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

J. Biol. Chem., Vol. 277, Issue 32, 28601-28608, August 9, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/32/28601    most recent M203618200v1 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 Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Naraba, H. Articles by Tanabe, T. Search for Related Content PubMed PubMed Citation Articles by Naraba, H. Articles by Tanabe, T. Social Bookmarking                      What's this?

Transcriptional Regulation of the Membrane-associated Prostaglandin E₂ Synthase Gene

ESSENTIAL ROLE OF THE TRANSCRIPTION FACTOR Egr-1^*

Hiroaki Naraba, Chieko Yokoyama, Naomi Tago, Makoto Murakami^§, Ichiro Kudo^§, Mai Fueki^¶, Sachiko Oh-ishi^¶, and Tadashi Tanabe^**

From the  Department of Pharmacology, National Cardiovascular Center Research Institute, 5-7-1 Fujishiro-dai, Suita, Osaka 565-8565, Japan, the ^§ Department of Health Chemistry, School of Pharmaceutical Sciences, Showa University, 1-5-8 Hatanodai, Shinagawa-ku, Tokyo 142-8555, Japan, and the ^¶ Department of Pharmacology, School of Pharmaceutical Sciences, Kitasato University, Shirokane 5-9-1, Minato-ku, Tokyo 108-8641, Japan

Received for publication, April 15, 2002, and in revised form, May 23, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Membrane-associated prostaglandin (PG) E₂ synthase (mPGES) is an inducible terminal enzyme in the biosynthetic pathway for prostaglandin E₂, which participates in many biological processes. In this study, we investigated the molecular mechanism controlling the inducible expression of mPGES. The mouse mPGES gene consisted of three exons, and its 5'-proximal promoter contained consensus motifs for the binding of several transcription factors. Transgenic expression in mice of the mouse mPGES promoter flanked by a reporter gene resulted in stimulus-dependent induction of the reporter in tissues where mPGES was intrinsically induced. Deletion and site-specific mutation analyses of the 5'-flanking region demonstrated that stimulus-inducible expression of mouse and human mPGES required tandem GC boxes adjacent to the initiation site. The stimulus-induced GC box binding activity was present in nuclear extracts of cells, in which the proximal GC box was essential for binding. An 80-kDa stimulus-inducible nuclear protein that bound to this GC box was identified as the transcription factor Egr-1 (for early growth response-1). These results suggest that Egr-1 is a key transcription factor in regulating the inducible expression of mPGES.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Prostaglandin E₂ (PGE₂)¹ is an important mediator that plays a variety of roles in biological events, such as fever, pain, tumorigenesis, gastrointestinal protection, osteogenesis, and inflammation (1-5). PGE₂ production from arachidonic acid, which is released by phospholipase A₂ from membrane glycerophospholipids (6), is controlled by two rate-limiting steps. The first step is catalyzed by cyclooxygenase (COX), which converts arachidonic acid to the intermediate prostanoid PGH₂. Two isoforms of the COX enzyme have been identified: cyclooxygenase-1 (COX-1), which is a constitutive enzyme, and cyclooxygenase-2 (COX-2), which is induced by various stimuli, including cytokines, growth factors, lipopolysaccharide (LPS), and tumor promoters such as 12-O-tetradecanoylphorbol-13-acetate (TPA) (7). The second step is the terminal conversion reaction of PGH₂ to PGE₂, which is catalyzed by PGE₂ synthase (PGES). At least two forms of PGES have recently been identified (8-13). Cytosolic PGES, which is identical to the heat shock protein 90-associated protein p23, is expressed ubiquitously in a wide variety of cells and tissues and promotes COX-1-mediated immediate PGE₂ production (8). Membrane-associated PGES (mPGES), which was originally designated MGST1-L1 (for membrane-bound GST1-like-1), is an inducible enzyme, which is coordinately induced with COX-2 on the perinuclear membrane and is functionally coupled with COX-2 in marked preference to COX-1 (9-13).

It is important to understand the molecular mechanisms underlying the inducible expression of both the COX-2 and mPGES genes. A number of studies have reported the transcriptional regulation of COX-2 expression, in which several consensus sites in the COX-2 promoter, such as NF-B, NF-interleukin 6, CRE, and E-box, have been identified as regulatory sequences for COX-2 induction in response to various stimuli (14-18). Although the structure of the human mPGES gene, including a 632-bp 5'-flanking region, has been reported (19), transcription factors or cis elements required for gene expression of mPGES remain elusive.

In the present study, we have isolated the mouse mPGES gene with its promoter region and analyzed the regulatory mechanisms for inducible expression of the mPGES gene in mouse MC3T3-E1 or RAW264.7 cells. Our results indicate that the tandem GC box sequences in the mPGES promoter play a major role in regulating its inducible transcription. Furthermore, we demonstrate that Egr-1, an inducible zinc finger protein that recognizes the GC-rich consensus DNA sequence 5'-GCG(T/G)GGGCG-3' (20-23), binds to the proximal GC box in the mPGES promoter region and facilitates inducible transcription of the mPGES gene.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Genomic Cloning-- A mouse 129SvJ genomic library in -FixII was kindly provided by Drs. Nobuya Sasaki and Makoto Taketo (Kyoto University, Kyoto, Japan). The primary genomic library was screened at 2 × 10⁵ plaque-forming units/145-mm plate. The -FixII library was screened with an oligonucleotide complementary from +1 to +817 of the mouse mPGES cDNA that was labeled with [-³²P]dCTP by random priming (Amersham Biosciences). The phages were transferred to Hybond-N membranes (Amersham Biosciences), which were then prehybridized at 60 °C and hybridized with the labeled oligonucleotide in 50% formamide, 5× Denhardt's, and 0.5% SDS at 42 °C for 18 h. Membranes were washed in 2× SSC (300 mM NaCl, 30 mM sodium citrate) and 0.1% SDS at 65 °C for 30 min, washed in 0.1× SSC and 0.1% SDS at 65 °C for 30 min, dried, and then exposed to Bio-Max MS film (Eastman Kodak) overnight at 80 °C. After the first screening of ~1 × 10⁶ plaques, positive phages were subjected to two additional rounds of screening. Of the multiple clones isolated, three of them, which contained different sizes of the insert, were purified (Qiagen) and characterized in detail. Pure positive phages were stored in 0.1 M NaCl, 8 mM MgSO₄, 50 mM Tris-HCl, pH 7.5, and 0.01% (w/v) gelatin with chloroform. PCR analysis using pfu Turbo DNA polymerase (Stratagene) and a combination of vector primers and mPGES cDNA-based primers revealed the complete structure of the mouse mPGES gene. All primers and oligonucleotides used here and in subsequent experiments were obtained from Amersham Biosciences.

Plasmid Constructs-- The genomic DNA fragment of the mouse mPGES gene was amplified by PCR using a -genomic clone as a template and Expand High Fidelity DNA/polymerase mixture (Roche Molecular Biochemicals). By subcloning and sequencing, an ~1.8-kb fragment upstream of the translation start site was isolated (DDBJ accession number AB083340). Human mPGES genomic DNA was also amplified by PCR using genomic DNA from human leukocytes as a template and specific primers designed on the basis of the reported genomic sequence of human mPGES (9). The chimeric constructs for transfection experiments were prepared by ligating different lengths of the 5'-flanking region (FL) plus the 33-bp untranslated region of the mouse mPGES gene with the luciferase gene (luc) into the pGL3-basic vector (Promega) at the XhoI and HindIII sites as follows: 1814/+33/luc (1814-bp 5'-FL, 33-bp untranslated region and luc), 1530/+33/luc, 1198/+33/luc, 930/+33/luc, 700/+33/luc, 655/+33/luc, 360/+33/luc, 150/+33/luc, and 70/+33/luc. The chimeric 5'-deletion constructs of the human mPGES promoter and luc were 650/luc, 329/luc, 312/luc, 158/luc, 99/luc, where the numbers represent 5'-upstream sites in the promoter region considering the translation initiation site as +1. Mutations were introduced into the GC box-related sequences at 70/77 (GC box-2, 5'-GGGCGGGG-3') and 81/88 (GC box-1, 5'-GGGCGGGG-3') of the mouse 5'-FL, and at 101/108 (GC box-2, 5'-GGGCGGGG-3') and 112/119 (GC box-1, 5'-CGTGGGCGGTG-3') of the human 5'-FL by using the manufacturer's protocol, as described previously (10). The pGL3/150GC box-2 mutant (mt) (GC box-2 in the mouse mPGES 5'-FL was mutated to 5'-GTTCGGGG-3'), pGL3/150GC box-1-mt (GC box-1 in the mouse mPGES 5'-FL was mutated to 5'-GTTCGGGG-3'), pGL3/150GC box-1 and 2-mt (both GC boxes in the mouse mPGES 5'-FL were mutated to 5'-GTTCGGGGAGTGTTCGGGG-3'), pGL3/158GC box 2-mt (GC box-2 in the human mPGES 5'-FL was mutated to 5'GTTCGGGG-3'), pGL3/158GC box-1-mt (GC box-1 in the human mPGES 5'-FL was mutated to 5'-GTTCGGTG-3'), and pGL3/158GC box-1 and 2-mt (both GC boxes in the human mPGES 5'-FL were mutated to 5'-GTTCGGGGCGTGTTCGGTG-3'), in which mutated sites are underlined. The PCR primers were designed on the basis of the sequences of the 5'-FL of the mouse and human mPGES genes, and all inserts derived from PCR amplification were cloned into the pGL3-basic vector and were sequenced on both strands. For preparation of the LacZ promoter gene construct, pGL3/1814/+33/luc was digested at the XhoI and HindIII sites, and the 1891-bp promoter fragment was subcloned into the XhoI/HindIII sites in the pgal-basic vector (CLONTECH) to generate pgal/1891.

Development of Transgenic Mice-- A 6658-bp insert containing the chimeric mPGES promoter and the lacZ gene, which was cut off from pgal/1891 by digestion with SacI and XmnI, was gel-purified with the Microspin column (Qiagen). Microinjection of the DNA and generation of chimeric mice were performed in the transgenic mouse facilities at the Central Laboratories for Experimental Animals (Kawasaki, Japan). Transgenic mice were identified by Southern blot analysis of genomic DNA from tail biopsies after hybridization with a lacZ gene probe. Transgenic mice were bred to C57BL/6 mice to generate F1 progeny. The copy number of the integrated transgene was estimated as three to six copies. -Galactosidase reporter gene activity in various tissues of F1 mice before or after intraperitoneal injection of 5 mg/kg lipopolysaccharide (LPS, Escherichia coli O111: B4, Sigma) was assessed by using the luminescent -galactosidase galactosidase detection kit II (CLONTECH). Tissue homogenate was prepared by sonication in a lysis solution containing 100 mM KHPO₄, 0.2% (v/v) Triton X-100, 1 mM dithiothreitol (DTT), 0.2 mM phenylmethylsulfonyl fluoride, and 5 µg/ml leupeptin. After centrifugation for 10 min, 25 µl of the supernatant was assayed by incubation with 200 µl of the reaction buffer at room temperature for 60 min, and luminescence was measured with a luminometer according to the manufacturer's instructions. Luminescence was normalized to protein concentration in each sample.

Cell Culture-- Mouse MC3T3-E1 osteoblastic cells and mouse RAW264.7 macrophage-like cells were obtained from the RIKEN Cell Bank (Tsukuba Science City, Japan) and maintained in -minimal essential medium and Dulbecco's modified Eagle's medium, respectively. Human embryonic kidney (HEK) 293 cells were purchased from Dainippon Pharmaceutical Co. and maintained in RPMI 1640. All media used in this study was from Invitrogen and supplemented with 10% fetal bovine serum (Atlanta Biologics), penicillin G, and streptomycin (Nacalai Tesque). The cells were cultured in a humidified 5% CO₂ and 95% air incubator at 37 °C.

Transfection and Luciferase Assay-- Transient transfection was performed with SuperFect reagent (Qiagen) according to the manufacturer's protocol for adherent cells, as described previously (10). MC3T3-E1 cells (3 × 10⁵ cells/dish), RAW264.7 cells (4 × 10⁵ cells/dish), and HEK293 cells (2 × 10⁵ cells/dish) were plated on 35-mm plastic dishes with 2 ml of culture medium 1 day before transfection. The cells were transfected with 1 µg of several firefly luciferase reporter vectors and 0.1 µg of the Renilla luciferase control reporter vector (pRL-TK, Promega) with the use of 10 µl of the SuperFect reagent. 1 day after transfection, the cells were incubated for 16 h with or without 100 nM TPA (Sigma), 5 ng/ml interleukin-1 (Genzyme) plus tumor necrosis factor  (Genzyme) or 100 ng/ml LPS. The luciferase assay was performed with the dual luciferase assay kit (Promega) and a luminometer according to the manufacturer's instructions. Lysates (10-µl aliquots) were assayed first for firefly luciferase and then for Renilla luciferase activities. The activity of firefly luciferase was normalized to that of Renilla luciferase. Data represent the average ± S.E. of four independent experiments, each of which was done in triplicate.

Electrophoretic Mobility Shift Assay (EMSA)-- The MC3T3-E1 and RAW264.7 cell nuclear extracts were prepared as previously described (24). EMSA was performed as described previously (24). Briefly, double-stranded oligonucleotides were radiolabeled with [-³²P]dCTP at their 3'-ends with Klenow fragment (Takara) and then purified with MicroSpin columns (Amersham Biosciences). The sense sequences of synthesized oligonucleotides used were as follows: 5'-GGTGGGCGGGGAGTGGGCGGGG-3' for native GC boxes in the mouse mPGES 5'-FL, 5'-GGTGTTCGGGGAGTGGGCGGGG-3' for mutation in the GC box-2 (mutated sites underlined), 5'-GGTGGGCGGGGAGTGTTCGGGG-3' for mutation in the GC box-1, and 5'-GGTGTTCGGGGAGTGTTCGGGG-3' for mutation in both GC boxes.

MC3T3-E1 cells (1 × 10⁵ cells/dish) and RAW264.7 cells (2 × 10⁵ cells/dish) were plated on 35-mm plastic dishes with 2 ml of culture medium, and after 3 days of culture the supernatant was replaced with 2 ml of fresh medium containing 0.1% fetal bovine serum. After 2 h, cells were incubated with or without TPA for various periods. The cells were suspended in 200 µl of ice-cold buffer A (10 mM HEPES, pH 7.9, 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, 1 mM DTT, 0.5 mM phenylmethylsulfonyl fluoride, 2 µg/ml leupeptin, 2 µg/ml pepstatin, and 2 µg/ml aprotinin) and incubated for 10 min on ice. Nonidet P-40 was added to the suspension at a final concentration of 0.4% (v/v) and the mixtures were mixed vigorously for 10 s. After centrifugation at 15,000 × g for 1 min, the resulting pellets were suspended in 30 µl of buffer C (20 mM HEPES, pH 7.9, 0.4 M NaCl, 0.1 mM EDTA, 0.1 mM EGTA, 1 mM DTT, 0.5 mM phenylmethylsulfonyl fluoride, 2 µg/ml leupeptin, 2 µg/ml pepstatin, and 2 µg/ml aprotinin) by vigorous mixing for 5 min and the resulting supernatants were used as the nuclear extracts.

The nuclear extracts were incubated with 1 ng of ³²P-labeled oligonucleotide in a binding buffer (10 mM HEPES, pH 7.9, 50 mM KCl, 0.5 mM EDTA, 5 mM DTT, 0.7 mM phenylmethylsulfonyl fluoride, 2 µg/ml leupeptin, 2 µg/ml pepstatin, 2 µg/ml aprotinin, 10% (v/v) glycerol, and 1 µg/ml poly(dI-dC)-poly(dI-dC)) for 30 min at 25 °C. For competition assay, a 100-fold excess of unlabeled oligonucleotide was added to the binding reaction prior to the addition of the radiolabeled probe. For supershift analysis, 1 µg of each antibody (Santa Cruz Biotechnology) was added to the nuclear extracts prior to the addition of the radiolabeled probe. These incubation mixtures were then electrophoresed in a 6% (w/v) native polyacrylamide gel in Tris borate-EDTA buffer. The gels were dried and analyzed with an image analyzer BAS2500 (Fuji Photo Film).

Northern Blotting-- Approximately equal amounts (30 µg) of the total RNA obtained from the various tissues were applied to separate lanes of 1.2% (w/v) formaldehyde-agarose gels, electrophoresed, and transferred to Hybond-N membranes. The resulting blots were then probed with mPGES cDNA probes that had been labeled with [-³²P]dCTP by random priming. Prehybridization and hybridization were carried out as described previously (25).

Southwestern Blotting-- MC3T3-E1 cells were stimulated with 100 nM TPA for 1 h, and then extracts of nuclear proteins were prepared as described above. The nuclear extracts and prestained molecular mass standards (Bio-Rad) were separated by SDS-PAGE on 10% (w/v) gels under reducing conditions and electroblotted onto polyvinylidene difluoride membranes (Millipore). The membrane-immobilized nuclear proteins were denatured in buffer (10 mM HEPES, pH 7.9, 50 mM KCl, 3 mM MgCl₂, 5 mM DTT) containing 6 M guanidine hydrochloride for 1 h at 4 °C and then slowly renatured. The renaturation process involved successive dilutions and incubations for 30 min at 4 °C in the same buffer, in which the concentrations of guanidine hydrochloride were reduced to 4, 2, 1, 0.5, 0.25, 0.125, and 0.065 M and finally to zero. Following renaturation, the membranes were incubated for 16 h at 4 °C in the same buffer containing 5% (w/v) nonfat dried milk, and then incubated overnight at 4 °C in the buffer containing 4 µg of poly(dI-dC)-poly(dI-dC) and the ³²P-labeled native GC box oligonucleotide. The membrane was then washed with six changes of the buffer, and signals were analyzed with BAS2500.

Western Blotting-- The membrane used for Southwestern blot analysis mentioned above was washed in 10 mM Tris-HCl, pH 7.4, 50 mM NaCl, 1 mM EDTA, and 1 mM DTT to remove the associated radioactivity. Then, the membrane was blocked for 2 h in 10 mM Tris-HCl, pH 7.4, containing 150 mM NaCl, 0.1% (v/v) Tween 20 (TBS-T), and 5% (w/v) skim milk. After washing the membrane with TBS-T, an antibody against Egr-1 was added at a 1:5,000 dilution in TBS-T containing 0.01% (w/v) bovine serum albumin and incubated for 1.5 h. After washing the membranes three times with TBS-T, horseradish peroxidase-conjugated goat anti-rabbit IgG (Zymed Laboratories Inc. was added at a 1:10,000 dilution in TBS-T containing 0.01% bovine serum albumin and incubated for 1 h. After six washes with TBS-T, protein bands were visualized with enhanced chemiluminescence Western blotting detection reagents (Amersham Biosciences) as described previously (25).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Structure of the Mouse mPGES Gene-- A Sv129J mouse genomic library was screened with a mouse mPGES cDNA probe. Three positive clones were isolated, each with inserts of different sizes and overlapped sequences. The structure of the mPGES gene was determined by combining the sequences of these three clones (Fig. 1A). The distances between exons were determined by long-PCR using exon-specific primers, and the numbers and sizes of exons and the positions of the exon-intron junctions were determined by comparing the sequence of the obtained genomic sequence with that of mPGES cDNA. The mouse mPGES gene was split into three exons (174, 80, and 567 bp) by two introns (2.3 and 6.5 kb) and spans a region of about 10 kb (Fig. 1A). The initiation and termination codons reside in exons 1 and 3, respectively. Overall, the organization of the mouse mPGES gene was similar to that of the human mPGES gene (19).

View larger version (45K): [in this window] [in a new window]   Fig. 1.   Genomic structure and promoter sequence of the mouse mPGES. A, the organization of the mouse mPGES gene. Exons are shown by boxes. Rectangle in the bottom represents the full-length protein. B, sequence of the mouse mPGES promoter region between 1814 and +54, considering the transcription start site as +1 (arrow), which was determined by 5'-rapid amplification of cDNA ends using mouse peritoneal macrophage cDNA. Consensus binding sites for transcription factors are indicated by underlines. The initiation codon ATG is shown in boldface.

The nucleotide sequence of the 5'-flanking region (~1.85 kb upstream of exon 1) was determined by oligonucleotide walking (Fig. 1B). Although no typical TATA box was found throughout the sequence, two tandem GC boxes were present in a region ~70 bp upstream of the transcription initiation site, which was often the case with the TATA-less genes. Moreover, the sequence around the transcription initiation site shows close similarity to the pyrimidine-rich initiator element, which encompasses the transcription initiation sites for a number of developmentally regulated genes with TATA-less promoters. In particular, the sequence ⁸CTCACTCT¹ in the mPGES gene matches exactly the initiator consensus sequence CTCANTCT, which was found in several other TATA-less genes (26, 27). Data base search for potential transcription factor-binding sites using the Transcription Element String Search (www.cbil.l.upenn.edu/tess), a web-based search engine, revealed the presence of several consensus cis-acting elements within 1.8 kb upstream of the initiator consensus sequence, including elements for C/EBP, AP-1, C/EBP, and CACCC-binding factors as well as putative glucocorticoid or progesterone response elements (Fig. 1B).

Generation of Transgenic Mice with the mPGES Promoter/LacZ Construct and Its Response to LPS in Vivo-- The expression of mPGES was increased markedly in various tissues of rats after administration of LPS (10, 12). To test whether the mPGES promoter obtained above could mimic tissue-specific and LPS-inducible expression of mPGES in vivo, we prepared a construct containing the mPGES promoter flanked by the lacZ gene and created mPGES promoter/LacZ transgenic mice that overexpressed -galactosidase in tissues under the control of the mPGES promoter region. F1 mice bearing the transgene were injected intraperitoneally with 5 mg/kg LPS or buffer as control, and mPGES expression and -galactosidase activity in various tissues were examined 16 h after treatment. The -galactosidase activity was increased markedly in several tissues such as thymus, lung, stomach, kidney, and spleen in LPS-treated mice (Fig. 2B), which occurred in parallel with the intrinsic LPS-dependent induction of mPGES mRNA (Fig. 2A). These results imply that critical elements that dictate tissue-specific and stimulus-inducible expression of mPGES were present in the 1814-bp 5'-flanking region of the mouse mPGES gene.

View larger version (48K): [in this window] [in a new window]   Fig. 2.   LPS-induced expression in mice of the transgene containing the mPGES promoter/LacZ reporter. A, expression of mPGES mRNA in various tissues of mice after injection of LPS or buffer (LPS) was assessed by RNA blotting. B, extracts prepared from F1 transgenic mice tissue after injection of LPS or buffer were assayed for -galactosidase activity as described under "Experimental Procedures." A representative result of three independent experiments is shown.

Promoter Activity of the 5'-Flanking Region of the Mouse mPGES Gene-- To assess which portions of the mPGES 5'-flanking region were involved in the regulation of inducible expression of mPGES, a series of chimeric 5'-deleted promoter-luciferase reporter constructs was developed. The structures and sizes of the chimeric constructs as well as the locations of the consensus sequences for various transcription factors found within the mPGES promoter region (Fig. 1B), are depicted in Fig. 3A. The resulting plasmids were then transfected into MC3T3-E1 cells (Fig. 3B) or RAW264.7 cells (Fig. 3C). After 24 h of transfection, the cells were stimulated with appropriate stimulants for 16 h, and relative luciferase activities were assessed. Whereas all constructs exhibited low basal luciferase activities, stimulation with TPA, cytokines (IL-1 plus TNF), or LPS greatly increased the luciferase activity in both cell types transfected with the full-length construct (1814). These stimulus-sensitive luciferase activities were unchanged when the 5'-end of the promoter region was deleted to positions 1530, 1198, 930, and 700, and further truncation up to 150 tended to increase the activity progressively. Thus, elements for several transcription factors including AP-1 and C/EBP and -, located between 1814 and 700 (Fig. 1B), were dispensable for inducible mPGES transcription, and there may be negative regulatory elements in a region between 700 and 150. Notably, the shortest construct, pGL3/70, which harbored only a 70-bp 5'-flanking sequence, showed no inducibility of luciferase activity even in the presence of stimuli, as in the case of the promoter-null control vector pGL3/basic. The plasmid pGL3/150rev, in which the 150-bp 5'-flanking sequence was inserted in an inverse orientation, showed minimal luciferase activity and did not respond to TPA stimulation in MC3T3-E1 cells (data not shown). These results suggest that the promoter region 150 to 70 contains positive regulatory elements essential for stimulus-induced mouse mPGES transcription in MC3T3-E1 and RAW264.7 cells.

View larger version (30K): [in this window] [in a new window]   Fig. 3.   The mouse mPGES promoter/luc reporter assay in MC3T3-E1 and RAW264.7 cells. A, structures of various 5'-deletion constructs of the mouse mPGES promoter fused to the luc reporter gene are shown. The consensus binding sites for transcription factors are shown by boxes. Luciferase activities in MC3T3-E1 (B) or RAW264.7 (C) cells transfected with these constructs were determined 16 h after treatment with or without stimulants. To calibrate the transfection efficiency, the cells were co-transfected with pRL-TK, and the luciferase activity associated with each construct was normalized to the activity of Renilla luciferase, as described under "Experimental Procedures." Results are the mean ± S.E. of four independent experiments.

As shown in Fig. 1B, two GC box-related sequences were tandemly present between 150 and 70 in the mPGES promoter region, suggesting that one or both of them may be involved in regulating the transcription of the mPGES gene. To determine their potential roles, we prepared a series of additional promoter constructs with site-specific mutations, as illustrated in Fig. 4A. MC3T3-E1 cells transfected with the chimeric plasmid containing either the distal GC box (GC box-1) or proximal GC box (GC box-2) mutant, respectively, in which two critical guanine residues were replaced with thymine residues, showed lower (albeit significant) luciferase activity in response to TPA or cytokines than did replicate cells transfected with the plasmid containing the two native GC boxes (Fig. 4B). Moreover, simultaneous mutations in both GC boxes further reduced luciferase expression, indicating complete loss of promoter activity.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Effects of GC box mutations on the mouse mPGES promoter activity. A, mutations were introduced into the GC box-related sequences at 70/77 (GC box-1) and/or 81/88 (GC box-2) regions of the mouse mPGES promoter. B, each plasmid was co-transfected with pRL-TK into MC3T3-E1 cells, and luciferase activities were determined 16 h after stimulation with or without stimulants as described under "Experimental Procedures." The activity associated with each construct was normalized relative to Renilla luciferase activity. Results are the mean ± S.E. of four independent experiments.

Specific DNA-Protein Interactions within the GC Box Sequences of mPGES-- The reporter gene assay suggested that both GC boxes play a significant role in transcription of the mPGES gene. To determine whether these elements could bind nuclear proteins, we synthesized a 22-bp double-stranded oligonucleotide containing the tandem GC boxes. MC3T3-E1 or RAW264.7 cells were stimulated with TPA, and the nuclear extracts were analyzed by EMSA for binding to the ³²P-labeled oligonucleotide probe. Two distinct bands, I and III, appeared after incubation of the probe with nuclear extracts derived from unstimulated MC3T3-E1 (Fig. 5A) or RAW264.7 (Fig. 5B) cells. When these cells were stimulated with TPA for 0.5 h, a novel band II became detectable between DNA-protein complexes I and III. The band II peaked at 1 h and declined gradually thereafter in MC3T3-E1 cells (Fig. 5A), whereas it reached a maximal level and a plateau after 1 h in RAW264.7 cells (Fig. 5B). This inducible expression of DNA-protein complex II depended on the concentrations of TPA in MC3T3-E1 (Fig. 5C) and RAW264.7 (Fig. 5D) cells.

View larger version (52K): [in this window] [in a new window]   Fig. 5.   EMSA analysis of the GC boxes in the mouse mPGES promoter. MC3T3-E1 (A and C) or RAW264.7 (B and D) cells were stimulated for the indicated periods with or without 100 nM TPA (A and B) or for 1 h with various concentrations of TPA (C and D), and then extracts of nuclear proteins were prepared as described under "Experimental Procedures." The extracts were incubated with a 32P-labeled DNA fragment containing the two GC box sequences, and protein/DNA interactions were tested by EMSA. Binding reactions were resolved by 6% acrylamide gel electrophoresis using 0.25× TBE (22 mM Tris base, 22 mM boric acid, 0.5 mM EDTA) as a buffer. Arrows indicate the three protein-DNA complexes (I, II, and III). A representative result of three independent experiments is shown.

To test whether the tandem GC boxes were required for binding of the nuclear proteins, we performed EMSA with oligonucleotide probes containing mutations in either or both of the two GC box sequences. Mutation in the GC box-2 alone or both GC boxes completely abolished all three bands in TPA-stimulated MC3T3-E1 (Fig. 6A) and RAW264.7 cells (Fig. 6B), whereas mutation in the GC box-1 alone had no effect. Moreover, the formation of these DNA-protein complexes was prevented by addition of a 100-fold molar excess of an unlabeled oligonucleotide harboring the GC box-1 alone or the two intact GC boxes, but not by an oligonucleotide harboring the GC box-2 mutation alone or mutation in both GC boxes (Fig. 6, A and B). These results suggest that the formation of these DNA-protein complexes was sequence-specific and that the proximal GC box-2 was essential.

View larger version (44K): [in this window] [in a new window]   Fig. 6.   EMSA analysis of sequence-specific DNA-protein interactions in the GC boxes. MC3T3-E1 (A and C) or RAW264.7 (B and D) cells were incubated with or without 100 nM TPA for 1 h, and then the nuclear proteins were extracted. A and B, the extracts were incubated with 32P-labeled DNA fragments containing two intact GC boxes, GC box-1 mutant, GC box-2 mutant, or both GC box mutants as described under "Experimental Procedures." C and D, the extracts were incubated with a 32P-labeled DNA fragment containing the two intact GC boxes with or without a 100-fold molar excess of unlabeled competitor DNA. EMSA was performed under the same conditions as described for Fig. 5. Arrows indicate the three protein-DNA complexes (I, II, and III). A representative result of three independent experiments is shown.

Identification of a GC Box-binding Protein-- To determine the molecular sizes of the factor(s) that interacts with the GC box-2, the nuclear extract from MC3T3-E1 cells was examined for binding to the GC box oligonucleotide probe by Southwestern blot analysis. The nuclear extracts prepared from the cells incubated with or without TPA were resolved on SDS-polyacrylamide gels and transferred to polyvinylidene difluoride membranes. The proteins were then renatured and incubated with a ³²P-labeled GC box oligonucleotide probe. As shown in Fig. 7, several bands were detected with similar intensity in crude nuclear extracts of both unstimulated and TPA-stimulated cells, except for an 80-kDa band that was evident only after TPA treatment.

View larger version (98K): [in this window] [in a new window]   Fig. 7.   Identification of a GC box-binding protein. A, Southwestern blotting. MC3T3-E1 cells were incubated with or without 100 nM TPA for 1 h, and the nuclear extracts were separated by SDS-PAGE, transferred to polyvinylidene difluoride membrane, renatured, and probed with a 32P-labeled DNA fragment containing the two GC boxes. The arrow indicates a 80-kDa GC box-binding protein. B, supershift analysis. The nuclear extracts were incubated with the 32P-labeled DNA fragment containing the two GC boxes in the presence or absence of the indicated antibodies. Arrows indicate the three protein-DNA complexes (I, II, and III). S indicates a supershifted band (lane 3). C, Western blotting. The membrane used for the Southwestern blot analysis (A) was reprobed with an antibody against Egr-1. A representative result of three independent experiments is shown.

The Sp1 and Egr families are known to bind to GC-rich sequences (28, 29). To identify the factor(s) that binds to the GC box in the mPGES promoter, supershift analysis was carried out using polyclonal antibodies that specifically recognize Egr-1, Egr-2, Egr-3, Sp1, or Sp3. As shown in Fig. 7B, the inducible DNA-protein complex band II was retarded to the position of band S only by an anti-Egr-1 antibody, whereas none of the other antibodies induced such a supershift (Fig. 7B). It is notable that the molecular mass of Egr-1 was 80-kDa (20), which was similar to the size of the inducible protein detected by Southwestern blot analysis (Fig. 7A). Indeed, when the membrane used for the Southwestern blot analysis was reprobed with anti-Egr-1 antibody by Western blotting, an 80-kDa Egr-1 protein was detected in TPA-treated cells (Fig. 7C). These data suggest that the inducible GC box-binding protein in TPA-stimulated MC3T3-E1 cells was the nuclear transcription factor Egr-1.

Promoter Activity of the 5'-Flanking Region of the Human mPGES Gene-- The tandem GC box sequences were highly conserved in both the human and mouse mPGES gene promoters (Fig. 8A). To test whether the GC boxes also play a significant role in transcription of the human mPGES gene, a similar reporter gene assay, in which the human mPGES promoter was flanked by the luciferase reporter gene, was performed in human HEK293 cells. A 650-bp human mPGES promoter fragment was amplified by PCR with human genomic DNA as a template. The locations of the consensus sequences for transcription factors in the amplified product that contains the tandem GC boxes and upstream GATA-2- and N-myc-binding sites (Fig. 8B) and those in the 5'-end deleted mutants are depicted in Fig. 8C. The resulting promoter-luciferase reporter plasmids were each transfected into HEK293 cells, and the cells were then stimulated with TPA for 16 h. As shown in Fig. 8D, the promoter activities of pGL3/650, pGL3/392, pGL3/312, and pGL3/158 were elevated by TPA stimulation, whereas pGL3/99, which contained no GC box sequence, showed low basal and TPA-inducible luciferase activity. Moreover, HEK293 cells transfected with plasmids containing either GC box-1 or -2 mutants showed low luciferase activity in response to TPA, and mutations in both GC boxes further reduced the induction level compared with the plasmid containing the native GC box sequences (Fig. 8E). These results were compatible with the results from the analyses of the mouse mPGES promoter, implying that the GC boxes play a significant role in the transcription of both human and mouse mPGES genes.

View larger version (40K): [in this window] [in a new window]   Fig. 8.   The human mPGES promoter/luc reporter assay in HEK293 cells. A, comparison of the GC box sequences in the mouse and human mPGES gene promoters. B, structures of various 5'-deletion constructs of the human mPGES promoter fused to the luc reporter gene are shown. The consensus binding sites for transcription factors are indicated by boxes. C, mutations were introduced into the GC boxes at 101/108 (GC box-1) and/or 112/119 (GC box-2) in the human mPGES promoter. D and E, each plasmid was co-transfected with pRL-TK into HEK293 cells, and luciferase activity was determined 16 h after treatment with or without 100 nM TPA. The activity associated with each construct was normalized relative to Renilla luciferase activity. Results are the mean ± S.E. of four independent experiments.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Evidence is emerging that mPGES is an inducible enzyme, the expression of which is markedly increased in various cells and tissues following proinflammatory stimuli (9, 10, 30-33). Inducible genes contain particular nucleotide elements within their promoter regions that are responsible for regulated transcription. The aim of this study was to characterize the cis-acting regulatory elements and trans-acting nuclear factors that regulate inducible expression of the mPGES gene. We found that the binding of Egr-1, an inducible transcription factor, to the proximal GC box is an essential event that directs the regulatory expression of mPGES.

Transgenic expression of the chimeric mouse mPGES gene (5'-untranslated fragment up to 1858 bp upstream of the initiating ATG) flanked by the LacZ reporter gene in mice resulted in LPS-stimulated expression of -galactosidase in tissues in which mPGES was intrinsically induced, implying that the regulatory elements for stimulus-induced mPGES expression reside within this promoter region. On the basis of results from the promoter/reporter assay in mouse osteoblastic and macrophage-like cells, we identified the core promoter region controlling reporter expression to be situated between 70 and 150, in which two GC boxes were located tandemly. Moreover, mouse mPGES promoter activity was significantly increased when the promoter region was deleted at 150 to 655, indicating that this region adjacent to the GC boxes contains negative transcriptional regulatory elements. In this region, there are two consensus regulatory elements for glucocorticoids that have been shown to down-regulate mPGES expression in several cell types (10, 12, 33, 34). Although the mouse mPGES promoter region contains several other consensus binding sites for transcription factors involved in the expression of proinflammatory genes, such as C/EBP and -, AP-1, and the CACCC-binding factor, these sites may be nonessential for mPGES promoter activation under the conditions employed here. There are no binding sites for NF-B, CRE, and E-box that have been implicated in COX-2 induction (14-18), in the mPGES promoter region, implying that the mechanisms for inducible expression of COX-2 and mPGES are distinct. Indeed, even though both enzymes are stimulus-inducible and function sequentially in the same PGE₂-biosynthetic pathway, the kinetics of induction of these two enzymes are not entirely identical in several systems (30, 33, 34).

