Etomoxir, Sodium 2-[6-(4Chlorophenoxy)Hexyl]Oxirane-2-Carboxylate, Increases Uncoupling Protein-3 Mrna Levels in Primary Culture

Diabetes Publish Ahead of Print, published online April 28, 2008

PPARβ/δ prevents NF-κB activation in adipocytes

Activation of Peroxisome Proliferator-Activated Receptor β/δ (PPARβ/δ) Inhibits LPS-induced Cytokine Production in Adipocytes by Lowering NF-κB Activity via ERK1/2

Ricardo Rodríguez-Calvo1, Lucía Serrano1, Teresa Coll1, Norman Moullan2, Rosa M. Sánchez1, Manuel Merlos1, Xavier Palomer1, Juan C. Laguna1, Liliane Michalik2, Walter Wahli2 and Manuel Vázquez-Carrera1.

1Pharmacology Unit, Department of Pharmacology and Therapeutic Chemistry, Faculty of Pharmacy, University of Barcelona, IBUB (Institut de Biomedicina de la UB), and CIBERDEM-Instituto de Salud Carlos III, Diagonal 643, E-08028 Barcelona, Spain and 2Center for Integrative Genomics, National Research Center Frontiers in Genetics, University of Lausanne, CH-1015 Lausanne, Switzerland.

Corresponding author: Manuel Vázquez-Carrera Unitat de Farmacologia. Facultat de Farmàcia. Diagonal 643. E-08028 Barcelona. Spain E-mail: [email protected]

Received for publication 07 February 2008 and accepted in revised form 21 April 2008.

Objective: Chronic activation of the nuclear factor (NF)-κB in white adipose tissue leads to increased production of pro-inflammatory cytokines, which are involved in the development of insulin resistance. It is presently unknown whether Peroxisome Proliferator-Activated Receptor (PPAR)β/δ activation prevents inflammation in adipocytes.

Research Design and Methods and Results: Firstly, we examined whether the PPARβ/δ agonist GW501516 prevents LPS-induced cytokine production in differentiated 3T3-L1 adipocytes. Treatment with GW501516 blocked LPS-induced IL-6 expression and secretion by adipocytes and the subsequent activation of the STAT3-SOCS3 pathway. This effect was associated with the capacity of GW501516 to impede LPS- induced NF-κB activation. Secondly, in in vivo studies, white adipose tissue from Zucker Diabetic Fatty (ZDF) rats, compared to that of lean rats, showed reduced PPARβ/δ expression and PPAR DNA-binding activity, which was accompanied by enhanced IL-6 expression and NF-κB DNA-binding activity. Furthermore, IL-6 expression and NF-κB DNA-binding activity was higher in white adipose tissue from PPARβ/δ-null mice than in wild-type mice. Since mitogen-activated protein kinase (MAPK)–extracellular signal– related kinase (ERK)1/2 (MEK1/2) is involved in LPS-induced NF-κB activation in adipocytes, we explored whether PPARβ/δ prevented NF-κB activation by inhibiting this pathway. Interestingly, GW501516 prevented ERK1/2-phosphorylation by LPS. Further, white adipose tissue from animal showing constitutively increased NF-κB activity, such as ZDF rats and PPARβ/δ-null mice, also showed enhanced phospho-ERK1/2 levels.

Conclusions: These findings indicate that activation of PPARβ/δ inhibits enhanced cytokine production in adipocytes by preventing NF-κB activation via ERK1/2, an effect that may contribute to prevent insulin resistance.

PPARβ/δ prevents NF-κB activation in adipocytes

ccumulating evidence implicates a and MCP-1). Of note, NF-κB activation by low-grade chronic systemic LPS requires mitogen-activated protein A inflammatory response to nutrient kinase (MAPK)–extracellular signal– excess as a key mechanism that links related kinase (ERK)1/2 (MEK1/2) obesity to metabolic disorders, including activation, since inhibition of this pathway insulin resistance and cardiovascular reduces LPS-induced cytokine production disease (1). Thus, models of diet-induced in adipocytes (9). and genetic obesity show increased Recent evidence suggests that adipose tissue expression and content of inflammatory processes induced by pro-inflammatory cytokines (such as obesity and high-fat diet cause systemic tumor necrosis factor α [TNFα], insulin resistance via a mechanism interleukin [IL] 1, monocyte chemo- involving TLR4 (10). For instance, attractant protein-1 [MCP-1] and IL-6) (2- saturated free fatty acids (FFA) activate 4). Of these cytokines, IL-6 correlates TLR4-mediated inflammatory signaling in most strongly with insulin resistance and adipocytes and macrophages and this type 2 diabetes (5-7); its plasma levels effect is blunted in the absence of this are increased 2-3 fold in patients with receptor (10). These observations obesity and type 2 diabetes compared indicate that enhanced adipose tissue with lean control subjects (6). At the lipolysis observed in insulin-resistant cellular level, insulin resistance and states may release the endogenous enhanced expression of these cytokines ligand for TLR4 to induce inflammation by adipose tissue during obesity, and also (11). In addition, it has been under a high-fat diet have been linked to demonstrated that high-fat diets augment activation of the pro-inflammatory plasma LPS to a concentration sufficient transcription factor NF-κB (4). This to increase body weight, fasting glycemia nuclear factor is activated by surface and inflammation (12). Furthermore, LPS proteins that recognize foreign receptor-deleted mice (CD14 mutants) substances, the so-called pattern are hypersensitive to insulin, and the recognition receptors, such as toll-like development of insulin resistance, obesity receptor-4 (TLR4). This receptor is and diabetes in this animal model is expressed on virtually all human cells and delayed in response to a high-fat diet binds a wide spectrum of exogenous and (12). endogenous ligands, including bacterial In recent years Peroxisome LPS (8). In the presence of LPS, the Proliferator-Activated Receptor β/δ TLR4 complex (including CD-14 and an (PPARβ/δ) activation has been proposed accessory protein, MD-2), recruits the as a potential treatment for insulin adaptor protein, myeloid differentiation resistance (13). PPARs are members of factor-88 (MyD88), which in turn recruits the nuclear receptor superfamily of interleukin-1 receptor-associated kinase ligand-inducible transcription factors. (IRAK), leading to NF-κB activation and They form heterodimers with retinoid X enhanced expression of several receptors (RXRs) and bind to consensus inflammatory mediators (including IL-6 DNA sites composed of direct repeats PPARβ/δ prevents NF-κB activation in adipocytes

(DRs) of hexameric DNA sequences activity, and enhanced IL-6 expression separated by 1 bp (DR1) (14). Ligand and NF-κB DNA-binding activity in white binding induces a conformational change adipose tissue. Likewise, IL-6 expression in PPAR-RXR complexes, thereby and NF-κB DNA-binding activity was releasing co-repressors in exchange for higher in this tissue in PPARβ/δ-null mice co-activators, which leads to the than in wild-type mice. Since MAPK– recruitment of the basal transcription ERK1/2 (MEK1/2) is involved in NF-κB machinery and enhanced gene activation in adipocytes (9), we explored expression. In addition, PPARs may whether PPARβ/δ blocked NF-κB suppress inflammation through diverse activation by inhibiting this pathway. In mechanisms, such as reduced release of agreement with this possibility, inflammatory factors or stabilization of GW501516 prevented ERK1/2- repressive complexes at inflammatory phosphorylation by LPS. In contrast, gene promoters (15-18). Of the three animal models showing increased NF-κB PPAR isotypes found in mammals, activity in white adipose tissue, the ZDF PPARα (NR1C1) (19) and rat and the PPARβ/δ-null mice, showed PPARγ(NR1C3) are the targets for enhanced phospho-ERK1/2 levels. hypolipidemic (fibrates) and anti-diabetic Overall, on the basis of our findings, we (thiazolidinediones) drugs, respectively. propose that PPARβ/δ activation be Finally, activation of the third isotype, considered a molecular target to prevent PPARβ/δ (NR1C2), by high-affinity inflammation of adipose tissue and the ligands (including GW501516) enhances metabolic alterations associated with this fatty acid catabolism in adipose tissue process, such as insulin resistance. and skeletal muscle, thereby delaying weight gain (for review see (13)). RESEARCH DESIGN AND METHODS However, there is no information Materials. The PPARβ/δ ligand available on whether PPARβ/δ ligands GW501516 was from Biomol Research prevent inflammation in adipocytes. Here Labs Inc. (Plymouth Meeting, PA). Other we examined whether PPARβ/δ activation chemicals were from Sigma (St. Louis, by GW501516 prevents LPS-induced MO). inflammation in adipocytes. We found that this drug prevented LPS-induced IL-6 Cell culture. 3T3-L1 preadipocytes expression and secretion by adipocytes. (ATCC) were grown to confluence in This effect was associated with the Dulbecco’s Modified Eagle’s Medium capacity of the PPARβ/δ ligand to prevent (DMEM) supplemented with 10% bovine LPS-induced NF-κB activation. calf serum. Two 2 days after confluence Consistent with the role of PPARβ/δ in (day 0), differentiation of the 3T3-L1 cells blocking NF-κB-induced IL-6 expression, was induced in DMEM containing 10% a genetic model of obesity and insulin fetal bovine serum, resistance, the ZDF rat, showed reduced methylisobutylxanthine (500 µM), PPARβ/δ expression and DNA-binding dexamethasone (0.25 µM), and insulin 2

PPARβ/δ prevents NF-κB activation in adipocytes

(10 µg/ml) for 48 h. The cells were then The generation of PPARβ/δ null mice incubated in 10% FBS/DMEM with insulin was described previously (20). for 8 days. Medium was changed every 2 Measurements of mRNA. Levels of days. Fat droplets were observed in more mRNA were assessed by the reverse than 90% of cells after day 10. transcription-polymerase chain reaction Adipocytes were then incubated for 96 h (RT-PCR) as previously described (21). with 0.5 µM GW501516 and then with 100 Total RNA was isolated using the ng/ml LPS for either 1 or 24 h. After the Ultraspec reagent (Biotecx, Houston). The incubation, RNA, total proteins and total RNA isolated by this method is non- nuclear extracts were extracted from degraded and free of protein and DNA adipocytes as described below. Inhibitors contamination. The sequences of the were added 30 min before the incubation sense and antisense primers used for with LPS. Culture supernatants were amplification were: IL-6, 5’- collected, and the secretion of IL-6 was TCCAGCCAGTTGCCTTCTTGG-3’ and assessed by ELISA (Amersham 5’-TCTGACAGTGCATCATCGCTG-3’; Biosciences, Little Chalfont, UK). Mcp-1, 5’- GGGCCTGTTGTTCACAGTTGC-3’ and Animals. Male obese ZDF rats (ZDF/Gmi, 5’-GGGACACCTGCTGCTGGTGAT-3’; fa/fa) and lean litter mates (fa/+ or +/+) Pdk-4, 5’- were used. Both strains were maintained AGGTCGAGCTGTTCTCCCGCT-3’ and under standard light-dark cycle (12-h 5’-GCGGTCAGGCAGGATGTCAAT-3’; light/dark cycle) and temperature (21 ± Cpt-I, 5’- 1ºC) and fed with the Purina 5008 chow. TTCACTGTGACCCCAGACGGG-3’ and Epidydimal white adipose tissue was 5’-AATGGACCAGCCCCATGGAGA-3’; rapidly removed, frozen in liquid nitrogen Pparβ/δ 5’- and stored at -80ºC. Blood samples were GAGGAAGTGGCCACGGGTGAC-3’ and collected in EDTA tubes and plasma was 5’-CCACCTGAGGCCCCATCACAG-3’; obtained by centrifugation at 2200 g for 10 Socs-3 (Suppressor of cytokine signaling min at 4ºC. Plasma glucose (Sigma, St. 3) 5’-TTTTCGCTGCAGAGTGACCCC-3’ Louis, MO), triglycerides (Sigma), and and 5?-TGGAGGAGAGAGGTCGGCTCA- non-esterified fatty acids (Wako, Germany) 3’ and Aprt (adenosyl phosphoribosyl levels were determined with a colorimetric transferase), 5’- test. Insulin (Amersham Biosciences) was GCCTCTTGGCCAGTCACCTGA-3’ and determined by RIA. All procedures were 5’-CCAGGCTCACACACTCCACCA-3’. conducted in accordance with the Amplification of each gene yielded a single principles and guidelines established by band of the expected size (IL-6: 229 bp, the University of Barcelona Bioethics Mcp-1: 157 bp, Pdk-4: 167 bp, Cpt-I: 222 Committee, as stated in Law 5/1995, 21st bp, Pparβ/δ: 151 bp, Socs-3: 250 bp and July, passed by the Generalitat de Aprt: 329 bp). Preliminary experiments Catalunya. were carried out with various amounts of cDNA to determine non-saturating

PPARβ/δ prevents NF-κB activation in adipocytes

conditions of PCR amplification for all the consensus binding site of the NF-κB genes studied. Therefore, under these nucleotide (5'- conditions, relative quantification of mRNA AGTTGAGGGGACTTTCCCAGGC-3') was assessed by the RT-PCR method and PPAR (5’- used in this study (22). Radioactive bands CAAAACTAGGTCAAAGGTCA-3’). were quantified by video-densitometric Oligonucleotides were labeled in the scanning (Vilbert Lourmat Imaging). The following reaction: 2 μl of oligonucleotide results for the expression of specific (1.75 pmol/μl), 2 μl of 5x kinase buffer, 1 mRNAs are always presented relative to μl of T4 polynucleotide kinase (10 U./μl), the expression of the control gene (Aprt). and 2.5 μl of [γ-32P] ATP (3000 Ci/mmol at 10 mCi/ml) incubated at 37ºC for 1 h. Isolation of nuclear extracts. Nuclear The reaction was stopped by adding 90 μl extracts were isolated as previously of TE buffer (10 mM Tris-HCl pH 7.4 and described (21). Cells were scraped into 1.5 1 mM EDTA). To separate the labeled ml of cold phosphate-buffered saline, probe from the unbound ATP, the pelleted for 10 seconds and resuspended reaction mixture was eluted in a Nick in 400μl of cold Buffer A (10mM HEPES column (Amersham) following the pH 7.9 at 4ºC, 1.5mM MgCl2, 10mM KCl, manufacturer’s instructions. Eight 0.5mM DTT, 0.2mM PMSF, and 5μg/ml micrograms of crude nuclear protein was aprotinin) by flicking the tube. Cells were incubated for 10 min on ice in binding allowed to swell on ice for 10 min, and buffer (10 mM Tris-HCl pH 8.0, 25 mM then vortexed for 10 sec. Samples were KCl, 0.5 mM DTT, 0.1 mM EDTA pH 8.0, then centrifuged for 10 sec and the 5% glycerol, 5 mg/ml BSA and 50 μg/ml supernatant fraction was discarded. poly(dI-dC)), in a final volume of 15 μl. Pellets were resuspended in 50μl of cold Labeled probe (approximately 60,000 Buffer C (20mM HEPES-KOH pH 7.9 at cpm) was added and the reaction was 4ºC, 25% glycerol, 420mM NaCl, 1.5mM incubated for 15 min at 4ºC (NF-κB) or MgCl2, 0.2mM EDTA, 0.5mM DTT, 0.2mM room temperature (PPRE). Where PMSF, 5μg/ml aprotinin and 2μg/ml indicated, specific competitor leupeptin) and incubated on ice for 20 min oligonucleotide was added before the for high-salt extraction. Cellular debris was labeled probe and incubated for 10 min removed by centrifugation for 2 min at 4ºC on ice. p65 antibody was added 15 min and the supernatant fraction (containing before incubation with the labeled probe DNA-binding proteins) was stored at – at 4ºC. Protein-DNA complexes were 80ºC. Nuclear extract concentration was resolved by electrophoresis at 4ºC on a determined by the Bradford method. 5% acrylamide gel and subjected to autoradiography. Electrophoretic mobility shift assay (EMSA). EMSA was performed using Immunoblotting. To obtain total protein, double-stranded oligonucleotides cells and adipose tissue were (Promega, Madison, WI) for the homogenized in cold lysis buffer (5 mM 4

PPARβ/δ prevents NF-κB activation in adipocytes

Tris-HCl (pH 7.4), 1 mM EDTA, 0.1 mM PPARβ/δ activation prevents LPS- phenylmethylsulfonyl fluoride, 1 mM induced IL-6 expression and secretion in sodium orthovanadate, 5.4 μg/ml 3T3-L1 adipocytes aprotinin). The homogenate was Differentiated 3T3-L1 adipocytes were centrifuged at 16,700 g for 30 min at 4ºC. exposed for 24 h to 0.5 μM GW501516, a Protein concentration was measured by selective ligand for PPARβ/δ with an the Bradford method. Proteins (30 μg) apparent Kd value of 1nM and 1000-fold were resolved by SDS-PAGE on 10% higher affinity toward PPARβ/δ than separation gels and transferred to PPARα and PPARγ (23). Under these Immobilon polyvinylidene diflouride conditions, we evaluated the expression membranes (Millipore, Bedford, MA). of two well-known PPARβ/δ-target genes, Western blot analysis was performed pyruvate dehydrogenase kinase 4 (Pdk4) using antibodies against total and and carnitine palmitoyltransferase 1 (Cpt- phospho-ERK1/2 (Cell Signaling) and I) (24). Ligand treatment for 24 h caused total (Santa Cruz) and phospho-STAT3 a slight increase in the mRNA levels of (signal transducer and activator of 705 727 both Pdk4 (100%, P<0.01) and Cpt-I transcription 3) (Tyr and Ser ) (Cell (45%, P<0.05) (Figure 1A). When Signaling). Detection was achieved using adipocytes were treated for 96 h with 0.5 the EZ-ECL chemiluminescence kit μM GW501516, a greater increase was (Amersham). Size of detected proteins observed in the transcript levels of Pdk4 was estimated using protein molecular- (170%, P<0.001) and Cpt-I (300%, mass standards (Invitrogen, Barcelona, P<0.01) (Figure 1B). Therefore, we Spain). selected this longer exposure to the PPARβ/δ ligand in further experiments. Statistical Analyses. Results are No changes were observed either in the expressed as means ± S.D. of 5 separate expression of well-known PPARγ-target experiments. Significant differences were genes (i.e. Ob, data not shown), established by Student’s t-test or one-way confirming that the drug treatment was ANOVA, according to the number of selective for PPARβ/δ, or in the groups compared, using the GraphPad expression of markers of adipocyte Instat programme (GraphPad Software differentiation (i.e. Pref-1, data not V2.03) (GraphPad Softwware Inc., San shown), indicating that this latter process Diego, CA). In the latter case, when was unaffected. After a 96 h-incubation significant variations were found, the with GW501516, cells were exposed to Tukey-Kramer multiple comparisons test 100 ng/ml LPS for 24 h to induce was applied. Differences were considered inflammation. LPS activates NF-κB significant at P<0.05. through TLRs in adipocytes (9,25) and

induces insulin resistance (12,26). When we evaluated the mRNA levels of Mcp-1 and IL-6, two genes that are under the RESULTS control of the pro-inflammatory 5

PPARβ/δ prevents NF-κB activation in adipocytes

transcription factor NF-κB, we observed PPARβ/δ activation prevents LPS- that LPS treatment strongly increased the mediated NF-κB activation in 3T3-L1 former (22-fold induction, P<0.001), adipocytes whereas cells co-incubated with Activation of NF-κB plays a crucial role in GW501516 showed a significant IL-6 production in adipocytes, as reduction (51% reduction, P<0.01 vs. demonstrated by the effects of LPS-exposed cells) (Figure 2A). parthenolide. To test whether GW501516 Furthermore, when LPS-exposed cells prevented LPS-induced IL-6 expression were co-incubated with 10 μM by reducing NF-κB activity, we performed parthenolide, which specifically inhibits EMSA. NF-κB formed three complexes the activation of NF-κB (27), the effect of with nuclear proteins (complexes I to III) LPS was suppressed. Treatment with (Figure 3A). The specificity of the three LPS also led to enhanced IL-6 mRNA DNA-binding complexes was assessed in levels (4.2-fold induction, P<0.05), but competition experiments by adding an this increase was blocked in the presence excess of unlabeled NF-κB of GW501516 and parthenolide (Figure oligonucleotide. NF-κB binding activity, 2B). Likewise, incubation with LPS for 24 mainly of specific complex II, increased in h caused a 4.2-fold increase in the levels nuclear extracts from LPS-treated cells of IL-6 protein secreted into the culture (Figure 3B). In contrast, in the presence media (control 10.5±3.7 vs. LPS 55±11 of LPS and the PPARβ/δ ligand, the ng/ml, Figure 2C), whereas pre-treatment binding activity of NF-κB was reduced with GW501516 (56% reduction, (Figure 3A). Addition of antibody against P<0.001) and parthenolide (32% the p65 subunit of NF-κB completely reduction, p<0.05) significantly reduced supershifted complex II, thereby the secretion of this cytokine. Several indicating that this band comprised mainly cytokines, including IL-6, stimulate this subunit (Figure 3C). suppressor of cytokine signaling 3 (Socs-

3) expression in adipocytes (28), a Reduced PPARβ/δ activity and enhanced protein that inhibits insulin signaling in IL-6 expression in white adipose tissue of these cells (29). The elevated levels of an animal model of obesity and insulin mRNA encoding Socs-3 and the increase resistance in Stat-3 phosphorylation (Figures 2D and We hypothesized that a reduction in E) suggested that the increase in IL-6 release caused by LPS treatment might PPARβ/δ activity and the consequent have local actions (30). Interestingly, increase in NF-κB activity is responsible, GW501516 treatment prevented the at least in part, for the enhanced increase in Socs-3 expression and Stat-3 expression of IL-6 in white adipose tissue phosphorylation, a fact that is consistent of animal models of obesity and insulin with the reduction in IL-6 levels after drug resistance. To check this hypothesis we treatment. evaluated the expression of PPARβ/δ and its activity in white adipose tissue of ZDF (fa/fa) rats. Male ZDF and lean rats were 6

PPARβ/δ prevents NF-κB activation in adipocytes

sacrificed at 12 weeks of age. By this IL-6 expression was enhanced (4-fold age, ZDF rats presented established induction, P<0.05) (Figure 5C). In diabetes compared with lean animals, as agreement with the enhanced expression demonstrated by increased plasma levels of IL-6 in white adipose tissue of of glucose (505±67 vs. 167±37 mg/dl, PPARβ/δ-null mice, NF-κB binding activity P<0.0001), triglycerides (649±298 vs. was higher in this tissue compared to that 92±31 mg/dl, P<0.01), non-esterified fatty of wild-type mice (Figure 5D). acids (0.62±0.17 vs. 0.26±0.09 mEq/L, P<0.01) and insulin (19.3±7.9 vs. PPARβ/δ regulates ERK1/2 10.4±2.0 ng/ml, P<0.05). White adipose phosphorylation in adipocytes tissue of ZDF rats showed a significant Next we attempted to determine the reduction in Pparβ/δ transcripts (50% molecular mechanism by which PPARβ/δ reduction, P<0.05) (Figure 4A). Similarly, activation inhibits NF-κB. Although it has the mRNA levels of its target gene, Pdk4, been reported that PPARβ/δ activation were also reduced (41%, P<0.05) (Figure prevents LPS-induced degradation of 4B), suggesting that the activity of this IκBα, thus suppressing LPS-induced NF- PPAR isotype was reduced. In contrast, κB in cardiomyocytes (31), we did not the expression of IL-6 was significantly observe changes in the protein levels of enhanced (60%, P<0.05) (Figure 4C). this NF-κB inhibitor (data not shown) after EMSA showed that the DNA-binding GW501516 treatment. On the other hand, activity of PPAR was reduced in white ERK1/2 activation leads to NF-κB adipose tissue from ZDF rats compared activation in adipocytes (9). In agreement to lean animals (Figure 4D). In contrast, with this hypothesis, the adipose tissue of NF-κB DNA-binding activity was ZDF rats showed enhanced phospho- increased in ZDF rats compared to lean ERK1/2 levels in the adipose tissue of animals (Figure 4E). These data suggest ZDF rats compared to lean rats (Figure that the low-grade chronic systemic 6A). This observation supports a potential inflammatory associated with the relationship between activation of ERK1/2 development of obesity and insulin and NF-κB. To demonstrate the resistance involves a decreased involvement of the ERK-MAPK cascade PPARβ/δ activity and the subsequent in LPS-induced NF-κB activation in activation of NF-κB. adipocytes, we used U0126, a potent and To clearly demonstrate the specific ERK1/2 inhibitor, which binds to involvement of PPARβ/δ in the control of MEK, thereby inhibiting its catalytic NF-κB-mediated Il-6 expression in white activity as well as phosphorylation of adipose tissue, we used the PPARβ/δ-null ERK1/2. Cells exposed to LPS showed mice. The absence of Pparβ/δ in white increased mRNA levels of both Mcp-1 adipose tissue (Figure 5A) was (31-fold induction, P<0.001) and IL-6 (4- associated with a reduction in the mRNA fold induction, P<0.05), whereas co- levels of Pdk4 (65%, P<0.001) (Figure incubation with U0126 caused a 35% 5B) compared to wild-type mice, whereas (P<0.001) and 72% (P<0.05) reduction in 7

PPARβ/δ prevents NF-κB activation in adipocytes

the LPS-mediated induction of these two factor NF-κB as one of these signaling genes, respectively (Figures 6B and C). pathways (34) that links inflammation with As a control, we verified that U0126 obesity and type-2 diabetes. For instance, treatment inhibited LPS-induced ERK1/2 over-expression of the NF-κB activator activation (Figure 6D). IKKβ in mice results in increased Finally, we examined whether the inflammatory cytokine production and the PPARβ/δ ligand prevented LPS-induced onset of diabetes (35). Furthermore, in NF-κB activation by reducing ERK1/2 human adipose tissue, inhibition of NF-κB phosphorylation in adipocytes. suppresses the release of pro- Immunoblotting detection of total and inflammatory cytokines (36). Activation of phosphorylated ERK1/2 (Figure 6E) NF-κB can be achieved by either LPS or revealed that LPS treatment activated this FFA through TLR4 (26). Several studies pathway, whereas in the presence of the have proposed that LPS participates in PPARβ/δ agonist this pathway was initiating the subclinical inflammation blocked. These observations suggest that observed in obesity and type 2 diabetes activation of this transcription factor mellitus. Thus, circulating LPS levels are prevents NF-κB activation by inhibiting higher in type 2 diabetic patients than in ERK1/2 phosphorylation. To demonstrate lean healthy subjects and LPS stimulates that PPARβ/δ regulates the ERK1/2 the secretion of pro-inflammatory pathway, we explored the expression of cytokines in human adipocytes (37). this kinase in adipose tissue of the Here we demonstrate for the first time PPARβ/δ-null mice. These mice showed that PPARβ/δ activation in adipocytes enhanced phospho-ERK1/2 protein levels inhibits LPS-induced cytokine expression compared to wild-type mice (Figure 6F), and secretion by preventing NF-κB indicating that this nuclear receptor activation. We (38) and others (31) have regulates this pathway. previously proposed that PPARβ/δ activation inhibits NF-κB activation in DISCUSSION cardiac cells by either promoting a Obesity, insulin resistance and type 2 protein-protein interaction between diabetes are closely associated with low PPARβ/δ and the p65 subunit of NF-κB or grade of chronic inflammation by increasing the expression levels of the characterized by abnormal cytokine NF-κB inhibitor IκBα. In adipocytes, we production (32). The adipocyte plays a did not observe changes in either the crucial role in this process, since this cell PPARβ/δ-p65 interaction (data not is a source of cytokines (TNF-α, IL-6, shown) or in IκBα protein levels after MCP-1), which are secreted as a result of GW501516 treatment. These the activation of several signaling observations suggest that these cascades involved in obesity-induced mechanisms are not involved in the anti- insulin resistance (33). A number of inflammatory effects of this drug in studies have implicated chronic activation adipocytes. In contrast, it has been of the pro-inflammatory transcription previously reported that ERK1/2 8

PPARβ/δ prevents NF-κB activation in adipocytes

activation is crucial for the induction of which PPARβ/δ controls ERK-1/2 inflammatory changes in adipocytes (34) activation. and leads to enhanced NF-κB activity (9). Here we demonstrate a correlation In agreement with this role of ERK1/2 in between enhanced ERK1/2 inflammation in adipocytes, we observed phosphorylation and NF-κB activation in a reduction in the expression of pro- the white adipose tissue of a genetic inflammatory cytokines in these cells model of obesity and diabetes, the ZDF when exposed to LPS in the presence of rat. Given the low level of Pparβ/δ in the the MAPK pathway inhibitor, U0126. This white adipose tissue of these animals, we finding is consistent with previous studies hypothesize that a decrease in the reporting reduced cytokine expression expression of this nuclear receptor is one levels in the presence of this inhibitor in of the underlying mechanisms of this human adipocytes (9). These data correlation. On the basis of our data, we demonstrate that ERK1/2 inhibition propose that in this animal model, prevents NF-κB activation and the PPARβ/δ down-regulation leads to subsequent increase in cytokine enhanced ERK1/2 phosphorylation, which expression. In addition, we demonstrate in turn activates NF-κB in white adipose that PPARβ/δ activation inhibits ERK1/2 tissue and, as a result, induces pro- phosphorylation, which is consistent with inflammatory cytokine expression. This the reduction in NF-κB activation after hypothesis is supported by the GW501516 treatment. Furthermore, the observation that white adipose tissue increase in phospho-ERK1/2 levels from the PPARβ/δ-null mice shows observed in white adipose tissue from increased phospho-ERK1/2 levels and PPARβ/δ-null mice is consistent with NF-κB activity and higher expression of enhanced NF-κB activity. The IL-6 compared with wild-type mice. involvement of PPARβ/δ in ERK1/2 Although further studies are required to regulation has been previously reported demonstrate this relationship between in other cell types. In fact, Burdick et al. PPARβ/δ and NF-κB in white adipose (39) recently reported that the PPARβ/δ tissue, our data provide further insight agonist GW0742 reduces phospho- into the role of obesity-related insulin ERK1/2 levels in human keratinocytes. In resistance and low grade inflammation. addition, phospho-ERK1/2 levels are In summary, on the basis of our higher in the skin of PPARβ/δ-null mice findings in adipocytes, we propose that treated with 12-O-tetradecanoylphorbol- PPARβ/δ activation prevents LPS- 13-acetate (TPA) than in the skin of wild- induced NF-κB activation via ERK1/2, type animals (40). Overall, these data thereby reducing the production of pro- confirm that PPARβ/δ regulates phospho- inflammatory cytokines involved in the ERK1/2 levels in several tissues. development of insulin resistance. This Additional studies are required to action of PPARβ/δ may contribute to elucidate the molecular mechanism by preventing obesity-induced insulin resistance. 9

PPARβ/δ prevents NF-κB activation in adipocytes

ERDF funds and the Fundació Privada ACKNOWLEDGEMENTS Catalana de Nutrició i Lípids. Ricardo This study was partly supported by funds Rodríguez-Calvo was supported by a from the Swiss National Science grant from the Fundación Ramón Areces. Foundation, Generalitat de Catalunya Teresa Coll and Xavier Palomer (Juan de (SGROS-00833), Fundación Ramón la Cierva grant) were supported by grants Areces, the Spanish Ministerio de from the Spanish Ministerio de Educación Educación y Ciencia (SAF2006-01475), y Ciencia. We would like to thank the Instituto de Salud Carlos III-RETIC University of Barcelona’s Language RD06/0015/ERDF, European Union Advisory Service for its help.

PPARβ/δ prevents NF-κB activation in adipocytes

REFERENCES

1. Wellen,KE, Hotamisligil,GS: Inflammation, stress, and diabetes. J Clin Invest 115:1111- 1119, 2005 2. Hotamisligil,GS, Shargill,NS, Spiegelman,BM: Adipose expression of tumor necrosis factor-alpha: direct role in obesity-linked insulin resistance. Science 259:87-91, 1993 3. Weisberg,SP, McCann,D, Desai,M, Rosenbaum,M, Leibel,RL, Ferrante,AW, Jr.: Obesity is associated with macrophage accumulation in adipose tissue. J Clin Invest 112:1796-1808, 2003 4. Shoelson,SE, Lee,J, Goldfine,AB: Inflammation and insulin resistance. J Clin Invest 116:1793-1801, 2006 5. Pickup,JC, Mattock,MB, Chusney,GD, Burt,D: NIDDM as a disease of the innate immune system: association of acute-phase reactants and interleukin-6 with metabolic syndrome X. Diabetologia 40:1286-1292, 1997 6. Kern,PA, Ranganathan,S, Li,C, Wood,L, Ranganathan,G: Adipose tissue tumor necrosis factor and interleukin-6 expression in human obesity and insulin resistance. Am J Physiol Endocrinol Metab 280:E745-E751, 2001 7. Pradhan,AD, Manson,JE, Rifai,N, Buring,JE, Ridker,PM: C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus. JAMA 286:327-334, 2001 8. Akira,S, Uematsu,S, Takeuchi,O: Pathogen recognition and innate immunity. Cell 124:783-801, 2006 9. Chung,S, Lapoint,K, Martinez,K, Kennedy,A, Boysen,SM, McIntosh,MK: Preadipocytes mediate lipopolysaccharide-induced inflammation and insulin resistance in primary cultures of newly differentiated human adipocytes. Endocrinology 147:5340-5351, 2006 10. Shi,H, Kokoeva,MV, Inouye,K, Tzameli,I, Yin,H, Flier,JS: TLR4 links innate immunity and fatty acid-induced insulin resistance. J Clin Invest 116:3015-3025, 2006 11. Suganami,T, Mieda,T, Itoh,M, Shimoda,Y, Kamei,Y, Ogawa,Y: Attenuation of obesity- induced adipose tissue inflammation in C3H/HeJ mice carrying a Toll-like receptor 4 mutation. Biochem Biophys Res Commun 354:45-49, 2007 12. Cani,PD, Amar,J, Iglesias,MA, Poggi,M, Knauf,C, Bastelica,D, Neyrinck,AM, Fava,F, Tuohy,KM, Chabo,C, Waget,A, Delmee,E, Cousin,B, Sulpice,T, Chamontin,B, Ferrieres,J, Tanti,JF, Gibson,GR, Casteilla,L, Delzenne,NM, Alessi,MC, Burcelin,R: Metabolic endotoxemia initiates obesity and insulin resistance. Diabetes 56:1761-1772, 2007 13. Barish,GD, Narkar,VA, Evans,RM: PPAR delta: a dagger in the heart of the metabolic syndrome. J Clin Invest 116:590-597, 2006 14. Michalik,L, Auwerx,J, Berger,JP, Chatterjee,VK, Glass,CK, Gonzalez,FJ, Grimaldi,PA, Kadowaki,T, Lazar,MA, O'Rahilly,S, Palmer,CN, Plutzky,J, Reddy,JK, Spiegelman,BM, Staels,B, Wahli,W: International Union of Pharmacology. LXI. Peroxisome proliferator- activated receptors. Pharmacol Rev 58:726-741, 2006

PPARβ/δ prevents NF-κB activation in adipocytes

15. Lee,CH, Chawla,A, Urbiztondo,N, Liao,D, Boisvert,WA, Evans,RM, Curtiss,LK: Transcriptional repression of atherogenic inflammation: modulation by PPARdelta. Science 302:453-457, 2003 16. Pascual,G, Fong,AL, Ogawa,S, Gamliel,A, Li,AC, Perissi,V, Rose,DW, Willson,TM, Rosenfeld,MG, Glass,CK: A SUMOylation-dependent pathway mediates transrepression of inflammatory response genes by PPAR-gamma. Nature 437:759-763, 2005 17. Daynes,RA, Jones,DC: Emerging roles of PPARs in inflammation and immunity. Nature Reviews Immunology 2:748-759, 2002 18. Devchand,PR, Keller,H, Peters,JM, Vazquez,M, Gonzalez,FJ, Wahli,W: The PPARalpha- leukotriene B4 pathway to inflammation control. Nature 384:39-43, 1996 19. Auwerx,J, Baulieu,E, Beato,M, Becker-Andre,M, Burbach,PH, Camerino,G, Chambon,P, Cooney,A, Dejean,A, Dreyer,C, Evans,RM, Gannon,F, Giguere,V, Gronemeyer,H, Gustafson,JA, Laudet,V, Lazar,MA, Mangelsdorf,DJ, Milbrandt,J, Milgrom,E, Moore,DD, O'Malley,B, Parker,M, Parker,K, Perlmann,T, Pfahl,M, Rosenfeld,MG, Samuels,H, Schutz,G, Sladek,FM, Stunnenberg,HG, Spedding,M, Thummel,C, Tsai,MJ, Umesono,K, Vennstrom,B, Wahli,W, Weinberger,C, Willson,TM, Yamamoto,K: A unified nomenclature system for the nuclear receptor superfamily. Cell 97:161-163, 1999 20. Nadra,K, Anghel,SI, Joye,E, Tan,NS, Basu-Modak,S, Trono,D, Wahli,W, Desvergne,B: Differentiation of trophoblast giant cells and their metabolic functions are dependent on peroxisome proliferator-activated receptor beta/delta. Mol Cell Biol 26:3266-3281, 2006 21. Coll,T, Jove,M, Rodriguez-Calvo,R, Eyre,E, Palomer,X, Sanchez,RM, Merlos,M, Laguna,JC, Vazquez-Carrera,M: Palmitate-mediated downregulation of peroxisome proliferator-activated receptor-gamma coactivator 1alpha in skeletal muscle cells involves MEK1/2 and nuclear factor-kappaB activation. Diabetes 55:2779-2787, 2006 22. Freeman,WM, Walker,SJ, Vrana,KE: Quantitative RT-PCR: Pitfalls and potential. Biotechniques 26:112-+, 1999 23. Oliver,WR, Jr., Shenk,JL, Snaith,MR, Russell,CS, Plunket,KD, Bodkin,NL, Lewis,MC, Winegar,DA, Sznaidman,ML, Lambert,MH, Xu,HE, Sternbach,DD, Kliewer,SA, Hansen,BC, Willson,TM: A selective peroxisome proliferator-activated receptor delta agonist promotes reverse cholesterol transport. Proc Natl Acad Sci U S A 98:5306-5311, 2001 24. Tanaka,T, Yamamoto,J, Iwasaki,S, Asaba,H, Hamura,H, Ikeda,Y, Watanabe,M, Magoori,K, Ioka,RX, Tachibana,K, Watanabe,Y, Uchiyama,Y, Sumi,K, Iguchi,H, Ito,S, Doi,T, Hamakubo,T, Naito,M, Auwerx,J, Yanagisawa,M, Kodama,T, Sakai,J: Activation of peroxisome proliferator-activated receptor delta induces fatty acid beta-oxidation in skeletal muscle and attenuates metabolic syndrome. Proc Natl Acad Sci U S A 100:15924- 15929, 2003 25. Lin,Y, Lee,H, Berg,AH, Lisanti,MP, Shapiro,L, Scherer,PE: The lipopolysaccharide- activated toll-like receptor (TLR)-4 induces synthesis of the closely related receptor TLR-2 in adipocytes. J Biol Chem 275:24255-24263, 2000

PPARβ/δ prevents NF-κB activation in adipocytes

26. Song,MJ, Kim,KH, Yoon,JM, Kim,JB: Activation of Toll-like receptor 4 is associated with insulin resistance in adipocytes. Biochem Biophys Res Commun 346:739-745, 2006 27. Hehner,SP, Heinrich,M, Bork,PM, Vogt,M, Ratter,F, Lehmann,V, Schulze-Osthoff,K, Droge,W, Schmitz,ML: Sesquiterpene lactones specifically inhibit activation of NF-kappa B by preventing the degradation of I kappa B-alpha and I kappa B-beta. J Biol Chem 273:1288-1297, 1998 28. Shi,H, Tzameli,I, Bjorbaek,C, Flier,JS: Suppressor of cytokine signaling 3 is a physiological regulator of adipocyte insulin signaling. J Biol Chem 279:34733-34740, 2004 29. Shi,H, Cave,B, Inouye,K, Bjorbaek,C, Flier,JS: Overexpression of suppressor of cytokine signaling 3 in adipose tissue causes local but not systemic insulin resistance. Diabetes 55:699-707, 2006 30. Lagathu,C, Bastard,JP, Auclair,M, Maachi,M, Capeau,J, Caron,M: Chronic interleukin-6 (IL-6) treatment increased IL-6 secretion and induced insulin resistance in adipocyte: prevention by rosiglitazone. Biochem Biophys Res Commun 311:372-379, 2003 31. Ding,G, Cheng,L, Qin,Q, Frontin,S, Yang,Q: PPARdelta modulates lipopolysaccharide- induced TNFalpha inflammation signaling in cultured cardiomyocytes. J Mol Cell Cardiol 40:821-828, 2006 32. Hotamisligil,GS: Inflammation and metabolic disorders. Nature 444:860-867, 2006 33. Tataranni,PA, Ortega,E: A burning question: does an adipokine-induced activation of the immune system mediate the effect of overnutrition on type 2 diabetes? Diabetes 54:917- 927, 2005 34. Suganami,T, Tanimoto-Koyama,K, Nishida,J, Itoh,M, Yuan,X, Mizuarai,S, Kotani,H, Yamaoka,S, Miyake,K, Aoe,S, Kamei,Y, Ogawa,Y: Role of the Toll-like receptor 4/NF- kappaB pathway in saturated fatty acid-induced inflammatory changes in the interaction between adipocytes and macrophages. Arterioscler Thromb Vasc Biol 27:84-91, 2007 35. Cai,D, Yuan,M, Frantz,DF, Melendez,PA, Hansen,L, Lee,J, Shoelson,SE: Local and systemic insulin resistance resulting from hepatic activation of IKK-beta and NF-kappaB. Nat Med 11:183-190, 2005 36. Lappas,M, Yee,K, Permezel,M, Rice,GE: Sulfasalazine and BAY 11-7082 interfere with the nuclear factor-kappa B and I kappa B kinase pathway to regulate the release of proinflammatory cytokines from human adipose tissue and skeletal muscle in vitro. Endocrinology 146:1491-1497, 2005 37. Creely,SJ, McTernan,PG, Kusminski,CM, Fisher,M, Da Silva,NF, Khanolkar,M, Evans,M, Harte,AL, Kumar,S: Lipopolysaccharide activates an innate immune system response in human adipose tissue in obesity and type 2 diabetes. Am J Physiol Endocrinol Metab 292:E740-E747, 2007 38. Planavila,A, Laguna,JC, Vazquez-Carrera,M: Nuclear factor-kappaB activation leads to down-regulation of fatty acid oxidation during cardiac hypertrophy. J Biol Chem 280:17464-17471, 2005

PPARβ/δ prevents NF-κB activation in adipocytes

39. Burdick,AD, Bility,MT, Girroir,EE, Billin,AN, Willson,TM, Gonzalez,FJ, Peters,JM: Ligand activation of peroxisome proliferator-activated receptor-beta/delta(PPARbeta/delta) inhibits cell growth of human N/TERT-1 keratinocytes. Cell Signal 19:1163-1171, 2007 40. Kim,DJ, Murray,IA, Burns,AM, Gonzalez,FJ, Perdew,GH, Peters,JM: Peroxisome proliferator-activated receptor-beta/delta inhibits epidermal cell proliferation by down- regulation of kinase activity. J Biol Chem 280:9519-9527, 2005

PPARβ/δ prevents NF-κB activation in adipocytes

FIG. 1. The PPARβ/δ agonist GW501516 induces Pdk-4 and Cpt-I expression in 3T3-L1 adipocytes. Differentiated adipocytes were incubated in the presence or in the absence of 0.5 µM GW501516 for either 24 h (A) or 96 h (B). Total RNA was isolated and analyzed by RT-PCR. A representative autoradiogram and the quantification normalized to the Aprt mRNA levels are shown. Data are expressed as means ± S.D. of five independent experiments. *P<0.05, **P<0.01 and ***P<0.001 vs. control.

PPARβ/δ prevents NF-κB activation in adipocytes

FIG. 2. The PPARβ/δ agonist GW501516 prevents LPS-induced expression and secretion of pro-inflammatory cytokines. When indicated differentiated adipocytes were incubated with 0.5 µM GW501516 for 96 h. and subsequently exposed to 100 ng/ml LPS for 24 h. in the presence or in the absence of GW501516 or 10 µM of the NF-κB inhibitor parthenolide (PARTH). Effects of GW501516 on the expression of Mcp-1 (A) and Il-6 (B). Total RNA was isolated and analyzed by RT-PCR. A representative autoradiogram and the quantification normalized to the Aprt mRNA levels are shown. C, Effect of GW501516 on the secretion of Il-6 to the culture media, as determined by ELISA. D, Effect of GW501516 on the expression of SOCS-3. E, Analysis of STAT3 and phospho-STAT3 by immunoblotting of nuclear protein extracts from 3T3-L1 adipocytes treated with 100 ng/ml LPS for 3 h. in the presence or in the absence of 0.5 µM GW501516. Data are expressed as means ± S.D. of five independent experiments. *P<0.05, **P<0.01 and ***P<0.001 vs. control. #P<0.05, ##P<0.01, ###P<0.001 vs LPS- exposed cells.

PPARβ/δ prevents NF-κB activation in adipocytes

FIG. 3. The PPARβ/δ agonist GW501516 prevents LPS-induced NF-κB activation in 3T3-L1 adipocytes. When indicated, differentiated adipocytes were incubated with 0.5 µM GW501516 for 96 h and subsequently exposed to 100 ng/ml LPS for 1 h in the presence or in the absence of GW501516. A, Autoradiograph of EMSA performed with a 32P-labeled NF-κB nucleotide and crude nuclear protein extract (NE). Three specific complexes (I to III), based on competition with a molar excess of unlabeled probe (B), are shown. C, A supershift analysis performed by incubating NE with an antibody directed against the p65 subunit of NF-κB is also shown.

PPARβ/δ prevents NF-κB activation in adipocytes

FIG. 4. PPARβ/δ expression is reduced, whereas NF-κB activity is increased in white adipose tissue of ZDF rats. Analysis of the mRNA levels of Pparβ/δ (A), Pdk-4 (B) and Il-6 (C) in white adipose tissue of lean and ZDF rats. Total RNA was isolated and analyzed by RT-PCR. A representative autoradiogram and the quantification normalized to the Aprt mRNA levels are shown. Data are expressed as means ± S.D. of five independent experiments. *P<0.05. D, Autoradiograph of EMSA performed with a 32P-labeled PPRE nucleotide and crude nuclear protein extract (NE). Three specific complexes (I to III), based on competition with a molar excess of unlabeled probe (right panel), are shown. An analysis performed by incubating NE with an antibody directed against PPARβ/δ is also shown. This antibody does not shift the complex, but prevents its binding to the PPRE. E, Autoradiograph of EMSA performed with a 32P-labeled NF-κB nucleotide and NE. Two specific complexes, based on competition with a molar excess of unlabeled probe (right panel), are shown.

PPARβ/δ prevents NF-κB activation in adipocytes

FIG. 5. The PPARβ/δ-null mouse shows increased NF-κB activity in white adipose tissue. Analysis of the mRNA levels of Pparβ/δ (A), Pdk-4 (B) and Il-6 (C) in white adipose tissue of wild-type (wt) or PPARβ/δ-null (ko) mouse white adipose tissue. Total RNA was isolated and analyzed by RT-PCR. A representative autoradiogram and the quantification normalized to the Aprt mRNA levels are shown. Data are expressed as means ± S.D. of five independent experiments. *P<0.05. D, Autoradiograph of EMSA performed with a 32P-labeled NF-κB nucleotide and crude nuclear protein extract (NE). One specific complex, based on competition with a molar excess of unlabeled probe (right panel), is shown.

PPARβ/δ prevents NF-κB activation in adipocytes

FIG. 6. PPARβ/δ regulates ERK1/2 phosphorylation in adipocytes and white adipose tissue. A, The MEK1/2 inhibitor U0126 blocks NF- B activation by LPS. Analysis of the mRNA levels of Mcp-1 (A) and Il-6 (B) in 3T3-L1 adipocytes. Differentiated adipocytes were treated with 100 ng/ml LPS for 6h in the presence or in the absence of 10 µM of the MEK1/2 inhibitor U0126. Total RNA was isolated and analyzed by RT-PCR. A representative autoradiogram and the quantification normalized to the Aprt mRNA levels are shown. Data are expressed as means ± S.D. of five independent experiments. *P<0.05 and ***P<0.001 vs. control. #P<0.05, ###P<0.001 vs. LPS-exposed cells. Analysis of ERK and phospho-ERK by immunoblotting of total protein extracts from: C, 3T3-L1 adipocytes treated with 100 ng/ml LPS for 10 min in the presence or in the absence of 0.5 µM GW501516; D, white adipose tissue of lean and ZDF rats; E, wild-type (wt) or PPARβ/δ-null (ko) mouse white adipose tissue. 21