JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M107180200 on September 10, 2001

J. Biol. Chem., Vol. 276, Issue 49, 46260-46267, December 7, 2001
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
276/49/46260    most recent
M107180200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hatae, T.
Right arrow Articles by Tanabe, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hatae, T.
Right arrow Articles by Tanabe, T.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Prostacyclin-dependent Apoptosis Mediated by PPARdelta *

Toshihisa Hatae, Masayuki Wada, Chieko Yokoyama, Manabu Shimonishi, and Tadashi TanabeDagger

From the Department of Pharmacology, National Cardiovascular Center Research Institute, Fujishiro-dai, Suita, Osaka 565-8565, Japan

Received for publication, July 27, 2001, and in revised form, September 6, 2001

    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS AND DISCUSSION
REFERENCES

Prostacyclin (PGI2) plays important roles in hemostasis both as a vasodilator and an endogenous inhibitor of platelet aggregation. PGI2 functions in these roles through a specific IP receptor, a G protein-coupled receptor linked to Gs and increases in cAMP. Here, we report that intracellular prostacyclin formed by expressing prostacyclin synthase in human embryonic kidney 293 cells promotes apoptosis by activating endogenous peroxisome proliferator-activated receptor delta  (PPARdelta ). In contrast, treatment of cells with extracellular prostacyclin or dibutyryl cAMP actually reduced apoptosis. On the contrary, treatment of the cells with RpcAMP (adenosine 3',5'-cyclic monophosphothioate, Rp-isomer), an antagonist of cAMP, enhanced prostacyclin-mediated apoptosis. The expression of an L431A/G434A mutant of PPARdelta completely blocked prostacyclin-mediated PPARdelta activation and apoptosis. These observations indicate that prostacyclin can act through endogenous PPARdelta as a second signaling pathway that controls cell fate.

    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS AND DISCUSSION
REFERENCES

Prostaglandins (PGs)1 are a diverse family of oxygenated fatty acids derived from arachidonic acid (AA). AA is converted to an intermediate, prostaglandin endoperoxide H2 (PGH2) by two isoforms of cyclooxygenase, COX-1 and COX-2. The COX product PGH2 is converted to one of several biologically important prostanoids, including PGE2, PGD2, PGF2alpha , thromboxane A2 and prostacyclin (PGI2) by specific synthases (1, 2). PGs have wide-ranging effects in regulating aspects of homeostasis and pathogenesis. For example, PGE2 modulates inflammation, pain, and fertility whereas PGI2 is important in hemostasis and also exerts bronchiectasis and proliferation inhibitory effect (3-5). PGs, including PGI2, function through cell surface G protein-coupled receptors linked to different cytoplasmic signaling pathways (6-11) and also exert their effects by interacting with a nuclear hormone receptor, peroxisome proliferator-activated receptor (PPAR) (12, 13). Many of the functions of PPARs are associated with pathways of lipid transport and metabolism (14-16). Moreover, PPARs have important roles in cell replication, differentiation, tumorigenesis, and apoptosis. For example, cell growth is inhibited by activation of PPARgamma in fibroblasts and adipogenic cells (17). Differentiation of adipocyte requires not only PPARgamma (12, 18) but also PPARalpha (19) and PPARdelta (20). Tumorigenesis is promoted by PPARgamma in colorectal cells (21). The nonsteroidal anti-inflammatory drug sulindac antagonizes PPARdelta and suppresses colorectal tumorigenesis (22). On the other hand, apoptosis is promoted by activation of PPARalpha in vascular smooth muscle cells (23) and PPARgamma in human breast cancer cells (24), macrophages (25), and vascular endothelial cells (26); however, apoptosis is suppressed by activation of PPARgamma in cerebellar granule cells (27).

Although PGI2 and its analogs can activate PPARdelta (28), it has not been known whether intracellular PGI2 can actually function via PPARs. In our previous study (29), we found that overexpression of PGI2 synthase (PGIS) in the human embryonic kidney epithelial 293 (HEK-293) cell line induces cell death with apoptotic characteristics. Here, we report that intracellular PGI2 produced by expressing PGIS in HEK-293 cells promotes apoptosis by activating the endogenous PPARdelta .

    EXPERIMENTAL PROCEDURES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS AND DISCUSSION
REFERENCES

Reagents-- Iloprost, carbaprostacyclin (cPGI), arachidonic acid, and anti-COX-1 polyclonal antiserum were purchased from Cayman Chemical Co. (Ann Arbor, MI). RpcAMP (adenosine 3',5'-cyclic monophosphothioate, Rp-isomer), dbcAMP (dibutyryl cyclic AMP), and H-7 (1-(5-isoquinolinesulfonyl)-2-methyl-piperazine) were from Sigma Chemical Co. (St. Louis, MO). Texas Red-linked donkey anti-rabbit IgG was purchased from Amersham Pharmacia Biotech (Buckinghamshire, UK). Hoechst 33258 (bisbenzimide H33258 fluorochrome trihydrochloride) was obtained from Nacalai Tesque Co. (Kyoto, Japan). Anti-PPAPdelta polyclonal antibody and anti-COX-1 goat polyclonal IgG were products from Santa Cruz Biotechnology (Santa Cruz, CA). Hemagglutinating virus Japan (HVJ) and liposome were provided by Dr. Y. Kaneda (Osaka University Medical School). All culture media were purchased from Life Technologies, Inc. (Rockville, MD).

Anti-prostacyclin Synthase Polyclonal Antiserum Production and Immunofluorescence Staining-- Synthetic peptides, P1: PGEPPLDLGSIPWLGYALDC corresponding to amino acid residues 27-45 in human PGIS, or P4: LMQPEHDVPVRYRIRP corresponding to amino acids 485-500 coupled to keyhole limpet hemocyanin were prepared by the Peptide Institute Inc. (Osaka, Japan). Japanese white rabbits were immunized with 1 mg of the conjugated peptide in Freund's complete adjuvant. Both P1 and P4 antisera were useful for immunoblotting. In this study, P1 antiserum was used for immunoblotting, and P4 antiserum was used for immunofluorescence staining. Monolayers of HEK-293 cells were seeded (3.0 × 105 cells/60-mm dish) in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS), 2 mM L-glutamine, 100 units/ml penicillin, and 100 mg/ml streptomycin ("medium A"). After incubation for 24 h, the cells were transfected with 3 µg of the high expression vector for wild-type PGIS (pCMV/PGISWT) (29), the catalytically inactive mutant PGISC441A (pCMV/PGISC441A) (29), or the control vector (pCMV-7) (29) plus 0.3 µg of pVA (a plasmid encoding adenovirus-associated RNA1) (29) using 9.9 µl of LipofectAMINE (Life Technologies, Inc.) with serum-free DMEM, and cells were cultured for 5 h. Subsequently, DMEM containing 20% FBS ("medium B") was added, and cells were cultured with or without 100 µM aspirin or 100 µM U46619. After 36 h, the cells were rinsed with PBS, fixed in 3.7% formaldehyde for 10 min, and washed with PBS three times. The cells were then incubated with anti-PGIS antibody P4 for 2 h, and with anti-rabbit IgG-Texas Red for 1 h at 37 °C, washed three times with PBS containing 2% FBS, stained with 0.3 mM Hoechst 33258 and observed under fluorescent microscopy.

Immunoblotting-- Total cellular extracts from the transfected cells were prepared by lysing cells in 1% SDS, 5 mM EDTA, 5 mM EGTA, 1 mM dithiothreitol, 200 mM phenylmethanesulfonyl fluoride, and 100 mM leupeptin. Protein from 1.0 × 104 cells was separated by 10% SDS-polyacrylamide gel electrophoresis and electrotransferred onto Immobilon-P polyvinylidene difluoride transfer membranes (Millipore, Bedford, MA). Filters were blocked overnight with Tris-buffered saline (TBS) containing 5% skim milk (Bio-Rad, Hercules, CA) and 3% bovine serum albumin (Seikagaku Kogyo Co., Tokyo, Japan) at 4 °C. Immunostaining steps were performed in TBS containing 0.05% Tween 20 and 3% bovine serum albumin at room temperature. Filters were incubated with primary and secondary antibodies for 1 h each. Filters were washed in TBS containing 0.05% Tween 20 four times for 10 min between each step and were developed with ECL reagent (Amersham Pharmacia Biotech).

Cell Culture and Transfection-- HEK-293 (29), CV-1, and bovine aortic endothelial cells (BAEC) (30) (1.2 × 104 cells/cm2) were cultured in medium A. Caco-2 cells (1.0 × 105 cells/30-mm dish coated with collagen) were cultured in RPMI 1640 supplemented with 10% FBS, 100 units/ml penicillin, and 100 mg/ml streptomycin. For expression of PGISwt, PGISC441A, COX-1, and COX-2, plasmid vectors pCMV/PGISWT (29), pCMV/PGISC441A (29), pcDNACOX-1 (31), and pcDNACOX-2 (31) were used, respectively. HEK-293, Caco-2 and CV-1 cells were transfected with DNA using LipofectAMINE. BAEC were transfected using TransIT LT-1 (PanVera Co., Madison, WI). Transfections were performed in a ratio of 1 µg of DNA to 3.0 µl of reagent, and cells were incubated in serum-free medium at 37 °C for 5 h. Subsequently, medium B or RPMI 1640 containing 20% FBS was added, and cells were grown for 12-72 h before analysis of apoptosis.

Measurement of Cell Viability-- Cell viability was determined by trypan blue exclusion. 10 µl of a 0.5% solution of the dye was added to 100 µl of cell suspension (1.0 × 104 cells/100 µl). The suspension was then applied to a hemacytometer, and both viable and nonviable cells were counted. A minimum of 300 cells was counted for each data point.

Assay of DNA Fragments-- A cell death detection enzyme-linked immunosorbent assay (ELISA) (Roche Molecular Biochemicals, Indianapolis, IN) was performed to determine the apoptotic index by detecting nucleosome breakdown, the histone-associated DNA fragments (mono- and oligonucleosomes) generated by apoptotic cells. HEK-293 cells (8.0 × 104 cells/well in 12-well plates) were plated in medium A and grown for 12 h. Cells were washed with PBS, transfected with PGISwt expression vector or mock vector, and cultured for 72 h. The cells were collected along with the floating cells and used to prepare the cytosol fractions. A volume of 10 µl of these cytosolic fractions were incubated in anti-histone antibody-coated wells (96-well plates), and the histones of the DNA fragments were detected by using 2,2'-azino-di-(3-ethylbenzathiazoline sulfonate). The reaction products in each of the 96 wells were read using a Bio-Rad microplate reader (model 3550-UV). To assess DNA ladder formation, low molecular weight DNA was extracted from 1.8 × 106 cells (both floating and adherent) using ApoLadder EX kit (Takara Shuzo Co., Tokyo). The extracted DNA fragments were applied to a 1.5% agarose gel, separated electrophoretically, and visualized with ethidium bromide.

Measurement of Apoptosis-- Apoptotic cells were distinguished by their characteristic morphological changes such as membrane blebbing and cell body shrinkage. HEK-293, CV-1, or Caco-2 cells (2.0 × 105 cells/well in 6-well plates) were plated with the respective serum-supplemented media and grown for 12 h. Cells were washed with PBS, cotransfected with 1 µg of beta -galactosidase expression vector, and 2.5 µg of PGISwt or PGISC441A expression vector and cultured for 5 h in serum-free medium. Subsequently, medium B was added and cells were grown for 60 h. The cells were stained with X-gal (5-bromo-4-chloro-3-indolyl beta -galactoside) solution (PBS containing 5 mM X-gal, 1 mM MgCl2, 10 mM KCl, 5 mM K3Fe(CN)6, 5 mM K4Fe(CN)6-3H2O) at 37 °C for 3 h and scored for apoptotic cells. For measurement of effects of reagents on apoptosis, cells (2.0 × 105 cells/well in 6-well plates) cotransfected with 1 µg of beta -galactosidase and 2.5 µg of PGISwt or PGISC441A were cultured in serum-free medium containing several concentrations of iloprost, dbcAMP, H-7, or RpcAMP for 24 h at 37 °C, washed gently with PBS, and stained with X-gal. For analysis of effects of PPARdelta on apoptosis, HEK-293 cells (8.0 × 104 cells/well in 12-well plates) were cotransfected with 0.1 µg of beta -galactosidase expression vector and 0.7 µg of expression vector for PPARdelta or mutant PPARdelta and PGISwt, PGISC441A, or mock vector using LipofectAMINE. The total amount of DNA was kept at 2 µg by the addition of control DNA (pCMV-7). After 24 h of transfection, cells were stained with X-gal and apoptosis was measured. The cells underwent apoptosis were indicated by a percentage of the total number of the blue cells stained by X-gal. A minimum of 300 cells was counted for each data point.

Measurement of CPP-32 Activity-- Apoptosis was monitored with the ApoAlert caspase (CPP-32) assay kit using the (acetyl-L-aspartyl-L-glutamyl-L-valyl-L-aspartic acid alpha -(4-trifluoromethyl-coumacyl-7-amido)) substrate (CLONTECH, Palo Alto, CA). HEK-293 cells (3.0 × 105 cells/60-mm dish) were transfected with 22 µg of antisense or sense oligonucleotide for PPARdelta using HVJ liposome (5), incubated for 48 h at 37 °C, and cotransfected with 3 µg of expression vector for PGISwt, PGISC441A, or mock vector in the presence of additional 3 µg of antisense or sense oligonucleotide for PPARdelta . After 36 h, CPP-32 activity was determined according to the manufacturer's instructions using (acetyl-L-aspartyl-L-glutamyl-L-valyl-L-aspart-1-al)-pretreated lysate as a control.

Immunoprecipitation of Cox-1-- Monolayers of HEK-293 cells in a 225-cm2 flask (3.0 × 106 cells) were washed three times with PBS, pH 7.4, and suspended in 1 ml of PBS, pH 7.4, containing 10 mM EDTA, 0.1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 10 µM leupeptin, 10 µg/ml antipain, 10 µM pepstatin (buffer A), and sonicated three times at 150 watts for 5 s on ice. The homogenate was centrifuged at 10,000 × g for 10 min, and the supernatant was centrifuged at 100,000 × g at 4 °C for 1 h. Pellets were suspended in 0.5 ml of ice-cold buffer A containing 1% Tween 20, sonicated three times at 150 watts for 15 s on ice, and centrifuged at 100,000 × g at 4 °C for 1 h. The solubilized protein (800 µg) was incubated with protein G-Sepharose (150 µl) preincubated with control IgG for 2 h at 4 °C as a preclearing step. After pelleting the protein G-Sepharose, COX-1 polyclonal IgG or control IgG coupled to protein G-Sepharose was added to a half of resulting supernatant and incubated for 3 h at 4 °C. Protein G-Sepharose was then washed five times with 1 ml of the same buffer, mixed with 50 µl of sample buffer under nonreducing condition, heated for 5 min at 100 °C, subjected to 10% SDS-polyacrylamide gel electrophoresis, and transferred onto Immobilon-P as described above. Membrane was then subjected to immunoblotting analysis using COX-1 polyclonal antiserum.

Assay of 6-Keto-PGF1alpha and cAMP-- PGI2 production was determined in HEK-293 cells (1.0 × 106/100-mm dish) cotransfected with expression vector for PGISwt, PGISC441A, or mock vector with or without COX-1 or COX-2 vector. After 24 h of transfection, medium was removed and serum-free medium containing 0 or 10 µM arachidonic acid was added. After 48 h, incubation media were collected and stored at -20 °C before assay. Amounts of 6-keto-PGF1alpha were measured using an ELISA kit (Cayman Chemical Co.).

For cAMP assay, HEK-293 or BAEC (2.0 × 105 cells/well in 6-well plates) were washed once in serum-free DMEM, followed by incubation with several concentrations of iloprost or cPGI for 24 h at 37 °C. Reactions were terminated by aspiration of the medium and addition of 2 ml of 5% trichloroacetic acid. The cells were cooled to 4 °C, cAMP was extracted with extraction buffer (65% ethanol containing 5 mM isobutylmethylxanthine, Sigma), and lysates were cleared by spinning at 10,000 × g for 5 min and dried under vacuum. Samples were reconstituted in assay buffer, and cAMP was quantified using an ELISA kit (Amersham Pharmacia Biotech).

Distribution of [3H]Iloprost in HEK-293 Cells-- Distribution of PGI2 analog, iloprost, in HEK-293 cells was measured using [3H]iloprost. HEK-293 cells (1.0 × 106 cells/100-mm dish) were cultured in medium A containing 10 nM [3H]iloprost (0.05 µCi/ml) for 0-180 min. Cells were washed with ice-cold PBS, harvested, and homogenized using a Dounce homogenizer in three volumes of homogenization buffer (10 mM Hepes, pH 7.5, 250 mM sucrose, 0.1 mM EDTA, 1.5 mM dithiothreitol, 10 µg/ml trypsin inhibitor, 10 µg/ml leupeptin, 2 µg/ml aprotinin, and 1.0 mg/ml phenylmethylsulfonyl fluoride). Homogenates were centrifuged at 1450 × g for 10 min. The supernatant was used as the cytoplasm-containing fraction. Nuclei and plasma membranes were isolated from the resulting pellets. The pellet was resuspended in ice-cold homogenization buffer. The solution containing the resuspended pellet was adjusted to a final concentration of 1.6 M sucrose by addition of homogenization buffer with a high density sucrose solution. A two-layer step gradient was set up by layering 250 mM sucrose homogenization buffer over the 1.6 M sucrose suspension and centrifuged at 70,900 × g for 60 min. The band at the gradient interface (250 mM and 1.6 M sucrose) was collected and used as the plasma membrane fraction. The pellet produced after the gradient centrifugation containing the nuclear fraction was collected and used for the nuclei fraction. The [3H]iloprost-derived radioactivity of each fraction or whole cells was determined by liquid scintillation counting.

Oligonucleotides and Luciferase Assay-- Oligonucleotides dS, 5'-AAGAGGAGGAGAAAGAGGA-3' corresponding to the human PPARdelta cDNA (32) sense sequence (nucleotide residues 38-56); dAS, 5'-TCCTCTTTCTCCTCCTCTT-3', the antisense sequence; aS, 5'-CTCGGTGACTTATCCTGTG-3' corresponding to the human PPARalpha cDNA sense sequence (nucleotide residues 237-255); and the antisense sequence aAS, 5'-CACAGGATAAGTCACCGAG-3', were synthesized. These oligonucleotides were transfected into cells using the HVJ liposome method (3). Each oligonucleotide (22 µg) was mixed with a nuclear protein, high mobility group-1. HVJ liposomes were prepared by mixing dried lipids (phosphatidylserine/phosphatidylcholine/cholesterol, 1:4.8:2, w/w/w) with UV light-inactivated HVJ virus. After an incubation and sucrose gradient centrifugation, the top layer was collected and used for transfection. After 48 h of transfection the cells were used for apoptosis assay and PPAR-responsive element (PPRE) reporter assay. The PPREx3-luciferase reporter vector, which contains three copies of PPRE for hydroxymethylglutaryl-CoA reductase (33), was constructed by synthesis of the sense oligonucleotide, 5'-CGCGTAAAAACTGGGCCAAAGGTCTAAAAACTGGGCCAAAGGTCTAAAAACTGGGCCAAAGGTCTC-3' and the antisense oligonucleotide 5'-TCGAGAGACCTTTGGCCCAGTTTTTAGACCTTTGGCCCAGTTTTTAGACCTTTGGCCCAGTTTTTA-3'. Both oligonucleotides were annealed and subcloned into the MluI-XhoI site of the pGL3-promoter vector (Promega, Madison, WI). BAEC (8.0 × 104 cells/well in 12-well plates) were cotransfected with the 0.2 µg of PPREx3-luciferase reporter vector and 0.1 µg of beta -galactosidase expression vector using TransIT LT-1. After 12 h of transfection, cells were washed with PBS, incubated in serum-free DMEM containing several concentrations of iloprost or cPGI for 24 h at 37 °C, and subjected to luciferase assay. HEK-293 cells (8.0 × 104 cells/well in 12-well plates) treated with antisense or sense oligonucleotide were cotransfected with 0.2 µg of PPREx3-luciferase reporter vector, 0.1 µg of beta -galactosidase expression vector, and 0.7 µg of expression vector for PGISwt, PGISC441A, or mock vector using LipofectAMINE. Nontreated HEK-293 cells (8.0 × 104 cells/well in 12-well plates) were cotransfected with 0.2 µg of PPREx3-luciferase reporter vector and 0.1 µg of beta -galactosidase expression vector with 0.7 µg of expression vector for PPARdelta or mutant PPARdelta and PGISwt, PGISC441A, or mock vector using LipofectAMINE. The total amount of DNA was kept at 2 µg by the addition of control DNA (pCMV-7). After another 24 h, the cells were harvested and the luciferase and beta -galactosidase activities were quantified. The luciferase activity of the extract was normalized with beta -galactosidase activity.

Construction of Expression Vectors for PPARdelta and PPARdelta L431A/G434A-- The cDNA for PPARdelta was isolated by PCR amplification. The poly(A)+ RNA was extracted from 2.0 × 106 cells of mouse vascular smooth muscle-derived SVS30 cells using a FastTrack 2.0 kit (Promega). The poly(A)+ RNA was reverse-transcribed by extension with Superscript reverse transcriptase (Life Technologies, Inc.) and a specific antisense primer complementary to the sequence located in the 3'-region of the cDNA for PPARdelta : 5'-TTAGTACATGTCCTTGTAGATTTC-3'; the reverse-transcribed cDNA was used for the following PCR amplification. The PPARdelta cDNA was amplified using the primers Pdw5 (5'-GAAAGCTTGTCGACCCACCATGGAACAGCCACAG-3') and Pdw3 (5'-GTCTAGAGGATCCTTAGTACATGTCCTTGTAGAT-3'). Pdw5 or Pdw3 has HindIII or BamHI restriction site (underlined, respectively). PCR was performed for 35 cycles with a temperature profile of 10 s at 94 °C, 30 s at 55 °C, and 4 min at 72 °C using KOD polymerase (Toyobo, Osaka, Japan). The amplification products were purified in 1% agarose gel using a Qiagen gel extraction kit. The isolated DNA was inserted into the HindIII-BamHI site of pBluescript II vector (Stratagene, La Jolla, CA) and sequenced using an Applied Biosystems model 310 genetic analyzer. For construction of an expression vector for PPARdelta , PCR-derived cDNA for PPARdelta was excised from the pBluescript vector with HindIII and BamHI and religated into the pcDNAIII vector (Invitrogen, Groningen, The Netherlands). The PPARdelta L431A/G434A double mutant was generated by site-directed mutagenesis of wild-type receptor. The oligonucleotide used in mutagenesis was Pdm3 (5'-GTACATGTCCTTGTAGATCGCCTGGAGCGCGGGGTGCAGC-3'), and the construct was verified by sequencing. Pdm3 has two mutated sites (underlined). The products were purified, subcloned into the HindIII-BamHI site of pBluescript II, and inserted into the HindIII-BamHI site of pcDNAIII vector.

    RESULTS AND DISCUSSION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS AND DISCUSSION
REFERENCES

An expression vector encoding human PGIS (29) or control vector was transfected into HEK-293 cells, and cell viability was determined. As shown in Fig. 1A, we observed a significant decrease in the viability of the cells transfected with the PGIS expression vector, whereas no change in viability was seen in the cells transfected with a mock vector. Moreover, transfection of the PGIS expression vector into HEK-293 cells resulted in significant morphological changes, including membrane blebbing and cell body shrinkage (Fig. 1B, left panel), features typical of apoptosis (34). When the histone-associated DNA fragments (mono- and oligonucleosomes) in cytosol fractions of the cells were measured by ELISA, the amounts of DNA fragments in the cells transfected with the PGIS expression vector increased gradually after 36 h of transfection (Fig. 1C). However, cell morphology and contents of DNA fragments were not changed in cells transfected with the mock vector (Fig. 1, B and C). These data indicate that transfection with the PGIS expression vector causes a decline in cell viability and an increase in apoptotic cell death in HEK-293 cells. There is no detectable endogenous PGIS in HEK-293 cells (29). A significant amount of endogenous COX-1, which synthesizes PGH2, a substrate for PGIS, was detected in the HEK-293 cells by immunoprecipitation (Fig. 1D), although it has been reported that expression of both COX-1 and -2 in the cells is undetectable by immunoblotting (35). As shown in Fig. 1E, overexpression of COX-1 or COX-2 alone in the HEK-293 cells did not produce apoptotic changes, suggesting that expression of PGIS is required for the apoptotic morphological changes in these cells. In contrast, overexpression of either COX-1 or COX-2 in bovine aortic endothelial cells (BAEC), which constitutively express PGIS at relatively high levels, increased apoptotic cell death significantly. It has been reported that overexpression of COX-2 in a gastrointestinal epithelial cell line, in which PGE2 is the major metabolite of PGH2 (36), is associated with inhibition of apoptosis (37). Overexpression of COX-2 in immortalized vascular endothelial cells, in which PGI2 is a major metabolite of PGH2, slows growth and increases cell death (38). COX-1 overexpression in the immortalized vascular endothelial cells inhibited the growth rate and enhanced apoptosis induced by TNF, which induces expression of COX-2 and PGIS (30, 37). These lines of evidence all suggest that PGI2 induces apoptotic cell death. The relationship between apoptosis and PGIS gene expression, however, has been unknown.


View larger version (14K):
[in this window]
[in a new window]
  		Fig. 1.   Time course of viability and apoptotic changes in HEK-293 cells transfected with a human PGIS expression vector. The human PGIS expression plasmid (pCMV/PGISWT) (closed circles) or control vector (pCMV-7) (open circles) was transfected into HEK-293 cells. Thereafter, cell viability was determined by trypan blue exclusion (A). HEK-293 cells were cotransfected with beta -galactosidase and PGISwt expression vectors (B, left panel) or mock vector (right panel), stained with X-gal after 60 h, and observed by light microscopy (magnification, × 100). The nucleosome breakdown of the cells was measured using an enzyme immunoassay for cytoplasmic histone-associated DNA fragments at the times indicated (C). Endogenous COX-1 protein expressed in intact HEK-293 cells was immunoprecipitated and detected by immunoblot using anti-COX-1 IgG (D). HEK-293 cells or BAEC were cotransfected with beta -galactosidase expression vector and expression vector for PGIS, COX-1, COX-2, or mock vector (MOC), and stained with X-gal after 60 h of transfection. X-gal-positive cells that underwent an apoptotic morphological change were counted (E). Results represent the mean ± S.D. of three experiments.

To examine the possibility that PGIS is involved in inducing apoptosis, the expression vector for wild-type human PGIS (PGISwt) or the catalytically inactive mutant PGISC441A (29) was transfected into HEK-293 cells under the same conditions. Immunofluorescence staining was performed using an anti-PGIS polyclonal antibody to confirm expression of these PGISs in the cells. At the same time, the condensation of chromatin, a typical morphological change associated with apoptosis, was measured using the fluorochrome bisbenzimide (Hoechst 33258) dye staining method (39). As shown in Fig. 2, A and B, the cells expressing PGISwt were stained coincidentally and specifically by Hoechst 33258, and the genomic DNA extracted from these cells showed laddering (Fig. 2G). Although the mutant PGISC441A protein was expressed at levels similar to those of the native enzyme, the PGIS-positive cells expressing PGISC441A protein were not stained by Hoechst 33258 (Fig. 2, C and D). The increase in the number of apoptotic Hoechst 33258-positive cells was time-dependent (Fig. 2H). The number of Hoechst 33258-positive cells relative to the number of cells expressing native PGIS was decreased by 32% when the cells were treated with 100 µM U46619, an inhibitor of PGIS (40), or by 48% when cells were treated with 100 µM aspirin, a COX inhibitor (Fig. 2H). These partial inhibitions may reflect the apoptosis-inducible activities by U46619 and aspirin (41, 42). Additionally, when the cells were transfected with increasing amounts of the PGISwt expression vector, both the amount of PGIS protein expressed in the cells and the numbers of apoptotic cells were increased concomitantly (Fig. 2I); in contrast, when the mutant PGISC441A expression vector was transfected into HEK-293 cells under the same conditions, the number of apoptotic cells did not increase, although the level of expression of PGISC441A protein was the same as that of PGISwt. Essentially identical results were obtained with CV-1 and Caco-2 cells (data not shown). These results establish that the increase of apoptosis of HEK-293 cells requires expression of catalytically active PGIS.


View larger version (20K):
[in this window]
[in a new window]
  		Fig. 2.   Effects of PGISwt and PGISC441A expression on apoptosis in HEK-293 cells. Monolayers of HEK-293 cells were transfected with either the PGISwt (panels A, B), PGISC441A (panels C, D), or mock vector (panels E, F). 36 h later, the cells were fixed in 3.7% formaldehyde for 10 min, incubated with anti-PGIS antibody P4, incubated with anti-rabbit IgG-Texas Red-conjugated second antibody, stained with 0.3 mM Hoechst 33258 for 15 min at room temperature, and observed under fluorescence microscopy (magnification, × 200). G, DNA laddering of the cells transfected with PGISwt or PGISC441A expression vector was analyzed 60 h after transfection. Lane 1, marker; lane 2, extracted DNA from PGISC441A-expressing cells; lane 3, extracted DNA from PGISwt-expressing cells. H, the number of apoptotic Hoechst 33258-positive cells expressing PGISwt or PGISC441A protein was counted (open circle, HEK-293 transfected with PGISwt; open square, cells transfected with PGISwt and treated with 100 µM U46619; closed square, cells transfected with PGISwt and treated with 100 µM aspirin; closed circle, cells transfected with PGISC441A). I, the HEK-293 cells were transfected with increasing amounts (0, 1, 2, and 3 µg) of the PGISwt (WT) or PGISC441A (C441A) expression vector, and apoptotic cells were measured. Expressed PGISs were detected by immunoblotting using anti-PGIS polyclonal antibody P1 (upper panel). Results represent the mean ± S.D. of three experiments.

PGI2 production by the cells transfected with PGISwt expression vector was measured using an ELISA. The amount of 6-keto-PGF1alpha , the stable hydrolysis product of PGI2, released into the culture medium was correlated with the frequency of apoptosis (Fig. 3, A and B). Cotransfection of COX-1 plasmid with PGISwt into the cells and/or treatment of the cells with arachidonic acid resulted in increase of PGI2 production and apoptosis. Furthermore, cotransfection of COX-2 into the cells with the PGISwt also increased PGI2 production and apoptosis. Addition of arachidonic acid to the cells enhanced both PGI2 production and apoptosis. When the PGISC441A expression vector was used in a parallel experiment, neither PGI2 production nor apoptosis was increased in the cells.


View larger version (22K):
[in this window]
[in a new window]
  		Fig. 3.   Correlation between PGI2 production and increase of apoptosis in HEK-293 cells. HEK-293 cells transfected with expression vectors for PGISwt, PGISC441A, or mock vector were cotransfected with COX-1 or COX-2 expression vector and cultured for 48 h in the presence or absence of arachidonic acid (AA). The amounts of PGI2 in the culture media were measured as 6-keto-PGF1alpha (A), and apoptosis was measured as described under "Experimental Procedures" (B). Results represent the mean ± S.D. of three experiments.

Overexpression of COX-1 or COX-2 in BAEC, which express PGIS constitutively, increased apoptosis (Fig. 4). There was essentially no difference in induction of apoptosis between coexpression with COX-1 or COX-2. These findings also indicate that PGI2 produced by PGIS is involved in the process of apoptosis.


View larger version (22K):
[in this window]
[in a new window]
  		Fig. 4.   Effect of COX gene transfection on induction of apoptosis in BAEC. BAEC were transfected with PGISwt, PGISC441A, COX-1, or COX-2 expression vector. After 60 h of transfection, apoptosis of each cell was measured as described under "Experimental Procedures." Mock vector was used as a control. Results represent the mean ± S.D. of three experiments.

The next question we addressed is which signaling cascade contributes to apoptosis induced by PGI2. We examined the effect of PGI2 on the cells using iloprost, a stable PGI2 analog. When the cells were incubated with [3H]iloprost, the radioactivity was recovered exclusively in the plasma membrane fractions (Fig. 5). Treatment of the cells expressing PGISwt or PGISC441A with various concentrations of iloprost (0, 1, 10, 100 µM) did not induce or enhance apoptosis (Fig. 6A) but did cause increases in cAMP at higher concentrations (Fig. 6B). Both extracellular iloprost and PGI2 probably raise the intracellular concentrations of cAMP by activating the PGI2 receptor IP and/or perhaps the prostaglandin E receptors EP2 (43) and EP4 (44). Essentially no expression of IP mRNA was detected in HEK-293 cells by reverse transcriptase-PCR, however, it has been reported that EP receptors are expressed in HEK-293 cells and that PGE1 raises the intracellular concentration of cAMP (45). Thus, it is likely that high concentrations of iloprost can stimulate cAMP accumulation through EP2 and/or EP4 receptors in the cells. Nevertheless, the number of cells undergoing apoptosis did not increase in response to iloprost nor did treatment of the cells with dibutyryl cyclic AMP (dbcAMP) promote apoptosis (Fig. 6C). In fact, both iloprost and dbcAMP inhibited apoptosis of these cells somewhat (Fig. 6, A and C).


View larger version (18K):
[in this window]
[in a new window]
  		Fig. 5.   Distribution of [3H]iloprost in HEK-293 cells. Distribution of the PGI2 analog, iloprost, was measured using [3H]iloprost. HEK-293 cells were cultured in medium A containing 10 nM [3H]iloprost for 0-180 min. At the indicated times, cells were harvested and plasma membrane (open circle), cytosol (closed circle), and nuclei (open square) fractions of the cells were prepared. The [3H]iloprost-derived radioactivity of each fraction or whole cells was determined by scintillation counting. Each count of fractions was quantified as a percentage of the total count found in whole cells. Results represent the mean ± S.D. of three experiments.


View larger version (35K):
[in this window]
[in a new window]
  		Fig. 6.   Effects of iloprost, dbcAMP, H-7, and RpcAMP on PGI2-mediated apoptosis. HEK-293 cells cotransfected with the beta -galactosidase expression vector and PGISwt (WT) or PGISC441A (C441A) expression vectors were treated with iloprost, dbcAMP, H-7, or RpcAMP for 24 h and stained with X-gal. The apoptotic cells are indicated by a percentage of the total number of blue cells stained with X-gal. Effects of concentrations (0, 1, 10, 100 µM) of iloprost on apoptosis (A) and the accumulation of cAMP (B) in the HEK-293 cells are indicated. Data are expressed as cAMP per milligram of total protein. HEK-293, CV-1, or Caco-2 cells transfected with PGISwt or PGISC441A vector were treated with dbcAMP (0, 1, 10, 100 µM), H-7 (0, 0.1, 0.5, 1 µM), or RpcAMP (0, 0.5, 1, 10 µM), and the effects of dbcAMP, H-7, or RpcAMP on apoptosis were measured (C). Results represent the mean ± S.D. of three experiments.

Apoptosis is usually associated with activation of protein kinases (46). To examine the correlation between protein kinase pathways and PGI2-mediated apoptosis, cells were treated with the protein kinase inhibitor H-7, which is an equipotent inhibitor of cAMP-dependent protein kinase, cGMP-dependent protein kinase, and lipid-dependent protein kinase C (47). H-7 did not block but actually enhanced somewhat PGI2-mediated apoptosis in HEK-293, CV-1, and Caco-2 cells (Fig. 6C) suggesting that all of these three types of protein kinases are involved in the induction of PGI2-induced apoptosis. Moreover, RpcAMP, an antagonist of cAMP, significantly enhanced PGI2-mediated apoptosis in these cells. Our data suggest that there is a novel PGI2 signaling pathway that induces apoptosis independent of IP and EP receptor signaling.

PPARs are nuclear hormone receptors and, as such, are ligand-activated transcription factors (15). Ligands for these receptors include hypolipidemic agents such as clofibrate, various fatty acids, and some arachidonic acid metabolites. PPARs activate their target genes upon binding to PPAR-response elements (PPREs) in promoters of target genes. PPARalpha is highly expressed in hepatocytes, cardiomyocytes, enterocytes, and proximal tubule cells of the kidney (48). PPARalpha can inhibit apoptosis in hepatocytes (49) or can promote apoptosis in human macrophages activated with tumor necrosis factor-alpha (25). PPARdelta is expressed ubiquitously and often at higher levels than either PPARalpha or PPARgamma (50). Iloprost, carbaprostacyclin, and PGI2 can activate recombinant PPARdelta when it is overexpressed in CV-1 (51) or colorectal cancer-derived U2OS cells (28). PPARdelta plays a central role in fatty acid-controlled differentiation of preadipose cells (20) and in embryo implantation and decidualization (30). Moreover PPARdelta , with its broad binding specificity, may have wide-ranging activities associated with the maintenance of molecular and cellular homeostasis such as body size, myelination of the corpus callosum, and epidermal cell proliferation (52). It has also been reported that inhibition of PPARdelta activity by nonsteroidal anti-inflammatory drugs suppressed tumorigenesis (22). However, the relationship between apoptosis and PPARdelta activated by endogenous PGI2 has not been resolved.

To determine if PPARs are involved in apoptosis induced by intracellular PGI2, the effects of antisense oligonucleotides for PPARalpha and PPARdelta on PGI2 signaling were examined. When a PPARalpha antisense oligonucleotide was transfected into HEK-293 cells expressing PGISwt, expression of PPARalpha was suppressed (Fig. 7A) and apoptosis was not inhibited (Fig. 7B, left panel). The cells transfected with the PPARalpha sense oligonucleotide also showed no marked changes in apoptosis. These results suggest that PPARalpha may play an important role in maintaining the viability of HEK-293 cells, and that PGI2-mediated apoptosis is not promoted through PPARalpha . These findings suggested to us that endogenous PPARdelta might be a second PGI2 receptor that acts as the key signaling protein in PGI2-mediated apoptosis. To investigate this directly, we examined HEK-293 cells transfected with a PPARdelta antisense oligonucleotide. After 48 h, suppression of the PPARdelta protein expression was confirmed by immunoblotting (Fig. 7A). The morphology of the cells treated with the PPARdelta antisense oligonucleotide was normal, and the number of cells was increased. When PGISwt expression vector was cotransfected along with the antisense oligonucleotide for PPARdelta , PGI2-mediated apoptosis was reduced significantly (Fig. 7B, right panel), as evidenced by a decrease in apoptotic cells bearing the antisense oligonucleotide. As shown in Fig. 7C, the luciferase activity of the PPRE-Luc reporter vector coexpressed with PGISwt in the HEK-293 cells, which were treated with the PPARdelta antisense oligonucleotide, was decreased significantly. Moreover, the decrease in luciferase activity was correlated with the decrease in caspase activity (Fig. 7D).


View larger version (30K):
[in this window]
[in a new window]
  		Fig. 7.   Suppression of endogenous PPARdelta using antisense oligonucleotide inhibits PGI2-mediated apoptosis. Endogenous expression of PPARalpha or PPARdelta protein in HEK-293 cells was suppressed using the antisense oligonucleotide (A) (-, nontreated; S, treated with sense oligonucleotide; AS, treated with antisense oligonucleotide). The HEK-293 cells pretreated with sense or antisense oligonucleotides of PPARalpha (alpha sense or alpha antisense) or PPARdelta (delta sense or delta antisense) were transfected with expression vectors for PGISwt or PGISC441A or a mock vector. The effects of the sense oligonucleotide for PPARalpha or PPARdelta or the antisense oligonucleotide for PPARalpha or PPARdelta on PGI2-mediated apoptosis induced in HEK-293 were measured (B) as described under "Experimental Procedures." The effects of the antisense oligonucleotide for PPARdelta on luciferase activity of PPREx3 reporter vector (C) and caspase (CPP-32) activity (D) in the HEK-293 were measured. Results represent the mean ± S.D. of three experiments.

To test whether PPARdelta activation by intracellular PGI2 is necessary for PGI2-mediated apoptosis, a mutant PPARdelta was generated by site-directed mutagenesis. The conserved hydrophobic and charged residues (Leu431 and Glu434) in the putative ligand-binding domain of PPARs (53) were both mutated to alanine (Fig. 8A). It has been reported that a dominant-negative mutant of PPARgamma was obtained by double mutation at positions Leu468 and Glu471 (53), which correspond to Leu431 and Glu434 in PPARdelta . As shown in Fig. 8B, the mutant PPARdelta exhibited markedly reduced transactivation, thus, this mutant functions as a dominant-negative inhibitor of wild-type receptor. As shown in Fig. 8, B and C, expression of PPARdelta L431A/G434A with PGISwt antagonized the PGI2-mediated PPARdelta activation and apoptosis was blocked completely. These results were consistent with the concept that PGI2-mediated apoptosis depends on activation of PPARdelta by PGI2.


View larger version (25K):
[in this window]
[in a new window]
  		Fig. 8.   PPARdelta L431A/E434A inhibits PGI2-mediated apoptosis. The conserved hydrophobic and charged residues Leu431 and Glu434 of PPARdelta corresponding to Leu468 and Glu471 of PPARgamma (53) were mutated to alanine (A). HEK-293 cells were cotransfected with the PPRE reporter vector and an expression vector for PGISwt or PGISC441A or a mock vector and with a wild-type PPARdelta (delta WT) or PPARdelta L431A/E434A (delta L431A/E434A) expression vector. Transcriptional activity in response to PGISwt expression was measured by luciferase assays 24 h after transfection (B). The PGI2-mediated apoptosis of HEK-293 cells cotransfected with the expression vector for wild-type PPARdelta or PPARdelta L431A/E434A and PGISwt or PGISC441A or a mock vector was measured as described under "Experimental Procedures" (C). Results represent the mean ± S.D. of three experiments.

We have shown that activation of endogenous PPARdelta by intracellular PGI2 or a metabolite of PGI2 results in activation of the apoptosis pathway. This, in turn, suggests that PGI2 can interact with both a nuclear PPARdelta and a cell surface IP (or EP) receptor coupled to increases in cAMP. These signaling pathways have opposing biological effects on cell apoptosis and/or viability, respectively. One question arises from our studies, Why don't endothelial cells and vascular smooth muscle cells expressing PGIS endogenously undergo apoptosis? One possibility is that the IP receptor/cAMP/protein kinase pathway serves to protect these cells from PGI2-mediated apoptosis primarily, because vascular endothelial cells and vascular smooth muscle cells express IP receptors, which can, in turn, lead to increases in cAMP levels at low concentrations of PGI2. In contrast, cells such as HEK-293 cells that lack the IP receptor easily undergo PGI2-mediated apoptosis. To verify this possibility, HEK-293 and BAEC were treated with a specific cAMP antagonist, RpcAMP. As shown in Fig. 9A, BAEC were highly sensitive to RpcAMP, and apoptosis was increased significantly in the cells treated with RpcAMP. However, HEK-293 cells did not show any in the degree of apoptosis when exposed to the same conditions. As shown in Fig. 9B, a membrane-permeable PGI2 analog, carbaprostacyclin (cPGI), activated PPRE-Luc activity (left panel) with a concomitant accumulation of intracellular cAMP (right panel) in BAEC. In contrast, iloprost did not increase luciferase activity, although the intracellular concentration of cAMP was increased. As shown in Fig. 9C, cPGI and iloprost did not induce apoptosis in BAEC, but cPGI strongly enhanced apoptosis in the presence of RpcAMP, which was not observed with membrane-impermeable iloprost. These data support the possibility that cAMP produced by the PGI2-IP-cAMP pathway serves to protect vascular endothelial cells from intracellular PGI2-PPARdelta -mediated apoptosis. In addition, there may be intricate mechanisms for controlling cell fate involving PGI2, PPARdelta , and the G protein-coupled PGI2 receptors. Further characterization of the PGI2 signaling cascade, including the PPARdelta pathway, is underway in our laboratory.


View larger version (23K):
[in this window]
[in a new window]
  		Fig. 9.   Effect of cAMP antagonist RpcAMP on apoptosis in BAEC. BAEC or HEK-293 cells (8.0 × 104 cells/well in 12-well plates) were cultured in serum-free DMEM containing RpcAMP (0, 0.1, 1, 10, 100 µM) for 24 h at 37 °C, and the nucleosomes released during apoptosis of the cells were measured as described under "Experimental Procedures" (A). BAEC transfected with PPRE-luciferase reporter vector were cultured in serum-free DMEM containing cPGI (0, 1, 10, 100, 500 nM) or iloprost (0, 1, 10, 100, 500 nM) for 24 h at 37 °C, and luciferase activity was measured (B, left panel). The intracellular cAMP accumulation in BAEC cultured in serum-free DMEM containing cPGI (0, 1, 10, 100, 500 nM) or iloprost (0, 1, 10, 100, 500 nM) for 24 h at 37 °C was measured (B, right panel). BAEC were treated with cPGI (0, 1, 10, 100, 500 nM) or iloprost (0, 1, 10, 100, 500 nM) in the presence or absence of 100 nM RpcAMP, and apoptosis was measured (C). Results represent the mean ± S.D. of three experiments.


    ACKNOWLEDGEMENTS

We thank Tamiko Sugimoto, Setsuko Bandoh, Kyoko Sasagawa, and Mari Yamada for their technical assistances and Drs. Y. Kaneda and M. Aoki for providing the HVJ virus and liposomes.

    FOOTNOTES

* This study was supported by grants from the Ministry of Health, Welfare and Labor and the Ministry of Education, Culture, Sports, Science and Technology of Japan and by the Takeda Medical Research Foundation.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.

Dagger To whom correspondence should be addressed: Tel.: 81-6-6833-5012 (ext. 2514); Fax: 81-6-6872-8090; E-mail: tanabe@jsc.ri.ncvc.go.jp.

Published, JBC Papers in Press, September 10, 2001, DOI 10.1074/jbc.M107180200

    ABBREVIATIONS

The abbreviations used are: PGs, prostaglandins; PGI2, prostacyclin; PGIS, PGI2 synthase; COX, cyclooxygenase; HEK-293, human embryonic kidney 293 cells; PPAR, peroxisome proliferator-activated receptor; PPRE, PPAR-response element; BAEC, bovine aortic endothelial cells; dbcAMP, dibutyryl cyclic AMP; PBS, phosphate-buffered saline; TBS, Tris-buffered saline; DMEM, Dulbecco's modified Eagle's medium; FBS, fetal bovine serum; ELISA, enzyme-linked immunosorbent assay; X-gal, 5-bromo-4-chloro-3-indolyl beta -galactoside; PCR, polymerase chain reaction; AA, arachidonic acid; cPGI, carbaprostacyclin; RpcAMP, adenosine 3',5'-cyclic monophosphothioate, Rp-isomer; H-7, 1-(5-isoquinolinesulfonyl)-2-methyl-piperazine; HVJ, hemagglutinating virus Japan; PGH2, prostaglandin endoperoxide H2.

    REFERENCES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS AND DISCUSSION
REFERENCES

1. Smith, W. L., DeWitt, D. L., and Garavito, R. M. (2000) Annu. Rev. Biochem. 69, 145-182[CrossRef][Medline] [Order article via Infotrieve]
2. Tanabe, T., and Ullrich, V. (1995) J. Lipid Mediat. Cell Signal. 12, 243-255[CrossRef][Medline] [Order article via Infotrieve]
3. Sugimoto, Y., Narumiya, S., and Ichikawa, A. (2000) Prog. Lipid Res. 39, 289-314[CrossRef][Medline] [Order article via Infotrieve]
4. Samuelsson, B., Goldyne, M., Granstrom, E., Hamberg, M., Hammarstrom, S., and Malmsten, C. (1978) Annu. Rev. Biochem. 47, 997-1029[CrossRef][Medline] [Order article via Infotrieve]
5. Todaka, T., Yokoyama, C., Yanamoto, H., Hashimoto, N., Nagata, I., Tsukahara, T., Hara, S., Hatae, T., Morishita, R., Aoki, M., Ogihara, T., Kaneda, Y., and Tanabe, T. (1999) Stroke 30, 419-426[Abstract/Free Full Text]
6. Narumiya, S., Sugimoto, Y., and Ushikubi, F. (1999) Physiol. Rev. 79, 1193-1226[Abstract/Free Full Text]
7. Moncada, S., and Vane, J. R. (1979) N. Engl. J. Med. 17, 1142-1147
8. Smith, E. M., Nestor, P. V., and FitzGerald, G. A. (1996) J. Biol. Chem. 271, 33698-33704[Abstract/Free Full Text]
9. Vane, J. R., and Botting, R. M. (1995) Am. J. Cardiol. 75, 3A-10A[CrossRef][Medline] [Order article via Infotrieve]
10. Dohlman, H. G. (1993) Nature 366, 307-308[Medline] [Order article via Infotrieve]
11. Ushikubi, F., Hirata, M., and Narumiya, S. (1995) J. Lipid Mediat. Cell Signal. 12, 343-359[CrossRef][Medline] [Order article via Infotrieve]
12. Forman, B. M., Tontonoz, P., Chen, J., Brun, R. P., Spiegelman, B. M., and Evans, R. M. (1995) Cell 83, 803-812[CrossRef][Medline] [Order article via Infotrieve]
13. Prescott, S. M., and White, R. L. (1996) Cell 87, 783-786[CrossRef][Medline] [Order article via Infotrieve]
14. Lowell, B. B. (1999) Cell 99, 239-242[CrossRef][Medline] [Order article via Infotrieve]
15. Kersten, S., Desvergne, B., and Wahli, W. (2000) Nature 405, 421-424[CrossRef][Medline] [Order article via Infotrieve]
16. Lazar, M. A. (2001) Nat. Med. 7, 23-24[CrossRef][Medline] [Order article via Infotrieve]
17. Altiok, S., Xu, M., and Spiegelman, B. M. (1997) Genes Dev. 11, 1987-1998[Abstract/Free Full Text]
18. Spiegelman, B. M., and Flier, J. S. (1996) Cell 87, 377-389[CrossRef][Medline] [Order article via Infotrieve]
19. Valmaseda, A., Carmona, M. C., Barbera, M. J., Vinas, O., Mampel, T., Iglesias, R., Villarroya, F., and Giralt, M. (1999) Mol. Cell. Endocrinol. 154, 101-109[CrossRef][Medline] [Order article via Infotrieve]
20. Bastie, C., Luquet, S., Holst, D., Jehl-Pietri, C., and Grimaldi, P. A. (2000) J. Biol. Chem. 275, 38768-38773[Abstract/Free Full Text]
21. Saez, E., Tontonoz, P., Nelson, M. C., Alvarez, J. G., Ming, U. T., Baird, S. M., Thomazy, V. A., and Evans, R. M. (1998) Nat. Med. 4, 1058-1061[CrossRef][Medline] [Order article via Infotrieve]
22. He, T. C., Chan, T. A., Vogelstein, B., and Kinzler, K. W. (1999) Cell 99, 335-345[CrossRef][Medline] [Order article via Infotrieve]
23. Diep, Q. N., Touyz, R. M., and Schiffrin, E. L. (2000) Hypertension 36, 851-855[Abstract/Free Full Text]
24. Elstner, E., Muller, C., Koshizuka, K., Williamson, E. A., Park, D., Asou, H., Shintaku, P., Said, J. W., Heber, D., and Koeffler, H. P. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 8806-8811[Abstract/Free Full Text]
25. Chinetti, G., Griglio, S., Antonucci, M., Torra, I. P., Delerive, P., Majd, Z., Fruchart, J. C., Chapman, J., Najib, J., and Staels, B. (1998) J. Biol. Chem. 273, 25573-25580[Abstract/Free Full Text]
26. Bishop-Bailey, D., and Hla, T. (1999) J. Biol. Chem. 274, 17042-17048[Abstract/Free Full Text]
27. Heneka, M. T., Feinstein, D. L., Galea, E., Gleichmann, M., Wullner, U., and Klockgether, T. (1999) J. Neuroimmunol. 100, 156-168[CrossRef][Medline] [Order article via Infotrieve]
28. Gupta, R., Tan, J., Krause, W. F., Geraci, M. W., Willson, T. M., Dey, S. K., and DuBois, R. N. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 13275-13280[Abstract/Free Full Text]
29. Hatae, T., Hara, S., Yokoyama, C., Yabuki, T., Inoue, H., Ullrich, V., and Tanabe, T. (1996) FEBS Lett. 389, 268-272[CrossRef][Medline] [Order article via Infotrieve]
30. Hara, S., Miyata, A., Yokoyama, C., Inoue, H., Brugger, R., Lottspeic, F., Ullrich, V., and Tanabe, T. (1994) J. Biol. Chem. 269, 19897-19903[Abstract/Free Full Text]
31. Kinoshita, T., Takahashi, Y., Sakashita, T., Inoue, H., Tanabe, T., and Yoshimoto, T. (1999) Biochim. Biophys. Acta 1438, 120-130[Medline] [Order article via Infotrieve]
32. Schmidt, A., Vogel, R. L., Witherup, K. M., Rutledge, S. J., Pitzenberger, S. M., Adam, M., and Rodan, G. A. (1992) Mol. Endocrinol. 6, 1634-1641[Abstract]
33. Juge-Aubry, C. (1997) J. Biol. Chem. 272, 25252-25259[Abstract/Free Full Text]
34. Yang, X., Khosravi-Far, R., Chang, H. Y., and Baltimore, D. (1997) Cell 89, 1067-1076[CrossRef][Medline] [Order article via Infotrieve]
35. Murakami, M., Kambe, T., Shimbara, S., and Kudo, I. (1999) J. Biol. Chem. 274, 3103-3115[Abstract/Free Full Text]
36. Tsujii, M., Kawano, S., Tsuji, S., Sawaoka, H., Hori, M., and DuBois, R. N. (1998) Cell 93, 705-716[CrossRef][Medline] [Order article via Infotrieve]
37. Tsujii, M., and DuBois, R. N. (1995) Cell 83, 493-501[CrossRef][Medline] [Order article via Infotrieve]
38. Narko, K., Ristimaki, A., MacPhee, M., Smith, E., Haudenschild, C. C., and Hla, T. (1997) J. Biol. Chem. 272, 21455-21460[Abstract/Free Full Text]
39. Diaz-Meco, M. T., Municio, M. M., Frutos, S., Sanchez, P., Lozano, J., Sanz, L., and Moscat, J. (1996) Cell 86, 777-786[CrossRef][Medline] [Order article via Infotrieve]
40. Zou, M., Martin, C., and Ullrich, V. (1997) Biol. Chem. 378, 707-713
41. Beauchamp, M. H., Martinez-Bermudez, A. K., Gobeil, F., Jr, Marrache, A. M., Hou, X., Speranza, G., Abran, D., Quiniou, C., Lachapelle, P., Roberts, J., 2nd, Almazan, G., Varma, D. R., and Chemtob, S. (2001) J. Appl. Physiol. 90, 2279-2288[Abstract/Free Full Text]
42. Bellosillo, B., Pique, M., Barragan, M., Castano, E., Villamor, N., Colomer, D., Montserrat, E., Pons, G., and Gil, J. (1998) Blood 92, 1406-1414[Abstract/Free Full Text]
43. Kedzie, K. M., Donello, J. E., Krauss, H. A., Regan, J. W., and Gil, D. W. (1998) Mol. Pharmacol. 54, 584-590[Abstract/Free Full Text]
44. Watanabe, A., Sugimoto, Y., Honda, A., Irie, A., Namba, T., Negishi, M., Ito, S., Narumiya, S., and Ichikawa, A. (1993) J. Biol. Chem. 268, 20175-20178[Abstract/Free Full Tex