin, B. R.,and Lichtman, A. H. (2001) Proc. Natl. Acad. Sci. U.S.A. 98, 9371-9376

8. Cadas, H., di Tomaso, E., and Piomelli, D. (1997) J. Neurosci. 17, 1226-12429. Sugiura, T., Kondo, S., Sukagawa, A., Tonegawa, T., Nakane, S., Yamashita, A., and

Waku, K. (1996) Biochemical and Biophysical Research Communications 218, 113-11710. Jin, X. H., Okamoto, Y., Morishita, J., Tsuboi, K., Tonai, T., and Ueda, N. (2007) J Biol

Chem 282, 3614-362311. Okamoto, Y., Morishita, J., Tsuboi, K., Tonai, T., and Ueda, N. (2004) J. Biol. Chem.

279, 5298-530512. Leung, A., Saghatelian, A., Simon, G. M., and Cravatt, B. F. (2006) Biochemistry 45,

4720-472613. Liu, J., Wang, L., Harvey-White, J., Osei-Hyiaman, D., Razdan, R., Gong, Q., Chan, A.

C., Zhou, Z., Huang, B. X., Kim, H. Y., and Kunos, G. (2006) Proc Natl Acad Sci U S A103, 13345-13350

14. Simon, G. M., and Cravatt, B. F. (2006) J. Biol. Chem. 281, 26465-2647215. Hansen, H. H., Hansen, S. H., Schousboe, A., and Hansen, H. S. (2000) Journal of

Neurochemistry 75, 861-87116. Yanaka, N. (2007) Biosci Biotechnol Biochem 71, 1811-181817. Patton-Vogt, J. (2007) Biochim Biophys Acta 1771, 337-34218. Nogusa, Y., Fujioka, Y., Komatsu, R., Kato, N., and Yanaka, N. (2004) Gene 337, 173-

17919. Zheng, B., Chen, D., and Farquhar, M. G. (2000) Proc Natl Acad Sci U S A 97, 3999-

400420. Okamoto, Y., Wang, J., Morishita, J., and Ueda, N. (2007) Chem Biodivers 4, 1842-185721. Liu, J., Wang, L., Harvey-White, J., Huang, B. X., Kim, H.-Y., Luquet, S., Palmiter, R.

D., Krystal, G., Rai, R., Mahadevan, A., Razdan, R. K., and Kunos, G.Neuropharmacology 54, 1-7

22. Sun, Y. X., Tsuboi, K., Okamoto, Y., Tonai, T., Murakami, M., Kudo, I., and Ueda, N.(2004) Biochem. J. 380, 749-756

23. Tommassen, J., Eiglmeier, K., Cole, S. T., Overduin, P., Larson, T. J., and Boos, W.(1991) Mol Gen Genet 226, 321-327

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

10

24. Almaguer, C., Fisher, E., and Patton-Vogt, J. (2006) Curr Genet 50, 367-37525. van der Rest, B., Boisson, A. M., Gout, E., Bligny, R., and Douce, R. (2002) Plant

Physiol 130, 244-25526. Rao, M., and Sockanathan, S. (2005) Science 309, 2212-221527. Zheng, B., Berrie, C. P., Corda, D., and Farquhar, M. G. (2003) PNAS 100, 1745-175028. Bifulco, M., Grimaldi, C., Gazzerro, P., Pisanti, S., and Santoro, A. (2007) Mol

Pharmacol 71, 1445-1456

FOOTNOTES

This work was supported by NIH (DA015197, DA017259) and GMS gratefully acknowledges supportfrom the ARCS foundation and a Koshland Graduate Fellowship in Enzyme Biochemistry.

The abbreviations used are: Abh4, alpha/beta-hydrolase 4; AEA, arachidonoyl ethanolamine; FAAH,fatty acid amide hydrolase; GDE1, glycerophosphodiesterase 1; GP-NAE, glycerophospho-N-acylethanolamine; GP-NArE, glycerophospho-N-arachidonoylethanolamine; LC, liquid chromatography;MAFP, methoxy arachidonoylfluorophosphonate; MIR16, membrane interacting protein of RGS16; MS,mass spectrometry; NAE, N-acyl ethanolamine; NAPE, N-acyl phosphatidylethanolamine; PC,phosphatidylcholine; PE, phosphatidylethanolamine; PLD, phospholipase D.

