Fatty acids cause endothelial dysfunction involving increased ROS (reactive oxygen species) and reduced NO (nitric oxide) bioavailability. We show that in MAECs (mouse aortic endothelial cells), the PPARβ/δ (peroxisome- proliferator-activated receptor β/δ) agonist GW0742 prevented the decreased A23187-stimulated NO production, phosphorylation of eNOS (endothelial nitric oxide synthase) at Ser1177 and increased intracellular ROS levels caused by exposure to palmitate in vitro. The impaired endothelium-dependent relaxation to acetylcholine in mouse aorta induced by palmitate was restored by GW0742. In vivo, GW0742 treatment prevented the reduced aortic relaxation, phosphorylation of eNOS at Ser1177, and increased ROS production and NADPH oxidase in mice fed on a high-fat diet. The PPARβ/δ antagonist GSK0660 abolished all of these protective effects induced by GW0742. This agonist enhanced the expression of CPT (carnitine palmitoyltransferase)-1. The effects of GW0742 on acetylcholine- induced relaxation in aorta and on NO and ROS production in MAECs exposed to palmitate were abolished by the CPT-1 inhibitor etomoxir or by siRNA targeting CPT-1. GW0742 also inhibited the increase in DAG (diacylglycerol), PKCα/βII (protein kinase Cα/βII) activation, and phosphorylation of eNOS at Thr495 induced by palmitate in MAECs, which were abolished by etomoxir. In conclusion, PPARβ/δ activation restored the lipid-induced endothelial dysfunction by up-regulation of CPT-1, thus reducing DAG accumulation and the subsequent PKC-mediated ROS production and eNOS inhibition.

CLINICAL PERSPECTIVES

  • The present study was carried out to examine the protective effects of PPARβ/δ activation on endothelial dysfunction and vascular inflammation associated with lipid metabolic disorders.

  • Activation of PPARβ/δ restored lipid-induced endothelial dysfunction, independently of changes in body weight and plasma levels of lipids or glucose. This protective effect was associated with the reduction in DAG accumulation, as a result of increased fatty acid β-oxidation in endothelial cells via up-regulation of CPT-1, and the subsequent PKC-mediated ROS production and eNOS inhibition.

  • These results provide a novel therapeutic target, modulation of CPT-1 in endothelial cells, to prevent vascular oxidative stress and lipid-induced endothelial dysfunction, an early marker of atherosclerosis.

INTRODUCTION

Endothelial cell dysfunction is an early event involved in the development of atherosclerosis. An HFD (high-fat diet) impairs endothelial function in animals and humans [1,2]. Elevated circulating concentrations of saturated NEFAs [non-esterified (‘free’) fatty acids] are implicated in the mechanism underlying endothelial dysfunction associated with lipid metabolic disorders. Potential mechanisms whereby exposure to saturated NEFAs induces changes in the endothelium include impairment of eNOS (endothelial nitric oxide synthase) activity, and increased ROS (reactive oxygen species) such as superoxide (O2•−) production [3], that can be generated by both mitochondrial electron transport [4] and cytosolic enzymes such as the NOX family of NADPH oxidases [5]. These enzymes transfer electrons from NADPH across cell membranes and are a major source of cytoplasmic ROS.

Decreasing lipotoxicity may be a key component to prevent and treat cardiovascular complications of lipid metabolic disorders. Accumulation of fatty acid derivatives can be attenuated by their metabolism via mitochondrial β-oxidation. The rate-limiting step for β-oxidation of long-chain fatty acids is their transport into mitochondria via CPT (carnitine palmitoyltransferase)-1. Activation of fatty acid oxidation by overexpressing CPT-1 in skeletal muscle improves lipid-induced insulin resistance [68].

The PPARs (peroxisome-proliferator-activated receptors) PPARα, PPARβ/δ and PPARγ are members of the nuclear hormone receptor superfamily. PPARβ/δ is expressed in multiple cell types, including adipose and endothelial cells [9]. PPARβ/δ activation promotes fatty acid β-oxidation in adipocytes and skeletal muscle, decreases lipid accumulation and reduces obesity [10]. PPARβ/δ also regulates glucose homoeostasis in endothelial cells, preventing glucose-induced impairment of endothelial insulin signalling in diabetic rodents [11,12]. Lipid metabolism in endothelial cells also involves conventional pathways, with functional rates much lower than in hepatocytes or in cardiomyocytes [13], which might be regulated by PPARβ/δ activation, as in other metabolic tissues [7,8,14,15].

In the present study, we hypothesized that PPARβ/δ activation would protect against endothelial dysfunction and vascular inflammation induced by lipids as a result of increased fatty acid β-oxidation in blood vessels. As a major component of dietary saturated fat and 20% of the total serum NEFAs, palmitic acid is often used to induce endothelial dysfunction [1618]. In the present study, we investigated the effects of PPARβ/δ activation on endothelial dysfunction in vitro, induced by palmitate in isolated vessels and cultured endothelial cells, and in vivo, in mice fed on an HFD.

MATERIALS AND METHODS

Ex vivo culture of mouse aortic rings

Mouse thoracic aortic rings (2 mm in length) were dissected in sterile PBS and incubated in DMEM (Dulbecco's modified Eagle's medium) supplemented with 10% (v/v) FBS, plus 100 units/ml penicillin and 100 μg/ml streptomycin. Lipid-containing media were prepared by conjugation of palmitic acid with fatty-acid-free BSA, using a method modified from that described previously [19]. Briefly, palmitic acid was dissolved in ethanol and diluted 1:100 in DMEM containing 2% (w/v) fatty-acid-free BSA. Aortic rings were incubated for 24 h in serum-free DMEM containing 2% fatty-acid-free BSA (control) or palmitate (100 μM)-conjugated BSA in either the presence or the absence of GW0742 (1 μM), or with GSK0660 (1 μM), or with the CPT-1 inhibitor etomoxir (40 μM). After the incubation period, the rings were transferred to a chamber filled with fresh Krebs solution and mounted in a myograph for measurement of changes in isometric force or used in Western blot analysis.

Primary culture of MAECs (mouse aortic endothelial cells)

MAECs were isolated from mouse thoracic aortic rings using a previously reported method with several modifications [20]. The cells were cultured (medium 199 plus 20% FBS, 2 mM penicillin/streptomycin, 2 mM amphotericin B, 2 mM glutamine, 10 mM Hepes, 30 μg/ml endothelial cell growth supplement and 100 mg/ml heparin) under 5% CO2 at 37°C. MAECs were incubated with the PPARβ/δ agonist GW0742 (1 μM) for 15 h, and in the last 3 h in serum-free medium 199 containing 2% fatty acid-free BSA (control cells) or palmitate (100 μM)-conjugated BSA. In some experiments, cells were co-incubated with the PPAR-β/δ antagonist GSK0660 (1 μM) 1 h before the addition of GW0742, or with etomoxir (40 μM) during palmitate incubation. Cells were then used to measure NO (nitric oxide) or ROS production, and gene and protein expression.

Transfection of CPT-1 siRNAs

Confluent MAECs were transfected with control or CPT-1-specific siRNA (pooled, validated siRNA from Dharmacon) using Lipofectamine RNAiMAX (Invitrogen) for 48 h, essentially as described previously [11].

Quantification of NO released by DAF-2 (diaminofluorescein-2)

Quantification of NO released by MAECs was performed using the NO-sensitive fluorescent probe DAF-2 as described previously [11]. Briefly, cells were incubated as mentioned above. After this period, cells were washed with PBS and then pre-incubated with 100 μM L-arginine in PBS for 5 min at 37°C. Subsequently, 0.1 μM DAF-2 was incubated for 2 min and then 1 μM calcium ionophore calimycin (A23187) was added for 30 min. Then the fluorescence intensity (in arbitrary units, AU) was measured using a spectrofluorimeter (Fluorostart, BMG Labtechnologies). The autofluorescence was subtracted from each value. In some experiments, 100 μM L-NAME (NG-nitro-L-arginine methyl ester) was added 15 min before the addition of L-arginine. The difference between fluorescence signal without and with L-NAME was considered NO production.

Measurement of intracellular ROS concentrations

Endothelial ROS production was measured using two independent methods. (i) The fluorescent probe CM-H2DCFDA [5-(and-6)-chloromethyl-2′,7′-dichlorodihydrofluorescein diacetate] was used to determine the intracellular generation of ROS in endothelial cells. Confluent MAECs in 96-well plates, incubated as described above, were then incubated with 5 μmol/l CM-H2DCFDA for 30 min at 37°C. The fluorescence intensity was measured using a spectrofluorimeter (Fluorostart, BMG Labtechnologies). (ii) The O2 production in intact endothelial cells was also measured by DHE (dihydroethidium) fluorescence HPLC assay as described previously [21,22]. After incubation as described above, MAECs were washed with PBS/DTPA (diethylenetriaminepenta-acetic acid) two to three times. All subsequent steps were carried out under dim light. PBS/DTPA and DHE (50 μM) were added to each well and plates were kept in an incubator (at 37°C under 5% CO2) for 30 min. Thereafter, cells were washed two to three times with PBS/DTPA. Acetonitrile (500 μl) was added to each well, and the cells were immediately harvested and the lysate was kept over ice. The lysates were centrifuged at 12000 g for 10 min at 4°C, and supernatants were dried under vacuum. For HPLC analysis, pellets were resuspended in 120 μl of PBS/DTPA, and a volume of 50 μl was injected. Samples were analysed by HPLC (Varian 920 LC series), using a 4 μm C18 reverse-phase column (Synergi 150 mm×4.6 mm, Phenomenex) and a gradient of solutions A (pure acetonitrile) and B (water/10% acetonitrile/0.1% trifluoroacetic acid, by vol.) at a flow rate of 1 ml/min, and run as described in [22]. Ethidium and 2-OH-E+ (2-hydroxyethidium) were monitored by fluorescence detection with excitation at 510 nm and emission at 595 nm. The 2-OH-E+ peak reflects the amount of O2 formed in the cells during the incubation. Values were normalized per μg of protein and the increase in the 2-OH-E+ peak is represented as an increase (n-fold) compared with control. To optimize the HPLC analysis, 50 μmol/l DHE was incubated with xanthine/xanthine oxidase (0–50 μmol/l and 0.1 unit/ml respectively) in KHS (Krebs–Henseleit solution)/Hepes/DTPA at 37°C for 30 min. 2-OH-E+ was separated by HPLC as described above. A representative HPLC trace of 2-OH-E+ measurements is shown in Supplementary Figure S3.

Animals and experimental groups

All procedures conformed to the Guide for the Care and Use of Laboratory Animals (National Institutes of Health publication no. 85-23, revised 1996) and were approved by our Institutional Committee for the ethical care of animals. Five-week-old male C57BL/6J mice were obtained from Janvier (St Berthevin Cedex, France). Mice were divided into four groups (n=10): control, HFD, HFD-treated and HFD+GSK0660-treated. Control mice received standard chow diet (13% calories from fat, 20% calories from protein, 67% calories from carbohydrate) (Global diet 2014, Harlan Laboratories), whereas HFD mice were fed a Western-type diet in which 60% of its caloric content was derived from fat (Purified diet 230 HF, Scientific Animal Food & Engineering). The PPARβ/δ antagonist GSK0660 was administered with 0.1 ml at 1 mg/kg per day i.p. (intraperitoneally) [23]. GSK0660 was diluted first in DMSO and later in 0.9% NaCl (DMSO less than 1%). Treated mice received GW0742 by oral gavage 3 mg/kg per day, mixed in 0.1 ml of 1% methylcellulose. Mice were treated with the corresponding oral and intraperitoneal vehicles of the substances given to the other groups. GW0742 treatment was followed for 3 weeks. A short-term study period of 3 weeks was selected to study the specific effects of lipid-induced endothelial dysfunction on the vessels, while avoiding other confounding factors such as obesity and established insulin resistance. Body weight, food and water intake were controlled regularly. All mice had free access to tap water and chow during the experimental period.

SBP (systolic blood pressure) was determined once a week, in the morning, 18–20 h after administration of the drugs in conscious pre-warmed restrained rats by tail-cuff plethysmography (Letica LE 5001 digital pressure meter) [24]. At 1 day before the mice were killed, a glucose tolerance test was performed on mice that had been starved for 18 h, as described previously [25]. At the end of the treatment, mice were killed under isoflurane anaesthesia. Blood samples were chilled on ice and centrifuged at 5000 g for 20 min at 4°C, and the plasma was frozen at −70°C. Plasma glucose, triacylglycerols, HDL (high-density lipoprotein) and total cholesterol concentrations were measured by colorimetric methods using Spinreact kits. Plasma NEFA concentration was determined using a Wako NEFA C test kit (Wako Chemicals). The heart, kidney and epididymal and mesenteric fat were excised, cleaned and weighed. The heart and kidney weight indices were calculated by dividing the heart and kidney weight by the tibia length. All tissue samples were frozen in liquid nitrogen and then stored at −80°C.

Vascular reactivity studies

Descending thoracic aortic rings were suspended in a wire myograph (model 610M, Danish Myo Technology) for isometric tension measurement as described previously [25]. In endothelium-intact aorta, concentration–relaxation response curves to acetylcholine (10−9–10−5 M) were performed in intact rings pre-contracted by 10−8 M U46619 in control or 100 μM L-NAME-treated aortic rings. The relaxant responses to sodium nitroprusside (10−9–10−5 M) were studied in the dark in aortic rings, with endothelium removed by gently rubbing with the forceps. In some experiments, responses to acetylcholine were studied after incubation with 0.1 μM of the mitochondrial antioxidant mitoQ (mitoquinone) (generously given by Dr M.P. Murphy, Medical Research Council Mitochondrial Biology Unit, Cambridge, U.K.) or 10 μM of the NADPH oxidase inhibitor apocynin for 60 min before the addition of U46619. Relaxant responses to acetylcholine and sodium nitroprusside were expressed as a percentage of pre-contraction values.

In situ detection of vascular ROS content

Unfixed thoracic aortic rings were cryopreserved in 0.1 mol/l PBS plus 30% sucrose for 1–2 h, included in OCT (optimum cutting temperature) compound medium (Tissue-Tek; Sakura Finetechnical), frozen (−80°C) and 10 μm cross sections were obtained in a cryostat (Microm International Model HM500 OM). Sections were incubated for 30 min in Hepes-buffered solution, containing 10−5 M DHE, counterstained with the nuclear stain DAPI at 3×10−7 M and in the following 24 h examined using a fluorescence microscope (Leica DM IRB,). Sections were photographed and fluorescence was quantified using ImageJ (version 1.32j, NIH, http://rsb.info.nih/ij/). All parameters (pinhole, contrast, gain and offset) were held constant for all sections from the same experiment. ROS production was estimated from the ratio of ethidium/DAPI fluorescence [24]. DHE fluorescence was also analysed in aortic sections from the HFD group incubated at 37°C for 30 min in the presence of 25 units/ml PEG–SOD (superoxide dismutase), before DHE addition.

NADPH oxidase activity

NADPH-enhanced O2•− release in homogenates from cultured MAECs or intact aortic rings was quantified by lucigenin-enhanced chemiluminescence, as described previously [26]. Briefly, cells were incubated as described above. After this period, cells were homogenized. Then NADPH (100 μM) was added to the buffer containing the MAEC homogenate suspension (30 μg of protein in 500 μl) or aortic rings, and 5 μM lucigenin was injected automatically. NADPH oxidase activity was calculated by subtracting the basal values from those in the presence of NADPH and expressed as RLU (relative light units)/min per μg of protein for cells or RLU/min per mg of tissue for aortic rings. The NADPH oxidase activity in intact MAECs was also measured using a DHE fluorescence assay in the microplate reader, as described previously [21]. Confluent MAECs grown in six-well dishes (well area of 9.6 cm2) incubated as previously described were washed with ice-cold PBS, harvested, homogenized in lysis buffer composed of 50 mM Tris/HCl (pH 7.4) containing 0.1 mM EDTA, 0.1 mM EGTA, 10 μg/ml aprotinin, 10 μg/ml leupeptin and 1 mM PMSF, and sonicated (10 s of three cycles at 8 W). Fresh homogenates (10 μg of protein) were incubated with DHE (10 μM) and DNA (1.25 μg/ml) in phosphate buffer (100 mM), pH 7.4, containing 100 μM DTPA with the addition of NADPH (50 μM), at a final volume of 120 μl. Incubations were performed for 30 min at 37°C, in the absence or the presence of 25 units/ml PEG–SOD in the dark. Total fluorescence was followed in a microplate reader using a rhodamine filter (excitation 490 nm and emission 590 nm) in a spectrofluorimeter (Fluorostart, BMG Labtechnologies).

RT (reverse transcription)–PCR analysis

For RT–PCR analysis, total RNA was extracted from aorta or MAECs by homogenization and converted into cDNA by standard methods. PCR was performed with a Techne Techgene thermocycler. A quantitative real-time RT–PCR technique was used to analyse mRNA expression of caveolin-1, eNOS, NOX1, NOX4, p22phox, p47phox, IL (interleukin)-1β, IL-6, TNFα (tumour necrosis factor α), GPx1 (glutathione peroxidase 1), HO-1 (haem oxygenase 1), Cu/Zn-SOD, Mn-SOD, CPT-1 and UCP2 (uncoupling protein 2). The sequences of the sense and antisense primers used for amplification are described in Table 1. Relative quantification of mRNA was assessed by RT–PCR. Quantification was performed using the ∆∆CT method. The housekeeping genes GAPDH (glyceraldehyde-3-phosphate dehydrogenase) and RPL13a (ribosomal protein L13a) were used for internal normalization.

Table 1
Oligonucleotides for real-time RT–PCR
mRNA targetsDescriptionSense (5′→3′)Antisense (5′→3′)
p47phox p47phox subunit of NADPH oxidase ATGACAGCCAGGTGAAGAAGC CGATAGGTCTGAAGGCTGATGG 
p22phox p22phox subunit of NADPH oxidase GCGGTGTGGACAGAAGTACC CTTGGGTTTAGGCTCAATGG 
caveolin-1 Caveolin-1 TCTACAAGCCCAACAACAAGG AGGAAAGAGAGGATGGCAAAG 
eNOS Endothelial nitric oxide synthase ATGGATGAGCCAACTCAAGG TGTCGTGTAATCGGTCTTGC 
NOX-1 NOX-1 subunit of NADPH oxidase TCTTGCTGGTTGACACTTGC TATGGGAGTGGGAATCTTGG 
NOX-4 NOX-1 subunit of NADPH oxidase ACAGTCCTGGCTTACCTTCG TTCTGGGATCCTCATTCTGG 
IL-6 Interleukin-6 GATGGATGCTTCCAAACTGG AGGAGAGCATTGGAAGTTGG 
TNFα Tumour necrosis factor α ACGATGCTCAGAAACACACG CAGTCTGGGAAGCTCTGAGG 
IL-1β Interleukin-1β GTCACTCATTGTGGCTGTGG GCAGTGCAGCTGTCTAATGG 
Cu/Zn SOD Copper/zinc superoxide dismutase CAATGTGACTGCTGGAAAGG AATCCCAATCACTCCACAGG 
MN-SOD Manganese superoxide dismutase ACGTTTCTTTGGCTCATTGG GCGCCTCTCAGATAAACAGG 
CPT-1 Carnitine palmitoyltransferase-1 GGACCGTGAAGAGATCAAGC CGAGGATTCTCTGGAACTGC 
UCP-2 Uncoupling protein 2 GCCACTTCACTTCTGCCTTC GAAGGCATGAACCCCTTGTA 
GPX-1 Glutathione peroxidase GCTGCTCATTGAGAATGTCG CAGGTCGGACGTACTTGAGG 
HO-1 Haem oxygenase-1 ACAGCCCCACCAAGTTCAAA GCCAGGCAAGATTCTCCCTT 
RPL13 Ribosomal protein L13 CCTGCTGCTCTCAAGGTTGTT TGGTTGTCACTGCCTGGTACTT 
GADPH Glyceraldehyde-3-phosphate dehydrogenase TGCACCACCAACTGCTTAGC GGATGCAGGGATGATGTTCT 
mRNA targetsDescriptionSense (5′→3′)Antisense (5′→3′)
p47phox p47phox subunit of NADPH oxidase ATGACAGCCAGGTGAAGAAGC CGATAGGTCTGAAGGCTGATGG 
p22phox p22phox subunit of NADPH oxidase GCGGTGTGGACAGAAGTACC CTTGGGTTTAGGCTCAATGG 
caveolin-1 Caveolin-1 TCTACAAGCCCAACAACAAGG AGGAAAGAGAGGATGGCAAAG 
eNOS Endothelial nitric oxide synthase ATGGATGAGCCAACTCAAGG TGTCGTGTAATCGGTCTTGC 
NOX-1 NOX-1 subunit of NADPH oxidase TCTTGCTGGTTGACACTTGC TATGGGAGTGGGAATCTTGG 
NOX-4 NOX-1 subunit of NADPH oxidase ACAGTCCTGGCTTACCTTCG TTCTGGGATCCTCATTCTGG 
IL-6 Interleukin-6 GATGGATGCTTCCAAACTGG AGGAGAGCATTGGAAGTTGG 
TNFα Tumour necrosis factor α ACGATGCTCAGAAACACACG CAGTCTGGGAAGCTCTGAGG 
IL-1β Interleukin-1β GTCACTCATTGTGGCTGTGG GCAGTGCAGCTGTCTAATGG 
Cu/Zn SOD Copper/zinc superoxide dismutase CAATGTGACTGCTGGAAAGG AATCCCAATCACTCCACAGG 
MN-SOD Manganese superoxide dismutase ACGTTTCTTTGGCTCATTGG GCGCCTCTCAGATAAACAGG 
CPT-1 Carnitine palmitoyltransferase-1 GGACCGTGAAGAGATCAAGC CGAGGATTCTCTGGAACTGC 
UCP-2 Uncoupling protein 2 GCCACTTCACTTCTGCCTTC GAAGGCATGAACCCCTTGTA 
GPX-1 Glutathione peroxidase GCTGCTCATTGAGAATGTCG CAGGTCGGACGTACTTGAGG 
HO-1 Haem oxygenase-1 ACAGCCCCACCAAGTTCAAA GCCAGGCAAGATTCTCCCTT 
RPL13 Ribosomal protein L13 CCTGCTGCTCTCAAGGTTGTT TGGTTGTCACTGCCTGGTACTT 
GADPH Glyceraldehyde-3-phosphate dehydrogenase TGCACCACCAACTGCTTAGC GGATGCAGGGATGATGTTCT 

Western blot analysis

Aortic or MAEC homogenates were separated by SDS/PAGE. Proteins were transferred on to PVDF membranes, incubated with primary monoclonal mouse anti-eNOS, rabbit anti-p-eNOS (Ser1177), rabbit anti-p-PKB (protein kinase B)/Akt (Ser473), rabbit anti-PKB/Akt (Transduction Laboratories), bovine anti-p-eNOS (Thr495) (Millipore), goat polyclonal anti-CPT-1 (Santa Cruz Biotechnology), polyclonal rabbit anti-p-PKC (protein kinase C)α/βII (Thr638/Thr641) and rabbit anti-PKCα (Cell Signaling Technology) antibodies overnight and with the corresponding secondary horseradish peroxidase-conjugated antibodies. Antibody binding was detected by an ECL system (GE Healthcare) and densitometric analysis was performed using Scion Image-Release Beta 4.02 software (http://www.scioncorp.com) [11]. Aortic or MAEC samples were re-probed for expression of α-actin or β-actin, respectively.

Measurement of DAG (diacylglycerol)

DAG levels in MAECs were measured by the DAG kinase assay as described previously [7].

Statistical analysis

Results are expressed as means±S.E.M. Statistical analyses were performed using Graph Pad Prism 5 software. A two-factor ANOVA was used to test for drug or group interactions. When a significant interaction was detected, a one-way ANOVA with a Student–Newman–Keuls post-hoc test was used to discern individual differences between groups. Significance was accepted at P<0.05.

RESULTS

PPARβ/δ activation restores the palmitate-induced impairment of endothelium-dependent vasodilatation and NO production

Incubation of mice isolated aortae with palmitate inhibited the endothelium-dependent relaxation in response to acetylcholine (Figure 1A) and the phosphorylation of eNOS at the activation site Ser1177 (Figure 1B). In MAECs, palmitate reduced the Ca2+ ionophore A23187-stimulated NO production (Figure 1C) and phosphorylation of eNOS at Ser1177 (Figure 1D). The PPARβ/δ agonist GW0742 (1 μM) restored the aortic relaxation in response to acetylcholine, eNOS phosphorylation at Ser1177 and the level of NO production induced by A23187 in aorta and MAECs exposed to palmitate. All the effects of GW0742 were abolished by co-incubation with the PPARβ/δ antagonist GSK0660 (1 μM). GW0742 did not modify either the aortic relaxation in response to acetylcholine or A23187-stimulated NO production in MAECs incubated under control conditions (results not shown).

The PPARβ/δ agonist prevents palmitate-induced endothelial dysfunction in vitro

Figure 1
The PPARβ/δ agonist prevents palmitate-induced endothelial dysfunction in vitro

Vascular relaxant responses induced by acetylcholine (Ach) (A) in mouse aortas pre-contracted by U46619 (10−8 M), and phosphorylation of eNOS at Ser1177 (B) after incubation for 24 h in the absence (control) or the presence of palmitate (Pal, 100 μM), GW0742 (1 μM) or GSK0660 (1 μM). Results are means±S.E.M., derived from nine to 11 different rings. Immunoblots from seven separate experiments were quantified. NO release (C) and phosphorylation of eNOS at Ser1177 (D) in MAECs incubated with the PPARβ/δ agonist GW0742 (1 μM) for 15 h, and in the last 3 h in the absence (control cells) or presence of palmitate (100 μM). In some experiments, cells were co-incubated with the PPARβ/δ antagonist GSK0660 (1 μM, 1 h before the addition of GW0742). NO release was estimated from the area under the curve (AUC) of the fluorescent signal of DAF-2 for 30 min in MAECs stimulated with the calcium ionophore calimycin (A23187, 1 μM). Results are means±S.E.M. (n=7–15). Immunoblots from three separate experiments were quantified. *P<0.05 and **P<0.01 palmitate compared with control; #P<0.05 and ##P<0.01 GW0742+palmatiate compared with palmitate; †P<0.05 and ††P<0.01 GSK0660+GW0742+palmitate compared with GW0742+palmitate.

Figure 1
The PPARβ/δ agonist prevents palmitate-induced endothelial dysfunction in vitro

Vascular relaxant responses induced by acetylcholine (Ach) (A) in mouse aortas pre-contracted by U46619 (10−8 M), and phosphorylation of eNOS at Ser1177 (B) after incubation for 24 h in the absence (control) or the presence of palmitate (Pal, 100 μM), GW0742 (1 μM) or GSK0660 (1 μM). Results are means±S.E.M., derived from nine to 11 different rings. Immunoblots from seven separate experiments were quantified. NO release (C) and phosphorylation of eNOS at Ser1177 (D) in MAECs incubated with the PPARβ/δ agonist GW0742 (1 μM) for 15 h, and in the last 3 h in the absence (control cells) or presence of palmitate (100 μM). In some experiments, cells were co-incubated with the PPARβ/δ antagonist GSK0660 (1 μM, 1 h before the addition of GW0742). NO release was estimated from the area under the curve (AUC) of the fluorescent signal of DAF-2 for 30 min in MAECs stimulated with the calcium ionophore calimycin (A23187, 1 μM). Results are means±S.E.M. (n=7–15). Immunoblots from three separate experiments were quantified. *P<0.05 and **P<0.01 palmitate compared with control; #P<0.05 and ##P<0.01 GW0742+palmatiate compared with palmitate; †P<0.05 and ††P<0.01 GSK0660+GW0742+palmitate compared with GW0742+palmitate.

PPARβ/δ activation in vivo improves endothelial dysfunction in aortae from HFD mice

To examine the effects of PPARβ/δ activation in vivo, GW0742 was given by oral gavage (3 mg/kg per day, for 3 weeks) to mice fed on an HFD. GW0742 treatment was unable to prevent the changes induced by HFD to body weight (Supplementary Figure S1A), energy intake (Supplementary Figure S1B), heart and kidney weight, and the higher plasmatic fasting glucose, cholesterol, triacylglycerols, HDL, the glucose tolerance test, and NEFAs (Table 2). However, GW0742 significantly reduced fat accumulation, an effect that was inhibited by co-administration with the PPARβ/δ antagonist GSK0660 (Table 2).

Table 2
Morphological and plasma determinations in all experimental groups

Results are shown as means±S.E.M. All parameters were assessed in mice fed on a control diet (control) or an HFD treated with vehicle or GW0742 (3 mg/kg per day, oral gavage) or GW0742 plus GSK0660 (1 mg/kg per day i.p.). *P<0.05 and **P<0.001 compared with control; #P<0.05 and ##P<0.01 compared with HFD; †P<0.05 compared with HFD-GW group.

Control (n=9)HFD (n=10)HFD-GW (n=10)HFD-GW-GSK (n=10)
Body weight (g) 23.4±0.5 27.2±0.7** 25.9±0.5 26.1±0.4 
Heart weight/tibia length (mg/cm) 60.5±1.5 62.2±1.3 62.7±2.0 63.1±1.7 
Mesenteric fat/body weight (%) 0.12±0.3 0.93±0.08** 0.31±0.04# 0.48±0.04† 
Epididymal weight/body weight (%) 1.01±0.17 2.58±0.22** 1.50±0.07## 1.94±0.19† 
Kidney weight/tibia length (mg/cm) 66.6±1.3 76.0±2.4** 71.4±2.0 70.8±2.2 
Fasting glucose (mg/dl) 102.3±8.2 140.7±14.9* 149.7±9.9 168.7±10.0 
Total cholesterol (mg/dl) 69.2±4.3 86.6±4.0* 93.0±5.8 112.0±6.2 
Triacylglycerol 71.8±4.9 93.0±7.6* 99.6±8.8 96.5±5.7 
HDL (mg/dl) 73.7±3.7 83.5±4.4 95.5±5.7 95.1±10.6 
Glucose tolerance test (mg/dl/min)  237±15  318±15**  290±16  287±16 
Free fatty acid (mmol/l) 0.36±0.03 0.46±0.04* 0.45±0.03 0.47±0.04 
Control (n=9)HFD (n=10)HFD-GW (n=10)HFD-GW-GSK (n=10)
Body weight (g) 23.4±0.5 27.2±0.7** 25.9±0.5 26.1±0.4 
Heart weight/tibia length (mg/cm) 60.5±1.5 62.2±1.3 62.7±2.0 63.1±1.7 
Mesenteric fat/body weight (%) 0.12±0.3 0.93±0.08** 0.31±0.04# 0.48±0.04† 
Epididymal weight/body weight (%) 1.01±0.17 2.58±0.22** 1.50±0.07## 1.94±0.19† 
Kidney weight/tibia length (mg/cm) 66.6±1.3 76.0±2.4** 71.4±2.0 70.8±2.2 
Fasting glucose (mg/dl) 102.3±8.2 140.7±14.9* 149.7±9.9 168.7±10.0 
Total cholesterol (mg/dl) 69.2±4.3 86.6±4.0* 93.0±5.8 112.0±6.2 
Triacylglycerol 71.8±4.9 93.0±7.6* 99.6±8.8 96.5±5.7 
HDL (mg/dl) 73.7±3.7 83.5±4.4 95.5±5.7 95.1±10.6 
Glucose tolerance test (mg/dl/min)  237±15  318±15**  290±16  287±16 
Free fatty acid (mmol/l) 0.36±0.03 0.46±0.04* 0.45±0.03 0.47±0.04 

Aortas from mice fed on an HFD showed significantly reduced endothelium-dependent vasodilator responses to acetylcholine compared with aortas from the control group (Figure 2A). GW0742 administration produced a significant increase in the endothelium-dependent relaxation induced by acetylcholine in mice fed on an HFD, being without effect in control mice. The increased relaxation in response to acetylcholine induced by GW0742 treatment in HFD-fed mice was abolished by co-administration with GSK0660 (1 mg/kg per day i.p.). The relaxant response induced by acetylcholine was almost fully inhibited by the eNOS inhibitor L-NAME in all experimental groups (Supplementary Figure S2A) involving eNOS activation, and the endothelium-independent vasodilator responses to nitroprusside were not different between groups (Supplementary Figure S2B). SBP significantly increased in mice fed on an HFD. GW0742 prevented the rise in SBP and this effect was abolished by GSK0660 (Figure 2B). Phosphorylation of eNOS at Ser1177 and PKB/Akt at Ser473 were reduced in the aortae from mice fed on an HFD. GW0742 restored the phosphorylation of both targets and these effects were suppressed by GSK0660 (Figure 2C).

The PPARβ/δ agonist prevents HFD-induced endothelial dysfunction in vivo

Figure 2
The PPARβ/δ agonist prevents HFD-induced endothelial dysfunction in vivo

(A) Vascular relaxant responses induced by acetylcholine (Ach) (A) in intact aortas pre-contracted by U46619 (10−8 M) and (B) SBP in mice fed on a control diet (control), or an HFD treated with vehicle, GW0742 (HFD-GW, 3 mg/kg per day, oral gavage) or GW0742 plus GSK0660 (HFD-GW-GSK, 1 mg/kg per day i.p.). Results are means±S.E.M. (n=6–8 rings from different mice). Phosphorylation of eNOS and PKB/Akt in aorta (C) from all experimental groups. Immunoblots from six separate experiments were quantified. *P<0.05 and **P<0.01 compared with control; #P<0.05 and ##P<0.01 compared with HFD; †P<0.05 and ††P<0.01 compared with HFD-GW0742 group. *P<0.05 and **P<0.01 HFD compared with control; #P<0.05 and ##P<0.01 GW0742+HFD compared with HFD; †P<0.05 and ††P<0.01 GSK0660+GW0742+HFD compared with GW0742-HFD group.

Figure 2
The PPARβ/δ agonist prevents HFD-induced endothelial dysfunction in vivo

(A) Vascular relaxant responses induced by acetylcholine (Ach) (A) in intact aortas pre-contracted by U46619 (10−8 M) and (B) SBP in mice fed on a control diet (control), or an HFD treated with vehicle, GW0742 (HFD-GW, 3 mg/kg per day, oral gavage) or GW0742 plus GSK0660 (HFD-GW-GSK, 1 mg/kg per day i.p.). Results are means±S.E.M. (n=6–8 rings from different mice). Phosphorylation of eNOS and PKB/Akt in aorta (C) from all experimental groups. Immunoblots from six separate experiments were quantified. *P<0.05 and **P<0.01 compared with control; #P<0.05 and ##P<0.01 compared with HFD; †P<0.05 and ††P<0.01 compared with HFD-GW0742 group. *P<0.05 and **P<0.01 HFD compared with control; #P<0.05 and ##P<0.01 GW0742+HFD compared with HFD; †P<0.05 and ††P<0.01 GSK0660+GW0742+HFD compared with GW0742-HFD group.

PPARβ/δ activation reduces the increased ROS production induced by palmitate

To examine whether ROS are involved in endothelial dysfunction induced by palmitate in mouse aorta, we analysed the endothelium-dependent relaxant response to acetylcholine in the presence of the mitochondrial antioxidant mitoQ or the NADPH oxidase inhibitor apocynin. Both mitoQ (Figure 3A) and apocynin (Figure 4A) significantly improved the impaired aortic relaxation in response to acetylcholine induced by palmitate. Intracellular ROS production measured by CM-H2DCFDA in MAECs was increased after palmitate incubation (Figure 3B). The increase in intracellular O2 levels induced by palmitate incubation in MAECs was then confirmed by determination of the 2-OH-E+ level by HPLC (Figure 3C). This ROS increase was inhibited by mitoQ and GW0742. The effect of GW0742 was abolished by blockade of PPARβ/δ by GSK0660. NADPH oxidase activity measured by both lucigenin-enhanced chemiluminescence (Figure 4B) and DHE fluorescence measured using a microplate reader (Figure 4C) was also increased by palmitate and apocynin and GW0742 inhibited this increase. Again, GSK0660 suppressed the effect of GW0742. Similarly, this PPARβ/δ agonist abolished the up-regulation of NOX4 mRNA in MAECs incubated with palmitate, but not in those co-incubated with GSK0660 (Figure 4D).

Role of ROS production in palmitate-induced endothelial dysfunction

Figure 3
Role of ROS production in palmitate-induced endothelial dysfunction

Vascular relaxant responses induced by acetylcholine (Ach) (A) in intact aortas pre-contracted by U46619 (10−8 M), after incubation for 24 h in the absence (control) or presence of palmitate (Pal) (100 μM). In some experiments, responses to acetylcholine were studied after incubation with mitoQ (0.1 μM) (A), incubated for 60 min before the addition of U46619. Results are means±S.E.M., derived from nine to 11 different rings. (B) ROS measured by fluorescence in CM-H2DCFDA-loaded cells, and (C) O2 production detected by 2-OH-E+ fluorescence HPLC. Left: representative example. Right: quantification of 2-OH-E+ peak, expressed as area under curve (AUC)/μg of protein. (B and C) MAECs were incubated with the PPARβ/δ agonist GW0742 (1 μM) for 15 h, and in the last 3 h in the absence (control cells) or presence of palmitate (100 μM). In some experiments, cells were co-incubated with mitoQ (0.1 μmol/l) added with palmitate, or incubations were performed for 30 min in the absence or presence of PEG–SOD (25 units/ml). Results are means±S.E.M. (n=6–15). *P<0.05 and **P<0.01 palmitate compared with control; #P<0.05 and ##P<0.01 compared with palmitate; †P<0.05 and ††P<0.01 compared with palmitate+GW0742 group.

Figure 3
Role of ROS production in palmitate-induced endothelial dysfunction

Vascular relaxant responses induced by acetylcholine (Ach) (A) in intact aortas pre-contracted by U46619 (10−8 M), after incubation for 24 h in the absence (control) or presence of palmitate (Pal) (100 μM). In some experiments, responses to acetylcholine were studied after incubation with mitoQ (0.1 μM) (A), incubated for 60 min before the addition of U46619. Results are means±S.E.M., derived from nine to 11 different rings. (B) ROS measured by fluorescence in CM-H2DCFDA-loaded cells, and (C) O2 production detected by 2-OH-E+ fluorescence HPLC. Left: representative example. Right: quantification of 2-OH-E+ peak, expressed as area under curve (AUC)/μg of protein. (B and C) MAECs were incubated with the PPARβ/δ agonist GW0742 (1 μM) for 15 h, and in the last 3 h in the absence (control cells) or presence of palmitate (100 μM). In some experiments, cells were co-incubated with mitoQ (0.1 μmol/l) added with palmitate, or incubations were performed for 30 min in the absence or presence of PEG–SOD (25 units/ml). Results are means±S.E.M. (n=6–15). *P<0.05 and **P<0.01 palmitate compared with control; #P<0.05 and ##P<0.01 compared with palmitate; †P<0.05 and ††P<0.01 compared with palmitate+GW0742 group.

Role of NADPH oxidase in palmitate-induced endothelial dysfunction

Figure 4
Role of NADPH oxidase in palmitate-induced endothelial dysfunction

Vascular relaxant responses induced by acetylcholine (Ach) (A) in intact aortas pre-contracted by U46619 (10−8 M), after incubation for 24 h in the absence (control) or the presence of palmitate (Pal, 100 μM). In some experiments, responses to acetylcholine were studied after incubation with apocynin (10 μM) (A) incubated for 60 min before the addition of U46619. Results are means±S.E.M., derived from nine to 11 different rings. NADPH oxidase activity measured by lucigenin-enhanced chemiluminescence (B), or DHE fluorescence measured in the microplate reader (C), and mRNA levels of NOX4 (D) in MAECs incubated with the PPARβ/δ agonist GW0742 (1 μM) for 15 h, and in the last 3 h in the absence (control cells) or presence of palmitate (100 μM). In some experiments, cells were co-incubated with apocynin (10 μM) added with palmitate, or incubations were performed for 30 min in the absence or presence of PEG–SOD (25 units/ml). Results are means±S.E.M. (n=6–15). *P<0.05 and **P<0.01 palmitate compared with control; #P<0.05 and ##P<0.01 compared with palmitate; †P<0.05 and ††P<0.01 compared with palmitate+GW0742 group.

Figure 4
Role of NADPH oxidase in palmitate-induced endothelial dysfunction

Vascular relaxant responses induced by acetylcholine (Ach) (A) in intact aortas pre-contracted by U46619 (10−8 M), after incubation for 24 h in the absence (control) or the presence of palmitate (Pal, 100 μM). In some experiments, responses to acetylcholine were studied after incubation with apocynin (10 μM) (A) incubated for 60 min before the addition of U46619. Results are means±S.E.M., derived from nine to 11 different rings. NADPH oxidase activity measured by lucigenin-enhanced chemiluminescence (B), or DHE fluorescence measured in the microplate reader (C), and mRNA levels of NOX4 (D) in MAECs incubated with the PPARβ/δ agonist GW0742 (1 μM) for 15 h, and in the last 3 h in the absence (control cells) or presence of palmitate (100 μM). In some experiments, cells were co-incubated with apocynin (10 μM) added with palmitate, or incubations were performed for 30 min in the absence or presence of PEG–SOD (25 units/ml). Results are means±S.E.M. (n=6–15). *P<0.05 and **P<0.01 palmitate compared with control; #P<0.05 and ##P<0.01 compared with palmitate; †P<0.05 and ††P<0.01 compared with palmitate+GW0742 group.

PPARβ/δ activation in vivo reduces vascular ROS levels in HFD-fed mice by reducing NADPH oxidase activity and mitochondrial sources and by up-regulation of antioxidant genes

To characterize and localize ROS levels within the vascular wall, red fluorescence was analysed in sections of aorta incubated with DHE. Rings from mice fed on an HFD showed marked increased staining in adventitial, medial and endothelial cells compared with control mice which was significantly reduced by GW0742 administration. The increased red fluorescence induced by HFD was abolished by PEG–SOD, indicating its specificity for O2 (Figure 5A). GW0742 treatment also prevented the rise in vascular NADPH oxidase activity (Figure 5B) and the up-regulation of the NADPH oxidase subunits NOX4, NOX1, p47phox and p22phox (Figure 5C) induced by the HFD. The increase in the pro-oxidant NADPH oxidase subunits induced by the HFD was accompanied by reduced mRNA levels of mitochondrial and cytoplasmic antioxidant enzymes (Figure 6), such as Mn-SOD, UCP2 and GPx1. The mRNA levels of HO-1 and CPT-1 were increased, and Cu/Zn-SOD was not significantly affected by the HFD. GW0742 administration prevented the HFD-induced changes in Mn-SOD and GPx1 and did not affect Cu/Zn-SOD. The PPARβ/δ agonist increased UCP2, CPT-1 and HO-1 mRNA levels above control levels. PPARβ/δ blockade with GSK0660 inhibited the effects of GW0742 on aortic ROS levels, NADPH oxidase activity and changes in gene expression (Figures 5 and 6).

Role of ROS production in HFD-induced endothelial dysfunction

Figure 5
Role of ROS production in HFD-induced endothelial dysfunction

(A) Right-hand images show arteries incubated in the presence of DHE which produces a red fluorescence when oxidized to ethidium by ROS. Left-hand images show blue fluorescence of the nuclear stain DAPI (×400 magnification). The histogram shows red ethidium fluorescence intensity normalized to the blue DAPI fluorescence (means±S.E.M., n=5–7 rings from different mice). (B) NADPH oxidase activity measured by lucigenin ECL (n=7–9), and (C) expression of NADPH oxidase subunits NOX4, NOX1, p47phox and p22phox at the level of mRNA by RT–PCR. Results are means±S.E.M. All parameters were assessed in aortas from mice fed on a control diet (control), or on an HFD treated with vehicle or GW0742 (3 mg/kg per day, oral gavage) or GW0742 plus GSK0660 (1 mg/kg per day i.p.). *P<0.05 compared with control; #P<0.05 compared with HFD; †P<0.05 and ††P<0.01 compared with the HFD-GW0742 group.

Figure 5
Role of ROS production in HFD-induced endothelial dysfunction

(A) Right-hand images show arteries incubated in the presence of DHE which produces a red fluorescence when oxidized to ethidium by ROS. Left-hand images show blue fluorescence of the nuclear stain DAPI (×400 magnification). The histogram shows red ethidium fluorescence intensity normalized to the blue DAPI fluorescence (means±S.E.M., n=5–7 rings from different mice). (B) NADPH oxidase activity measured by lucigenin ECL (n=7–9), and (C) expression of NADPH oxidase subunits NOX4, NOX1, p47phox and p22phox at the level of mRNA by RT–PCR. Results are means±S.E.M. All parameters were assessed in aortas from mice fed on a control diet (control), or on an HFD treated with vehicle or GW0742 (3 mg/kg per day, oral gavage) or GW0742 plus GSK0660 (1 mg/kg per day i.p.). *P<0.05 compared with control; #P<0.05 compared with HFD; †P<0.05 and ††P<0.01 compared with the HFD-GW0742 group.

Effects of GW0742 on gene expression

Figure 6
Effects of GW0742 on gene expression

Mn-SOD (A), UCP2 (B), CPT-1 (C), GPx1 (D), HO-1 (E) and Cu/Zn-SOD (F) mRNA expression. Results are means±S.E.M. in aortas from mice fed on a control diet (control), or an HFD treated with vehicle or GW0742 (3 mg/kg per day, oral gavage) or GW0742 plus GSK0660 (1 mg/kg per day i.p.). *P<0.05 HFD compared with control; #P<0.05 and ##P<0.01 GW0742-HFD compared with HFD; †P<0.05 and ††P<0.01 GSK0660-GW0742-HFD compared with GW0742-HFD.

Figure 6
Effects of GW0742 on gene expression

Mn-SOD (A), UCP2 (B), CPT-1 (C), GPx1 (D), HO-1 (E) and Cu/Zn-SOD (F) mRNA expression. Results are means±S.E.M. in aortas from mice fed on a control diet (control), or an HFD treated with vehicle or GW0742 (3 mg/kg per day, oral gavage) or GW0742 plus GSK0660 (1 mg/kg per day i.p.). *P<0.05 HFD compared with control; #P<0.05 and ##P<0.01 GW0742-HFD compared with HFD; †P<0.05 and ††P<0.01 GSK0660-GW0742-HFD compared with GW0742-HFD.

Since oxidative stress induced inflammation and vice versa, the mRNA expression of the pro-inflammatory cytokines TNFα, IL-1β and IL-6 was analysed. We found that in aortic homogenates, the expression of these cytokines was higher in the HFD group compared with the control group (Supplementary Figure S3). As expected, GW0742 administration significantly reduced the mRNA levels of these pro-inflammatory cytokines in mice fed on an HFD, and this effect was prevented by GSK0660.

Role of CPT-1 in the preventive effects of PPARβ/δ activation on palmitate-induced endothelial dysfunction

GW0742 induced a marked increase in the mRNA and protein levels of CPT-1 in MAECs which was sensitive to PPARβ/δ blockade and independent of the presence of palmitate (Figure 7A). The preventive effects induced by GW0742 on the impaired A23187-stimulated NO production (Figure 7B), the increased ROS generation (Figure 7C) in MAECs and the reduced acetylcholine relaxation in aorta (Figure 7D) incubated with palmitate, were abolished by the CPT-1 irreversible inhibitor etomoxir. Moreover, siRNA targeting CPT-1 in MAECs, which effectively down-regulated CPT-1 mRNA and protein (Figure 7E), also blunted the increase in NO production (Figure 7F), and the reduction in ROS generation (Figure 7G) induced by GW0742 in MAECs incubated with palmitate.

Role of CPT-1 in the preventive effects of PPARβ/δ activation on palmitate-induced endothelial dysfunction

Figure 7
Role of CPT-1 in the preventive effects of PPARβ/δ activation on palmitate-induced endothelial dysfunction

(A) mRNA and protein expression of CPT-1 in MAECs incubated with the PPARβ/δ agonist GW0742 (1 μM) for 15 h, and in the last 3 h in the absence (control cells) or presence of palmitate (100 μM). In some experiments, cells were co-incubated with the PPARβ/δ antagonist GSK0660 (1 μM), 1 h before the addition of GW0742. Results are means±S.E.M. (n=7–15). Immunoblots from four separate experiments were quantified. (B) A23187-stimulated NO production and (C) CM-H2DCFDA-detected intracellular ROS in MAECs (incubated as in A) with or without etomoxir (40 μM) during palmitate incubation. (D) Vascular relaxant responses induced by acetylcholine (ACh) in mouse aortas pre-contracted by U46619 (10−8 M), after incubation for 24 h in the presence of palmitate (Pal, 100 μM), GW0742 (1 μM) or etomoxir (40 μM). Results are means±S.E.M., derived from nine to 11 different rings. (E) mRNA and protein levels of CPT-1 after CPT-1-specific siRNA transfection in MAECs for 48 h, (F) A23187-stimulated NO production, and (G) CM-H2DCFDA-detected intracellular ROS in MAECs incubated with GW0742 (1 μM) for 15 h, and in the last 3 h in the absence (control cells) or the presence of palmitate (100 μM) in control siRNA and siRNA-CPT-1-transfected cells. *P<0.05 and **P<0.01 compared with control; #P<0.05 and ##P<0.01 compared with palmitate; †P<0.05 and ††P<0.01 compared with palmitate+GW0742 group.

Figure 7
Role of CPT-1 in the preventive effects of PPARβ/δ activation on palmitate-induced endothelial dysfunction

(A) mRNA and protein expression of CPT-1 in MAECs incubated with the PPARβ/δ agonist GW0742 (1 μM) for 15 h, and in the last 3 h in the absence (control cells) or presence of palmitate (100 μM). In some experiments, cells were co-incubated with the PPARβ/δ antagonist GSK0660 (1 μM), 1 h before the addition of GW0742. Results are means±S.E.M. (n=7–15). Immunoblots from four separate experiments were quantified. (B) A23187-stimulated NO production and (C) CM-H2DCFDA-detected intracellular ROS in MAECs (incubated as in A) with or without etomoxir (40 μM) during palmitate incubation. (D) Vascular relaxant responses induced by acetylcholine (ACh) in mouse aortas pre-contracted by U46619 (10−8 M), after incubation for 24 h in the presence of palmitate (Pal, 100 μM), GW0742 (1 μM) or etomoxir (40 μM). Results are means±S.E.M., derived from nine to 11 different rings. (E) mRNA and protein levels of CPT-1 after CPT-1-specific siRNA transfection in MAECs for 48 h, (F) A23187-stimulated NO production, and (G) CM-H2DCFDA-detected intracellular ROS in MAECs incubated with GW0742 (1 μM) for 15 h, and in the last 3 h in the absence (control cells) or the presence of palmitate (100 μM) in control siRNA and siRNA-CPT-1-transfected cells. *P<0.05 and **P<0.01 compared with control; #P<0.05 and ##P<0.01 compared with palmitate; †P<0.05 and ††P<0.01 compared with palmitate+GW0742 group.

PPARβ/δ activation prevents DAG accumulation and PKCα/βII and phosphorylation of eNOS at Thr495 in palmitate-exposed MAECs

Because exposure to palmitate increased ROS production, which was concomitant with increases in the DAG level and PKC activity in endothelial cells [5], we next assessed the capacity of GW0742 to prevent DAG accumulation and the role of the PKC pathway. As expected, MAECs exposed to palmitate showed enhanced DAG levels compared with control cells (Figure 8A). When palmitate-exposed cells were treated with the PPARβ/δ-specific agonist DAG levels decreased. The mechanism by which GW0742 prevented DAG accumulation appears to involve increased fatty acid oxidation because the effect of this PPARβ/δ agonist was blunted by etomoxir (Figure 8A). The non-selective PKC inhibitor chelerythrine reversed the impaired A23187-stimulated NO production (Figure 8B) and the increased ROS production (Figure 8C) induced by palmitate in MAECs. Exposure to palmitate increased the phosphorylated levels of PKCα/βII (Figure 8D). GW0742 and the selective PKC inhibitor chelerythrine prevented the increase in PKCα/βII phosphorylation. The effects of GW0742 were prevented by etomoxir. Moreover, eNOS phosphorylation at Thr495, which reduces eNOS activity, was increased by palmitate and reversed after GW0742 or chelerythine pre-treatment (Figure 8E). Again, the effect of GW0742 was suppressed by co-incubation with etomoxir.

Role of PKC on the protective effects of PPARβ/δ activation in endothelial dysfunction induced by palmitate

Figure 8
Role of PKC on the protective effects of PPARβ/δ activation in endothelial dysfunction induced by palmitate

(A) DAG accumulation in MAECs incubated with the PPARβ/δ agonist GW0742 (1 μM) for 15 h, and in the last 3 h in the absence (control cells) or presence of palmitate (100 μM) or with etomoxir (40 μM) during palmitate incubation. (B) A23187-stimulated NO production, and (C) CM-H2DCFDA-detected intracellular ROS in MAECs incubated with GW0742 (1 μM) for 3 h in the absence (control cells) or presence of palmitate (100 μM) or with chelerythrine (1 μM). PKCα/βII (D) and Thr495 eNOS (E) phosphorylation in MAECs (incubated as above). Results are means±S.E.M. (n=7–15). Immunoblots from seven separate experiments were quantified. *P<0.05 and **P<0.01 compared with control; #P<0.05 and ##P<0.01 compared with palmitate; †P<0.05 compared with palmitate+GW0742 group.

Figure 8
Role of PKC on the protective effects of PPARβ/δ activation in endothelial dysfunction induced by palmitate

(A) DAG accumulation in MAECs incubated with the PPARβ/δ agonist GW0742 (1 μM) for 15 h, and in the last 3 h in the absence (control cells) or presence of palmitate (100 μM) or with etomoxir (40 μM) during palmitate incubation. (B) A23187-stimulated NO production, and (C) CM-H2DCFDA-detected intracellular ROS in MAECs incubated with GW0742 (1 μM) for 3 h in the absence (control cells) or presence of palmitate (100 μM) or with chelerythrine (1 μM). PKCα/βII (D) and Thr495 eNOS (E) phosphorylation in MAECs (incubated as above). Results are means±S.E.M. (n=7–15). Immunoblots from seven separate experiments were quantified. *P<0.05 and **P<0.01 compared with control; #P<0.05 and ##P<0.01 compared with palmitate; †P<0.05 compared with palmitate+GW0742 group.

DISCUSSION

Activation of PPARβ/δ has been previously reported to have beneficial effects on endothelial function in diabetic rats [11] and mice [12]. In the present study, we provide the first evidence that PPARβ/δ activation restores the lipid-induced impairment of endothelium-dependent vasodilation and NO production, acting directly on endothelial cells via up-regulation of CPT-1. Moreover, the present study points to CPT-1 as a novel therapeutic target in endothelial dysfunction.

We used GW0742, a highly potent and selective PPARβ/δ agonist with a 200-fold higher affinity towards PPARβ/δ than towards other PPAR isotypes [27]. All of the effects of GW0742 reported in the present paper were inhibited by GSK0660, a selective inhibitor of PPARβ/δ, confirming the specificity of this drug for this nuclear receptor. Previous studies found that the protective effects on endothelial function in diabetic animals were related to changes in the metabolic profile (reduced plasma triacylglycerols and insulin resistance, and increased plasma HDL) [11,12]. In contrast, the effects of GW0742 in the present study were independent of changes in body weight and plasma levels of lipids or glucose. Moreover, the PPARβ/δ agonist prevented endothelial dysfunction both in vitro as induced by palmitate and in vivo as induced by an HFD.

NO is the main endothelium-dependent vasodilator. The activity of eNOS is regulated by multiple post-transcriptional mechanisms including the phosphorylation of Ser1177, which is mainly regulated by PKB/Akt and results in increased enzyme activity, and the phosphorylation of Thr495, which involves PKC among other kinases and inhibits the catalytic activity [28]. NO bioavailability is also strongly dependent on the production of ROS in vascular cells, particularly superoxide, which rapidly reacts with and inactivates NO.

The relaxant response to the activator of soluble guanylate cyclase nitroprusside was similar in aorta from control and HFD-fed mice, indicating that alterations in the NO sensitivity of vascular smooth muscle does not seem to contribute to endothelial dysfunction induced by lipids. Palmitate and oleic acid inhibit the PKB/Akt-dependent phosphorylation of eNOS at Ser1177, reducing NO production, which may contribute to endothelial dysfunction and the occurrence of atherosclerosis ([29,30] and results of the present study). Additionally, aorta from mice fed on an HFD also showed reduced phosphorylation of both eNOS at Ser1177 and Akt at Ser473 and reduced acetylcholine-induced relaxation. These results indicate reduced basal and stimulated eNOS activity induced by an HFD. However, when the rings were incubated with a high concentration (>50-fold IC50) of the eNOS inhibitor L-NAME, this relaxation was abolished in rings from all experimental groups, showing that it is completely dependent of eNOS-derived NO. GW0742 treatment in vitro and in vivo prevented the altered responses to acetylcholine, eNOS phosphorylation at Ser1177 and NO production induced by lipids, indicating a protective role for this drug in both basal and agonist-induced NO bioactivity.

Endothelial dysfunction induced by palmitate and an HFD was also associated with increased production of ROS in vascular cells. In fact, the mitochondrial antioxidant mitoQ and the NADPH oxidase inhibitor apocynin partially prevented the palmitate-induced increase in ROS and the impairment of endothelium-dependent relaxation. The increased NADPH oxidase activity correlated with increased expression of its NOX1, NOX4, p22phox and p47phox subunits. Increased ROS may also result from reduced antioxidant enzymes GPx1, Mn-SOD and UCP2. Activation of PPARβ/δ with GW0742 prevented the increased ROS levels in vitro and in vivo. This effect was due to a reduction in both mitochondria- and NADPH oxidase-derived ROS production in the vascular wall. The normalization of ROS also seems to contribute to the restoration of endothelial function.

CPT-1 transports fatty acids into the mitochondria promoting its metabolism via β-oxidation. Up-regulation of CPT-1 seems to play a key role in the protective effects induced by PPARβ/δ activation in endothelial dysfunction evoked by lipids. This is supported by several findings. First, GW0742 produced a large increase in the expression of CPT-1 in vitro (MAEC and aorta) and in vivo. Secondly, the effects of GW0742 on acetylcholine relaxation in isolated aortae, and on NO production and ROS levels in MAECs were also abolished by the CPT-1 inhibitor etomoxir. Thirdly, down-regulation of CPT-1 also suppressed the effects of GW0742 in NO and ROS levels in MAECs.

Accumulation of fatty acid derivatives, such as DAG, induces endothelial dysfunction in a PKC-dependent manner [3134]. PKCα/βII activation leads to endothelial dysfunction by increasing phosphorylation of eNOS at Thr495, thus reducing NO production [27], and by increasing ROS production from the mitochondria [35] and NADPH oxidase [5], reducing NO bioavailability. Likewise, we found that palmitate increased DAG and activation of PKCα/βII. Moreover, PKC inhibition with chelerythrine prevented the reduction in NO and the increased ROS in MAECs incubated with palmitate. Interestingly, our findings demonstrate that PPARβ/δ activation by GW0742 prevents palmitate-induced DAG accumulation in MAECs. Our data implicate increased CPT-1 activity in this effect, because DAG levels were restored by the presence of etomoxir. Accordingly, overexpression of CPT-1 in cultured skeletal muscle cells affords protection against lipid-induced accumulation of DAG [7,8]. Consistent with the reduction in DAG, GW0742 prevented the increase in PKCα/βII activation and phosphorylation of eNOS at Thr495 in palmitate-exposed cells. Taken together, these data indicate that, as a result of increased mitochondrial transport and oxidation of NEFAs induced by GW0742, their accumulation in the form of DAG would be reduced. Our data confirm the key role of CPT-1 in the prevention of fatty-acid-induced endothelial dysfunction and support the use of GW0742 as a pharmacological tool to prevent DAG accumulation and reduce the vascular alterations derived from this process.

In conclusion, the present study demonstrates that, in lipid-induced endothelial dysfunction, PPARβ/δ activation improves endothelium-dependent relaxation, essentially by preserving the NO-mediated component. This protective effect may be attributable to an up-regulation of CPT-1, which prevented DAG accumulation, thereby impeding PKCα/βII activation and the subsequent eNOS inactivation and ROS production (Figure 9).

Scheme representing the mechanisms involved in the palmitate-induced impairment of endothelial function and the proposed mechanisms (red arrows) for the protective effect of PPARβ/δ activation

Figure 9
Scheme representing the mechanisms involved in the palmitate-induced impairment of endothelial function and the proposed mechanisms (red arrows) for the protective effect of PPARβ/δ activation
Figure 9
Scheme representing the mechanisms involved in the palmitate-induced impairment of endothelial function and the proposed mechanisms (red arrows) for the protective effect of PPARβ/δ activation

AUTHOR CONTRIBUTION

Marta Toral, Miguel Romero, Ayman Moawad Mahmoud, Emma Barroso, Manuel Gómez-Guzmán, Ana García-Redondo, Ana Briones and Manuel Sánchez performed experiments. Rosario Jiménez, Ángel Cogolludo and Manuel Vázquez-Carrera performed experiments and contributed to discussion. Francisco Pérez-Vizcaíno and Juan Duarte designed experiments and wrote the paper.

FUNDING

This work was supported by the Comisión Interministerial de Ciencia y Tecnología [grant numbers SAF2011-28150, SAF2010-22066-C02-01, SAF2010-22066-C02-02 and SAF2014-55523-R], Junta de Andalucía (Proyecto de excelencia) [grant number P12-CTS-2722] and the Ministerio de Ciencia e Innovación, Instituto de Salud Carlos III [Red de Investigación Cardiovascular (RIC) grant number RD12/0042/0011 and RD12/0042/0024], Spain. M.S. is a postdoctoral fellow of Red HERACLES, M.R. is a postdoctoral fellow of RIC, and A.M.B. was supported by the Ramón y Cajal programme [grant number RYC-2010-06473].

Abbreviations

     
  • CM-H2DCFDA

    5-(and-6)-chloromethyl-2′,7′-dichlorodihydrofluorescein diacetate

  •  
  • CPT

    carnitine palmitoyltransferase

  •  
  • DAF-2

    diaminofluorescein-2

  •  
  • DAG

    diacylglycerol

  •  
  • DHE

    dihydroethidium

  •  
  • DMEM

    Dulbecco’s modified Eagle’s medium

  •  
  • DTPA

    diethylenetriaminepenta-acetic acid

  •  
  • eNOS

    endothelial nitric oxide synthase

  •  
  • GPx1

    glutathione peroxidase 1

  •  
  • HDL

    high-density lipoprotein

  •  
  • HFD

    high-fat diet

  •  
  • HO-1

    haem oxygenase 1

  •  
  • IL

    interleukin

  •  
  • i.p.

    intraperitoneally

  •  
  • MAEC

    mouse aortic endothelial cell

  •  
  • mitoQ

    mitoquinone

  •  
  • L-NAME

    NG-nitro-L-arginine methyl ester

  •  
  • NEFA

    non-esterified (‘free’) fatty acid

  •  
  • 2-OH-E+

    2-hydroxyethidium

  •  
  • PKB

    protein kinase B

  •  
  • PKC

    protein kinase C

  •  
  • PPAR

    peroxisome-proliferator-activated receptor

  •  
  • RLU

    relative light units

  •  
  • ROS

    reactive oxygen species

  •  
  • RT

    reverse transcription

  •  
  • SBP

    systolic blood pressure

  •  
  • TNFα

    tumour necrosis factor α

  •  
  • UCP2

    uncoupling protein 2

References

References
1
Vogel
 
R.A.
Corretti
 
M.C.
Plotnick
 
G.D.
 
Effect of a single high-fat meal on endothelial function in healthy subjects
Am. J. Cardiol.
1997
, vol. 
79
 (pg. 
350
-
354
)
[PubMed]
2
Esposito
 
K.
Marfella
 
R.
Ciotola
 
M.
Di Palo
 
C.
Giugliano
 
F.
Giugliano
 
G.
D’Armiento
 
M.
D’Andrea
 
F.
Giugliano
 
D.
 
Effect of a Mediterranean-style diet on endothelial dysfunction and markers of vascular inflammation in the metabolic syndrome: a randomized trial
JAMA
2004
, vol. 
292
 (pg. 
1440
-
1446
)
[PubMed]
3
Bedard
 
K.
Krause
 
K.H.
 
The NOX family of ROS-generating NADPH oxidases: physiology and pathophysiology
Physiol. Rev.
2007
, vol. 
87
 (pg. 
245
-
313
)
[PubMed]
4
Nishikawa
 
T.
Araki
 
E.
 
Impact of mitochondrial ROS production in the pathogenesis of diabetes mellitus and its complications
Antioxid. Redox Signal.
2007
, vol. 
9
 (pg. 
343
-
353
)
[PubMed]
5
Inoguchi
 
T.
Li
 
P.
Umeda
 
F.
Yu
 
H.Y.
Kakimoto
 
M.
Imamura
 
M.
Aoki
 
T.
Etoh
 
T.
Hashimoto
 
T.
Naruse
 
M.
, et al 
High glucose level and free fatty acid stimulate reactive oxygen species production through protein kinase C-dependent activation of NAD(P)H oxidase in cultured vascular cells
Diabetes
2000
, vol. 
49
 (pg. 
1939
-
1945
)
[PubMed]
6
Sebastián
 
D.
Herrero
 
L.
Serra
 
D.
Asins
 
G.
Hegardt
 
F.G.
 
CPT I overexpression protects L6E9 muscle cells from fatty acid-induced insulin resistance
Am. J. Physiol. Endocrinol. Metab.
2007
, vol. 
292
 (pg. 
E677
-
E686
)
[PubMed]
7
Coll
 
T.
Alvarez-Guardia
 
D.
Barroso
 
E.
Gómez-Foix
 
A.M.
Palomer
 
X.
Laguna
 
J.C.
Vázquez-Carrera
 
M.
 
Activation of peroxisome proliferator-activated receptor-δ by GW501516 prevents fatty acid-induced nuclear factor-κB activation and insulin resistance in skeletal muscle cells
Endocrinology
2010
, vol. 
151
 (pg. 
1560
-
1569
)
[PubMed]
8
Bruce
 
C.R.
Hoy
 
A.J.
Turner
 
N.
Watt
 
M.J.
Allen
 
T.L.
Carpenter
 
K.
Cooney
 
G.J.
Febbraio
 
M.A.
Kraegen
 
E.W.
 
Overexpression of carnitine palmitoyltransferase-1 in skeletal muscle is sufficient to enhance fatty acid oxidation and improve high-fat diet-induced insulin resistance
Diabetes
2009
, vol. 
58
 (pg. 
550
-
558
)
[PubMed]
9
Piqueras
 
L.
Reynolds
 
A.R.
Hodivala-Dilke
 
K.M.
Alfranca
 
A.
Redondo
 
J.M.
Hatae
 
T.
Tanabe
 
T.
Warner
 
T.D.
Bishop-Bailey
 
D.
 
Activation of PPARβ/δ induces endothelial cell proliferation and angiogenesis
Arterioscler. Thromb. Vasc. Biol.
2007
, vol. 
27
 (pg. 
63
-
69
)
[PubMed]
10
Wang
 
Y.X.
Lee
 
C.H.
Tiep
 
S.
Yu
 
R.T.
Ham
 
J.
Kang
 
H.
Evans
 
R.M.
 
Peroxisome-proliferator-activated receptor δ activates fat metabolism to prevent obesity
Cell
2003
, vol. 
113
 (pg. 
159
-
170
)
[PubMed]
11
Quintela
 
A.M.
Jiménez
 
R.
Piqueras
 
L.
Gómez-Guzmán
 
M.
Haro
 
J.
Zarzuelo
 
M.J.
Cogolludo
 
A.
Sanz
 
M.J.
Toral
 
M.
Romero
 
M.
, et al 
PPARβ activation restores the high glucose-induced impairment of insulin signalling in endothelial cells
Br. J. Pharmacol.
2014
, vol. 
171
 (pg. 
3089
-
3102
)
[PubMed]
12
Tian
 
X.Y.
Wong
 
W.T.
Wang
 
N.
Lu
 
Y.
Cheang
 
W.S.
Liu
 
J.
Liu
 
L.
Liu
 
Y.
Lee
 
S.S.
Chen
 
Z.Y.
, et al 
PPARδ activation protects endothelial function in diabetic mice
Diabetes
2012
, vol. 
61
 (pg. 
3285
-
3293
)
[PubMed]
13
Héliès-Toussaint
 
C.
Gambert
 
S.
Roller
 
P.
Tricot
 
S.
Lacour
 
B.
Grynberg
 
A.
 
Lipid metabolism in human endothelial cells
Biochim. Biophys. Acta
2006
, vol. 
1761
 (pg. 
765
-
774
)
[PubMed]
14
Tanaka
 
T.
Yamamoto
 
J.
Iwasaki
 
S.
Asaba
 
H.
Hamura
 
H.
Ikeda
 
Y.
Watanabe
 
M.
Magoori
 
K.
Ioka
 
R.X.
Tachibana
 
K.
, et al 
Activation of peroxisome proliferator-activated receptor δ induces fatty acid β-oxidation in skeletal muscle and attenuates metabolic syndrome
Proc. Natl. Acad. Sci. U.S.A.
2003
, vol. 
100
 (pg. 
15924
-
15929
)
[PubMed]
15
Alvarez-Guardia
 
D.
Palomer
 
X.
Coll
 
T.
Serrano
 
L.
Rodríguez-Calvo
 
R.
Davidson
 
M.M.
Merlos
 
M.
El Kochairi
 
I.
Michalik
 
L.
Wahli
 
W.
Vázquez-Carrera
 
M.
 
PPARβ/δ activation blocks lipid-induced inflammatory pathways in mouse heart and human cardiac cells
Biochim. Biophys. Acta
2011
, vol. 
1811
 (pg. 
59
-
67
)
[PubMed]
16
Kim
 
F.
Tysseling
 
K.A.
Rice
 
J.
Pham
 
M.
Haji
 
L.
Gallis
 
B.M.
Baas
 
A.S.
Paramsothy
 
P.
Giachelli
 
C.M.
Corson
 
M.A.
Raines
 
E.W.
 
Free fatty acid impairment of nitric oxide production in endothelial cells is mediated by IKKβ
Arterioscler. Thromb. Vasc. Biol.
2005
, vol. 
25
 (pg. 
989
-
994
)
[PubMed]
17
Maloney
 
E.
Sweet
 
I.R.
Hockenbery
 
D.M.
Pham
 
M.
Rizzo
 
N.O.
Tateya
 
S.
Handa
 
P.
Schwartz
 
M.W.
Kim
 
F.
 
Activation of NF-κB by palmitate in endothelial cells: a key role for NADPH oxidase-derived superoxide in response to TLR4 activation
Arterioscler. Thromb. Vasc. Biol.
2009
, vol. 
29
 (pg. 
1370
-
1375
)
[PubMed]
18
Contreras
 
C.
Sánchez
 
A.
García-Sacristán
 
A.
Martínez
 
M.C.
Andriantsitohaina
 
R.
Prieto
 
D.
 
Preserved insulin vasorelaxation and up-regulation of the Akt/eNOS pathway in coronary arteries from insulin resistant obese Zucker rats
Atherosclerosis
2011
, vol. 
217
 (pg. 
331
-
339
)
[PubMed]
19
Chavez
 
J.A.
Summers
 
S.A.
 
Characterizing the effects of saturated fatty acids on insulin signaling and ceramide and diacylglycerol accumulation in 3T3-L1 adipocytes and C2C12 myotubes
Arch. Biochem. Biophys.
2003
, vol. 
419
 (pg. 
101
-
109
)
[PubMed]
20
Kobayashi
 
M.
Inoue
 
K.
Warabi
 
E.
Minami
 
T.
Kodama
 
T.
 
A simple method of isolating mouse aortic endothelial cells
J. Atheroscler. Thromb.
2005
, vol. 
12
 (pg. 
138
-
142
)
[PubMed]
21
Fernandes
 
D.C.
Wosniak
 
J.
Pescatore
 
L.A.
Bertoline
 
M.A.
Liberman
 
M.
Laurindo
 
F.R.
Santos
 
C.X.
 
Analysis of DHE-derived oxidation products by HPLC in the assessment of superoxide production and NADPH oxidase activity in vascular systems
Am. J. Physiol. Cell Physiol.
2007
, vol. 
292
 (pg. 
C413
-
C422
)
[PubMed]
22
Laurindo
 
F.R.
Fernandes
 
D.C.
Santos
 
C.X.
 
Assessment of superoxide production and NADPH oxidase activity by HPLC analysis of dihydroethidium oxidation products
Methods Enzymol.
2008
, vol. 
441
 (pg. 
237
-
260
)
[PubMed]
23
Sanchez-Siles
 
A.A.
Ishimura
 
N.
Rumi
 
M.A.
Tamagawa
 
Y.
Ito
 
S.
Ishihara
 
S.
Nabika
 
T.
Kinoshita
 
Y.
 
Administration of PPARβ/δ agonist reduces copper-induced liver damage in mice: possible implications in clinical practice
J. Clin. Biochem. Nutr.
2011
, vol. 
49
 (pg. 
42
-
49
)
[PubMed]
24
Zarzuelo
 
M.J.
Jiménez
 
R.
Galindo
 
P.
Sánchez
 
M.
Nieto
 
A.
Romero
 
M.
Quintela
 
A.M.
López-Sepúlveda
 
R.
Gómez-Guzmán
 
M.
Bailón
 
E.
, et al 
Antihypertensive effects of peroxisome proliferator-activated receptor-β activation in spontaneously hypertensive rats
Hypertension
2011
, vol. 
58
 (pg. 
733
-
743
)
[PubMed]
25
Toral
 
M.
Gómez-Guzmán
 
M.
Jiménez
 
R.
Romero
 
M.
Sánchez
 
M.
Utrilla
 
M.P.
Garrido-Mesa
 
N.
Rodríguez-Cabezas
 
M.E.
Olivares
 
M.
Gálvez
 
J.
Duarte
 
J.
 
The probiotic Lactobacillus coryniformis CECT5711 reduces the vascular pro-oxidant and pro-inflammatory status in obese mice
Clin. Sci.
2014
, vol. 
127
 (pg. 
33
-
45
)
[PubMed]
26
Sanchez
 
M.
Lodi
 
F.
Vera
 
R.
Villar
 
I.C.
Cogolludo
 
A.
Jimenez
 
R.
Moreno
 
L.
Romero
 
M.
Tamargo
 
J.
Perez-Vizcaino
 
F.
Duarte
 
J.
 
Quercetin and isorhamnetin prevent endothelial dysfunction, superoxide production, and overexpression of p47phox induced by angiotensin II in rat aorta
J. Nutr.
2007
, vol. 
137
 (pg. 
910
-
915
)
[PubMed]
27
Graham
 
T.L.
Mookherjee
 
C.
Suckling
 
K.E.
Palmer
 
C.N.
Patel
 
L.
 
The PPARδ agonist GW0742X reduces atherosclerosis in LDLR−/− mice
Atherosclerosis
2005
, vol. 
181
 (pg. 
29
-
37
)
[PubMed]
28
Rafikov
 
R.
Fonseca
 
F.V.
Kumar
 
S.
Pardo
 
D.
Darragh
 
C.
Elms
 
S.
Fulton
 
D.
Black
 
S.M.
 
eNOS activation and NO function: structural motifs responsible for the posttranslational control of endothelial nitric oxide synthase activity
J. Endocrinol.
2011
, vol. 
210
 (pg. 
271
-
284
)
[PubMed]
29
Zanetti
 
M.
Barazzoni
 
R.
Stebel
 
M.
Roder
 
E.
Biolo
 
G.
Baralle
 
F.E.
Cattin
 
L.
Guarnieri
 
G.
 
Dysregulation of the endothelial nitric oxide synthase–soluble guanylate cyclase pathway is normalized by insulin in the aorta of diabetic rat
Atherosclerosis
2005
, vol. 
181
 (pg. 
69
-
73
)
[PubMed]
30
Guo
 
W.X.
Yang
 
Q.D.
Liu
 
Y.H.
Xie
 
X.Y.
Wang-Miao
 
Niu
 
R.C.
 
Palmitic and linoleic acids impair endothelial progenitor cells by inhibition of Akt/eNOS pathway
Arch. Med. Res.
2008
, vol. 
39
 (pg. 
434
-
442
)
[PubMed]
31
Shiba
 
T.
Inoguchi
 
T.
Sportsman
 
J.R.
Heath
 
W.F.
Bursell
 
S.
King
 
G.L.
 
Correlation of diacylglycerol level and protein kinase C activity in rat retina to retinal circulation
Am. J. Physiol.
1993
, vol. 
265
 (pg. 
E783
-
E793
)
[PubMed]
32
Naruse
 
K.
Rask-Madsen
 
C.
Takahara
 
N.
Ha
 
S.W.
Suzuma
 
K.
Way
 
K.J.
Jacobs
 
J.R.
Clermont
 
A.C.
Ueki
 
K.
Ohshiro
 
Y.
, et al 
Activation of vascular protein kinase C-β inhibits Akt-dependent endothelial nitric oxide synthase function in obesity-associated insulin resistance
Diabetes
2006
, vol. 
55
 (pg. 
691
-
698
)
[PubMed]
33
Chiasson
 
V.L.
Quinn
 
M.A.
Young
 
K.J.
Mitchell
 
B.M.
 
Protein kinase CβII-mediated phosphorylation of endothelial nitric oxide synthase threonine 495 mediates the endothelial dysfunction induced by FK506 (tacrolimus)
J. Pharmacol. Exp. Ther.
2011
, vol. 
337
 (pg. 
718
-
723
)
[PubMed]
34
Mugabo
 
Y.
Mukaneza
 
Y.
Renier
 
G.
 
Palmitate induces C-reactive protein expression in human aortic endothelial cells: relevance to fatty acid-induced endothelial dysfunction
Metabolism
2011
, vol. 
60
 (pg. 
640
-
648
)
[PubMed]
35
Wang
 
Y.
Biswas
 
G.
Prabu
 
S.K.
Avadhani
 
N.G.
 
Modulation of mitochondrial metabolic function by phorbol 12-myristate 13-acetate through increased mitochondrial translocation of protein kinase Cα in C2C12 myocytes
Biochem. Pharmacol.
2006
, vol. 
72
 (pg. 
881
-
892
)
[PubMed]

Author notes

1

These authors contributed equally to this work.

Supplementary data