HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 42, 32783-32792, October 20, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/42/32783    most recent M003505200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Murakami, M. Articles by Kudo, I. Search for Related Content PubMed PubMed Citation Articles by Murakami, M. Articles by Kudo, I. Social Bookmarking                      What's this?

Regulation of Prostaglandin E₂ Biosynthesis by Inducible Membrane-associated Prostaglandin E₂ Synthase That Acts in Concert with Cyclooxygenase-2^*

Makoto Murakami^§, Hiroaki Naraba^§¶, Toshihiro Tanioka^§, Natsuki Semmyo, Yoshihito Nakatani, Fumiaki Kojima^¶, Tomomi Ikeda^¶, Mai Fueki^¶, Akinori Ueno^¶, Sachiko Oh-ishi^¶, and Ichiro Kudo

From the  Department of Health Chemistry, School of Pharmaceutical Sciences, Showa University, 1-5-8 Hatanodai, Shinagawa-ku, Tokyo 142-8555 and the ^¶ Department of Pharmacology, School of Pharmaceutical Sciences, Kitasato University, Shirokane 5-9-1, Minato-ku, Tokyo 108-0072, Japan

Received for publication, April 24, 2000, and in revised form, May 25, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Here we report the molecular identification of membrane-bound glutathione (GSH)-dependent prostaglandin (PG) E₂ synthase (mPGES), a terminal enzyme of the cyclooxygenase (COX)-2-mediated PGE₂ biosynthetic pathway. The activity of mPGES was increased markedly in macrophages and osteoblasts following proinflammatory stimuli. cDNA for mouse and rat mPGESs encoded functional proteins that showed high homology with the human ortholog (microsomal glutathione S-transferase-like 1). mPGES expression was markedly induced by proinflammatory stimuli in various tissues and cells and was down-regulated by dexamethasone, accompanied by changes in COX-2 expression and delayed PGE₂ generation. Arg¹¹⁰, a residue well conserved in the microsomal GSH S-transferase family, was essential for catalytic function. mPGES was functionally coupled with COX-2 in marked preference to COX-1, particularly when the supply of arachidonic acid was limited. Increased supply of arachidonic acid by explosive activation of cytosolic phospholipase A₂ allowed mPGES to be coupled with COX-1. mPGES colocalized with both COX isozymes in the perinuclear envelope. Moreover, cells stably cotransfected with COX-2 and mPGES grew faster, were highly aggregated, and exhibited aberrant morphology. Thus, COX-2 and mPGES are essential components for delayed PGE₂ biosynthesis, which may be linked to inflammation, fever, osteogenesis, and even cancer.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The two kinetically distinct prostaglandin (PG)¹ biosynthetic responses, the immediate and delayed phases, imply the recruitment of different sets of the biosynthetic enzymes whose expression and activation are tightly regulated by post-receptor transmembrane signaling. In immediate PG biosynthesis, which occurs within several minutes after stimulation with agonists that increase cytoplasmic Ca²⁺ levels, cytosolic phospholipase A₂ (cPLA₂) is a prerequisite for supplying arachidonic acid (AA) to the constitutive cyclooxygenase (COX) isozyme, COX-1 (1-6). Delayed PG biosynthesis, which proceeds gradually over a long term period after a proinflammatory stimulus, is accompanied by de novo induction of COX-2, which is an absolute requirement irrespective of the coexistence of COX-1 (2-4, 6-10). cPLA₂ and several inducible secretory phospholipase A₂ isozymes cooperatively contribute to supplying AA to COX-2 (1, 2, 4-9, 11-13). The preference of COX-2 over COX-1 in the delayed response is explained, at least in part, by the ability of COX-2 to metabolize lower levels of AA to PGH₂ than those required for COX-1-directed catalysis (2, 14, 15). When cells are first treated with proinflammatory stimuli and subsequently exposed to Ca²⁺ mobilizers, the inducible COX-2 can also promote the immediate response (priming or induced immediate response) (7, 16).

Understanding COX-2-dependent biological responses has received much attention in the past few years, because numerous pharmacological, biological and genetic studies have suggested that this inducible COX isozyme is involved in various human diseases, including inflammation and cancer (10, 17-24). In many cells, the main PG species produced during the delayed response is PGE₂. Indeed, among the several PGs produced by macrophages, only the level of PGE₂ was increased during the delayed response (7, 25, 26). Moreover, in vivo studies have shown that COX-2 inhibitors reduce PGE₂ more profoundly than other PGs (17). Thus, the COX-2-dependent pathway may be more selectively linked to the terminal PGE₂ synthase (PGES). More importantly, several recent studies have suggested that PGES activity is increased during the period when COX-2-dependent delayed PGE₂ generation is ongoing (7, 26).

PGES activity has been detected in both cytosolic and membrane-associated fractions of various cells and tissues (27-30). In most cases the enzyme requires glutathione (GSH) for catalytic activity (27, 28, 30). In an effort to identify PGES isoforms, we have succeeded in identification of the GSH-dependent cytosolic PGES (cPGES/p23), as shown in an accompanying paper (31). However, this enzyme is constitutively expressed in a wide variety of cells and tissues and shows preferential functional coupling with COX-1. The linkage between the three constitutive enzymes of the biosynthetic cascade (i.e. cPLA₂, COX-1, and cPGES/p23) implies that this pathway is crucial for the production of the PGE₂ required for maintenance of tissue homeostasis.

In view of the fact that PGE₂ is often produced via the COX-2-dependent pathway (1, 2, 3, 7-9, 11, 12, 16, 17, 26), we looked for another PGES that is induced by proinflammatory stimuli and shows selective coupling with COX-2. While this study was under way, Jakobsson et al. (30) reported that human microsomal GST-like 1 (MGST-L1), a member of the MAPEG (membrane-associated proteins involved in eicosanoid and GSH metabolism) superfamily (32), exhibits significant PGES activity. Moreover, the expression of this enzyme has been shown to increase after stimulation with interleukin (IL)-1 in A549 cells (30). In the present study, we show that MGST-L1 is identical to the membrane-associated PGES (mPGES), which we have originally detected in lipopolysaccharide (LPS)-stimulated macrophages (7). mPGES/MGST-L1 expression is strongly induced in several cells and tissues related to the inflammatory response in vitro and in vivo. Coexpression experiments clearly demonstrate that mPGES/MGST-L1 is preferentially linked with COX-2, promoting delayed and induced immediate PGE₂ biosynthesis. Furthermore, sustained expression of both COX-2 and mPGES/MGST-L1 leads to aberrant cell growth. Our results indicate the presence of two segregated PGE₂-biosynthetic routes, the cPLA₂-COX-1-cPGES/p23 and cPLA₂-COX-2-mPGES/MGST-L1 pathways, in the same cell.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Animals-- Harlan Sprague-Dawley rats and C57BL/6 mice were purchased from Japan SLC (Hamamatsu, Japan). Rabbits (New Zealand White, 1-kg body weight, female) were from Saitama Experimental Animal Supply (Saitama, Japan).

