Molecular Characterization of a Phospholipase D Generating Anandamide and Its Congeners*

Anandamide (N-arachidonoylethanolamine) is known to be an endogenous ligand of cannabinoid and vanilloid receptors. Its congeners (collectively referred to as N-acylethanolamines) also show a variety of biological activities. These compounds are principally formed from their corresponding N-acyl-phosphatidylethanolamines by a phosphodiesterase of the phospholipase D-type in animal tissues. We purified the enzyme from rat heart, and by the use of the sequences of its internal peptides cloned its complementary DNAs from mouse, rat, and human. The deduced amino acid sequences were composed of 393–396 residues, and showed that the enzyme has no homology with the known phospholipase D enzymes but is classified as a member of the zinc metallohydrolase family of the β-lactamase fold. As was overexpressed in COS-7 cells, the recombinant enzyme generated anandamide and other N-acylethanolamines from their corresponding N-acyl-phosphatidylethanolamines at comparable rates. In contrast, the enzyme was inactive with phosphatidylcholine and phosphatidylethanolamine. Assays of the enzyme activity and the messenger RNA and protein levels revealed its wide distribution in murine organs with higher contents in the brain, kidney, and testis. These results confirm that a specific phospholipase D is responsible for the generation of N-acylethanolamines including anandamide, strongly suggesting the physiological importance of lipid molecules of this class.

It is accepted that N-acylethanolamines are principally biosynthesized in animal tissues from membrane phospholipid by two steps of enzyme reactions (1)(2)(3)(4)12): 1) N-acylation of phosphatidylethanolamine (PE) 1 to generate N-acyl-phosphatidylethanolamine (NAPE) by an acyltransferase, and 2) subsequent release of N-acylethanolamine from NAPE by a phosphodiesterase of the phospholipase D (PLD) type. In this article, we refer to the latter enzyme as NAPE-hydrolyzing PLD (NAPE-PLD).
Catalytic properties of NAPE-PLD have been reported by the use of crude preparations and partially purified enzymes (13)(14)(15)(16)(17)(18)(19). The enzyme was suggested to be distinct from the known PLD enzymes based on substrate specificity (15,16). However, to date its molecular characterization has not been performed.
Enzyme Purification-One hundred and fifty adult Wistar ST rats (SLC, Japan) were anesthetized with diethyl ether and sacrificed by cervical dislocation. Hearts were removed, cut into small pieces, and then homogenized in 5 times the volume (v/w) of 20 mM Tris-HCl (pH 7.4) containing 0.32 M sucrose by a Polytron homogenizer. The homogenates were centrifuged at 800 ϫ g for 15 min. The supernatant was further centrifuged at 105,000 ϫ g for 55 min, and the resultant pellet was suspended in phosphate-buffered saline (pH 7.4). After freezing and thawing, the sample was centrifuged at 105,000 ϫ g for 55 min. The pellet was suspended in 20 mM Tris-HCl (pH 7.4) containing 1% (w/v) octyl glucoside and centrifuged again at 105,000 ϫ g for 55 min. The resultant clear supernatant was frozen at Ϫ80°C. After thawing slowly at 4°C, the sample (50 mg of protein) was loaded on a HiTrap SP HP cation-exchange column (bed volume, 5 ml) pre-equilibrated with 20 mM Tris-HCl (pH 7.4) containing 1% (w/v) CHAPS (buffer A). After washing the column with 20 ml of buffer A, the enzyme was eluted with 28 ml of buffer A containing 150 mM NaCl. Active fractions (11 mg of protein) were diluted 3-fold (v/v) with buffer A and then loaded on a HiTrap Q anion-exchange column (1 ml). After washing the column with 5 ml of 50 mM NaCl and 10 ml of 200 mM NaCl, the enzyme was eluted with 4 ml of 300 mM NaCl. Active fractions were diluted 3-fold with buffer A and loaded on a HiTrap Blue affinity column (1 ml). After washing the column with 5 ml of 50 mM NaCl and 5 ml of 200 mM NaCl, the enzyme was eluted with 6 ml of 500 mM NaCl. Active fractions were finally applied onto a Bio-Gel HTP hydroxyapatite column (1 ml). After washing the column with 5 ml of buffer A and 3 ml of buffer A containing 50 mM potassium phosphate, the enzyme was eluted with 6 ml of 100 mM potassium phosphate. Protein concentration was determined by the method of Bradford (20) with bovine serum albumin as standard. When the particulate fractions of rat heart were stored at Ϫ80°C for 1 month, 4°C for 1 week, 4°C for 24 h, or 25°C for 3 h, the enzyme activity was lost by 8, 86, 45, or 25%, respectively. By the same treatment, the enzyme activity of the octyl glucoside-solubilized proteins was lost by 9, 78, 50, or 26%, respectively.
Microsequencing and Bioinformatics-To isolate internal peptides from the purified rat heart NAPE-PLD, active fractions from the hydroxyapatite column were subjected to SDS-PAGE on a 10% gel. Band A stained with Coomassie ( Fig. 1) was excised from the gel and subjected to in-gel digestion using trypsin. The digest mixture was separated by reverse-phase high performance liquid chromatography using a TSKgel ODS-80Ts column (2.0 ϫ 250 mm, TOSOH) at a flow rate of 200 l/min. The mobile phase used was a 0 -90% gradient of acetonitrile in 0.1% trifluoroacetic acid, and fractions were collected every 1 min. Selected peptide peaks were subjected to microsequencing with a Procise 494 cLC protein sequencer (Applied Biosystems, Foster City, CA). By this method, the sequences of five peptides were determined. By the use of data base, the three peptide sequences (KLHDEEIQELQA, LH-DEEIQELQAQ, and LLAELEQLK) were found to be contained in rat vimentin, and the other two sequences (ELPVLKPY(F/V)VSD and YGLKSEDFFILK) were found to be contained in a single cDNA clone of mouse (GenBank TM accession number, XM144214), which was assumed to be NAPE-PLD (see "Results"). The BLAST program, non-redundant data bases, and EST data bases in GenBank TM were searched for rat and human homologous sequences of the mouse cDNA (XM144214). As the result, one rat putative slug protein (XP231294), two rat EST sequences (AI013914 and CB773576), two human putative slug proteins (XP168636 and XP168592), and seven human EST sequences (AI857635, AL564333, AW293283, BG977438, BQ224588, BU507368, and BX411424) were found to be highly homologous to the mouse cDNA at the protein level.
cDNA Cloning-The mouse, rat, and human cDNAs containing the putative full-length coding region of NAPE-PLD were generated by PCR using the forward primers 5Ј-GGATCCATGGATGAGTATGAGG-ACAGCCAG-3Ј (mouse), 5Ј-GGTACCATGGATGAAAATGAGAACAGC-CAG-3Ј (rat), and 5Ј-ATGGATGAAAATGAAAGCAACCAG-3Ј (human) and the reverse primers 5Ј-CTCGAGTCATGTTTCTTCAAAAGCTCTA-TC-3Ј (mouse), 5Ј-GCGGCCGCTCATGTTTCCTCAAAGGCTTTGTC-3Ј (rat), and 5Ј-TTAAAAGTTTTCATCATCATTATT-3Ј (human). These primers were designed on the basis of the abovementioned cDNA sequences (XM144214 and its homologous sequences of rat and human). cDNAs used as templates of PCR were synthesized from total RNA (5 g) of mouse brain, rat brain, and human megakaryoblastic leukemia cells (CMK cells) by the use of Moloney murine leukemia virus-reverse transcriptase and the Superscript first-strand synthesis system, respectively. PCR amplification was performed at a denaturing temperature of 94°C for 30 s followed by annealing at 58°C for 30 s and extension at 72°C for 2 min (30 cycles). The full-length cDNA was subcloned into pCR2.1-TOPO vector, and sequencing of the inserts was performed with the aid of an ABI 377 DNA sequencer (Applied Biosystems, Foster City, CA).
Expression of NAPE-PLD in COS-7 Cells-The 1.2-kb NAPE-PLD cDNAs were ligated into the eukaryotic expression vector pcDNA3.1(ϩ). COS-7 cells were grown at 37°C to 70% confluency in a 100-mm dish containing Dulbecco's modified Eagle's medium with 10% fetal calf serum in a humidified 5% CO 2 , 95% air incubator. The cells were then treated with 8 g of NAPE-PLD-pcDNA3.1(ϩ) and Lipo-fectAMINE, and cultured at 37°C for 48 h, with one change of medium at 12 h. The harvested cells were sonicated 3 times each for 3 s in 20 mM Tris-HCl (pH 7.4) and used as homogenates. Control COS-7 cells were prepared in the same way, except that the insert-free pcDNA3.1(ϩ) vector was used for transfection. The homogenates were centrifuged at 105,000 ϫ g at 4°C for 15 min, and the obtained pellet (membrane fraction) was resuspended in 20 mM Tris-HCl (pH 7.4) containing 1% (w/v) octyl glucoside, followed by further centrifugation at 105,000 ϫ g at 4°C for 15 min. The resultant clear supernatant was referred to as the octyl glucoside-solubilized proteins.
Enzyme Assay-NAPE-PLD was incubated with 100 M N-[ 14 C]acyl-PE or N-[ 14 C]acyl-lyso-PE (10,000 cpm in 5 l of ethanol) in 100 l of 50 mM Tris-HCl (pH 7.5) at 37°C for 10 min. An activator (0.1% Triton X-100, 10 mM MgCl 2 , or 10 mM CaCl 2 ) was also contained in the reaction mixture. A mixture of chloroform/methanol (2:1, v/v, 0.3 ml) was added to the reaction mixture to terminate the reaction. After centrifugation, 100 l of the lower layer was spotted on a silica gel thin layer plate (10 cm height) and developed in chloroform, methanol, 28% ammonium hydroxide (80:20:2, v/v) at 4°C for 20 min. Distribution of radioactivity on the plate was quantified by a BAS1500 bioimaging analyzer (Fujix, Tokyo, Japan).
Preparation of Anti-NAPE-PLD Antiserum-Rabbit antiserum was raised against a hexahistidine-tagged mouse NAPE-PLD protein. A mouse NAPE-PLD cDNA lacking stop codon was generated by PCR using a forward primer 5Ј-GGATCCGATGGATGAGTATGAGGACCA-G-3Ј and a reverse primer 5Ј-CTCGAGTGTTTCTTCAAAAGCTCTAT-CATC-3Ј. This cDNA was ligated into a prokaryotic expression vector pTrcHis2B to generate a C-terminally hexahistidine-tagged NAPE-PLD protein. The construct was confirmed by sequencing in both directions and was introduced chemically into competent E. coli TOP10 FЈ cells as host. Cultures were induced with 1 mM isopropyl-␤-D-thiogalactopyranoside at an A 600 of 0.6 -0.7, allowed to grow at 37°C for 5 h, and pelleted at 2,000 ϫ g for 30 min. After freezing and thawing, the cells were resuspended in 50 mM Tris-HCl (pH 7.4) containing 100 mM NaCl (buffer B) and were sonicated on ice 10 times each for 20 s at a high intensity with a 1-min cooling period between each burst. The cell homogenates were then treated with 1% Triton X-100 on ice for 30 min and were centrifuged at 10,000 ϫ g for 30 min. The pellet was suspended in buffer B containing 0.3% SDS and centrifuged again at 105,000 ϫ g for 30 min. The resultant clear supernatant was diluted 3-fold with buffer B and was applied onto a nickel-nitrilotriacetic acidagarose column (2 ml of bed volume). After washing the column with 20 ml of buffer B containing 10 mM imidazole and 0.05% Triton X-100, the enzyme was eluted with 9 ml of buffer B containing 100 mM imidazole and 0.05% Triton X-100. Purification of the recombinant protein with a molecular mass of about 50 kDa was monitored by SDS-PAGE stained with Coomassie and by Western blotting with anti-hexahistidine antibody (1:10,000 dilution).
The purified NAPE-PLD (0.4 -0.5 mg protein in 0.75 ml) was emulsified with an equal volume of Freund's adjuvant, and the emulsion was injected subcutaneously at multiple sites along the back of a New Zealand White rabbit (SLC, Japan) every 2 weeks 4 times in total. The animal was bled 7 days after the final injection, and the serum fraction was prepared. The antiserum was titrated by Western blotting.
Western Blotting-After separation by SDS-PAGE, proteins were electrotransferred to a hydrophobic polyvinylidene difluoride membrane (Hybond P). The membrane was blocked with phosphate-buffered saline containing 0.1% Tween 20 and 5% dried milk (buffer C) and then incubated with anti-NAPE-PLD antiserum (1:2,000 dilution) in buffer C at room temperature for 1 h, followed by the incubation with the horseradish peroxidase-labeled secondary antibody (1:2,000 dilution) in buffer C at room temperature for 1 h. Finally, NAPE-PLD was visualized using enhanced chemiluminescence and analyzed by a LAS1000plus lumino-imaging analyzer (Fujix, Tokyo, Japan).

RESULTS
Identification and Cloning of NAPE-PLD cDNA-In earlier studies we solubilized the NAPE-PLD from the 105,000 ϫ g pellet of rat heart homogenates, and partially purified it by ion-exchange column chromatographies (13,14). We further purified the rat heart enzyme by the use of a combination of four column chromatographies to a specific activity of 406 nmol/min/mg protein at 37°C (Table I). As analyzed by SDS-PAGE, the final preparation was not completely pure, and several major protein bands were still detected (Fig. 1). However, since we noted that the intensity of Band A with a molecular mass of about 46 kDa changed in agreement with the NAPE-hydrolyzing activity through the hydroxyapatite chromatography, we presumed that Band A was the enzyme protein.
The protein of Band A was digested with trypsin. The resulting peptides were separated by reverse-phase high performance liquid chromatography and were microsequenced. By this procedure, we could determine five peptide sequences. By the use of data base, the three peptide sequences (KLHDEE-IQELQA, LHDEEIQELQAQ, and LLAELEQLK) were found to be contained in rat vimentin. Vimentins are class-III intermediate filaments found in various non-epithelial cells, and their molecular mass is 56 kDa (21). Hence, this protein did not seem to be NAPE-PLD but a contaminant. The other two sequences (ELPVLKPY(F/V)VSD and YGLKSEDFFILK) were found to be contained in a single cDNA clone of mouse (GenBank TM accession number XM144214), the function of which has not been reported. However, because the data base search with RPS-BLAST suggested that this clone belongs to the zinc metallohydrolase family of the ␤-lactamase fold (22,23), we assumed the clone as NAPE-PLD. We amplified the putative coding region of this cDNA by RT-PCR with RNA of mouse brain, and determined its nucleotide sequence. The sequence was completely identical with that of XM144214. The deduced amino acid sequence is shown in Fig. 2.
When the BLAST program, non-redundant data bases, and EST data bases in GenBank TM were also searched for rat and human homologous sequences of the mouse cDNA, one rat putative slug protein (XP231294), two rat EST sequences (AI013914 and CB773576), two human putative slug proteins (XP168636 and XP168592), and seven human EST sequences (AI857635, AL564333, AW293283, BG977438, BQ224588, BU507368, and BX411424) were found to be highly homologous to the mouse cDNA at the protein level. We prepared PCR primers based on these nucleotide sequences, and we cloned rat and human homologues from rat brain and human megakaryoblastic leukemia cells (CMK cells) by RT-PCR (Fig. 2). The amino acid sequences deduced from the cDNAs of three animal species were composed of 396 (mouse and rat) and 393 (human) residues, and their molecular weights were calculated to be 45,816 (mouse), 45,737 (rat) and 45,596 (human). Their amino acid identity was 95.5 (between mouse and rat), 89.1 (between mouse and human), and 90.4% (between rat and human). The sequences showed no homology with those of the reported PLDs such as mammalian PLD1 (24) and PLD2 (25), Saccharomyces cerevisiae SPO14/PLD1 (26), Streptomyces antibioticus PLD (27), and glycosylphosphatidylinositol-specific PLD (28).
The zinc metallohydrolase family of the ␤-lactamase fold is a large superfamily of proteins including a wide variety of hydrolases such as class B ␤-lactamase, glyoxalase II, arylsulfatase, cytidine monophosphate-N-acetylneuraminic acid hydroxylase, and cAMP phosphodiesterase (22,23). Proteins belonging to this superfamily are characterized by an HX(E/H)XD(C/R/S/ H)X 50 -70 HX 15-30 (C/S/D)X 30 -70 H motif that is highly conserved among members of this family. This motif is presumed to participate in zinc coordination and hydrolysis reaction (22,23). As shown in Fig. 3, this motif was completely conserved in the putative NAPE-PLD of the three animal species, suggesting that the catalytic activity of NAPE-PLD is correlated with the zinc content.
Overexpression and Characterization of NAPE-PLD-We overexpressed the mouse cDNA in COS-7 cells by the lipofection method. The homogenates of the transfected cells were   (Fig. 4, lane 3). In contrast, the enzyme activity in the homogenates of COS-7 cells transfected with the insert-free vector was very low (0.02 Ϯ 0.004 nmol/

FIG. 2. Deduced amino acid sequences from mouse, rat, and human NAPE-PLD cDNAs.
Sequence identity shared by three NAPE-PLD proteins is shown by asterisks. Underlines denote sequences corresponding with the peptide fragments obtained by tryptic digestion of the rat heart enzyme.

FIG. 3. Comparison of NAPE-PLD to representative members of the zinc metallohydrolase family of the ␤-lactamase fold.
Alignment over the highly conserved segments among representative members of the zinc metallohydrolase family of the ␤-lactamase fold is shown. Highly conserved residues are shown in boldface type and are boxed. GenBank TM accession numbers of Bacillus cereus ␤-lactamase, human glyoxalase II, Desulfovibrio gigas ROO (rubredoxin oxygen:oxidoreductase), human ENAC1 (aryl sulfatase), E. coli PhnP (phosphonate uptake and biodegradation), mouse CMP-NeuAc hydroxylase (cytidine monophosphate-N-acetylneuraminic acid hydroxylase), and S. cerevisiae PDE1 (cAMP phosphodiesterase) are M19530, NM005326, AF218053, AF308695, AE000482, NM007717, and P22434, respectively. min/mg protein), and an excessive amount of the homogenates gave a faint band of the product (Fig. 4, lane 2). The rat and human homologues were also expressed in COS-7 cells by the same method, and the homogenates of the transfected cells exhibited the NAPE-PLD activities.
As analyzed by Western blotting with polyclonal antibody raised against a mouse recombinant NAPE-PLD, the homogenates of COS-7 cells transfected with mouse or rat cDNA revealed an immunoreactive band around 46 kDa (Fig. 5, lanes 3  and 4). In contrast, the homogenates of COS-7 cells transfected with the insert-free vector did not show the immunoreactive protein band (lane 2). The antibody also stained a 46-kDa band with the purified native NAPE-PLD from rat heart (lane 1), demonstrating immunological identity of the recombinant enzyme with the native enzyme.
When the cell homogenates were subjected to ultracentrifugation, the enzyme activity was mostly recovered in the membrane fraction (31 Ϯ 1 nmol/min/mg protein) with a much lower activity in the cytosol (4 Ϯ 1 nmol/min/mg protein). Again, the membrane fraction of COS-7 cells transfected with the insertfree vector showed a very low activity (0.05 Ϯ 0.005 nmol/ min/mg protein). Its localization in the membrane fraction was in agreement with that of the native enzyme (15). The recombinant mouse enzyme was stimulated 1.7-fold by 0.1% Triton X-100. The enzyme could be solubilized from the membrane with 1% octyl glucoside, and the soluble enzyme was activated 1.9-and 1.8-fold by 10 mM CaCl 2 and MgCl 2 , respectively.
We next examined substrate specificity of the recombinant enzyme with various N-[ 14 C]acyl-PE as substrates (Table II) (Table II). In addition, we found that the native enzyme from rat heart showed similar K m values for N-palmitoyl-PE and N-palmitoyl-lyso-PE to those of the recombinant enzyme (Table II).
Moreover, we investigated whether or not PC and PE were also substrates for NAPE-PLD. However, although Streptomyces sp. PLD generated phosphatidic acid from PC and PE, even a large amount of the recombinant enzyme could not hydrolyze PC or PE (Fig. 6).
Organ Distribution of NAPE-PLD-The organ distribution of the NAPE-PLD activity was examined using microsomes from various murine organs (Fig. 8A). The N-palmitoyl-PE hydrolyzing activity was widely distributed with higher specific activities in brain, kidney, and testis. As shown in Fig. 8, B and C, the distribution of murine NAPE-PLD at mRNA and protein levels was not completely identical but similar to that of the  enzyme activity. These results suggest that the enzyme identified by us is principally responsible for the NAPE-PLD activity in various organs.

DISCUSSION
The major biosynthetic pathway of anandamide and other bioactive N-acylethanolamines is composed of two enzyme reactions: 1) transfer of an acyl group from the sn-1 position of glycerophospholipid to the amino group of PE resulting in the formation of NAPE, and 2) subsequent hydrolysis of NAPE to N-acylethanolamine and phosphatidic acid (1)(2)(3)(4). The former reaction is catalyzed by a Ca 2ϩ -dependent N-acyltransferase and the latter reaction by a phosphodiesterase of the PLD type (NAPE-PLD). However, molecular characterization of enzymes involved in this pathway has not been performed. Here for the first time we report the molecular cloning and expression of NAPE-PLD that specifically generates N-acylethanolamines including anandamide from their corresponding NAPEs. The cDNA cloning revealed that the enzyme belongs to the zinc metallohydrolase family of the ␤-lactamase fold (Fig. 3) and is structurally unrelated to the known PLD enzymes. We cloned NAPE-PLD from mouse, rat, and human, and the three enzymes were found to be highly conserved with more than 89% identity at the amino acid level (Fig. 2).
The present studies have several important and interesting implications. First, the sequence analysis assigned NAPE-PLD to a member of the zinc metallohydrolase family of the ␤-lactamase fold (Fig. 3). The rapidly increasing numbers of proteins classified in this superfamily catalyze a variety of diverse reactions (22,23,30). The members of this family that have been functionally and structurally characterized include ␤-lactamase (31), glyoxalase II (32), and rubredoxin:oxygen oxidoreductase (33). Five segments conserved in the enzymes of this family contain several highly conserved histidine and aspartic acid residues and are thought to participate in zinc coordination and hydrolysis reaction (Fig. 3) (22,23,30). These amino acid residues were completely conserved in NAPE-PLD of the three animal species, suggesting that the catalytic activity of NAPE-PLD is correlated with the zinc content. Detailed  (n ϭ 3). B, distribution of the NAPE-PLD and GAPDH mRNA was analyzed by RT-PCR as described under "Experimental Procedures." C, distribution of the NAPE-PLD protein was analyzed by Western blotting as described under "Experimental Procedures." The 105,000 ϫ g pellet (50 g of protein) from the homogenates of each mouse organ were used. Lane R, recombinant mouse PLD used as a positive control. mutational studies using the recombinant enzyme will be required to elucidate catalytic mechanisms of NAPE-PLD. Because NAPE-PLD is classified as a phosphodiesterase based on the type of reaction, we should note that Vibrio fischeri and S. cerevisiae cAMP phosphodiesterases, and E. coli phosphodiesterase (Elac) are included in this family (22,23,34,35).
Second, NAPE-PLD was catalytically distinguishable from the known PLD. Previous reports showed that microsomes of dog brain and rat heart had a low PLD activity toward PC and PE in addition to the NAPE-PLD activity (15,16). However, our present studies revealed that the recombinant NAPE-PLD had no activity with PC or PE (Fig. 6). It should be noted that PC is a common substrate of mammalian PLD1 and PLD2 (29). We also confirmed that the enzyme does not catalyze transphosphatidylation to generate phosphatidyl alcohol in the presence of primary alcohols (Fig. 7). The lack of transphosphatidylation was in agreement with the previous results (18) with microsomes of rat brain and heart having the NAPE-PLD activity. These catalytic properties of NAPE-PLD indicate that the enzyme is not only structurally but also functionally distinct from PLD members of the HKD/phosphatidyltransferase gene family (29).
Third, our results demonstrated that not only N-arachidonoyl-PE (a precursor of anandamide) but also other NAPEs with different N-acyl groups are good substrates of the recombinant NAPE-PLD (Table II). It has been a matter of debate whether or not an anandamide-specific biosynthetic pathway exists in animal tissues because most of endogenous N-acylethanolamines except anandamide do not function as cannabinoid receptor ligands (4). Sugiura et al. (17) reported previously that rat brain microsomes showed the PLD activity for all of NAPEs with different N-acyl groups (16:0, 18:0, 18:1, 18:2, and 20 :4) and that N-arachidonoyl-PE was less active than other NAPEs. Our results with the recombinant enzyme confirm this earlier finding and explain why the composition of naturally occurring N-acylethanolamines tends to resemble the N-acyl composition of the precursor NAPE (4). Some exceptional tissues such as uterus (36) and some tumor tissues (37) in which anandamide is relatively abundant may possess other enzymes or pathways to contribute to the selective generation of anandamide.
Fourth, we detected the N-palmitoyl-PE hydrolyzing activity in almost all the mouse organs with higher specific activities in brain, kidney, and testis (Fig. 8A). Earlier, Schmid et al. (16) examined the organ distribution of NAPE-PLD with rat organ homogenates and reported that the heart exhibited the highest activity, followed by brain, testis, kidney, spleen, liver, and lung. We also showed a similar distribution of the enzyme in rat with the proteins solubilized from rat organ microsomes (14). Petersen et al. (38) reported that among bovine organs the brain exhibited the highest specific activity, followed by kidney, spleen, lung, heart, and liver. From these results, the brain consistently displays a high NAPE-PLD activity over animal species. In contrast, remarkable species difference of the NAPE-PLD activity has been found with the heart (39). Furthermore, we revealed that relative amounts of the mRNA and protein of NAPE-PLD in different organs showed similar patterns to potency of the NAPE-PLD activity (Fig. 8, B and C). The results suggest that the enzyme identified by us is principally responsible for the NAPE-PLD activity in various organs, although we cannot rule out the possibility that isozymes or different biosynthetic pathways also participate in the generation of N-acylethanolamines. Future studies using specific inhibitors and gene-disrupted animals of NAPE-PLD will elucidate this question. Age-dependent change of the NAPE hydrolyzing activity was also reported with rat brain (40). Thus, it will be of interest to investigate the regulatory mechanism for the expression levels of mRNA and protein of NAPE-PLD.
Previously, we showed that NAPE-PLD partially purified from rat heart was markedly stimulated with Triton X-100 and divalent cations including Ca 2ϩ and Mg 2ϩ (13). Marked stimulatory effects of Triton X-100, Ca 2ϩ , and Mg 2ϩ were also observed with the native enzyme of mouse brain (data not shown). However, the recombinant mouse enzyme was stimulated only 2-fold by these activators. The reason for this difference remains unclarified, and we cannot rule out the possibility that there is a subtle structural difference between the native enzyme and recombinant enzyme. Further examination on the cofactor requirement will be necessary with the purified recombinant enzyme.
Recent studies with gene-disrupted mice of the anandamidedegrading enzyme (fatty acid amide hydrolase (41)) emphasize physiological importance of anandamide as a neuromodulator (42)(43)(44). Not only anandamide but also other N-acylethanolamines were dramatically accumulated in the gene-disrupted mice (44). However, physiological significance of cannabinoid receptor-inactive N-acylethanolamines remains unclarified. Identification of the gene of NAPE-PLD enables us to generate model animals in which NAPE-PLD is knocked out or overexpressed. In addition, the structural analysis of the recombinant NAPE-PLD protein should help us to elucidate the catalytic and regulatory mechanisms of NAPE-PLD in biologic responses and to develop its selective inhibitors. These future studies will contribute to better understanding of the physiological and pathophysiological significance of anandamide and other bioactive N-acylethanolamines in mammals.