Microsomal Prostaglandin E Synthase-1 Deficiency Is Associated with Elevated Peroxisome Proliferator-activated Receptor γ

mPGES-1 (microsomal PGE synthase-1) is an inducible enzyme that acts downstream of cyclooxygenase (COX) and specifically catalyzes the conversion of prostaglandin (PG) H2 to PGE2 under basal as well as inflammatory conditions. In this study, using mouse embryo fibroblasts (MEFs) isolated from mice genetically deficient for the mPges-1 gene, we show basal elevation of peroxisome proliferator-activated receptor γ (PPARγ) expression (protein and mRNA) and transcriptional activity associated with reduced basal PGE2. We further show that basal mPGES-1-derived PGE2 suppresses the expression of PPARγ through a cAMP-independent pathway involving phosphatidylinositol 3-kinase and Akt signaling. Using specific PPARγ agonist (rosiglitazone), PPARγ ligand (15-deoxy-Δ12,14-PGJ2), and PPARγ inhibitor (GW9662), we confirm that activation of PPARγ blocks interleukin-1β-induced up-regulation of COX-2, mPGES-1, and their derived PGE2. Furthermore, we demonstrate that up-regulation of PPARγ upon genetic deletion of mPGES-1 is responsible for reduced COX-2 expression under basal as well as interleukin-1β-stimulated conditions. This study provides evidence for the first time that mPGES-1 deletion not only decreases proinflammatory PGE2 but also up-regulates anti-inflammatory PPARγ, which has the ability to suppress COX-2 and mPGES-1 expression and PGE2 production. Thus, mPGES-1 inhibition may limit inflammation by multiple mechanisms and is a potential therapeutic target.

diate substrate, PGH 2 , which is further metabolized by terminal synthases to generate specific PGs (1,2). mPGES-1, originally known as microsomal glutathione S-transferase 1-like 1 (MGST1-L1), is an inducible enzyme that acts downstream of COX and specifically catalyzes the conversion of PGH 2 to PGE 2 (3), most prominently in inflammatory conditions (4,5). However, we have recently shown that mPGES-1 is critical for PGE 2 production under basal as well as inflammatory conditions (6). mPGES-1 is coordinately induced with COX-2 by inflammatory stimuli in a variety of cells and tissues (4,7). PGE 2 is the most abundant PG associated with inflammatory conditions, and overproduction of PGE 2 coincides with increased COX-2 and mPGES-1 expression (4,7). PGE 2 exerts the majority of its actions through a family of G protein-coupled receptors, including EP1, EP2, EP3, and EP4 (8). The effects of PGE 2 via these receptors are mediated through various downstream signaling pathways, including cAMP-dependent protein kinase, mitogen-activated protein kinase (MAP kinase), phosphatidylinositol 3-kinase (PI 3-kinase), and Akt (8 -10).
Inhibition of PGE 2 production and action is associated with reduction of the pain and inflammation associated with a wide variety of diseases. Nonselective and COX-2-selective nonsteroidal anti-inflammatory drugs (NSAIDs) block PGE 2 production by inhibiting the activity of COX and are extensively used to treat arthritis and other inflammatory conditions. However, side effects associated with the inhibition of COX-2 (11)(12)(13) have revived efforts to develop safer anti-inflammatory drugs. mPGES-1 is an attractive target to achieve more specific inhibition of PGE 2 production associated with inflammatory disorders while preserving production of other PGs. Specific inhibitors of mPGES-1 are yet not available; however, studies using mice genetically deficient in mPGES-1 have demonstrated that this enzyme is a key mediator of inflammation, pain, angiogenesis, fever, bone metabolism, and tumorigenesis (14 -17). Our studies also demonstrate that mPGES-1 expression is increased in tissues and cells of various inflammatory conditions, including rheumatoid arthritis and osteoarthritis (4,5,18,19). Previ-ous studies have also shown that mPGES-1 null mice are resistant to arthritis in the models of collagen-induced arthritis and collagen antibody-induced arthritis (14,15).
Peroxisome proliferator-activated receptor ␥ (PPAR␥) is a member of nuclear hormone receptor superfamily of ligandactivated transcription factors that have been shown to regulate inflammatory responses and assist in the resolution of inflammation (20 -24). Recent studies have shown a close relationship between PPAR␥ and PGs in the regulation of inflammation (25)(26)(27). However, until now, no study has evaluated the potential role of mPGES-1 in the regulation of PPAR␥. We created mouse embryo fibroblast (MEF) cell lines derived from mPGES-1 null mice and wild type (WT) littermates to facilitate these studies. This study demonstrates for the first time that mPGES-1 deficiency and reduced PGE 2 lead to elevation of PPAR␥ under basal conditions. This study further identifies key downstream signaling targets responsible for PPAR␥ regulation by mPGES-1-derived PGE 2 .

EXPERIMENTAL PROCEDURES
Animals-mPGES-1 heterozygous mice on a DBA1 lac/J background were obtained from Pfizer (15). Mice were housed in microisolator cages in a pathogen-free barrier facility, and all experiments were performed under the approved IACUC and institutional guidelines.
Preparation and Activation of Mouse Embryo Fibroblasts-Embryos were harvested from mPGES-1 (DBA1 lac/J) pregnant heterozygous females (E12.5) who had been mated with heterozygous males. Whole embryos were minced and placed into culture DMEM containing 10% FBS, 100 units/ml penicillin, and 100 g/ml streptomycin at 37°C under an atmosphere of 5% CO 2 . At confluence, the cells were detached and passaged, and 3-4 passage cells were used in all experiments. MEFs were plated into the wells of a 6-well plate at a density of 3 ϫ 10 5 cells/well in DMEM containing 10% FBS. Cells were starved for 72 h in DMEM containing 1% FBS and then incubated with or without 1 ng/ml IL-1␤ in the presence or absence of various treatments for 12 h. Cell viability was determined by measuring mitochondrial NADH-dependent dehydrogenase activity with WST-1 assay (Dojindo Laboratories, Kumamoto, Japan).
Reverse Transcription (RT)-PCR-RNA from the cells was extracted with TRIpure reagent according to the manufacturer's instructions. Reverse transcription was performed according to the manufacturer's instructions using a SuperScript preamplification system (Invitrogen) with 1 g of total RNA from the cells as a template. Subsequent amplifications of the cDNA fragments by PCR with HotStar Taq polymerase (Qiagen, Valencia, CA) were performed using 0.5 l of the reverse-transcribed mixture as a template with specific oligonucleotide primers and cycle numbers as follows: mouse PPAR␥ (31 cycles), sense 5Ј-CCT CTC CGT GAT GGA AGA CC-3Ј and antisense 5Ј-GCA TTG TGA GAC ATC CCC AC-3Ј; mouse GAPDH (20 cycles), sense 5Ј-GGG GTG AGG CCG GTG CTG AGT AT-3Ј and antisense 5Ј-CAT TGG GGG TAG GAA CAC GGA AGG-3Ј. After initial denaturation at 95°C for 15 min, PCR involved amplification cycles of 30 s at 95°C, 30 s at 56°C, and 45 s at 72°C, followed by elongation for 5 min at 72°C. The amplified cDNA fragments were resolved by electrophoresis on 2% (w/v) agarose gel and were visualized under UV light using a Chemidoc apparatus (Bio-Rad) after staining of the gel with ethidium bromide.
Western Blotting-Cells were lysed in Tris-buffered saline (TBS) containing 0.1% SDS, and the protein content of the lysates was determined using bicinchoninic acid (BCA) protein assay reagent (Pierce) with bovine serum albumin as the standard. Cell lysates were adjusted to equal equivalents of protein and then were applied to SDS-polyacrylamide gels (10 -20%) for electrophoresis. Next, the proteins were electroblotted onto polyvinylidene difluoride membranes. After the membranes were blocked in 10 mM TBS containing 0.1% Tween 20 (TBS-T) and 5% skim milk, the membranes were probed for 1.5 h with the respective antibodies (1:1000 for mPGES-1, COX-2, PPAR␥, and ␤-actin) in TBS-T for 1.5 h. After washing the membranes with TBS-T, the membranes were incubated with horseradish peroxidase-conjugated anti-rabbit (for mPGES-1 and COX-2) or horseradish peroxidase-conjugated anti-mouse (for PPAR␥ and ␤-actin) IgG (1:10,000 dilution in TBS-T containing 5% skim milk) overnight at 4°C. After further washing with TBS-T, protein bands were visualized with an ECL Western blot analysis system using a Chemidoc apparatus (Bio-Rad).
PPAR␥ Transcriptional Activity Assay-MEFs were plated into the T-75 flasks at a density of 2 ϫ 10 6 cells/flask in DMEM containing 10% FBS. Cells were starved for 72 h in DMEM containing 1% FBS and then incubated with or without 1 ng/ml IL-1␤ in the presence/absence of various treatments. Nuclear extraction and PPAR␥ transcriptional activity assays were performed using the PPAR␥ transcription assay kit (Cayman Chemicals, Ann Arbor, MI), and procedures were followed according to the manufacturer's recommendation.
Measurement of PGE 2 in Culture Medium-MEFs were incubated for 12 h in the presence or absence of IL-1␤ (1 ng/ml). In experiments involving treatment with indomethacin and NS-398, these compounds were added 72 h before IL-1␤ stimulation. The culture supernatant was harvested, and concentration of PGE 2 was measured by ELISA (Cayman Chemical, Ann Arbor MI). Assays were performed according to the manufacturer's recommendation.
Statistical Analysis-The data are expressed as mean Ϯ S.E. Statistical analysis was performed using Student's t test. p Ͻ 0.05 was considered statistically significant.

Effect of mPGES-1 Genetic Deletion on PPAR␥ Expression
(mRNA and Protein) and Transcriptional Activity in mPGES-1 WT and Null MEFs-mPGES-1 WT and null MEFs, in the presence/absence of IL-1␤ stimulation, were analyzed for PPAR␥ mRNA expression (Fig. 1a), protein expression (Fig. 1b), and transcriptional activity (Fig. 1c). Results showed that under basal conditions, PPAR␥ mRNA and protein levels were significantly higher (p Ͻ 0.05) in the mPGES-1 null MEFs compared with their WT counterparts. Similarly PPAR␥ transcriptional activity was significantly higher (p Ͻ 0.01) in mPGES-1 null MEFs compared with WT MEFs under basal conditions. IL-1␤ stimulation decreased the levels of PPAR␥ mRNA, protein, and transcriptional activity by near 50% in both cell types; however, PPAR␥ expression and transcriptional activity levels were still significantly higher (p Ͻ 0.05) in the mPGES-1 null MEFs compared with WT MEFs. These results show that genetic deletion of mPGES-1 leads to elevation of PPAR␥ expression and transcriptional activity levels. Furthermore, proinflammatory signals reduce PPAR␥ expression and transcriptional activity, an effect mitigated in the absence of mPGES-1.
Effect of mPGES-1 Genetic Deletion on PGE 2 Levels in mPGES-1 WT and Null MEFs-PGE 2 levels were significantly (p Ͻ 0.05) higher in mPGES-1 WT MEFs compared with null MEFs under unstimulated conditions (Fig. 2). Upon stimulation with IL-1␤, a significant (p Ͻ 0.01) increase in the levels of PGE 2 was observed in WT MEFs; however, no change in the PGE 2 levels was observed in mPGES-1 null MEFs. These results clearly demonstrate that mPGES-1 is critical for PGE 2 production at basal as well as stimulated conditions even though we have shown previously that both cytosolic PGES (cPGES) and mPGES-2 are expressed in these MEFs (6).
Our recent study using mPGES-1 WT and null MEFs also showed that genetic deletion of mPGES-1 not only blocks PGE 2 production but also results in the elevation of 6-keto-PGF 1␣ (stable breakdown product of prostacyclin; PGI 2 ) under basal as well as cytokine-stimulated conditions, suggesting a shunting phenomenon within the arachidonic acid metabolic pathway upon deletion of mPGES-1 (6). In this study, we further showed that mPGES-1 genetic deletion did not have any effect on the production pattern of other PGs such as PGD 2 and thromboxane B 2 , which remained unaltered in mPGES-1 WT and null MEFs under basal as well as cytokine-stimulated conditions.

Effect of Exogenous PGE 2 and Carbacyclin (Prostacyclin Analogue) on PPAR␥ Expression and Transcriptional Activity in mPGES-1 WT and Null
MEFs-To determine whether PGE 2 or an alternate change induced by mPGES-1 deletion suppresses PPAR␥ expression and transcriptional activity in MEFs, exogenous PGE 2 treatment of mPGES-1 WT and null MEFs in the presence or absence of IL-1␤ stimulation was performed. Addition of PGE 2 under basal conditions resulted in a significant decrease (p Ͻ 0.05) in the PPAR␥ expression both at mRNA and protein levels in mPGES-1 null MEFs with little or no change observed in WT MEFs (Fig. 3, a and b). In the presence of IL-1␤, a further decrease in the PPAR␥ mRNA expression levels was observed in both mPGES-1 WT and null MEFs. PPAR␥ protein expression did not show any further decrease when PGE 2 was used in combination with IL-1␤.
Because our previous studies using mPGES-1 WT and null MEFs showed elevation of 6-keto-PGF 1␣ upon genetic deletion of mPGES-1 in MEFs (6), we therefore investigated the effects of carbacyclin (PGI 2 analogue) to determine the contribution of PGI 2 versus PGE 2 toward regulation of PPAR␥ expression in MEFs. However, we did not observe any changes in the protein expression levels of PPAR␥ in mPGES-1 WT and null MEFs upon treatment with carbacyclin (Fig. 3c).
Addition of PGE 2 in the absence of IL-1␤ stimulation significantly (p Ͻ 0.01) decreased PPAR␥ transcriptional activity levels in mPGES-1 null MEFs bringing the levels similar to mPGES-1 WT MEFs (Fig. 3d). In the presence of IL-1␤ stimulation, a further significant decrease in the levels of both mPGES-1 WT (p Ͻ 0.05) and null MEFs (p Ͻ 0.01) was observed. Treatment with carbacyclin did not have any effect on PPAR␥ transcriptional activity in mPGES-1 WT and null MEFs. These results suggest that deletion of mPGES-1 and a subsequent decrease in PGE 2 levels (and not increased PGI 2 ) play a key role in the differential regulation of PPAR␥ in MEFs. In addition, these results also suggest that IL-1␤ has the ability to itself regulate the expression and transcriptional activity of PPAR␥ and may exert other intrinsic effects on PPAR␥ expression and activity in addition to its ability to increase PGE 2 production.
Effect of PGE 2 Inhibition by NSAIDs (Indomethacin and NS-398) on PPAR␥ Expression and Transcriptional Activity in mPGES-1 WT and Null MEFs-To further confirm the contribution of PGE 2 toward differential regulation of PPAR␥ expression and transcriptional activity, mPGES-1 WT and null MEFs were treated with NSAIDs, including indomethacin (nonselective COX inhibitor) and NS-398 (selective COX-2 inhibitor). Addition of indomethacin and NS-398 significantly increased the levels of PPAR␥ protein expression (Fig. 4a) and transcriptional activity (Fig. 4b) in mPGES-1 WT MEFs only, raising the levels similar to that of mPGES-1 null MEFs. These results further confirm that PGE 2 is the key mediator involved in differential regulation of PPAR␥ in mPGES-1 WT and null MEFs.
Effect of mPGES-1 Deletion on cAMP Levels in mPGES-1 WT and Null MEFs-PGE 2 has been shown to mediate some of its downstream effects via cAMP-dependent pathways (8). Therefore, we investigated the effect of mPGES-1 deletion and the subsequent decrease in the levels of PGE 2 on cAMP levels in mPGES-1 WT and null MEFs. Results showed that mPGES-1 deletion did not have any significant effect on the levels of cAMP in MEFs under basal conditions (Fig. 5a). We further investigated the effects of exogenous PGE 2 and forskolin (a direct adenyl cyclase activator) on the levels of cAMP in mPGES-1 WT and null MEFs. Treatment with PGE 2 (p Ͻ 0.05) and forskolin (p Ͻ 0.01) significantly increased cAMP levels in both mPGES-1 WT and null MEFs to a similar extent. These results show that the cAMP machinery is intact in both mPGES-1 WT and null MEFs.
Effect of Forskolin on PPAR␥ Expression and Transcription Activity Levels in mPGES-1 WT and Null MEFs-To delineate the mechanism by which PGE 2 regulates PPAR␥ expression and transcriptional activity, we treated mPGES-1 WT and null MEFs with forskolin in the presence/absence of PGE 2 , and we assessed its effect on PPAR␥ protein expression and transcriptional activity. Forskolin did not have any effect on PPAR␥ protein expression (Fig. 5b) and transcriptional activity (Fig. 5c) in mPGES-1 WT and null MEFs in the absence of PGE 2 . However, PPAR␥ protein expression and transcriptional activity were significantly (p Ͻ 0.01) decreased when forskolin was used in combination with PGE 2 , an effect that seems to be solely elicited by PGE 2 and not forskolin. These results suggest that even though cAMP pathways are activated by PGE 2 and forskolin in MEFs, PGE 2 regulates PPAR␥ expression and transcriptional activity via a signaling pathway independent of cAMP.
Effect of PI 3-Kinase Inhibitor, Akt Inhibitor, p38 Inhibitor, and MEK Inhibitor on PPAR␥ Expression in mPGES-1 WT and Null MEFs-PGE 2 is also known to mediate some of its biological responses through MAP kinase and PI 3-kinase/Akt pathways (8). Therefore, to delineate the mechanism with which PGE 2 regulates PPAR␥ expression in mPGES-1 WT and null  Addition of PI 3-kinase or Akt inhibitors elevated the protein expression levels of PPAR␥ in mPGES-1 WT MEFs only, raising the levels similar to that observed in mPGES-1 null MEFs (Fig. 6, a and b). PI 3-kinase and Akt inhibitors in the presence of PGE 2 also showed elevation in the PPAR␥ expression in both mPGES-1 WT and null MEFs, thus reversing the inhibitory effects of PGE 2 on PPAR␥ expression. Addition of p38 MAP kinase and MEK inhibitors showed no effect on PPAR␥ protein expression in mPGES-1 WT and null MEFs in the presence or absence of PGE 2 (Fig.  6c). These results clearly suggest that PI 3-kinase and Akt signaling are key downstream pathways by which PPAR␥ expression is suppressed by PGE 2 in MEFs.
Because previous reports have shown that rosiglitazone, 15-deoxy-PGJ 2 , and GW9662 have the ability to affect cell viability, we therefore first determined the effects of various concentrations of these reagents on cell viability of MEFs using WST-1 assay. Concentrations of rosiglitazone (10 M), 15-deoxy-PGJ 2 (10 M), and GW9662 (3 M) were chosen for the subsequent experiments as these concentrations did not have any effect on cell viability of MEFs (data not shown).
Low levels of COX-2 protein were observed in unstimulated mPGES-1 WT MEFs (Fig. 7a). Upon stimulation with IL-1␤, a significant elevation (p Ͻ 0.05) in the protein expression of COX-2 was observed. Pretreatment with rosiglitazone and 15-deoxy-PGJ 2 blocked up-regulated expression of COX-2 in IL-1␤stimulated mPGES-1 WT. To confirm that rosiglitazone and 15-deoxy-PGJ 2 reduced COX-2 expression via the PPAR␥ pathway, we incubated mPGES-1 MEFs with rosiglitazone and 15-deoxy-PGJ 2 in the presence or absence of PPAR␥ inhibitor (GW9662). GW9662 completely reversed the inhibitory effect of rosiglitazone on COX-2 expression but showed only partial recovery on the inhibitory effect of 15-deoxy-PGJ 2 on COX-2 expression in mPGES-1 WT and null MEFs. Some effects of 15-deoxy-PGJ 2 have been shown previously to be mediated via alternate mechanistic pathways independent of PPAR␥ (26). In this study, partial recovery of COX-2 expression by GW9662 in the 15-deoxy-PGJ 2 group further suggests the involvement of an alternative mechanistic pathway in addition to the PPAR␥ pathway by which 15-deoxy-PGJ 2 blocks COX-2 expression in MEFs.
We further investigated the effects of rosiglitazone and 15-deoxy-PGJ 2 on mPGES-1 expression in mPGES-1 WT MEFs. Low level of mPGES-1 protein was observed in unstimulated MEFs (Fig. 7a). Upon stimulation with IL-1␤, significant elevation (p Ͻ 0.05) in the protein expression of mPGES-1 was seen in MEFs as expected. Treatment with rosiglitazone and 15-deoxy-PGJ 2 blocked the increased expression of mPGES-1 in IL-1␤-stimulated mPGES-1 WT and null MEFs. To confirm that rosiglitazone and 15-deoxy-PGJ 2 reduced mPGES-1 expression via the PPAR␥ pathway, we incubated mPGES-1 WT MEFs with rosiglitazone and 15-deoxy-PGJ 2 in the pres-ence or absence of PPAR␥ inhibitor (GW9662). GW9662 significantly (p Ͻ 0.05) reversed the inhibitory effects of rosiglitazone and 15-deoxy-PGJ 2 on mPGES-1 expression. These results show that increased PPAR␥ expression results in downregulation of COX-2 and mPGES-1 expression under basal as well as inflammatory conditions. A significant elevation (p Ͻ 0.05) in the levels of PGE 2 was observed in MEFs upon stimulation with IL-1␤ compared with unstimulated MEFs (Fig. 7b). Treatment with rosiglitazone and 15-deoxy-PGJ 2 significantly (p Ͻ 0.05) blocked the increased production of PGE 2 in IL-1␤-stimulated mPGES-1 WT MEFs. This inhibition was recovered when rosiglitazone and 15-deoxy-PGJ2 were used in the presence of GW9662 in IL-1␤-stimulated mPGES-1 WT MEFs. These results show that increased PPAR␥ expression results in decreased PGE 2 production under basal as well as inflammatory conditions.
Because our results showed that genetic deletion of mPGES-1 and resultant decrease in PGE 2 production increased the levels of PPAR␥ in MEFs, we expected that mPGES-1 null MEFs would have low levels of COX-2 compared with WT MEFs. Indeed, we observed significantly higher levels of COX-2 protein in mPGES-1 WT MEFs compared with null MEFs under basal conditions (Fig. 7c). These results suggest that increased PPAR␥ expression as a result of genetic deletion of mPGES-1 could lead to decreased COX-2 expression in mPGES-1 null MEFs under basal conditions.

DISCUSSION
This tudy using MEFs isolated from mPGES-1-deficient mice clearly presents three major conclusions. First, genetic deletion of mPGES-1 results in up-regulation of PPAR␥ expression and transcriptional activity under basal conditions. Second, mPGES-1-derived PGE 2 is responsible for regulation of PPAR␥ expression through a cAMP-independent pathway involving PI 3-kinase and Akt signaling. Third, specific activation of PPAR␥ blocks Differential Regulation of PPAR␥ in mPGES-1 Null MEFs-mPGES-1 is an inducible enzyme that acts downstream of COX and specifically catalyzes the conversion of PGH 2 to PGE 2 . Using MEFs isolated from mPGES-1-deficient mice, we have shown previously that in the absence of mPGES-1, low levels of PGE 2 are produced under basal as well as cytokine-stimulated conditions (6). In this study, we demonstrate for the first time that genetic deletion of mPGES-1 leads to increased PPAR␥ expression and transcriptional activity by eliminating the suppressive effects of PGE 2 . PPAR␥ is a member of the nuclear hormone receptor superfamily of ligand-activated transcription factors that have been shown to initiate mechanisms that inhibit inflammatory responses and assist in the resolution of inflammation (20,21,29). Thus inhibition of mPGES-1 not only blocks proinflammatory PGE 2 production but also results in the up-regulation of anti-inflammatory PPAR␥ in MEFs.

PGE 2 Regulates PPAR␥ Expression via Non-cAMP Pathway Involving PI 3-Kinase and Akt
Signaling-Genetic deletion of specific genes can be associated with unforeseen phenotypic changes. However, we show here that PGE 2 is responsible for the differential regulation of PPAR␥ expression and transcriptional activity in MEFs. Exogenous PGE 2 decreased and NSAIDs (indomethacin, a nonselective COX inhibitor, and NS-398, a selective COX-2 inhibitor) increased PPAR␥ under basal conditions confirming that the presence of PGE 2 down-regulates PPAR␥, whereas PGE 2 depletion results in the up-regulation of PPAR␥.
The next aim of this study was to delineate underlying signaling pathways by which mPGES-1-derived PGE 2 regulates PPAR␥ expression and transcriptional activity in MEFs. PGE 2 exerts the majority of its actions through a family of G protein-coupled receptors, including EP1, EP2, EP3, and EP4 via downstream signaling pathways, including cAMP-dependent protein kinase, MAP kinase, and PI 3-kinase/Akt signaling (8, 30 -32).
In this study we first confirmed that the cAMP pathway was intact in mPGES-1 WT and null MEFs. However, forskolin, a direct adenyl cyclase activator, did not lead to any change in PPAR␥ expression and transcriptional activity in either mPGES-1 WT or null MEFs in the presence or absence of PGE 2 . These data suggest that regulation of PPAR␥ by mPGES-1-derived PGE 2 occurs via a cAMP-independent pathway. No changes in PPAR␥ expression were observed following inhibition of MEK or p38 MAP kinase. However, our results clearly show that PI 3-kinase and Akt inhibitors significantly up-regulated PPAR␥ expression in mPGES-1 WT MEFs, raising the levels similar to mPGES-1 null MEFs. PI 3-kinase and its immediate downstream effector Akt are therefore the key downstream signaling pathways by which PPAR␥ expression is suppressed by PGE 2 . Various other studies using fibroblasts and other cell types have also shown that PGE 2 mediates its regulatory effects on various downstream targets via PI 3-kinase and Akt signaling pathways (33,34).
Negative Regulation of Proinflammatory COX-2, mPGES-1, and Their Derived PGE 2 by PPAR␥-PPAR␥ is an endogenous regulator known to mediate its anti-inflammatory effects by down-regulation of proinflammatory mediators (20,21,29). In contrast COX-2, mPGES-1, and their derived PGE 2 are key proinflammatory mediators involved during initiation of inflam-mation (1). Using specific PPAR␥ agonist (rosiglitazone), PPAR␥ ligand (15-deoxy-PGJ 2 ), and PPAR␥ inhibitor (GW9662), we observed that specific activation of PPAR␥ blocked IL-1␤-induced up-regulation of proinflammatory COX-2, mPGES-1, and their derived PGE 2 in WT MEFs. Because of the elevated PPAR␥ levels observed upon genetic deletion of mPGES-1, we therefore expected that mPGES-1 null MEFs would have lower levels of COX-2 compared with WT MEFs. Indeed, our results showed significantly lower levels of COX-2 protein in mPGES-1 null MEFs compared with WT MEFs under basal conditions. These results further confirm the endogenous antiinflammatory properties of PPAR␥ in down-regulating key proinflammatory signals.
Significance of mPGES-1 Inhibition-PGE 2 is a key proinflammatory mediator of inflammation associated with various disease states, and increased PGE 2 requires both COX-2 and mPGES-1 (4,5,18,19,(35)(36)(37)(38). Nonselective and COX-2selective NSAIDs reduce PGE 2 production by inhibiting COX-2 activity and are extensively used for reducing inflammation, pain, and fever (39). COX-2-specific NSAIDs were developed with improved gastrointestinal safety (40 -42). However, recent clinical trials using selective COX-2 inhibitors suggest that specific inhibition of COX-2 is associated with increased incidence of cardiovascular events (11)(12)(13). Specific COX-2 inhibition results in loss of anti-thrombotic prostacyclin (PGI 2 ) derived from endothelial COX-2, which plays a key role in the regulation of thrombogenesis (43) and is a possible factor associated with cardiovascular side effects observed with the use of specific COX-2 inhibitors.
shown that genetic deletion of mPGES-1 results in diversion of prostaglandin production from predominant PGE 2 toward PGI 2 . In addition, we have recently shown that genetic deletion of mPGES-1 results in up-regulation of nitrite levels (stable metabolic product of nitric oxide (NO)) (6). PGI 2 and NO are key mediators involved in maintaining vascular homeostasis (46). A recent in vivo study by Fitzgerald and co-workers (45) also showed that mPGES-1 deletion depressed PGE 2 and increased PGI 2 , with no effect on thrombogenesis or blood pressure in mice. Thus, these observations suggest that inhibition of mPGES-1 may avoid the cardiovascular side effects seen with inhibition of COX-2. In addition, this study shows that genetic deletion of mPGES-1 elevates anti-inflammatory PPAR␥. In vitro and in vivo studies suggest that PPAR␥ has the ability to stimulate anti-inflammatory responses and assist in the resolution of inflammation by inhibiting a broad range of proinflammatory mediators, including IL-1␤, tumor necrosis factor-␣, and nuclear transcription factor B (20,23,28,47). In this study we also demonstrate that PPAR␥ has the ability to down-regulate proinflammatory COX-2, mPGES-1, and their derived PGE 2 (Fig. 8).
Although the biology of mPGES-1 and the consequences of blocking its activity have yet to be completely delineated, mPGES-1 remains a viable target. It is not known if the efficacy of such a therapeutic strategy would equal inhibition of COX and all downstream PGs; however, data in mPGES-1 null mice and the efficacy of monoclonal antibodies against PGE 2 in an arthritis model (48) offer promise. Studies to date in mPGES-1 null mice and MEFs suggest that inhibiting mPGES-1 may be associated with downstream changes, such as increased NO (6) and PPAR␥, which would promote efficacy and potentially limit adverse effects associated with pharmacological inhibition of mPGES-1. mPGES-1 deletion decreases proinflammatory PGE 2 and its downstream signaling and as a result elevates anti-inflammatory PPAR␥, which has the ability to suppress proinflammatory COX-2, mPGES-1, and their derived PGE 2 . Ϫ represents down-regulation; ϩ represents elevation; dotted arrows represent decrease in PGE 2 signaling via EP receptors and its downstream signaling pathways.