Materials-- The goat anti-human COX-2 and rabbit anti-human cPLA₂ antibodies were purchased from Santa Cruz. The rabbit anti-rat hematopoietic PGD₂ synthase (hPGDS) antibody was a generous gift from Dr. Y. Urade (Osaka Bioscience Institute, Osaka, Japan). cDNA probes for human COX-1, human COX-2, and mouse COX-2 were described previously (2, 3). Human cPGES/p23 cDNA was described in the accompanying paper (31). Rat thromboxane synthase (TXS) cDNA was obtained by reverse transcriptase-polymerase chain reaction (RT-PCR) using rat platelet mRNA as a template using 5'- and 3'-primers corresponding the N- and C-terminal 23-base pair nucleotide sequences. The touchdown PCR condition was 94 °C for 30 s and then 30 cycles of 94 °C for 5 s and 68 °C for 4 min using Advantage cDNA polymerase mix (CLONTECH). Superscript II RNase H-reverse transcriptase, RNase H, Taq DNA polymerase, dNTP mixture, LipofectAMINE Plus reagent, Opti-MEM, RPMI 1640 medium, and TRIzol reagent were obtained from Life Technologies, Inc.. Bacterial LPS (E. coli O111:B4), dexamethasone, fetal calf serum (FCS), GSH, and mouse anti-FLAG epitope monoclonal antibody were purchased from Sigma. Freund's complete and incomplete adjuvants, thioglycollate, and bactopeptone were from Difco Laboratories. PGH₂, rabbit anti-human COX-1, and anti-mouse COX-2 polyclonal antibodies, and the enzyme immunoassay kits for PGE₂ and TXB₂ were from Cayman Chemical. AA was purchased from NuChek Prep. Oligonucleotide primers were from Amersham Pharmacia Biotech. The plasmid pGEM-T easy was purchased from Promega. Geneticin, hygromycin, zeocin, and the mammalian expression vectors pCR3.1, pCDNA3.1/hyg(+), and pCDNA3.1/zeo(+) were from Invitrogen. A23187 was purchased from Calbiochem. Human and mouse interleukin (IL)-1s and mouse tumor nectoris factor (TNF)  were from Genzyme. Fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse IgG, FITC-rabbit anti-goat IgG, and FITC-goat anti-rabbit IgG antibodies, and horseradish peroxidase-conjugated anti-rabbit and mouse IgGs were purchased from Zymed Laboratories Inc. Cy3-conjugated donkey anti-rabbit IgG antibody was from Chemicon. Other reagents were obtained from Wako Pure Chemical Industries. Computational analysis on the isolated cDNAs and related sequences were performed using the GENETYX program (Software Development). Culture of human embryonic kidney (HEK) 293 cells (1, 2), mouse osteoblastic MC3T3-E1 cells (8) and rat calvaria osteoblasts (33) was described previously.

Preparation and Activation of Rat and Mouse Peritoneal Macrophages-- To prepare macrophages, 5% (w/v) Bactopeptone in saline (5 ml/100 g of body weight) was injected intraperitoneally into Harlan Sprague-Dawley rats and 4% thioglycollate (w/v) solution was injected intraperitoneally into C57BL/6 mice (1 ml/20 g body weight). The peritoneal exudate cells of rats and mice were collected on day 4 by washing the cavity with ice-cold Ca²⁺/Mg²⁺-free Hanks' balanced salt solution. The cells were washed twice and plated onto six-well plastic plates (Corning) at a density of 4.5 × 10⁶ cells/well in 2 ml of RPMI 1640 medium containing 10% (v/v) FCS. After 2 h of incubation at 37 ^?C in a humidified atmosphere of 5% CO₂ and 95% air, non-adherent cells were removed by rinsing. Then RPMI 1640 medium containing 10% FCS was added to the adherent cells and used as macrophages. The cells were incubated in the medium with or without 10 µg/ml LPS for up to 24 h. After incubation, PGE₂ and TXB₂ accumulated in the supernatants were measured by the enzyme immunoassay kits, and PGD₂ was quantified by high performance liquid chromatography, as described previously (26). In some experiments, the cells were incubated with LPS in the presence of 10 µM dexamethasone.

Molecular Cloning of Mouse and Rat mPGESs-- Total RNA was extracted from mouse and rat peritoneal macrophages incubated with LPS for 12 h by using TRIzol reagent. The RT reaction was carried out by using the Superscript preamplification system (Life Technologies) according to the manufacturer's instructions. RNA (1 µg) was mixed with 1 µl of 50 µg/ml random hexamer oligonucleotides and 200 units of reverse transcriptase in a total volume of 20 µl, and incubated for 50 min at 42 °C. The PCR primers 5'-ATC AAG ATG TAC GTG GTG GC-3' (sense) and 5'-GAG CTG GGC CAG GGT GTA GG-3' (antisense), designed on the basis of the reported cDNA sequence of human mPGES (MGST-L1) (30), were used for PCR amplification. PCR was performed by adding both primers (0.2 µM for each) and an appropriate amount of template DNA to 25 µl of PCR buffer (20 mM Tris-HCl, pH 8.4, 50 mM KCl, 1.5 mM MgCl₂, and 0.05% W-1) containing 0.5 units of Taq DNA polymerase and 0.2 mM dNTP. The reaction was carried out with 36 cycles of 30 s of denaturation at 94 °C, 30 s of annealing at 58 °C, and 45 s of extension at 72 °C using a DNA thermal cycler (PerkinElmer Life Sciences). The amplified DNA fragments were directly subcloned into the TA cloning vector pGEM and sequenced by an autofluorometric DNA sequencer DSQ-1000L (Shimadzu) using the Thermo Sequenase cycle sequencing kit (Amersham Pharmacia Biotech).