Further studies using mutation constructs and EMSA confirmed that the two GC box sequences located between 70 and 150 are essential for the activation of the mouse mPGES gene in response to stimuli. Increase in reporter gene expression was reduced by mutating either of the two GC boxes, whereas TPA-stimulated DNA/protein binding activity was completely abolished only by mutation of the GC box-2, but not the GC box-1. These results suggest that even though the nuclear GC box-binding protein (Egr-1) binds only to the proximal GC box-2, both GC boxes are required for full promoter activity. Although the role of the distal GC box-1 is unclear, it may contribute to full promoter activity through association with coactivators or transcription initiation complex, which may act in synergy with Egr-1. Our EMSA experiments indicated that two other proteins, designated as I and III, also interacted with the GC box-2 even in unstimulated cells. It is possible that Egr-1 and the two other GC box-binding proteins share an overlapping site, although it is currently unclear whether these interactions are mutually exclusive or act in concert. The molecular identities of the proteins responsible for bands I and III detected in the EMSA analysis are under investigation.

It has been reported that tandem GC box sequences exist in several gene promoters, such as prostacyclin synthase (35), thromboxane receptor (36), hemopoietic cell kinase (37), N-methyl-D-aspartate receptor 1 (38), and thrombomodulin (39, 40). These tandem GC boxes represent binding sites for members of the Sp1 transcription factor family (41) that play an important role in the control of transcription of the aforementioned genes. However, our EMSA studies using specific antibodies showed that the GC box in the mPGES promoter region does not bind Sp1 or Sp3. Instead, the transcription factor Egr-1 was found to bind the proximal GC box sequence in the nuclear extract of TPA-stimulated cells. Egr-1 (also known as Krox-24), which was first identified as an immediate-early gene induced by mitogenic stimulation (23), contains three zinc finger motifs that regulate transcription through binding to the GC-rich consensus sequence 5'-GCG(T/G)GGGCG-3' (20). The Egr-1 gene is rapidly and transiently induced by a variety of stimuli or cellular stresses, including serum, TPA, and other growth factors (21, 22). Egr-1-binding sites have been identified in the promoters of cyclin D1 (42), platelet-derived growth factor (43), and p53 (44), where this inducible transcription factor is proposed to mediate the activation of downstream genes that play crucial roles in cellular proliferation and differentiation. We have previously shown that sustained production of PGE₂ by HEK293 cells stably co-transfected with COX-2 and mPGES is associated with aggressive cell growth and morphological change (9). Furthermore, the present study provides unequivocal evidence that Egr-1 binds to the GC box in the mPGES promoter region and regulates inducible transcription of the mPGES gene. Thus, transcriptional regulation of mPGES expression by Egr-1 appears to be closely correlated with cellular proliferation and carcinogenesis. We speculate that Egr-1 may regulate cellular proliferation and differentiation at least in part through up-regulation of mPGES gene transcription, particularly in tissues and cells where PGE₂ plays a regulatory role in these events. The functional link between Egr-1 and mPGES expression is further supported by recent observations that cytokine-induced expression of mPGES is suppressed by a p38 mitogen-activated protein kinase inhibitor (45) and that Egr-1 induction is mediated by the p38 mitogen-activated protein kinase pathway (46). Indeed, pharmacological inhibition of p38 attenuates Egr-1-mediated mPGES expression in our preliminary reporter assay.²

Although the sequences of the mouse and human mPGES promoters (1 to 640) were only 48% homologous, the homology between them around the tandem GC boxes (70 to 124) is relatively high (78%). Of importance, our studies revealed that the tandem GC boxes were also critical for transcriptional activation of the human mPGES gene. Moreover, a similar regulatory mechanism is common to several distinct cell types (osteoblasts, macrophages, and kidney cells) following various stimuli (TPA, cytokines, and LPS). It is therefore likely that Egr-1-mediated regulation of the mPGES gene represents the common regulatory machinery in various situations.

	ACKNOWLEDGEMENTS

We are grateful to Drs. Nobuya Sasaki and Makoto Taketo (Kyoto University, Kyoto) for helpful advice with the initial genomic cloning experiments.

	FOOTNOTES

^* This work was supported by grants from the Ministry of Health, Welfare and Labor, the Ministry of Education, Culture, Sports, Science and Technology of Japan, and the Promotion of Fundamental Studies in Health Science of the Organization for Pharmaceutical Safety and Research (OPSR) of Japan.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 DDBJ/GenBank^TM/EBI Data Bank with accession number(s) AB083340.

Present address: Kitasato Institute, Tokyo 108-8642, Japan.

^** To whom correspondence should be addressed. Tel.: 81-6-6833-5012; Fax: 81-6-6872-8090; E-mail: tanabe@ri.ncvc.go.jp.

Published, JBC Papers in Press, May 28, 2002, DOI 10.1074/jbc.M203618200

² H. Naraba, C. Yokoyama, N. Tago, and T. Tanabe, unpublished observation.

	ABBREVIATIONS

The abbreviations used are: PGE₂, prostaglandin E₂; mPGES, membrane-bound prostaglandin E₂ synthase; COX, cyclooxygenase; Egr, early growth response; Sp, specificity protein; EMSA, electrophoretic mobility shift assay; TPA, 12-O-tetradecanoylphorbol-13-acetate; LPS, lipopolysaccharide; TNF, tumor necrosis factor; HEK, human embryonic kidney; FL, flanking region; luc, luciferase; DTT, dithiothreitol; CRE, cAMP-response element-binding protein; mt, mutant; PGES, prostaglandin E₂ synthase.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati