Down-regulation of Microsomal Prostaglandin E2 Synthase-1 in Adipose Tissue by High-fat Feeding


  • Pierre-Olivier Hétu,

    1. Department of Biochemistry and Molecular Biology, Merck Frosst Centre for Therapeutic Research, Kirkland, Quebec, Canada and Department of Biochemistry, Université de Montréal, Montreal, Quebec, Canada.
    Search for more papers by this author
  • Denis Riendeau

    Corresponding author
    1. Department of Biochemistry and Molecular Biology, Merck Frosst Centre for Therapeutic Research, Kirkland, Quebec, Canada and Department of Biochemistry, Université de Montréal, Montreal, Quebec, Canada.
      Department of Biochemistry and Molecular Biology, Merck Frosst Centre for Therapeutic Research, 16711 Trans-Canada Hwy, Kirkland, Quebec, Canada H9H 3L1. E-mail:
    Search for more papers by this author

Department of Biochemistry and Molecular Biology, Merck Frosst Centre for Therapeutic Research, 16711 Trans-Canada Hwy, Kirkland, Quebec, Canada H9H 3L1. E-mail:


Objective: Prostaglandin (PG)E2 is a lipid mediator implicated in inflammatory diseases and in the regulation of lipolysis and adipocyte differentiation. This work was, thus, undertaken to study the regulation of the various PGE2 synthases (PGESs) in obesity.

Research Methods and Procedures: C57Bl/6 mice were subjected to a high-fat or regular diet for 12 weeks. The levels of PGE2 in white adipose tissue (WAT) of lean and obese mice were quantified by liquid chromatography-mass spectrometry, and the change in expression of the three major PGES caused by diet-induced obesity was characterized by Western blotting. Human preadipocytes and 3T3-L1 cells were used to assess the expression of microsomal prostaglandin E2 synthase-1 (mPGES-1) during adipogenesis.

Results: mPGES-1, mPGES-2, and cytosolic PGES proteins were all detected in WAT of lean animals. mPGES-1 was expressed at higher levels in WAT than in any other tissues examined and was more abundant (3- to 4-fold) in epididymal (visceral) compared with inguinal (subcutaneous) WAT. Expression of mPGES-1 was also detected in undifferentiated and differentiated 3T3-L1 cells and in human primary subcutaneous preadipocytes at all stages of adipogenesis. The mPGES-1 protein was substantially down-regulated in epididymal and inguinal WAT of obese mice, whereas mPGES-2 and cytosolic PGES remained relatively stable. Concordantly, the PGE2 levels in obese inguinal WAT were significantly lower than those of lean animals.

Discussion: These data suggest that mPGES-1 is the major form of PGESs contributing to the synthesis of PGE2 in WAT and that its down-regulation might be involved in the alterations of lipolysis and adipogenesis associated with obesity.


Obesity is a condition associated with low-grade chronic inflammation (1, 2, 3, 4, 5). The inflammatory condition in obesity is increasingly being recognized as an important contributor to the development of the metabolic syndrome and of its associated complications (4, 5, 6, 7, 8). Adipocytes can secrete pro-inflammatory factors, such as plasminogen activator inhibitor-1, tumor necrosis factor-α, interleukin-1β, interleukin-6, and prostaglandin (PG)E21 (5, 6, 9). Overall, the increase in adipose tissue during obesity is associated with elevated circulating levels of several markers of inflammation and of cytokines (5, 10, 11). Moreover, PGE2 has been shown to inhibit adipocyte lipolysis (12, 13, 14, 15). Treatment of adipocytes with nanomolar concentrations of PGE2 potently inhibits epinephrine- or isoproterenol-induced lipolysis in vitro (16, 17), and this effect is mediated through inhibition of cAMP accumulation (18), probably through EP3 receptor signaling (19). Additionally, PGE2 has been shown to negatively affect adipocyte differentiation in vitro (20).

PGs are synthesized by the concerted action of cyclooxygenases (COXs) and specific terminal synthases (21). Two major isoforms of COX have been identified: COX-1, which mainly synthesizes the PGs required for maintaining homeostasis, and COX-2, which is induced in inflammatory reactions to produce the high levels of PGs observed in those conditions (22). Both enzymes use arachidonic acid to synthesize PGH2, which is the common precursor for the synthesis of the various PGs (21). PGE2 is a major PG implicated in inflammatory reactions (23, 24, 25), and several proteins with the ability to convert PGH2 to PGE2 have been identified, notably microsomal prostaglandin E2 synthase-1 (mPGES-1) (26, 27), mPGES-2 (28), and cytosolic PGES (cPGES) (29). mPGES-2 and cPGES are constitutively expressed (29, 30, 31), whereas mPGES-1 is expressed constitutively in many tissues and at elevated levels in various inflammatory conditions (23, 24). In such, mPGES-1 represents a novel therapeutic target for the treatment of pain and inflammation (24, 31).

Although PGE2 is a well-characterized antilipolytic agent in vitro and a key regulator of inflammation, little is known about the contribution of PGE2 to the disease state and to the complications of obesity. Furthermore, previous studies with adipocytes isolated from lean and obese white adipose tissue (WAT) have shown that the ex vivo production of PGE2 is altered in adipocytes from obese tissues (9, 32). For these reasons, the following study was undertaken to examine the regulation of the major PGE2 synthases during diet-induced obesity.

Research Methods and Procedures


The PG standards used for liquid chromatography-mass spectrometry (LC-MS) quantification were purchased as solids from Cayman Chemicals (Ann Arbor, MI), as were the corresponding deuterated internal standards and indomethacin. The antibodies used were as follows: rabbit antihuman mPGES-1 (cat. no. PG15; Oxford Biomedical Research, Oxford, MI), rabbit antihuman mPGES-2 (cat. no. 160145; Cayman Chemicals), rabbit antihuman cPGES (cat. no. 160150; Cayman Chemicals), and donkey antirabbit IgG (cat. no. NA934V; GE Healthcare Bio-Sciences, Quebec, Canada). Complete protease inhibitor tablets were purchased from Roche Diagnostics (Mannheim, Germany). Human preadipocytes with related media and cell culture reagents were obtained from Cambrex (East Rutherford, NJ). 3T3-L1 cells were purchased from ATCC (Manassas, VA), and the insulin, dexamethasone, and isobutyl-methylxantine used to stimulate differentiation were from Sigma-Aldrich (Oakville, Ontario, Canada).


All animal procedures were approved by the Animal Care Committee at the Merck Frosst Centre for Therapeutic Research (Kirkland, Quebec, Canada) and followed the guidelines established by the Canadian Council on Animal Care. Lean and obese 18-week-old C57Bl/6 male mice were purchased from Taconic (Hudson, NY). These obese mice were obtained by feeding with a high-fat (36% fat, accounting for 59% of calories), high-carbohydrate (36%) diet for 12 weeks, starting at 6 weeks of age (the mice were subjected to high-fat diet by the vendor and were already obese when purchased). The high-fat diet administered by the vendor was Bio-Serv diet no. S3282, with lard as the source of fat, and, thus, contained mainly long-chain monounsaturated and saturated fatty acids (45% oleic acid, 25% palmitic acid, and 12% stearic acid), and linoleic acid (10%). Control mice were fed regular rodent diet (NIH no. 31M diet; Taconic) containing ∼5.3% fat.

Tissue Preparation for Protein Analysis

Mice were killed by CO2 overdose, and tissues were collected as fast as possible, rinsed in phosphate-buffered saline containing 10 μM indomethacin, and frozen in liquid nitrogen. Samples were kept at −80 °C until analysis. Frozen tissues were weighed and homogenized in five volumes of ice-cold homogenizing buffer (50 mM Tris-HCl, pH 7.4; supplemented with 150 mM NaCl, 1% Triton X-100, 1% octylglucoside, 0.1% sodium dodecyl sulfate (SDS), 10 μM indomethacin, and 1× complete protease inhibitor mixture) using a Brinkman Polytron PT 10/35 homogenizer (Fisher Scientific, Nepean, Ontario, Canada). The homogenates were subsequently sonicated on ice for 20 seconds (Cole-Parmer ultrasonic homogenizer, 40% output) before being subjected to centrifugation for 20 minutes at 1000g (4 °C). Supernatants (or soluble fraction between pellet and fat layer, when applicable) were collected with a syringe and transferred to new tubes. Total protein concentration in homogenates was determined by the DC-Lowry protein assay (BioRad Laboratories, Hercules, CA) per the manufacturer's instructions.

Western Blotting Analysis

Samples were diluted in homogenizing buffer to have the same total protein content, and an appropriate volume of 5× loading buffer was added. Equal quantities of total protein were run on tris-glycine gels and transferred to polyvinylidene fluoride membranes (Invitrogen, Ontario, Canada). Blots were blocked for 1 hour in 5% milk/PBST (phosphate-buffered saline + 0.1% Tween 20) and incubated with the appropriate primary antibody in 1% milk/PBST. The antibodies were diluted as follows: 1/1000 for mPGES-1, 1/500 for mPGES-2, and 1/250 for cPGES. The blots were subsequently washed with PBST and incubated with the secondary antibody (1/5000) for 1 hour. Protein bands were detected with LAS-1000plus CCD camera in an Intelligent DarkBox II (Fujifilm, Ontario, Canada) after incubation in SuperSignal West Femto Maximum Sensitivity Substrate (Pierce Biotechnology, Rockford, IL). Quantification of the intensity of the immunoreactive bands was accomplished with Multi Gauge 2.3 software (Fujfilm). The different proteins were detected on the same blot after stripping by incubating membranes for 30 minutes at 50 °C in 62.5 mM Tris-HCl (pH 6.8), 2% SDS, and 0.7% β-mercaptoethanol.

Cell Culture

3T3-L1 preadipocytes were grown in Dulbecco's modification of Eagle's medium containing 10% fetal bovine serum, 100 units/mL penicillin, and 100 μg/mL streptomycin. The cells were brought to confluence and stimulated to differentiate 2 days later by the addition of growth media supplemented with 10 mM N-[2-hydroxyethyl]piperazine-N′-[2-ethanesulfonic acid] (HEPES) (pH 7.0), 167 nM human recombinant insulin, 0.5 mM isobutyl-methylxanthine, and 1 μM dexamethasone. Two days after the start of differentiation, the differentiation media were replaced by growth media supplemented with HEPES and insulin. After 2 days (Day 4 of differentiation), the media were replaced with regular growth media, and cells were fed with fresh growth media every 2 days until differentiation was complete (10 days after induction). Subcutaneous human preadipocyte (1 × 106 cells) from Cambrex were seeded in a T-175 dish in 35 mL preadipocyte basal media (Cambrex, East Rutherford, NJ) supplemented with 10% fetal bovine serum, 2 mM l-glutamine, 100 units/mL penicillin, and 100 μg/mL streptomycin. Cells were cultured until ∼70% confluent and were trypsined, counted, diluted to 100,000 cells/mL, and seeded in 24-well plates (50,000 cells in 500 μL/well). Adipocyte differentiation was initiated on fully confluent preadipocytes by addition of 500 μL differentiating media (preadipocyte basal media supplemented with 20 μg/mL insulin, 2 μM dexamethasone, 1 mM isobutyl-methylxanthine, and 400 μM indomethacin) per the supplier's instructions. At different stages of adipogenesis, cells were washed with Dulbecco's phosphate-buffered saline and frozen on dry ice. For analysis of protein expression, cells were thawed briefly on ice (5 minutes), and 100 μL lysis buffer was added to each well for 60 minutes (20 mM HEPES, pH 7.0; supplemented with 100 mM NaCl, 1% Triton X-100, 1 mM EGTA, 1 mM EDTA, and 1× complete protease inhibitor mixture).

PGE2 Quantification by LC-MS

For the quantification of PGE2 levels, WAT of lean and obese mice was homogenized as described above but in phosphate-buffered saline supplemented with 10 μM indomethacin and 1× complete protease inhibitor mixture. An aliquot of homogenate was protein-precipitated with two volumes of acetonitrile containing deuterated prostaglandin internal standards (350 pg/mL of d4-PGE2). Samples were centrifuged for 20 minutes at 1200g (4 °C). The supernatants were collected, and ∼2.5 volumes of H2O were added. The samples were acidified to pH 3 with 1 N HCl, and the PGs were extracted by solid phase extraction on Sep-Pak C18 cartridges (Waters, Milford, MA). Briefly, samples were added to Sep-Pak cartridges that had been pre-rinsed with methanol and H2O. Cartridges were washed with hexane, and PGs were eluted with ethyl acetate. Solid phase extraction eluates were evaporated under N2 and resuspended in 150 μL of phosphate-buffered saline/acetonitrile (40:60, vol/vol). PGE2 concentration was quantified by LC-MS as reported previously (33).

Statistical Analysis

Densitometric data and PGE2 levels are presented as mean ± SE. Comparison between the different groups was done using Student's t test. Results were considered statistically different at the p < 0.05 level.


PGE2 Levels in Subcutaneous WAT Are Significantly Decreased in Obese Mice

Obesity is associated with chronic low-grade inflammation, and PGE2 is often produced at high levels in inflammatory conditions. However, previous studies have suggested that the ex vivo production of PGE2 by isolated adipocytes is reduced in obesity (9, 32). The levels of this PG were thus quantified in WAT of lean and obese mice to determine whether the in vivo levels of PGE2 in WAT vary during obesity. Six-week-old C57Bl/6 mice were subjected to high-fat or regular diet for 12 weeks, and after this period, mice on the regular diet were lean and of normal appearance, whereas those on the high-fat diet were obese, with a 40% increase in body weight (Table 1). The inguinal (subcutaneous, abdominal) and epididymal (visceral) WATs were collected from five lean and five obese animals for the measurement of PGE2 levels by LC-MS. PGE2 levels in inguinal fat were found to be significantly lower in obese tissues compared with lean tissues when normalized to tissue weight or protein content (76 ± 5% and 65 ± 6% inhibition, respectively, p < 0.05; Table 1). PGE2 levels in obese epididymal WAT could not be detected because of large and irregular unidentified contaminating peaks that coeluted with PGE2 on the LC-MS.

Table 1.  PGE2 levels in inguinal WAT of lean and obese animals
  • Levels of PGE2 were quantified by LC-MS in inguinal WAT of lean and obese mice. Values are expressed as mean ± SE of n = 5 animals per group.

  • *

    p< 0.05.

Animal weight (g)31.6 ± 0.244.0 ± 0.5*
Inguinal WAT  
 Tissue weight (mg)174 ± 58736 ± 40*
 PGE2 level (pg/mg tissue)1.28 ± 0.240.29 ± 0.03*
 PGE2 level (pg/mg protein)210.6 ± 22.774.2 ± 9.5*

Tissue Distribution Analysis of the Different PGE2 Synthases in Mice Reveals a Distinct Pattern for mPGES-1 Expression in WAT

The expression of the three major PGE2 synthases was determined in WAT and compared with that of different tissues in lean C57Bl/6 mice (Figure 1A). When equal amounts of total homogenate protein from different tissues were analyzed, mPGES-1 expression was much more pronounced in epididymal and inguinal WAT than in any other tissue examined (Figure 1A). The expression of mPGES-1 was also detected in seminal vesicles and in lung, kidney, and spleen, as previously reported (34). As a control, the gel was stained with GelCode Blue (Pierce Biotechnology, Rockford, IL) after the transfer to ascertain that the protein loading was equivalent for the different tissues analyzed (data not shown). The high expression level of mPGES-1 in WAT compared with the other tissues was in contrast to the broader pattern of expression observed for mPGES-2 and cPGES. To evaluate the capacity of adipocytes to express mPGES-1, mouse 3T3-L1 cells and isolated human subcutaneous preadipocytes were cultured and stimulated to differentiate into adipocytes. The data showed that mPGES-1 was expressed both in undifferentiated and differentiated 3T3-L1 cells (Figure 1B). The mPGES-1 protein was also detected in human primary subcutaneous preadipocytes and mature adipocytes (Figure 1B), as well as throughout the adipogenesis process (data not shown). The mPGES-2 protein was also expressed in both mouse and human adipocytes (data not shown).

Figure 1.

Expression of mPGES-1 in preadipocytes, adipocytes, and whole WAT. (A) Mouse WAT contains high levels of mPGES-1 as compared with other tissues. Equal amounts of protein (30 μg) from total tissue homogenates were analyzed by SDS-polyacrylamide gel electrophoresis (PAGE) and immunoblotted for the detection of mPGES-1 (17 kDa), mPGES-2 (33 kDa), and cPGES (21 kDa). The three PGE2 synthases were consecutively detected on the same blot by stripping the membrane between each detection. Purified human mPGES-1 (0.4 ng) was used as control. The distribution of the PGE2 synthases is representative of that observed for three different animals. (B) The mPGES-1 is expressed in preadipocytes and mature adipocytes. Mouse 3T3-L1 and human primary subcutaneous preadipocytes were differentiated into adipocytes in culture, and the expression of mPGES-1 in homogenates (30 and 4 μg of protein for 3T3-L1 and human primary cells, respectively) was measured by SDS-PAGE and immunoblotting.

Expression of mPGES-1 in WAT Is Depot-dependent

Interestingly, mPGES-1 was expressed at higher levels in epididymal (visceral) WAT compared with inguinal (subcutaneous) WAT (Figure 1A, lanes 2 and 3). This depot-dependent intensity of expression was further characterized by quantifying the level of expression of mPGES-1 in epididymal and inguinal WAT of four animals and by comparing with that of mPGES-2. As seen in Figure 2, the expression of mPGES-1 was consistently higher (3- to 4-fold, p < 0.05) in epididymal WAT than in inguinal WAT, whereas the expression of mPGES-2 was similar between the two depots (a 1.6-fold increase that did not reach statistical significance).

Figure 2.

The expression of mPGES-1 protein in WAT is dependent on fat depot. (A) Expression of mPGES-1 and mPGES-2 in epididymal and inguinal WAT homogenates. Equal amounts of protein (15 μg) from total WAT homogenates of four different lean mice were analyzed by SDS-PAGE and Western blotting. The first lane is 0.4 ng of purified human mPGES-1. (B) Densitometric quantification of the intensity of the immunoreactive bands (mean ± SE, n = 4). *Comparison between inguinal and epididymal WAT shows a significant difference for mPGES-1 (p < 0.05).

Expression of mPGES-1 Is Selectively Decreased in WAT of Obese Animals

Because PGE2 levels were found to be decreased after diet-induced obesity, the effect of high-fat feeding on the expression of the different PGE2 synthases was determined. In both inguinal and epididymal WAT depots, the expression of mPGES-1 was greatly decreased in obese tissues, whereas mPGES-2 and cPGES expression showed minor or no changes (Figure 3). Acetone extraction of the proteins in the fat cakes (35) followed by Western blotting analysis ruled out the possibility that a sequestration of mPGES-1 in the fat could explain the observed decrease in expression (data not shown). The mPGES-1 expression in obese inguinal WAT compared with the lean tissues was decreased by 65 ± 4% (p < 0.05), whereas mPGES-2 and cPGES expression did not vary significantly. The decrease of mPGES-1 was more pronounced in obese epididymal WAT (91 ± 1% decrease, p < 0.05) and was accompanied by a slight increase in mPGES-2 and cPGES levels compared with lean tissues (1.6- and 2.2-fold increase, respectively, p < 0.05). The results showed that mPGES-1 expression in WAT was greatly reduced during diet-induced obesity and that this modulation of expression was more pronounced than that of the other two major PGE2 synthases.

Figure 3.

Decrease in the expression of mPGES-1 in epididymal and inguinal WAT during diet-induced obesity in mice. (A) Six-week-old C57Bl/6 mice were subjected to either high-fat diet or regular diet for 12 weeks, after which the expression of mPGES-1, mPGES-2, and cPGES in WAT homogenates (15 μg of protein) was analyzed by SDS-PAGE and Western blotting. The first lane is 0.4 ng of purified human mPGES-1. (B) The relative intensities of the immunoreactive bands were quantified by densitometric analysis for each group (mean ± SE, n = 4). Significant differences between lean and obese tissues are indicated (*p < 0.05, **p < 0.01).


We report here that mPGES-1 is expressed at high levels in WAT and adipocytes. The level of expression of mPGES-1 is fat depot-dependent, being higher in epididymal (visceral) compared with inguinal (subcutaneous, abdominal) WAT. However, both WAT depots express considerably more mPGES-1 than all other tissues examined, including spleen and kidney, in which mPGES-1 has previously been shown to play a role in the synthesis of PGE2 (34). The mPGES-2 and cPGES were also detected in WAT but only within the same range of expression than that of the other tissues. Finally, both the levels of PGE2 and of mPGES-1 protein were found to be reduced in WAT during obesity, whereas mPGES-2 and cPGES levels remained relatively constant. These data indicate that mPGES-1 is playing a role in the regulation of PGE2 synthesis in the adipose tissue.

Obesity is associated with low-grade chronic inflammation. Although one might expect that macrophage infiltration in obese fat tissue (1, 3) could have contributed to an increase in PGE2 levels, our data show a decrease in PGE2 during diet-induced obesity caused by a down-regulation of mPGES-1. Although total PGE2 content in whole fad pad was not significantly different between lean and obese inguinal WAT, the mean concentrations of PGE2 in WAT were significantly reduced in obese tissues when normalized to tissue weight or protein content. This decreased PGE2 content per milligram protein correlates with the decreased mPGES-1 expression (on a protein basis and also relative to the other PGES) and suggests that PGE2 signaling in WAT is most probably decreased in obesity. Interestingly, these data are consistent with the recent report by Fain et al. (9) showing that adipocytes isolated from human subcutaneous and visceral WAT of obese individuals (BMI of 45 kg/m2) have a tendency to release lower amounts of PGE2 (86% decrease, p < 0.001, and 55% decrease, not significant, respectively) than those of leaner individuals. An overall decreased release rate of PGE2 by adipocytes isolated from obese Zucker rats (fa/fa) had also previously been reported by Gaskins et al. (32). Our data on the reduced PGE2 content in obese WAT are, thus, in agreement with the reported decreased capacity of obese adipocytes to synthesize PGE2 ex vivo.

The extent of obesity and potentially the synthesis of PGs could be affected by the composition of the high-fat diet administered to the animals. In this study, lard was the source of fat in the high-fat diet, and as such, the composition of the diet was mainly long-chain monounsaturated and saturated fatty acids. A recent paper by Buettner et al. (36) compared different high-fat diets administered to male Wistar rats and concluded that lard-based high-fat diet models are suitable to study obesity and insulin resistance. The lard-based high-fat diet significantly affected the plasma levels of free fatty acids, increasing oleic (monounsaturated) and stearic (saturated) acid and decreasing linoleic (ω-6 polyunsaturated) and linolenic (ω-3 polyunsaturated) acid (36). Lowering the levels of linoleic acid, which is the natural precursor of arachidonic acid, could potentially lead to a decreased availability of arachidonic acid for PG synthesis, but the level of arachidonic acid was found not to differ significantly between animals fed the lard-based high-fat diet and those on regular chow (36). However, the mechanism by which PGE2 levels and expression of mPGES-1 are reduced by the high-fat diet is unknown, and the effect of dietary fatty acids on the regulation of PGE2 synthesis deserve to be further characterized.

mPGES-1 is an inducible enzyme (23, 26, 37). Previous studies reported low or undetectable basal levels of mPGES-1 RNA in several rat (37, 38) and human (26) tissues, although some basal mPGES-1 protein expression has been detected in mice lung, spleen, kidney, and stomach (34). The much higher level of mPGES-1 protein expression in WAT compared with other tissues possibly reflects the importance of this enzyme and of PGE2 in regulating adipocyte or WAT biological processes. However, further studies are needed to elucidate the role of mPGES-1 in WAT in vivo and to determine the potential effects of pharmacological modulation of mPGES-1 activity or PGE2 signaling on WAT metabolism. Interestingly, mPGES-1 expression was consistently higher (3- to 4-fold) in lean epididymal compared with lean inguinal WAT, which suggests that there exists a depot-dependent regulation of mPGES-1 expression.

The mPGES-1 protein was detected both in undifferentiated and differentiated mouse 3T3-L1 adipocytes, and the expression of mPGES-1 was also detected at all stages of adipogenesis in isolated human primary subcutaneous preadipocytes. The data obtained from these two independent cell systems show that mPGES-1 is expressed in adipocytes of both mice and humans. However, our results differ from those of Xie et al. (39), who recently described an increase of mPGES-1 protein expression (compared with actin) during differentiation of mouse 3T3-L1 adipocytes. The explanation for this discrepancy is unknown, and the role of mPGES-1 activity during adipocyte differentiation remains to be studied. Nonetheless, the data of Xie et al. are in agreement with our results showing that mPGES-1 is expressed both in preadipocytes and in mature adipocytes.

Analysis of the expression of the three major PGE2 synthases in lean and obese fat revealed a pronounced and selective decrease of mPGES-1 protein expression in both visceral and subcutaneous WAT depots during diet-induced obesity. The mPGES-1 expression has been shown to be down-regulated by dexamethasone (38, 40) and phenobarbital (41, 42), but, to our knowledge, this is the first report of an in vivo drug-independent down-regulation of mPGES-1. The amplitude of modulation for mPGES-1 expression in obesity was much greater than that of the two other PGE2 synthases. Nevertheless, mPGES-2 and cPGES expression in epididymal WAT was significantly elevated in obese tissues, possibly reflecting a partial compensation for the decrease of mPGES-1. Partial compensation by mPGES-2 in myometrium of mPGES-1–deficient mice has previously been reported in a lipopolysaccharide pre-term delivery model (43).

PGE2 is a well-known potent antilipolytic agent in vitro (12, 13, 14) and inhibits catecholamine-induced lipolysis in the nanomolar range (16, 17). Neuronal or hormonal activation of lipolysis stimulates the release of prostaglandins from WAT (44), which have been proposed to be part of a negative feedback mechanism that controls lipolysis (16). However, the endogenous regulation of lipolysis by PGE2 has not been conclusively shown because the effects of nonsteroidal anti-inflammatory drugs on lipolysis are inconsistent. In fact, WAT produces mainly PGE2 and PGI2, which potently inhibit and stimulate lipolysis, respectively (14, 15), and the production of both these PGs is inhibited by nonsteroidal anti-inflammatory drugs. Some studies report stimulation (45) or no effect (46, 47) of nonsteroidal anti-inflammatory drugs on agonist-induced lipolysis. Nonetheless, a selective modulation of PG levels in WAT has the potential to affect lipolytic responses of adipocytes and, in such, the down-regulation of mPGES-1 in obese WAT could potentially contribute to the increased lipolytic activity associated with obesity. Moreover, different PGs have been shown to differentially regulate adipogenesis in vitro. PGF has been shown to inhibit differentiation of 3T3-L1, 3T3-F442A, Ob1771, and rat primary preadipocytes (48, 49). In contrast, carbacyclin, a stable analog of PGI2, has been shown to stimulate Ob1771 preadipocyte differentiation (50). Of interest, PGE2 has been shown to inhibit 3T3-L1 adipocyte differentiation through its interaction with the EP4 receptor (20). The net effects of PGs on adipogenesis are not completely understood, because arachidonic acid has been shown to either promote (49, 51) or inhibit (52) adipocyte differentiation in vitro and seems to vary depending on cellular cAMP levels (52). The role of PGs in adipogenesis regulation in vivo remains to be established, but if PGE2 negatively modulates adipocyte differentiation, the down-regulation of mPGES-1 could contribute to the increased adipogenesis caused by high-fat feeding.

In summary, the observed decreases in mPGES-1 protein expression and PGE2 levels in obese WAT might potentially contribute to the changes in lipolysis and adipogenesis regulation associated with diet-induced obesity.


The authors thank S. Desmarais (Merck Frosst) for providing the 3T3-L1 samples, M. Ouellet and K. Bateman (Merck Frosst) for technical advice and support for the LC-MS analysis, and D. Normandin, S. Wong, and the Comparative Medicine Department staff (Merck Frosst) for animal procedures. This work was supported, in part, by studentships (P.-O.H.) from the Natural Sciences and Engineering Research Council of Canada and from the Department of Biochemistry at the Université de Montréal.


  • 1

    Nonstandard abbreviations: PG, prostaglandin; COX, cyclooxygenase; mPGES, microsomal prostaglandin E2 synthase; cPGES, cytosolic PGE2 synthase; WAT, white adipose tissue; LC-MS, liquid chromatography-mass spectrometry; SDS, sodium dodecyl sulfate; PBST, phosphate-buffered saline and Tween; HEPES, N-[2-hydroxyethyl]piperazine-N′-[2-ethanesulfonic acid]; PAGE, polyacrylamide gel electrophoresis.

  • The costs of publication of this article were defrayed, in part, by the payment of page charges. This article must, therefore, be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.