The cDNA fragments flanking the 3'-end regions of mouse and rat PGES cDNAs were obtained by the 3'-RACE method. The first strand cDNA from the total RNA isolated from LPS-treated macrophages was synthesized by the RT reaction as follows. After denaturation at 70 °C for 10 min, 1.2 µg of RNA was mixed with 500 nM (dT)₁₇-adaptor primer (5'-GGC CAC GCG TCG ACT AGT AC(dT)₁₇-3') and 200 units of reverse transcriptase (Life Technologies, Inc.) and incubated for 50 min at 42 °C in 20 µl of reaction mixture (20 mM Tris, pH 8.4, 50 mM KCl, 2.5 mM MgCl₂, 0.5 mM dNTP, and 10 mM dithiothreitol). After heating at 70 °C for 15 min, the reaction mixture was further incubated with 2 units of RNase H at 37 °C for 20 min. The materials obtained were used for nested PCR amplification with the adapter primer (5'-GGC CAC GCG TCG ACT AGT AC-3') and the gene-specific primers for mouse (5'-TGT CAT CAC AGG CCA GAT-3') or rat (5'-TGT CAT CAC AGG CCA AGT-3') mPGES. These gene-specific primers were designed on the basis of the nucleotide sequence data of the partial cDNA fragments obtained above. 3'-RACE-PCR was performed by 36 cycles of 30 s of denaturation at 94 °C, 30 s of annealing at 54 °C, and 45 s of extension at 72 °C. The nested PCR products were subcloned into the pGEM vector and sequenced.

5'-RACE was conducted using the 5'-RACE system version 2.0 (Life Technologies, Inc) according to the manufacturer's instructions. The 5'-gene-specific primers used for this RACE were based on the nucleotide sequence data obtained above (5'-TCG ATT AAG GCG TGG GCT-3' for mouse and 5'-GGA GCG AAT GCG GGG-3' for rat). Total RNAs (0.5 µg) from LPS-treated mouse and rat macrophages were reverse-transcribed using 5'-gene-specific primers as described above for the 3'-RACE. The first strand products were isolated using a GlassMax DNA isolation spin cartridge (Life Technologies, Inc.). A 10-µl portion of cDNA was heated at 94 °C for 2 min and incubated with 0.4 units of terminal deoxynucleotidyltransferase at 37 °C for 10 min in 25 µl of reaction mixture (10 mM Tris, pH 8.4, 25 mM KCl, 1 mM MgCl₂, and 0.2 mM dCTP). The first PCR was carried out using the 5'-gene-specific primers for mouse (5'-TTG TCT CCA TGT CGT TGC-3') or rat (5'-TCG TCT CCA TGT CGT TGC-3') mPGES and the anchor primer (5'-CUA CUA CUA CUA GGC CAC GCG TCG ACT AGT ACG GGI IGG GII GGG IIG-3') under the same amplification condition as for the 3'-RACE. Subsequently, a 1-µl aliquot of the first PCR product was subjected to the second PCR amplification using the upstream 5'-gene-specific primers (5'-ATC TGG CCT GTG ATG ACA-3' for mouse and 5'-ACT TGG CCT GTG ATG ACA-3' for rat) and the universal amplification primer (5'-CUA CUA CUA CUA GGC CAC GCG TCG ACT AGT AC-3'). The PCR products were subcloned into the pGEM vector and sequenced.

The cDNA encoding the open reading frame of human PGES was amplified by RT-PCR as described above using total RNA obtained from human umbilical vein endothelial cells as a template and the following oligonucleotide primers designed from the human mPGES (MGST-L1) cDNA sequence (30): sense, 5'-ATG CCT GCC CAC AGC CTG-3'; and antisense, 5'-TCA CAG GTG GCG GGC CGC-3'. To obtain the C-terminally FLAG-epitope-tagged human mPGES cDNA, PCR was conducted using the sense primer (see above) and the antisense primer 5'-TCA CTT GTC ATC GTC GTC CTT GTA GTC CAG GTG GCG GGC CGC TTC-3' (the underlined sequence corresponds to the FLAG epitope). The amplified product was subcloned into pCR3.1 and sequenced.

Site-directed Mutagenesis-- Site-specific mutations were introduced by mismatched primer PCR reactions with Advantage cDNA polymerase mix using human mPGES cDNA as a template, as described previously (34). In order to obtain the R110S mutant, a product obtained from the PCR using the mutated sense primer 5'-CTC GTG GGC AGT GTG GCA CAC-3', in which Arg¹¹⁰ was replaced by Ser at the underlined site, and the C terminus antisense primer that was tagged with the FLAG epitope (underlined) 5'-TCA CTT GTC ATC GTC GTC CTT GTA GTC CAG GTG GCG GGC CGC T-3' (C-FLAG primer) was mixed with another product obtained from the PCR using the N terminus sense primer 5'-ATG CCT GCC CAC AGC CTG-3' (N-primer) and the mutated antisense primer 5'-GTG TGC CAC ACT GCC CAC GAG-3'. After annealing, the second PCR was carried out using the N-primer and C-FLAG primer. The mutants R70S (Arg⁷⁰ replaced by Ser) and Y117F (Tyr¹¹⁷ replaced by Phe) were prepared by using the same strategy. The sequences of the mutated primers for them were as follows: sense 5'-CGC TGC CTC AGC GCC CAC CGG-3' and antisense 5'-CCG GTG GGC GCT GAG GCA GCG-3' for R70S and sense 5'-ACC GTG GCC TTC CTG GGG AAG-3' and antisense 5'-CTT CCC CAG GAA GGC CAC GGT-3' for Y117F. Each PCR product was ligated into the pCR3.1 vector and sequenced.

Expression of mPGES in 293 Cells-- The cDNAs flanking the entire open reading frames of mouse and rat mPGESs were amplified by PCR using the Expand high fidelity PCR system (Roche Molecular Biochemicals) with cDNA reverse-transcribed from total RNAs obtained from mouse and rat peritoneal macrophages as templates and the appropriate combinations of the following oligonucleotide primers: mouse sense primer 5'-ATG CCT TCC CCG GGC CTG-3', rat sense primer 5-ATG ACT TCC CTG GGT TTG-3', and mouse and rat antisense primer 5'-TCA CAG ATG GTG GGC CAC-3'. The amplified cDNAs were directly subcloned into the mammalian expression vector pCR3.1 and sequenced.