FIGURE LEGENDS

Figure 1. Proposed biosynthetic pathway for N-acyl ethanolamines (NAEs) including anandamide(AEA). Both O-acyl chains are removed from N-arachidonoyl phosphatidylethanolamine (NArPE) byAbh4 yielding glycerophospho-N-arachidonoylethanolamine (GP-NArE). The phosphodiester bond ofGP-NArE is then hydrolyzed by a phosphodiesterase (identified herein as GDE1) to release freeanandamide and glycerol-3-phosphate.

Figure 2. Fragmentation and chromatography of endogenous GP-NAEs in mouse brain via LC-MS/MS.Fragmentation and chromatographic behavior is shown for endogenous N-palmitoyl (A - C) and N-oleoyl(D - F) GP-NAEs. Three major fragment ions are produced, a diagnostic NAE-phosphate fragment (378and 404 for palmitoyl- and oleoyl-GP-NAEs, respectively) and two fragments corresponding to glycerol-3-phosphate and its dehydrate at 171 and 153, respectively (A and D). Representative multiple reactionmonitoring (MRM) chromatographs of synthetic and endogenous GP-NAEs (B and E). Tandem mass-spectrometry (MS/MS) fragmentation spectra of synthetic and endogenous GP-NAEs (C and F).

Figure 3. Ex vivo accumulation of GP-NAEs in brain tissue via an MAFP-sensitive and EDTA-dependentenzymatic pathway. (A) GP-NAE levels in brain tissue homogenized in buffer (50mM Tris pH 8.0) alone(black and light grey bars), buffer with 10 mM EDTA (clear bars) or 10 mM EDTA and 20 mM MAFP(MAFP, dark-grey bars). Homogenates were incubated for 0 hours (black bars) or 4 hours (all other bars)and then subjected to organic extraction and LC-MS/MS analysis. The inset shows a magnification oflow-abundance GP-NAEs. Student’s t-test compared EDTA-treated to samples treated with EDTA andMAFP; asterisk (*) indicates p < 0.05, NS – not significant (n = 4 per group). (B) Fragmentation of N-arachidonoyl-GP-NAE produces a diagnostic NAE-phosphate fragment (m/z 426) and two fragmentscorresponding to glycerol-3-phosphate and its dehydrate at 171 and 153, respectively. Representativechromatographs (C) and tandem MS fragmentation (D) are shown for synthetic GP-NArE and natural GP-NArE which accumulates in the presence of EDTA.

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

11

Figure 4. Expression and GP-NAE phosphodiesterase activity of GDE-domain containing enzymes. (A)Myc-tagged GDE-domain containing enzymes were over-expressed in COS-7 cells. Membrane fractionswere treated with PNGase F and visualized by SDS-PAGE and western blotting with an anti-mycantibody (Cell Signal). (B) GP-NAE phosphodiesterase activity of membrane fractions of COS-7 cellsexpressing GDE-domain containing proteins was measured by monitoring release of N-C16:0 NAE viaLC-MS with 100 µM of N-C16:0 GP-NAE as substrate.

Figure 5. GDE1 and GP-NAE phosphodiesterase activity reside in the membrane. (A) Western blot ofmouse brain GDE1 showing that it resides exclusively in the membrane fraction and is a glycoprotein asevidenced by a ~5 kDa shift in migration upon treatment with PNGase F. (B) GP-NAEphosphodiesterase activity resides exclusively in the membrane fraction of mouse brain.

Figure 6. Mouse brain GP-NAE phosphodiesterase and GDE1 display similar biochemical properties.(A) pH-rate profile of brain membrane GP-NAE phosphodiesterase (open diamonds) and membranesfrom COS-7 cells transfected with GDE1 (filled squares) showing equivalent pH optima. Assays wereperformed in buffer containing 50 mM Tris, 50 mM HEPES, and 50 mM glycine and adjusted to thedesired pH with HCl/NaOH. (B) Mouse brain GP-NAE phosphodiesterase and GDE1 display similarcation sensitivity profiles. Mouse brain membranes (open bars) and GDE1-transfected COS-7 cellmembranes (filled bars) were treated with 2 mM of CaCl2, ZnCl2, MgCl2, or EDTA. Assays wereperformed with 100 µM [1’-14C]N-C16:0 GP-NAE substrate, where the release of [1’-14C]N-C16:0 NAEwas measured by thin layer radiochromatography.

Figure 7. Tissue distribution of GP-NAE phosphodiesterase activity and GDE1. (A) GP-NAEphosphodiesterase activity in membrane fractions of mouse tissues. Assays were performed with 100 µM[1’-14C]N-C16:0 GP-NAE substrate, where the release of [1’-14C]N-C16:0 NAE was measured by thinlayer radiochromatography. (B) Western blot of GDE1 expression in membrane fractions of mousetissues following treatment with PNGase F showing strong correlation with GP-NAE phosphodiesteraseactivity. (C) Reverse transcription-PCR analysis of GDE1 mRNA expression correlates with GDE1protein levels and GP-NAE phosphodiesterase activity. The ubiquitously expressed enzymeglyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a control to confirm integrity of tissuecDNA. Br – brain; Spc – spinal cord; Ht – heart; Lv – liver; Kd – kidney; Spln – spleen; Ts – testes.

Figure 8. N-acyl chain selectivity of GDE1. Membrane fractions from GDE1 transfected COS-7 cellswere tested for activity against a panel of GP-NAE substrates bearing saturated (C16:0, C20:0),monounsaturated (C18:1) and polyunsaturated (C20:4, C22:6) N-acyl chains. Assays were performedwith 100 µM of each substrate using LC-MS to measure release of NAE.

Supplemental Figure 1. Specificity of rabbit polyclonal anti-mGDE1 antibodies. GDE1- and mock-transfected COS-7 cell membranes were treated with or without PNGase F and probed with anti-mGDE1antibodies using standard western blotting procedures. A 35-40 kDa doublet was detected exclusively inGDE1-transfected cells that shifted to a single 35 kDa protein after PNGase F digestion.

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

12

Table 1. Endogenous levels of GP-NAEs in mouse brain as measured by LC-MS/MS. C15:1 GP-NAEwas included as an internal standard. Measurements were corrected for extraction-efficiency as describedin Experimental Procedures and represent average values ± standard errors (n = 4 per group).

pmol/g wet tissueC16:0 GP-NAE 13.17 ± 4.02C18:1 GP-NAE 3.18 ± 1.09C20:0 GP-NAE 0.07 ± 0.03C20:4 GP-NAE 0.16 ± 0.07C22:6 GP-NAE 0.54 ± 0.19

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

13

Supplemental Table 1. High resolution MS and MS/MS of endogenous GP-NAEs. Parent ions anddiagnostic fragment ions ([M-H-74]-) were measured using an Orbitrap mass spectrometer in negativemode run at a resolution of 30,000. C16:0 and C18:1 GP-NAEs, which are more abundant, weremeasured directly from brain tissue, while the C20:4 GP-NAE was measured following incubation withEDTA as described in Figure 2.

C16:0 GP-NAEObserved m/z Calculated m/z Δppm

Parent Mass 452.27805 452.27771 0.75[M-H-74]- 378.24066 378.24093 0.71

C18:1 GP-NAEObserved m/z Calculated m/z Δppm

Parent Mass 478.29288 478.29336 1.00[M-H-74]- 404.2562 404.25658 0.94

C20:4 GP-NAEObserved m/z Calculated m/z Δppm

Parent Mass 500.27707 500.27771 1.28[M-H-74]- 426.24103 426.24093 0.23

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

14

Figure 1.

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

15

Figure 2.

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

16

Figure 3.

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

17

Figure 4.

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

18

Figure 5.

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

19

Figure 6.

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

20

Figure 7.

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

21

Figure 8.

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from

22

Supplemental Figure 1.

at The S

cripps Research Institute on M

arch 5, 2008 w

ww

.jbc.orgD

ownloaded from