The antiinflammatory effect of laminar flow: the role of PPARgamma, epoxyeicosatrienoic acids, and soluble epoxide hydrolase

Abstract

We previously reported that laminar flow activates peroxisome proliferator-activated receptor gamma (PPARgamma) in vascular endothelial cells in a ligand-dependent manner that involves phospholipase A2 and cytochrome P450 epoxygenases. In this study, we investigated whether epoxyeicosatrienoic acids (EETs), the catalytic products of cytochrome P450 epoxygenases, are PPARgamma ligands. Competition and direct binding assays revealed that EETs bind to the ligand-binding domain of PPARgamma with K(d) in the microM range. In the presence of adamantyl-ureido-dodecanoic acid (AUDA), a soluble epoxide hydrolase (sEH)-specific inhibitor, EETs increased PPARgamma transcription activity in endothelial cells and 3T3-L1 preadipocytes. Inclusion of AUDA in the perfusing media enhanced, but overexpression of sEH reduced, the laminar flow-induced PPARgamma activity. Furthermore, laminar flow augmented cellular levels of EETs but decreased sEH at the levels of mRNA, protein, and activity. Blocking PPARgamma by GW9662 abolished the EET/AUDA-mediated antiinflammatory effect, which indicates that PPARgamma is an effector of EETs.

Fig. 1.

EETs are PPARγ ligands. (A) One microgram of GST-PPARγ-LBD fusion protein was incubated with 100 nM [³H]rosiglitazone (Rosi) in a total volume of 50 μl in the presence of vehicle (DMSO), unlabeled rosiglitazone, or various EETs at the indicated concentrations. (B) For each data point, 1 μg of GST-PPARγ-LBD was incubated with various amounts of [³H]EETs in a total volume of 50 μl. The free and bound ligands in A and B were separated on a Sephadex G-25 column, and the amount of bound [³H]rosiglitazone or [³H]EETs was then determined by liquid scintillation counting. The results in A are presented as mean ± SD from three sets of experiments. ^*, P < 0.05 compared with DMSO controls. The plot of the binding curve in B represents three independent experiments, and Scatchard analysis was performed by replotting the data shown in Insets. K_d was presented as mean ± SD of Scatchard analysis results from three independent experiments.

Fig. 2.

EETs, together with AUDA, activate the PPARγ-regulated transcription. (A) BAECs in 12-well plates were cotransfected with MH100×4-TK-Luc (0.25 μg), GAL-mPPARγ-LBD (0.25 μg), and CMV-Renilla-Luc (0.05 μg). The transfected cells were then incubated with EETs (1 μM) in the presence or absence of AUDA (1 μM), and the media were replaced every 2 h. After 8 h, cells were lysed for luciferase activity assays. The results represent the relative luciferase activity defined as the normalized luciferase activity of various experiments in reference to that of DMSO controls. (B) Total RNAs were isolated from HUVECs or 3T3-L1 cells incubated with EETs (1 μM) in the presence of AUDA (1 μM) for 12 h. The levels of fatty acid-binding protein 4 mRNA in HUVECs or adipocyte P2 mRNA in 3T3-L1 cells were determined by quantitative RT-PCR in which β-actin was an internal control. The relative mRNA level is defined as the levels in cells treated with various EETs in reference to that of DMSO set as 1. (C) BAECs in 12-well plates were cotransfected with MH100×4-TK-Luc, GAL-mPPARγ-LBD, and CMV-Renilla-Luc. The transfected cells were then incubated with DHETs or GW9662 (5 μM) in the presence or absence of rosiglitazone (1 μM) for 8 h and lysed for luciferase activity assays. The results represent the relative luciferase activity defined as the normalized luciferase activity of various experiments in reference to that of DMSO controls as 1. ^*, P < 0.05.

Fig. 3.

Inhibition of sEH in ECs enhances the response of PPARγ to laminar flow. (A) BAECs transfected with GAL-mPPARγ-LBD and MH100×4-Luc or PPRE×3-TK-Luc were kept as static controls or subjected to laminar flow for 8 h in the presence or absence of AUDA (1 μM), an sEH-specific inhibitor, or 1-cyclohexyl-3-ethyl urea (CEU, 1 μM), an AUDA analogue that does not inhibit sEH. (B) BAECs on glass slides were cotransfected with pCDNA3 (0.3 μg) or an expression plasmid encoding sEH (psEH, 0.3 μg) with GAL-mPPARγ-LBD (0.3 μg) and MH100×4-Luc (0.3 μg) or PPRE×3-TK-Luc (0.6 μg). The transfected cells were then kept as static controls or subjected to laminar flow for 8 h. Cells in both A and B were then lysed for luciferase activity assays. The bars represent the relative luciferase activity defined as the normalized luciferase activity of various experiments in reference to that of DMSO-treated static cells in A or pCDNA3-transfected cells under static conditions in B. ^*, P < 0.05.

Fig. 4.

Laminar flow increases EETs and down-regulates sEH in ECs. BAECs in A, D, and E and HUVECs in C were kept as static controls or subjected to a laminar flow at 12 dyne/cm² for the indicated times. (A) Total lipids were extracted from static BAECs or those exposed to laminar flow in the absence of FBS. EETs were quantified by LC/MS/MS. The amounts of EETs were normalized to the mass of the cell pellets (in grams). (B) BAECs in 12-well plates were cotransfected with MH100×4-TK-Luc, GAL-mPPARγ-LBD, and CMV-Renilla-Luc. The transfected cells were then incubated with the condition media collected from the flow experiments in A for 8 h and lysed for luciferase activity assays. The results represent the relative luciferase activity defined as the normalized luciferase activity of various experiments in reference to that of static medium controls as 1. (C) Total RNA was isolated from HUVECs, and the level of sEH mRNA was determined by quantitative RT-PCR with β-actin used as an internal control. (D) BAECs were lysed for immunoblotting with the use of polyclonal anti-sEH Ab. (E) Cytosolic supernatants obtained from BAECs were incubated with trans-[³H]stilbene oxide for sEH activity assays. ^*, P < 0.05, compared with static controls.

Fig. 5.

PPARγ mediates the antiinflammatory effect of EETs. BAECs were pretreated with AUDA (1 μM) for 2 h and then incubated with various EETs (1 μM) for 8 h. (B) Cells were cultured as in A, except that GW9662 (1 μM) was included in the indicated experiments during the 8-h incubation. All cells were then stimulated with TNF-α (10 ng/ml) for 30 min. The collected cell lysates were immunoblotted with anti-IκBα antibody. α-Tubulin was used as a loading control.