Human, mouse, and rat mPGES cDNAs subcloned into PCR3.1 were each transfected into 293 cells using LipofectAMINE Plus according to the manufacturer's instruction. Briefly, 1 µg of each plasmid was mixed with 4 µl of LipofectAMINE and 6 µl of Plus reagent in 200 µl of Opti-MEM, left for 15 min, and then added to cells that had attained 70% confluence in six-well plates (Corning) in 1 ml of Opti-MEM. After incubation for 4 h, 2 ml of fresh culture medium was added. After 18 h, the medium was replaced with 2 ml of fresh medium, and culture was continued for 3 days. In order to establish stable transfectants, cells transfected with each cDNA were cloned by limiting dilution in 96-well plates in culture medium supplemented with 800 µg/ml Geneticin. After culture for 2 weeks, wells containing a single colony were chosen, and the expression of PGES was assessed by PGES enzymatic activity and Northern blotting, as described below.

Cotransfection of mPGES and COXs in 293 Cells-- Establishment of 293 cells stably overexpressing human COX-1 or COX-2 has been described previously (2). These cells were transiently transfected with mouse or rat mPGES cDNAs in pCR3.1 together with the marker plasmid pGL-1 (Life Technologies, Inc.), which contains cDNA for green fluorescent protein, using LipofectAMINE Plus. Briefly, the plasmid containing mPGES and pGL-1 (0.5 µg for each) was mixed with 6 µl of LipofectAMINE and 10 µl of Plus reagent in 200 µl of Opti-MEM, left for 15 min, and then added to cells that had attained 70% confluence in six-well plates in 1 ml of Opti-MEM. After incubation for 2 h, 2 ml of fresh culture medium was added. After 16 h, the medium was replaced with 2 ml of fresh medium. After 36 h, the cells were used for the experiments. Green fluorescent protein fluorescence was measured in cell lysates using fluorometer at excitation wavelength of 475 nm.

In order to establish 293 transfectants stably coexpressing human mPGES and either of the COX isozymes, cells expressing each COX were subjected to the second transfection with human mPGES cDNA subcloned into pCDNA3.1/hyg(+). After selection in culture medium containing 50 µg/ml hygromycin in 96-well plates, single colonies were picked up and expanded. Expression of mPGES and each COX was assessed by Western and Northern blotting, as described below.

Activation of 293 Cells-- All procedures were described in our previous reports (1, 2). Briefly, 293 cells (5 × 10⁴/ml) were seeded into each well of 24- or 48-well plates in 1 and 0.5 ml of culture medium, respectively. After culture for 4 days, the cells were washed once with culture medium and then incubated with 250 µl (24-well plate) or 100 µl (48-well plate) of various concentrations of AA or 10 µM A23187 in medium containing 1% FCS for 30 min or 1 ng/ml IL-1 in medium containing 10% FCS for 4 h. The supernatants were subjected to the PGE₂ enzyme immunoassay.

Preparation of Antibody against mPGES Peptide-- A synthetic peptide (CRSDPDVERCLRAHRN, which corresponds to human mPGES 59-74) was covalently conjugated with poly-L-lysine-coated beads, a 500-µl portion of which (100-µg peptide equivalent) was mixed with an equal volume of Freund's complete adjuvant and injected into rabbits. Immunization was repeated every 3 weeks with the same amounts of the antigen mixed with an equal volume of Freund's incomplete adjuvant. Serum titers were checked by Western blotting using a lysate of 293 cells stably overexpressing human mPGES.

Western Blotting-- The cells were lysed in phosphate-buffered saline (PBS) containing 0.1% sodium dodecyl sulfate (SDS) at 1 × 10⁷ cells/ml, applied to SDS-polyacrylamide gels (10% for COXs and 15% for mPGES and hPGDS), and electrophoresed as reported previously (34). Then proteins were electroblotted onto nitrocellulose membranes with a semidry blotter (MilliBlot-SDE system; Millipore). The membranes were blocked for 1 h in 10 mM Tris-HCl, pH 7.4, containing 150 mM NaCl, 0.1% Tween 20 (TBS-T), and 3% skim milk. After washing the membranes with TBS-T, antibodies against cPLA₂, COX-1, COX-2, hPGDS, and mPGES were added at a 1:5,000, 1:10,000, 1:5,000, 1:5,000, and 1:1,000 dilution, respectively, in TBS-T and incubated for 2 h. After washing the membranes three times with TBS-T, horseradish peroxidase-conjugated goat anti-rabbit IgG (for cPLA₂, COX-1, hPGDS, and mPGES) or rabbit anti-goat IgG (for COX-2) was added at a 1:5,000 dilution in TBS-T and incubated for 1 h. After six washes with TBS-T, protein bands were visualized with enhanced chemiluminescence Western blotting detection reagents (Amersham Pharmacia Biotech).

RNA Blotting-- Approximately equal amounts (~10 µg) of the total RNAs obtained from the cells were applied to separate lanes of 1.2% (w/v) formaldehyde-agarose gels, electrophoresed, and transferred to Immobilon-N membranes (Millipore). The resulting blots were then probed with the respective cDNA probes that had been labeled with [³²P]dCTP (Amersham Pharmacia Biotech) by random priming (Takara Biomedicals). All hybridizations were carried out as described previously (34).

Assay of Enzymatic Activity of PGES-- PGES activities in cell lysates were measured by assessment of conversion of PGH₂ to PGE₂ as reported previously (7). The cells were scraped off from the dishes and disrupted by sonication (10 s, three times, 1-min interval) in 400 µl of 10 mM Tris-HCl, pH 8.0. After centrifugation of the sonicates at 1,700 × g for 10 min at 4 °C, the supernatants were used as the enzyme source. An aliquot of each lysate (100-µg protein equivalents) was incubated with 2 µg of PGH₂ for 30 s at 24 °C in 0.1 ml of 1 M Tris-HCl, pH 8.0, containing 2 mM GSH. After terminating the reaction by the addition of 100 mM FeCl₂, PGE₂ contents in the supernatants were quantified by use of the enzyme immunoassay kit.

Immunofluorescent Microscopic Analysis-- 293 cells coexpressing human mPGES-FLAG and either of the two COX isozymes were seeded onto collagen-coated cover glasses (Iwaki Glass) at 5 × 10⁴ cells/ml and cultured for 2 days. After removing the supernatants, the cells were fixed with 2% (w/v) paraformaldehyde in PBS for 30 min at 4 °C. The cells were then treated sequentially at room temperature with 1% (w/v) bovine serum albumin containing 1% (w/v) saponin for 30 min in PBS to block nonspecific binding and to permeabilize the membranes, appropriate first antibodies against FLAG epitope, COX-1 and COX-2 (1:500 dilution for each) in PBS containing 1% albumin for 2 h, and FITC- and/or Cy3-conjugated second antibodies (1:100 dilution for each) in PBS containing 1% albumin for 1 h. The coverslips were mounted on glass slides using Perma Fluor (Japan Tanner) and examined using a Fluoview laser fluorescence microscope (Olympus).

Statistical Analysis-- Data were analyzed by Student's t test. Results are expressed as the mean ± S.E., with p = 0.05 as the limit of significance.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Molecular Cloning of Mouse and Rat mPGESs-- We have previously demonstrated that, when rat peritoneal macrophages were cultured with LPS, COX-2-dependent generation of PGE₂, but not of TXB₂ and PGD₂, from endogenous AA proceeded over 3-12 h, accompanied by increase in PGES activity (7, 26). The inducible PGES activity in LPS-stimulated macrophages was tightly associated with membrane, whereas PGES activity in the cytosol was unchanged (Fig. 1A). The membrane-associated activity showed strict dependence on GSH and, unlike cPGES/p23 (31), was almost insensitive to 1-chloro-2,4-dinitrobenzene (Fig. 1B). Increase in PGES activity was also observed in LPS-stimulated mouse macrophages (data not shown) and IL-1-stimulated mouse osteoblastic MC3T3-E1 cells (Fig. 1C), accompanied by delayed PGE₂ generation.

View larger version (28K): [in this window] [in a new window]   Fig. 1.   Detection of inducible mPGES. A, PGES activity in the cytosol and membrane fractions of rat macrophages after incubation with or without 10 µg/ml LPS for 12 h. B, PGES activity in lysates of rat macrophages treated for 12 h with LPS was measured in the presence or absence of 1 mM GSH and 1 mM chloro-2,4-dinitrobenzene (CDNB). C, changes in the activity of PGES in the lysates (left) and accumulation of PGE2 in the culture supernatants (right) of MC3T3-E1 osteoblasts after culture for the indicated periods with 5 ng/ml IL-1. A representative result of three independent experiments is shown.

Because of the stimulus inducibility and GSH dependence, membrane-associated PGES detected in macrophages and osteoblasts are likely to be orthologs of human MGST-L1, which has been shown to exhibit PGES activity (30). In order to confirm this, we attempted to clone mouse and rat MGST-L1 ortholog cDNAs from LPS-stimulated macrophages. RT-PCR was performed using mRNA prepared from LPS-stimulated mouse and rat macrophages as templates with several sets of primers designed on the basis of the reported sequence of human MGST-L1 (30). Using one set of the primers described under "Experimental Procedures," a 280-base pair fragment, which showed a high degree of sequence identity to the corresponding portion of human MGST-L1 (84% for mouse and 82% for rat), was amplified from both species. Based on these sequences, 5'- and 3'-RACE PCRs were carried out to obtain the full-length cDNAs (see "Experimental Procedures"). The deduced mouse (GenBank^TM accession no. AB041997) and rat (GenBank^TM accession no. AB041998) cDNAs had 820 and 710 nucleotides, respectively, each of which contained an open reading frame coding for the predicted protein of 153 amino acids (Fig. 2). Since the proteins encoded by these cDNAs were functionally active (see below), we hereafter designate them as mouse and rat mPGESs (m stands for membrane-bound).

View larger version (73K): [in this window] [in a new window]   Fig. 2.   Amino acid sequences of mPGES. Alignment of amino acid sequences of mouse, rat, and human mPGES is shown. Conserved amino acids are boxed.

Alignment of the mPGES proteins of the three species revealed a high degree of sequence homology; mouse and rat mPGES proteins showed 79 and 80% sequence identity to human mPGES, respectively, and mouse and rat mPGESs were highly homologous (94%) (Fig. 2). A search of the DNA and protein data bases revealed up to 38% homology between the cloned mouse and rat mPGESs and the family of MGSTs, as well as several EST genes that had been obtained from the human genome project or cDNA libraries from specific tissues and tumors. An EST gene (GenBank^TM accession no. AA178132) contained an entire open reading frame for mouse mPGES, although no function was described for this gene product.

Enzymatic Characterization of mPGES-- To ascertain that the cDNA clones thus obtained encode functional mouse and rat mPGES proteins, their cDNAs, as well as human mPGES cDNA, were subcloned into a mammalian expression vector and transfected into 293 cells. PGES activities in the lysates of cells transfected with mouse (Fig. 3A), rat (Fig. 3B), and human (Fig. 3C) cDNAs were increased markedly in the presence of GSH compared with parental cells. This activity was tightly associated with the membrane fraction (data not shown), and was insensitive to 1-chloro-2,4-dinitrobenzene (Fig. 3C). No other PGs were formed under the same conditions (data not shown).

View larger version (13K): [in this window] [in a new window]   Fig. 3.   Enzymatic characterization of recombinant mPGES. cDNAs for mouse (A), rat (B), and human (C) mPGES subcloned into pCR3.1 and empty vector were transfected into HEK293 cells. After 3 days, PGES activity in cell lysates was measured in the presence of 2 mM GSH (A and B) or in the presence or absence of 2 mM GSH and 1 mM chloro-2,4-dinitrobenzene (CDNB) (C).

Alignment of the known MAPEG family members, which are subdivided into four classes (32), demonstrates that Arg¹¹⁰ in mPGES is strictly conserved (Fig. 4A). Replacement Arg¹¹⁰ by Ser abrogated the catalytic function of mPGES (Fig. 4B), implying an essential role of this residue. Arg⁵¹ in LTCS is crucial for catalytic function, being presumed to open the epoxide ring of leukotriene A₄ for the conjugation with the thiolate anion of the reduced GSH (36). This speculation is supported by the fact that 5-lipoxygenase-activating protein FLAP, which is most similar to LTCS in overall structures but does not exhibit enzymatic function (37), lacks Arg in this position (Fig. 4A). However, replacement of Arg⁷⁰ (corresponding to Arg⁵¹ in LTCS (Fig. 4A)) by Ser did not change the activity of mPGES, suggesting that the catalytic mechanisms between the class I and II MAPEG enzymes are distinct.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Site-directed mutagenesis of human mPGES. A, alignment of amino acid sequences of two conserved regions of the MAPEG family members, which are subdivided into four subgroups (I-IV). Residues well conserved among the members are indicated by bold letters. Asterisks show the residues replaced by site-directed mutagenesis. B, catalytic activity of mPGES mutants. Wild-type (WT) mPGES and its mutants R70S, R110S, and Y117F were transfected into HEK293 cells and PGES activity in cell lysates was measured. Expression levels of mPGES-WT and mutants were comparable as assessed by immunoblotting (data not shown).

The Tyr residue is essential for catalytic activity of cytosolic GSTs (38), hPGDS (belonging to the  class of cytosolic GST) (39), and cPGES/p23 (31), where a Tyr residue near the N terminus acts as the GSH binding site. Tyr⁹³ in LTCS plays such a role (36), yet this residue is not present in the corresponding position of the class I MAPEG enzymes. Since Tyr¹¹⁷ in mPGES is conserved in all known members of the mammalian MAPEG enzymes, we expected that this Tyr might be crucial. However, replacement of Tyr¹¹⁷ by Phe did not alter the catalytic activity of mPGES (Fig. 4B). Consistently, this Tyr residue is dispensable for the function of MGST-1 (40) and does not exist in the class IV bacterial MGSTs (Fig. 4A) (32). Moreover, all Tyr to Phe substitutions in MGST-1 results in mutants with activities similar to that of the native enzyme (40), suggesting that the class I MAPEG enzymes may display an alternate stabilization of the thiolate anion of GSH other than through interaction with the phenolic hydroxyl group of Tyr residue.

Induction of mPGES Expression-- RNA blot analysis demonstrated that both mPGES and COX-2 transcripts in rat macrophages were barely detectable before culture, were markedly induced during 3-12 h, and then declined by 24 h following LPS stimulation (Fig. 5A). Immunoblotting using an antibody raised against a human mPGES-derived synthetic peptide showed that the kinetic change in mPGES protein expression correlated with that of mPGES mRNA (Fig. 5A) and that of PGES enzymatic activity in cell lysates (Fig. 5B). The levels of TXS, as assessed by Northern blotting, and hPGDS, as assessed by Western blotting, were almost constant during culture with LPS (Fig. 5A), consistent with the fact that the production of TXB₂ and PGD₂ was changed only minimally (7, 26). mPGES mRNA was also detected in mouse macrophages (data not shown) and rat calvaria osteoblasts (Fig. 5C), in which mPGES was markedly up-regulated by proinflammatory stimuli. More sensitive RT-PCR analysis revealed that mPGES mRNA was expressed in various cell types, where it also underwent stimulus-dependent induction (data not shown). Although the expression of mPGES mRNA was barely detectable in normal rat tissues as assessed by RNA blotting, it increased markedly in brain, lung, spleen, stomach, kidney, and testis 6-12 h after LPS administration, and then declined (Fig. 5D).

View larger version (57K): [in this window] [in a new window]   Fig. 5.   Induction of mPGES expression by proinflammatory stimuli. A, expression of mRNAs for COX-2, mPGES, cPGES/p23 and TXS, as assessed by RNA blotting, and proteins for mPGES and hPGDS, as assessed by immunoblotting, in rat peritoneal macrophages after stimulation with LPS for the indicated periods. Ribosomal RNA (rRNA) was visualized by ethidium bromide staining. B, kinetic changes in PGES activity in macrophages with (close circles) or without (open circles) LPS treatment. C, expression of mPGES and cPGES mRNAs in rat calvaria osteoblasts after 24 h of culture with or without 1 ng/ml IL-1 and 100 units/ml TNF. D, expression of mPGES mRNA in various tissues of rats before and after LPS injection was assessed by RNA blotting. A representative result of three independent experiments is shown.

Effect of Dexamethasone-- The increase of PGE₂ generation from endogenous AA (Fig. 6A) and that of PGES activities (Fig. 6B) in rat macrophages, assessed 12 h after LPS stimulation, were suppressed almost completely by dexamethasone. Dexamethasone not only suppressed the induced expression of COX-2 mRNA (Fig. 6C), as demonstrated by several investigators (3, 10, 41-43), it also markedly reduced that of mPGES mRNA (Fig. 6C). The effects of dexamethasone on COX-2 and mPGES expression were already evident 3-6 h after the addition of LPS (data not shown). Expression of cPGES/p23, which was constant during LPS stimulation (Fig. 5A), was not affected by dexamethasone (Fig. 6C). Similarly, dexamethasone markedly suppressed the induced expression of mPGES in mouse macrophages and osteoblasts (data not shown).

View larger version (21K): [in this window] [in a new window]   Fig. 6.   Effect of dexamethasone (Dex) on mPGES expression in rat macrophages. Effect of 10 µM Dex on accumulation of PGE2 in the supernatants (A), PGES activity in cell lysates (B), and expression of mRNAs for cPGES/p23, mPGES and COX-2 (C) after 12 h of culture with LPS. The means ± S.E. (A, B) and a representative result (C) of three independent experiments are shown.

Functional Coupling between mPGES and COXs-- That coordinate induction of COX-2 and mPGES was accompanied by accumulation of PGE₂ in LPS-stimulated macrophages (see above) argues that both enzymes are functionally linked. To explore this, we cotransfected mPGES and either of the two COX isozymes into HEK293 cells and their functional coupling was reconstituted.

Cotransfection of human mPGES and COX-2, the expression of which was confirmed by RNA blotting and immunoblotting (Fig. 7A), into HEK293 cells resulted in a dramatic increase in the conversion of exogenous AA to PGE₂, which reached a maximal level at 2-5 µM AA and then a plateau (Fig. 7B). An increase in PGE₂ production was also observed in cells cotransfected with human mPGES and COX-1, yet their coupling became apparent only when a high concentration (10 µM) of AA was added (Fig. 7B).

View larger version (35K): [in this window] [in a new window]   Fig. 7.   Functional coupling between COXs and mPGES. A, the expression levels of COX-1, COX-2, and human mPGES in HEK293 cells stably transfected with their cDNAs alone or in combination were assessed by RNA blotting and immunoblotting. B, conversion of exogenous AA to PGE2 by HEK293 transfectants. Control cells (open squares), cells expressing either COX alone (open circles), and cells coexpressing human mPGES and either COX (filled circles) were incubated for 30 min with the indicated concentrations of AA. C, the transfectants shown in panel A were stimulated for 30 min with 10 µM A23187 (left) or for 4 h with 1 ng/ml human IL-1 (right). D, the expression levels of cPLA2, COX-1, COX-2 and human mPGES in HEK293 stable transfectants were assessed by RNA blotting (mPGES) and immunoblotting (cPLA2 and COXs). E, the transfectants shown in panel D were stimulated for 30 min with A23187 (left) or for 4 h with IL-1 (right). F, conversion of exogenous AA to PGE2 by HEK293 cells stably transfected with either COX alone (open circles) and those with both mouse mPGES and either COX (filled circles). A representative result (B, C, and F) and the means ± S.E. (E) of more than three experiments are shown.

When these cells were stimulated for 30 min with A23187 (immediate response) or for 4 h with IL-1 (delayed response), a marked increase in PGE₂ generation was observed in COX-2/mPGES-coexpressing cells, whereas the increase in PGE₂ generation by COX-1/mPGES-coexpressing cells was only modest, albeit significant (Fig. 7C). Supplying more endogenous AA by further transfection of cPLA₂, the expression of which is shown in Fig. 7D, led to a marked increase in A23187-induced production of PGE₂ via COX-1 and mPGES (Fig. 7E), consistent with the fact that COX-1/mPGES coupling requires a high concentration of AA (Fig. 7B). In contrast to the A23187-dependent event, IL-1-dependent PGE₂ generation was increased only slightly in cPLA₂/COX-1/mPGES triple transfectants (Fig. 7E), probably because the amount of AA gradually released by cPLA₂ during the IL-1-initiated response is still insufficient to go through the COX-1/mPGES pathway. The cPLA₂/COX-2/mPGES triple transfectants produced more PGE₂ than did the double transfectants expressing cPLA₂/COX-2 or COX-2/mPGES in both the A23187- and IL-1-dependent responses (Fig. 7E).

COX-2 preference was not restricted to human mPGES, but was also observed with its mouse (Fig. 7F) and rat (data not shown) counterparts. Thus, 293 cells cotransfected with mouse mPGES and COX-2 produced a large amount of PGE₂ from exogenous AA, which was evident at 5 µM AA (Fig. 7F). In contrast, PGE₂ generation by cells coexpressing COX-1 and mouse mPGES was not increased significantly at 5 µM AA, and elevated only modestly at 10 µM (Fig. 7F). The expression levels of mPGES in both cells, assessed by PGES activity in cell lysates, were comparable (data not shown).

Subcellular Distribution-- In order to determine the subcellular localization of mPGES, confocal immunofluorescent microscopic analysis was performed. COX-1- or COX-2-expressing HEK293 cells were transiently transfected with the C-terminally FLAG-tagged human mPGES, and mPGES and each COX were visualized by anti-FLAG and anti-COX-1 or -COX-2 antibodies. As shown in Fig. 8, mPGES colocalized with the perinuclear COX-1 and COX-2.

View larger version (64K): [in this window] [in a new window]   Fig. 8.   Confocal microscopic analyses of subcellular localization of mPGES and either of the COX isozymes. HEK293 cells stably expressing COX-1 or COX-2 were transfected with C-terminally FLAG-tagged mPGES. After 3 days, the cells were subjected to indirect immunofluorescent microscopy as described under "Experimental Procedures." The yellow areas indicate the regions where COXs and mPGES colocalized.

Combination of COX-2 and mPGES Affects Cellular Growth and Morphology-- We unexpectedly found that 293 transfectants stably coexpressing COX-2 and mPGES grew faster than those expressing the single enzymes alone, those coexpressing COX-1 and mPGES, and the control cells (Fig. 9A, top). The rapid proliferation of the COX-2/mPGES cotransfectants was accompanied by accumulation of a large amount of PGE₂ in the culture supernatants (Fig. 9A, bottom), and correlated with the expression levels of COX-2 and mPGES (Fig. 9A, bottom). Moreover, the COX-2/mPGES stable cotransfectants were highly aggregated, piled up, and exhibited aberrant round-shape morphology (Fig. 9B, right), in contrast to those coexpressing COX-1 and mPGES, which, like control cells or cells coexpressing cPGES/p23 and COX-2 (data not shown), showed a typical fibroblastic shape (Fig. 9B, left).

View larger version (40K): [in this window] [in a new window]   Fig. 9.   Coexpression of COX-2 and mPGES affects cell proliferation and morphology. A, several independent HEK293 transfectants stably expressing with COX-1, COX-2, and human mPGES, alone or in combination, were cultured for 4 days, and cell number (upper panel) and PGE2 accumulated in the supernatants (lower panel) were assessed. Expression levels of each enzyme, assessed by Western and Northern blotting, is shown in the bottom. B, cellular morphology of the transfectants stably expressing COX-1/mPGES (left) and COX-2/mPGES (right).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Our current studies have suggested that the two COX isozymes, which exert segregated functions in different phases of the PGE₂-biosynthetic response, are functionally coupled with the two distinct downstream PGES enzymes (Fig. 10). cPGES/p23, a constitutive enzyme expressed in a wide variety of cells and tissues, is predominantly linked with COX-1 to promote the immediate response, during which relatively high concentrations of AA are released in a short period (31). Coupling with COX-1 indicates that the physiological role of cPGES/p23 may be to produce the PGE₂ required for the maintenance of homeostasis. mPGES, an inducible perinuclear enzyme, is preferentially coupled with the inducible COX-2 to promote delayed PGE₂ generation and, if COX-2 already exists in cells, also regulates immediate PGE₂ generation (likely to be a reflection of the priming response). The striking induction of mPGES in tissues and cells related to the inflammatory response suggests its crucial role in the process of inflammation, as well as in other chronic diseases in which COX-2 is involved. mPGES may also participate in fever and osteogenesis, where PGE₂ acts as an endogenous mediator (44, 45). Moreover, our present study has raised the intriguing possibility that mPGES, in concert with COX-2, is involved in cellular transformation.

View larger version (53K): [in this window] [in a new window]   Fig. 10.   A schematic model for functional coupling between the two COXs and two PGESs. In the immediate response elicited by Ca2+ agonists, AA rapidly released by cPLA2 is metabolized to PGE2 via the constitutive enzymes COX-1 and cPGES/p23. In the delayed response induced by proinflammatory stimuli, AA gradually released by cPLA2 is metabolized to PGE2 via the two inducible enzymes COX-2 and mPGES. mPGES is capable of producing PGE2 via COX-1 only when explosive activation of cPLA2 occurs.

Mouse, rat, and human mPGESs show a high degree of sequence homology to one another, and at least one residue, Arg¹¹⁰, which is conserved in the MGST family members (32), is essential for catalytic function. They exhibit a GSH-dependent catalytic activity, which is a common feature of this enzyme family (32). Coexpression experiments clearly demonstrated that mPGES is functionally coupled with COX-2 in marked preference to COX-1 irrespective of the source of AA (i.e. exogenous and endogenous) (Fig. 7). Coupling between COX-1 and mPGES occurs only when a large amount of AA is supplied exogenously, or if burst activation of cPLA₂ takes place, endogenously. The mechanisms whereby mPGES favors COX-2 over COX-1 cannot be explained simply by their subcellular localizations (Fig. 8), although the presence of microdomains, in which mPGES is located in closer proximity to COX-2 than to COX-1, within the perinuclear compartment cannot be ruled out. Nevertheless, since subcellular location of mPGES was examined only by overexpression experiments, it would be necessary to examine the localization of endogenous enzyme in a future study.

Of importance, mPGES is an inducible enzyme, the expression of which is markedly increased in various cells and tissues following proinflammatory stimuli (Fig. 5). The selective increase in PGE₂ relative to TXB₂ and PGD₂ in LPS-stimulated rat macrophages (7, 26) is probably because expression of mPGES, but not TXS and hPGDS, is up-regulated by LPS. The observations that the kinetic change in PGES activity was well correlated with that in mPGES mRNA and protein expression (Fig. 5) and that dexamethasone blunted both PGES activity and mPGES expression almost completely (Fig. 6) imply that most of the PGES activity detected in LPS-stimulated macrophages is ascribed to mPGES. Thus, preferential coupling between the two inducible biosynthetic enzymes, COX-2 and mPGES, represents a well controlled cellular system that facilitates ongoing PGE₂ biosynthesis in synergy. Furthermore, striking dexamethasone sensitivity of mPGES implies that all three enzymatic steps in the delayed PGE₂-biosynthetic pathway (PLA₂s (Refs. 46 and 47), COX-2 (Refs. 3, 10, and 41-43), and mPGES) are potential targets for the anti-inflammatory actions of glucocorticoids.

A number of pharmacological, biochemical, and genetic studies have documented the involvement of COX-2 in the development of cancer (19-22, 48-53). Non-steroidal anti-inflammatory drugs reduce the size and number of colorectal tumors (49-51). This action of these drugs has been reported to result from COX-2-dependent (21) and -independent (54, 55) mechanisms. In support of the former, COX-2 expression is elevated in tumors (52, 53) and inactivation of the COX-2 gene in mice markedly reduces the development of intestinal tumorigenesis (21). In the present study, we have shown that mPGES, when combined with COX-2, has the potential to affect cellular proliferation and morphology. This remarkable aspect was observed only in 293 cells stably, but not transiently, cotransfected with COX-2 and mPGES. In our preliminary studies, however, addition of indomethacin to the culture of COX-2/mPGES stable cotransfectants did not reverse their proliferation and morphology and addition of PGE₂ to parental 293 cells for at least 2 weeks did not affect their growth. We therefore speculate that sustained production of abnormal amounts of PGE₂ or some other additional metabolites via the COX-2-mPGES pathway in the restricted intracellular compartment over a long period may affect cellular states of proliferation and differentiation. In order to confirm the possible involvement of mPGES in cancer, several criteria on cellular transformation, such as growth factor-independent growth, loss of contact inhibition, anchorage-independent growth, and tumor formation when transplanted in nude mice are now under investigation.

Nevertheless, our data provide strong support for the hypothesis that functional segregation of the two COXs could be regulated by at least two steps. In agreement with previous reports (2, 14, 15), COX-1 requires higher concentrations of AA for its optimal function than does COX-2, implying that the amount of AA supplied by cPLA₂ critically influences which COX isozymes are utilized. Our preliminary study has shown that several inducible sPLA₂ isozymes are also capable of supplying AA to the COX-2/mPGES pathway through the transcellular route.² Moreover, each COX is selectively coupled with a distinct PGES that shows distinct tissue and subcellular distribution and transcriptional regulation. The factors that define the coupling selectivity between each COX and PGES need to be elucidated. Finally, the two PGESs will constitute novel targets for therapeutic and prophylactic drugs for inflammation, osteoporosis, and possibly cancer.

	FOOTNOTES

^* This work was supported by grants-in-aid for scientific research from the Ministry of Education, Science and Culture of Japan and special coordination funds for promoting science and technology from the Science and Technology Agency.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in this paper has been submitted to the GenBank^TM/EMBL Data Bank with accession number(s) AB041997 and AB041998.

^§ These authors contributed equally to this work.

To whom correspondence should be addressed. Tel.: 81-3-3784-8196; Fax: 81-3-3784-8245; E-mail: kudo@pharm.showa-u.ac.jp.

Published, JBC Papers in Press, June 26, 2000, DOI 10.1074/jbc.M003505200

² M. Murakami and I. Kudo, unpublished observation.

	ABBREVIATIONS

The abbreviations used are: PG, prostaglandin; PGES, prostaglandin E₂ synthase; mPGES, membrane-bound prostaglandin E₂ synthase; cPGES, cytosolic prostaglandin E₂ synthase; COX, cyclooxygenase; cPLA₂, cytosolic phospholipase A₂; TXS, thromboxane synthase; hPGDS, hematopoietic prostaglandin D₂ synthase; LTCS, leukotriene C₄ synthase; LPS, lipopolysaccharide; GSH, glutathione; GST, glutathione S-transferase; MGST, microsomal GST; MGST-L1, microsomal glutathione S-transferase-like 1; MAPEG, membrane-associated proteins involved in eicosanoid and GSH metabolism; AA, arachidonic acid; PBS, phosphate-buffered saline; IL, interleukin; TNF, tumor necrosis factor; FCS, fetal calf serum; HEK, human embryonic kidney; FITC, fluorescein isothiocyanate; RT, reverse transcriptase; PCR, polymerase chain reaction; RACE, rapid amplification of cDNA ends; TBS-T, Tris-buffered saline plus Tween 20.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES