Advertisement

LXR agonism improves TNF-α-induced endothelial dysfunction in the absence of its cholesterol-modulating effects

  • Frank Spillmann
    Affiliations
    Department of Cardiology and Pneumology, Charité – University Medicine Berlin, Campus Benjamin Franklin, Berlin, Germany
    Search for articles by this author
  • Sophie Van Linthout
    Affiliations
    Berlin-Brandenburg Center for Regenerative Therapies, Charité – University Medicine Berlin, Campus Virchow, Berlin, Germany
    Search for articles by this author
  • Kapka Miteva
    Affiliations
    Berlin-Brandenburg Center for Regenerative Therapies, Charité – University Medicine Berlin, Campus Virchow, Berlin, Germany
    Search for articles by this author
  • Mario Lorenz
    Affiliations
    Department of Cardiology and Angiology, Charité – University Medicine Berlin, Campus Mitte, Berlin, Germany

    DZHK (German Center for Cardiovascular Research), Berlin, Germany
    Search for articles by this author
  • Verena Stangl
    Affiliations
    Department of Cardiology and Angiology, Charité – University Medicine Berlin, Campus Mitte, Berlin, Germany

    DZHK (German Center for Cardiovascular Research), Berlin, Germany
    Search for articles by this author
  • Heinz-Peter Schultheiss
    Affiliations
    Department of Cardiology and Pneumology, Charité – University Medicine Berlin, Campus Benjamin Franklin, Berlin, Germany
    Search for articles by this author
  • Carsten Tschöpe
    Correspondence
    Corresponding author. Department of Cardiology & Pneumology, Charité – University Medicine Berlin, Campus Benjamin Franklin, Hindenburgdamm 30, 12200 Berlin. Tel.: +49 (0)30 8445 4780; fax: +49 (0)30 8445 4648.
    Affiliations
    Department of Cardiology and Pneumology, Charité – University Medicine Berlin, Campus Benjamin Franklin, Berlin, Germany

    Berlin-Brandenburg Center for Regenerative Therapies, Charité – University Medicine Berlin, Campus Virchow, Berlin, Germany

    DZHK (German Center for Cardiovascular Research), Berlin, Germany
    Search for articles by this author

      Highlights

      • T0901317 LXR agonism improves TNF-α-induced endothelial dysfunction.
      • T0901317 LXR agonism decreases oxidative stress in endothelial cells.
      • T0901317 LXR agonism improves NO metabolism in endothelial cells.
      • LXRβ-, but not LXRα-silencing, abrogates the anti-apoptotic effects of T0901317.
      • The LXR agonist GW3965 confirms the endothelial-protective effects of T0901317.

      Abstract

      Stimulation of the liver X receptor (LXR) is associated with anti-inflammatory and vascular-protective effects under hyperlipemic conditions. We examined whether LXR stimulation influences TNF-α-induced endothelial dysfunction under normolipemic conditions. Endothelium-dependent vasorelaxation of aortic rings was determined in an organ water bath. Human umbilical vein endothelial cells (HUVEC) were exposed to TNF-α (10 ng/ml) in the presence or absence of 5 μM of the LXR agonist T0901317 or GW3965 and changes in TNF-α-induced endothelial cell apoptosis, inflammation, oxidative stress, and NO metabolism were analyzed. T0901317 improved TNF-α-impaired endothelium-dependent relaxation of aortic rings in response to acetylcholine. T0901317 decreased the TNF-α-induced apoptosis and inflammation as indicated by a decrease in caspase 3/7 activity, VCAM-1 mRNA expression and subsequent mononuclear cell adhesion. Furthermore, T0901317 reduced the expression of the oxidative stress markers: AT1R, NOX4, and p22phox and normalized the TNF-α-induced NOX activity to basal levels. In line with the reduced AT1R expression, T0901317 impaired the Ang II responsiveness. T0901317 influenced NO metabolism as indicated by a decrease in TNF-α-upregulated arginase activity, a reversal of TNF-α-induced downregulation of argininosuccinate synthase mRNA expression and eNOS expression to basal levels and a raise in NO production. Furthermore, T0901317 decreased the TNF-α-induced superoxide and nitrotyrosine production, but did not upregulate the TNF-α-downregulated eNOS dimer/monomer ratio. Silencing of LXRβ, but not of LXRα, abrogated the anti-apoptotic effects of T0901317. We conclude that LXR agonism improves TNF-α-impaired endothelial function via its anti-apoptotic, anti-inflammatory, and anti-oxidative properties and its capacity to restore TNF-α-impaired NO bioavailability independent of its cholesterol-modulating effects.

      Keywords

      Abbreviations:

      HUVEC (human umbilical vein cells), LXR (Liver X receptor), NO (nitric oxide), eNOS (endothelial nitric oxide synthase), NOX (NAD(P)H oxidase), O2• (superoxide), ROS (reactive oxygen species), RXR (retinoid X receptor), TNF (tumor necrosis factor)

      1. Introduction

      Vascular endothelial dysfunction occurs at the early onset of atherogenesis. It is characterized by a reduction of the bioavailability of the major endothelium-derived vasoactive mediator, nitric oxide (NO) [
      • Deanfield J.E.
      • Halcox J.P.
      • Rabelink T.J.
      Endothelial function and dysfunction: testing and clinical relevance.
      ] and takes place in the absence of any structural changes of the vessel wall. NO is produced by the action of (endothelial) NO synthase (eNOS) which converts l-arginine to l-citrulline. Among others, deficit of its substrate l-arginine, due to increased arginase activity [
      • Gao X.
      • Xu X.
      • Belmadani S.
      • et al.
      TNF-alpha contributes to endothelial dysfunction by upregulating arginase in ischemia/reperfusion injury.
      ] or impaired recycling of l-citrulline to l-arginine [
      • Goodwin B.L.
      • Solomonson L.P.
      • Eichler D.C.
      Argininosuccinate synthase expression is required to maintain nitric oxide production and cell viability in aortic endothelial cells.
      ] results in impaired NO production.
      Inflammation is an important trigger of endothelial dysfunction [
      • Riad A.
      • Westermann D.
      • Van Linthout S.
      • et al.
      Enhancement of endothelial nitric oxide synthase production reverses vascular dysfunction and inflammation in the hindlimbs of a rat model of diabetes.
      ]. Among the cytokines, TNF-α is known to affect NO production by 1) reducing eNOS expression [
      • Yan G.
      • You B.
      • Chen S.P.
      • Liao J.K.
      • Sun J.
      Tumor necrosis factor-alpha downregulates endothelial nitric oxide synthase mRNA stability via translation elongation factor 1-alpha 1.
      ], 2) to increase the natural competitor of eNOS, arginase [
      • Gao X.
      • Xu X.
      • Belmadani S.
      • et al.
      TNF-alpha contributes to endothelial dysfunction by upregulating arginase in ischemia/reperfusion injury.
      ], and 3) to decrease the recycling of l-citrulline to l-arginine via the downregulation in argininosuccinate synthase expression [
      • Goodwin B.L.
      • Pendleton L.C.
      • Levy M.M.
      • Solomonson L.P.
      • Eichler D.C.
      Tumor necrosis factor-alpha reduces argininosuccinate synthase expression and nitric oxide production in aortic endothelial cells.
      ]. Furthermore, TNF-α induces reactive oxygen species (ROS) [
      • Kataoka H.
      • Murakami R.
      • Numaguchi Y.
      • Okumura K.
      • Murohara T.
      Angiotensin II type 1 receptor blockers prevent tumor necrosis factor-alpha-mediated endothelial nitric oxide synthase reduction and superoxide production in human umbilical vein endothelial cells.
      ], involving NAD(P)H oxidase (NOX) [
      • Gao X.
      • Zhang H.
      • Belmadani S.
      • et al.
      Role of TNF-alpha-induced reactive oxygen species in endothelial dysfunction during reperfusion injury.
      ] and uncoupled eNOS [
      • Pritchard Jr., K.A.
      • Groszek L.
      • Smalley D.M.
      • et al.
      Native low-density lipoprotein increases endothelial cell nitric oxide synthase generation of superoxide anion.
      ]. Consequently, TNF-α impairs endothelium-dependent relaxation [
      • Katz S.D.
      • Rao R.
      • Berman J.W.
      • et al.
      Pathophysiological correlates of increased serum tumor necrosis factor in patients with congestive heart failure. Relation to nitric oxide-dependent vasodilation in the forearm circulation.
      ].
      Liver X receptors (LXR) are ligand-activated transcription factors that form heterodimers with the retinoid X receptor (RXR) and belong to the nuclear hormone receptor superfamily. Oxysterols and oxidized cholesterol derivates are endogenous ligands of LXRs. LXRs regulate the expression of genes involved in lipid and glucose metabolism and reverse cholesterol transport [
      • Im S.S.
      • Osborne T.F.
      Liver X receptors in atherosclerosis and inflammation.
      ]. They also raise high-density lipoproteins (HDL) [
      • Miao B.
      • Zondlo S.
      • Gibbs S.
      • et al.
      Raising HDL cholesterol without inducing hepatic steatosis and hypertriglyceridemia by a selective LXR modulator.
      ], which are besides their important role in reverse cholesterol transport also known for their endothelial-protective effects [
      • Van Linthout S.
      • Spillmann F.
      • Lorenz M.
      • et al.
      Vascular-protective effects of high-density lipoprotein include the downregulation of the angiotensin II type 1 receptor.
      ,
      • Van Linthout S.
      • Spillmann F.
      • Graiani G.
      • et al.
      Down-regulation of endothelial TLR4 signalling after apo A-I gene transfer contributes to improved survival in an experimental model of lipopolysaccharide-induced inflammation.
      ,
      • Van Linthout S.
      • Spillmann F.
      • Riad A.
      • et al.
      Human apolipoprotein A-I gene transfer reduces the development of experimental diabetic cardiomyopathy.
      ]. In addition, synthetic LXR agonists, such as T0901317 have been shown to have anti-inflammatory [
      • Blaschke F.
      • Takata Y.
      • Caglayan E.
      • et al.
      A nuclear receptor corepressor-dependent pathway mediates suppression of cytokine-induced C-reactive protein gene expression by liver X receptor.
      ], anti-oxidative [
      • Imayama I.
      • Ichiki T.
      • Patton D.
      • et al.
      Liver X receptor activator downregulates angiotensin II type 1 receptor expression through dephosphorylation of Sp1.
      ], and anti-apoptotic [
      • Lei P.
      • Baysa A.
      • Nebb H.I.
      • et al.
      Activation of Liver X receptors in the heart leads to accumulation of intracellular lipids and attenuation of ischemia-reperfusion injury.
      ] features. Given these properties, LXRs have been suggested as potential targets for therapeutic intervention in human cardiovascular disorders [
      • Tontonoz P.
      • Mangelsdorf D.J.
      Liver X receptor signaling pathways in cardiovascular disease.
      ]. Recently, LXRs have been shown to reduce plaque formation and to improve vasomotor function in atherosclerotic apo E−/− mice [
      • Chen J.
      • Zhao L.
      • Sun D.
      • et al.
      Liver X receptor activation attenuates plaque formation and improves vasomotor function of the aortic artery in atherosclerotic ApoE(−/−) mice.
      ]. However, an impact of LXR agonism on endothelial function independent of its cholesterol lowering capacity has not been demonstrated before. Therefore, the aim of the present study was to evaluate whether LXR agonism improves TNF-α-induced endothelial dysfunction in aortic rings, independent of its cholesterol lowering capacity. Underlying mechanisms were analyzed by supplementation of the LXR agonist T0901317 on TNF-α-stimulated human umbilical vein cells (HUVEC).

      2. Methods

      2.1 Vasorelaxation studies in isolated rat aortic rings

      Thoracic aortae from anaesthesized male Wistar rats were rapidly excised, cleaned of connective tissue, and cut into rings 2–3 mm in length for organ-chamber experiments as described previously [
      • Stangl V.
      • Lorenz M.
      • Meiners S.
      • et al.
      Long-term up-regulation of eNOS and improvement of endothelial function by inhibition of the ubiquitin-proteasome pathway.
      ]. Before mounting on the organ-chamber, rings were cultured in the presence or absence of 25 pg/ml of TNF-α, with or without 5 μM T0901317 for 24 h in DMEM medium. Rings were mounted on platinum hooks in 10 ml jacketed organ baths containing modified Krebs–Henseleit solution (composition, in mmol/l: NaCl 144, KCl 5.9, CaCl2 1.6, MgSO4 1.2, KH2PO4 1.2, NaHCO3 25, and d-glucose 11.1) and 1 μmol/l diclofenac. Tension was gradually adjusted to 2 g over 1 h. The solution in the bath was maintained at 37 °C with a gas mixture of 5% CO2 and 95% O2. Following equilibration and submaximal pre-contraction with phenylephrine (0.05 μmol/l), relaxation to increasing concentrations (10 nmol/l to 10 μmol/l) of the endothelium-dependent vasodilator acetylcholine was performed to obtain cumulative concentration–response curves. Maintenance of smooth-muscle integrity was confirmed by evaluation of endothelium-independent vasodilation to sodium nitroprusside (SNP, 0.1–10 nM). Vasorelaxation is expressed as percentage of pre-contraction with phenylephrine.

      2.2 Cell culture

      HUVEC (7,500 or 10,000 and 150,000 per 96-well and 6-well, respectively) were cultured in endothelial basal medium and endothelial supplements (Promocell). After 24 h (h) of culture, medium was changed and cells were stimulated with or without 10 ng/ml of TNF-α in the presence or absence of 5 μM of the LXR agonist T0901317, which activates both LXRα and LXRβ. The concentration of 5 μM of T0901317 was determined in an experiment showing that among the evaluated concentrations (200 nM, 1 μM, 5 μM, and 10 μM), the most pronounced reduction in TNF-α-induced oxidative stress in HUVEC was reached by 5 μM of T0901317 (see Supplemental Fig. 1), a concentration also used in experiments with vascular smooth cells [
      • Cardellini M.
      • Menghini R.
      • Martelli E.
      • et al.
      TIMP3 is reduced in atherosclerotic plaques from subjects with type 2 diabetes and increased by SirT1.
      ] and aortic cells [
      • Hsu J.J.
      • Lu J.
      • Huang M.S.
      • et al.
      T0901317, an LXR agonist, augments PKA-induced vascular cell calcification.
      ]. Besides experiments with T0901317, HUVEC were also supplemented with TNF-α in the presence or absence of the LXR agonist GW3965 or l-NAME.

      2.3 Caspase 3/7 activity assay

      4 h after TNF-α stimulation in the presence or absence of 5 μM of T0901317 or 5 μM of GW3965, respectively, caspase 3/7 activity was determined with a Caspase Glo 3/7 activity kit (Promega) according to the manufacturer's protocol. Luminescence was measured in a microplate-reading luminometer (Mithras LB 940, Berthold Technologies GmbH & Co KG, Germany).

      2.4 Real-time PCR quantification

      Thirty minutes (min) after TNF-α stimulation, cells were collected in a lysis buffer (Miltenyi Biotech, Bergisch Gladbach, Germany). Next, mRNA was isolated and cDNA prepared with the MultiMACS One-step cDNA synthesis Kit (Miltenyi Biotech) according to the manufacturer's protocol. To analyze LXRα, LXRβ, RXRα, VCAM-1, AT1R, NOX4, p22phox, arginase II, and argininosuccinate synthase mRNA expression, quantitative real-time PCR (Eppendorf Mastercycler epgradient realplex, Hamburg, Germany) was performed. mRNA expression were normalized to 18S or L32 and relatively expressed with the control group set as 1. Commercial human NOX4, p22phox, arginase II, argininosuccinate synthase, and 18S primers (Applied Biosystems, Carlsbad, CA, USA) or self-designed human LXRα, LXRβ, RXRα, VCAM-1 (see Table 1 in Supplement), AT1R [
      • Van Linthout S.
      • Spillmann F.
      • Lorenz M.
      • et al.
      Vascular-protective effects of high-density lipoprotein include the downregulation of the angiotensin II type 1 receptor.
      ], and L32 [
      • Van Linthout S.
      • Spillmann F.
      • Lorenz M.
      • et al.
      Vascular-protective effects of high-density lipoprotein include the downregulation of the angiotensin II type 1 receptor.
      ] primers were used. Absolute mRNA expression levels of LXRα, LXRβ, and RXRα towards L32 were quantified via the use of plasmid standards containing the respective amplified PCR fragments, as described previously [
      • Van Linthout S.
      • Spillmann F.
      • Riad A.
      • et al.
      Human apolipoprotein A-I gene transfer reduces the development of experimental diabetic cardiomyopathy.
      ].

      2.5 Adhesion assay

      Thirty min after TNF-α and/or T0901317 supplementation, HUVEC (10,000/96-well) were stimulated with 0 or 5 μM Ang II. 4 h after the TNF-α and/or T0901317 stimulation, HUVEC were washed and 100,000 DiO-labeled mononuclear cells (MNCs) activated with phorbolmyristate acetate (50 ng/ml)/ionomycin (500 ng/ml) were added to HUVEC for 30 min. After three washes with PBS, absorbance was measured at 517 nM with a Berthold fluorometer (Mithras LB 940, Berthold Technologies GmbH & Co KG, Germany). Data depict the absorbance of bound MNCs to HUVEC minus the absorbance of HUVEC without MNC.

      2.6 NADPH oxidase activity

      As described previously [
      • Van Linthout S.
      • Spillmann F.
      • Lorenz M.
      • et al.
      Vascular-protective effects of high-density lipoprotein include the downregulation of the angiotensin II type 1 receptor.
      ], NOX activity of HUVEC was analyzed according to Griendling et al. [
      • Griendling K.K.
      • Minieri C.A.
      • Ollerenshaw J.D.
      • Alexander R.W.
      Angiotensin II stimulates NADH and NADPH oxidase activity in cultured vascular smooth muscle cells.
      ], slightly modified. In brief, 1 h after TNF-α and/or T0901317 supplementation, cells were washed in ice-cold PBS. Then, cells were scraped from the plate in 1 ml of ice-cold PBS and centrifuged for 10 min at 4 °C at 10,000 rpm. The supernatant was discarded and the pellet resuspended in 90 μl of cell lysis buffer (Invitrogen). NOX activity was measured by lucigenin-enhanced chemiluminescent detection of superoxide (O2•) with a luminometer (Mithras LB 940, Berthold Technologies GmbH & Co KG, Germany). The reaction buffer contained 1 mM EGTA, 150 mM sucrose, 500 μM lucigenin and 1 mM of NADPH (180 μl). The reaction was started by addition of 20 μl of protein homogenate. Luminescence was measured as the rate of photon counts per μg protein, after substraction of the counts obtained from a buffer blank.

      2.7 Reactive oxygen species analysis

      To determine the dose response effect of T0901317 and the impact of 5 μM of GW3965 on TNF-α-induced ROS production in HUVEC, HUVEC (7,500/96-well) were supplemented for 4 h with TNF-α with or without 200 nM, 1 μM, 5 μM, or 10 μM T0901317, or 5 μM GW3965, respectively. Next, cells were washed with PBS and incubated with 20 μM CM-H2DCFDA in serum-free medium for 45 min at 37 °C. After two washing steps with warm PBS for 10 min, cells were incubated with serum-free medium for 30 min. Thereafter, fluorescence intensity was read in a Berthold Mithras LB 940 reader at 485 nm excitation and 530 nm emission wavelength. Data represent the absorbance of the DCF-supplemented cells minus the absorbance of unstained cells.
      The impact of T0901317 on TNF-α-induced endothelial ROS production was further analyzed via flow cytometry. Therefore, 4 h after stimulation with TNF-α and/or T0901317, cells were collected in pre-warmed PBS containing 5 μM of CM-H2DCFDA (Life Technologies), and then incubated at 37 °C for 30 min. After washing with PBS, all the samples were analyzed via flow cytometry (BD FACSCantoII, BD FACSDiva™ software 6.1.3.). Data are expressed as DCF + cells (% gated).

      2.8 Arginase activity

      Arginase activity of HUVEC was analyzed according to Chandra et al. [
      • Chandra S.
      • Romero M.J.
      • Shatanawi A.
      • Alkilany A.M.
      • Caldwell R.B.
      • Caldwell R.W.
      Oxidative species increase arginase activity in endothelial cells through the RhoA/Rho kinase pathway.
      ], slightly modified. 2 h after stimulation with TNF-α, cells were washed with ice-cold PBS, lysed with lysis buffer (Invitrogen) containing proteinase inhibitors (Roche), and centrifuged at 14,000 g for 10 min at 4 °C. The supernatant was collected for assay. Next, 25 μl of 10 mM of MnCl2 (dissolved in 50 mM of Tris–HCl) was added to 25 μl of cell lysate in eppendorfs and incubated at 56 °C for 10 min for activation of the enzyme. Fifty μl of 0.5 M of l-Arginine (dissolved in 50 mM of Tris–HCl, pH 9.7) was added to these tubes, incubated at 37 °C for 1 h and the reaction was stopped using 400 μl of the acid mixture consisting of H2SO4:H3PO4:H2O at a ratio of 1:3:7. For colorimetric determination, 25 μl of α-ISPF (dissolved in 9% ethanol) was added to each tube. Next, the mixture was heated for 45 min at 100 °C and kept in dark for 10 min. Readings were taken spectrophotometrically at an absorbance of 540 nm. Sample blanks contained lysate without addition of MnCl2 or l-arginine to measure basal arginase activity, and these readings were subtracted from all respective samples. Values were divided by the protein content to determine the specific activity of arginase, and multiplied by 1000.

      2.9 Western blot and low temperature SDS-PAGE

      2 h after stimulation with TNF-α, cells were lysed in lysis buffer (Invitrogen) containing proteinase inhibitors (Roche). An equal amount of protein (20 μg) was loaded into a SDS-polyacrylamide gel. Total eNOS (BD Biosciences, San Diego, CA, USA), LXRα (Active Motif, Carlsbad, CA, USA), LXRβ (Active Motif), GAPDH (Biodesign, Memphis, TN, USA) and β tubulin (Santa Cruz Biotechnology, Santa Cruz, CA, USA) were detected with each specific antibody, followed by incubation with an IR dye secondary antibody (LI-COR Biosciences, Lincoln/Nebraska, USA). For the investigation of eNOS homodimer formation, non-boiled proteins were resolved by a 6% SDS-polyacrylamide gel at 4 °C. All blots were visualized with Odyssey (LI-COR Biosciences). Quantitative analysis of the intensity of the bands was performed with Odyssey V3.0 software.

      2.10 Nitric oxide measurement

      Intracellular NO was measured with DAF-FM diacetate (4-amino-5-methylamino-2′,7′-difluorofluorescein diacetate, Invitrogen). 2 h after stimulation with TNF-α and/or 5 μM of T0901713 or 5 μM of GW3965, HUVEC were incubated at 37 °C for 30 min in PBS containing 1 μM of DAF-FM diacetate. After loading, cells were rinsed two times with PBS and incubated with fresh PBS at 37 °C for 30 min. NO fluorescence was measured with Gemini Fluorescence Microplate Readers using excitation and emission wavelengths of 488 nm and 520 nm, respectively. By the GW3965 experiment, the NO fluorescence intensity was read in a Berthold Mithras LB 940 reader at 495 nm excitation and 515 nm emission wavelength.

      2.11 Nitrotyrosine measurement

      Nitrotyrosine was analyzed in 50 μg protein of HUVEC by the OxiSelect nitrotyrosine ELISA kit of Cell Biolabs (Cell Biolabs, San Diego, CA, USA) according to the manufacturer's protocol.

      2.12 Superoxide analysis

      2 h after TNF-α and/or T0901317 supplementation or in the experiment to evaluate TNF-α-mediated eNOS uncoupling, 2 h after TNF-α and/or 1 mM L-NAME (Sigma) supplementation, HUVEC (10,000 or 7500 cells/96-well, respectively) were washed and a 10 μM dihydroethidium solution containing 2 mM HEPES and 50 mM glucose was added for 30 min. After rinsing with Hanks balanced salt solution, Hanks balanced salt solution was supplemented to the cells and the excitation/emission was analyzed at a wavelength of 560/660 nm. Data represent the absorbance of the DHE-supplemented cells minus the absorbance of unstained cells.

      2.13 siRNA experiments

      HUVEC were transfected with Lipofectamine RNAiMAX (Life Technologies) with LXRα, LXRβ or scrambled siRNA (Life Technologies) following the manufacturer's protocol. Real-time PCR was performed to confirm knock down of respective receptors (Supplemental Fig. 6).

      2.14 Statistical analysis

      Statistical analysis was performed using GraphPad Instat 3.0a (GraphPad Software, Inc., La Jolla, USA). Assumption of Gaussian distribution was consistently tested by the method of Kolmogorov and Smirnov. Paired and unpaired Student's t tests were used for statistical analysis or as differently stated. When no Gaussian distribution was reached, a non-parametrical test was used. Data are presented as mean ± SEM. Differences were considered to be significant when the two-sided P-value was lower than 0.05.

      3. Results

      3.1 T0901317 improves TNF-α-impaired vascular reactivity

      Acetylcholine-induced endothelium-dependent relaxation was impaired in aortic rings incubated with TNF-α for 24 h (P < 0.05 vs basal). Co-incubation of aortic rings with the LXR agonist T0901317 improved vascular reactivity to acetylcholine to control levels (P < 0.05 vs TNF-α). Endothelium-independent sodium nitroprusside-induced relaxations did not differ between all groups (Fig. 1).
      Figure thumbnail gr1
      Fig. 1T0901317 improves TNF-α-impaired vasorelaxation. A. Acetylcholine-induced endothelial-dependent relaxation and B. sodium nitroprusside-induced endothelial-independent relaxation. *P < 0.05 TNF-α versus TNF-α + T0901317.

      3.2 LXRα, LXRβ, and RXRα expression in endothelial cells

      Before investigating the underlying mechanisms in HUVEC, how LXR agonism with T0901317 can improve TNF-α-induced endothelial dysfunction, the expression of LXRα, LXRβ, and RXRα in HUVEC was determined. LXRβ and RXRα were more abundantly expressed as LXRα in HUVEC, whereas the mRNA expression of neither of the genes was affected by TNF-α supplementation or treatment with the LXR agonist T0901317 (Supplemental Fig. 2). In contrast, LXRβ protein expression was significantly increased upon T0901317 stimulation (Supplemental Fig. 3).

      3.3 T0901317 reduces TNF-α-induced endothelial cell apoptosis and inflammation

      Supplementation of T0901317 to HUVEC decreased the TNF-α-induced endothelial cell apoptosis as indicated by a 1.2-fold (P < 0.05 vs TNF-α) decline in caspase 3/7 activity (Fig. 2). A reduction in TNF-α-induced caspase 3/7 activity could also be demonstrated for the LXR agonist GW3965 (Supplemental Fig. 4A). In addition, T0901317 downregulated the TNF-α-induced endothelial VCAM-1 expression by 2.3-fold (P < 0.05) (Fig. 3A), which was associated with a 1.4-fold (P < 0.05) decline in TNF-α-induced adhesion of MNCs to HUVEC (Fig. 4).
      Figure thumbnail gr2
      Fig. 2T0901317 decreases TNF-α-induced endothelial cell apoptosis. Bar graphs represent the mean ± SEM of caspase 3/7 activity (n = 5–6/group) in untreated (open bar graphs) or TNF-α-treated (black bar graphs) HUVEC in the presence or absence of the LXR agonist T0901317, as indicated.
      Figure thumbnail gr3
      Fig. 3T0901317 decreases TNF-α-induced endothelial VCAM-1 and AT1R mRNA expression. Bar graphs represent the mean ± SEM of A. VCAM-1 and B. AT1R mRNA expression (n = 4–6/group) in untreated (open bar graphs) or TNF-α-treated (black bar graphs) HUVEC in the presence or absence of the LXR agonist T0901317, as indicated.
      Figure thumbnail gr4
      Fig. 4T0901317 decreases the Angiotensin II responsiveness in TNF-α-stimulated endothelial cells. Bar graphs represent the mean ± SEM of the absorbance at 517 nm of DiO-labeled MNC, which have bound to HUVEC (n = 5/group) in untreated (open bar graphs) or Ang II-treated (gray bar graphs) HUVEC, stimulated with or without TNF-α in the presence or absence of the LXR agonist T0901317, as indicated.

      3.4 T0901317 decreases TNF-α-induced oxidative stress in endothelial cells

      With respect to markers of oxidative stress, T0901317 reduced the TNF-α-upregulated AT1R (Fig. 2B), NOX4, and p22phox mRNA expression by 5.4-fold (P < 0.05), 1.7-fold (P < 0.05), and 2.9-fold (P < 0.05) (Fig. 5A and B), respectively, and normalized the TNF-α-induced NOX activity by 1.5-fold (P < 0.01) to basal levels (Fig. 5C). Furthermore, T0901317-decreased the TNF-α-induced ROS production in HUVEC by 1.2-fold (P = 0.001) (Fig. 4D), an effect, which could also be found for the LXR agonist GW3965 (Supplemental Fig. 4B). In agreement, with the T0901317-mediated reduction in AT1R expression, LXR agonism with T0901317 diminished the responsiveness to Ang II, as indicated by the finding that MNC adhesion to HUVEC pretreated with TNF-α and T0901317 before Ang II stimulation, was significantly lower compared to HUVEC supplemented with TNF-α alone preceding Ang II supplementation (Fig. 4).
      Figure thumbnail gr5
      Fig. 5T0901317 decreases NOX activity and ROS production in TNF-α-stimulated endothelial cells. Bar graphs represent the mean ± SEM of A. NOX4, B. p22phox mRNA expression (n = 4–6/group), C. NOX activity (n = 4–6/group), and D. ROS production (n = 4/group) in untreated (open bar graphs) or TNF-α-treated (black bar graphs) HUVEC in the presence or absence of the LXR agonist T0901317, as indicated.

      3.5 T0901317 restores NO production in TNF-α-stimulated endothelial cells

      T0901317 decreased the TNF-α-upregulated arginase II mRNA expression and arginase activity by 1.2-fold (P < 0.05) and 4.2-fold (P < 0.01), respectively (Fig. 6A and B), whereas it reversed the TNF-α-downregulated argininosuccinate synthase mRNA expression by 1.4-fold (P < 0.05) to levels not different from basal conditions (Fig. 6C). Furthermore, LXR agonism abrogated the TNF-α-induced downregulation in eNOS expression leading to normalization of the ratio of eNOS to β-tubulin to levels not different from basal conditions (Fig. 7A). Accordingly, T0901317 raised the TNF-α-impaired NO production by 1.9-fold (P < 0.05 vs TNF-α) (Fig. 7B), an effect, which could also be found for the LXR agonist GW3965 (Supplemental Fig. 4C). TNF-α induced eNOS uncoupling, which is supported by the L-NAME-mediated decrease in TNF-α-induced O2• production (Supplemental Fig. 5) and the TNF-α-induced decline in eNOS dimer/monomer ratio (Fig. 7C). T0901317 did not upregulate the TNF-α-declined eNOS dimer/monomer ratio (Fig. 7C), indicating no impact of T0901317 on eNOS uncoupling. However, it dropped the 1.4-fold and 1.6-fold TNF-α-induced nitrotyrosine and O2• production (P < 0.05 vs TNF-α) (Fig. 8A and B), respectively.
      Figure thumbnail gr6
      Fig. 6T0901317 influences the use and recycling of l-arginine in TNF-α-stimulated endothelial cells. Bar graphs represent the mean ± SEM of A. arginase II mRNA expression (n = 4–6/group), of B. arginase activity (n = 4–5/group), and of C. argininosuccinate synthase (AS) mRNA expression (n = 4–6/group) in untreated (open bar graphs) or TNF-α-treated (black bar graphs) HUVEC in the presence or absence of the LXR agonist T0901317, as indicated.
      Figure thumbnail gr7
      Fig. 7T0901317 increases eNOS expression and NO production in TNF-α-stimulated endothelial cells. A. Upper panel: Representative Western blots of eNOS and β-tubulin, and lower panel: bar graphs representing the mean ± SEM of eNOS towards β-tubulin expression, depicted as % versus basal set as 100% (n = 8–10/group), B. bar graphs representing the mean ± SEM of intracellular NO production (n = 4–6/group) and C. Upper panel: Representative blots of eNOS dimer and monomers, and lower panel: bar graphs representing the mean ± SEM of eNOS dimer towards eNOS monomer, depicted as % versus basal set as 100% (n = 3–4/group), with all (A,B, and C) of untreated (open bar graphs) or TNF-α-treated (black bar graphs) HUVEC in the presence or absence of the LXR agonist T0901317, as indicated.
      Figure thumbnail gr8
      Fig. 8T0901317 decreases TNF-α-induced nitrotyrosine and superoxide production in TNF-α-stimulated endothelial cells. A. Bar graphs represent the mean ± SEM of nitrotyrosine (nM per μg protein) (n = 4/group) and B. of O2• production depicted as the absorbance of dihydroethidium-supplemented cells minus the absorbance of unstained cells (n = 7–9/group) in untreated (open bar graphs) or TNF-α-treated (black bar graphs) HUVEC in the presence or absence of the LXR agonist T0901317, as indicated.

      3.6 LXRβ is involved in the T0901317-mediated effects in HUVEC

      To elucidate the LXR involved in the T0901317-mediated endothelial-protective effects, LXRα or LXRβ were knocked down via siRNA and supplemented with TNF-α in the presence of absence of T0901317 (Supplemental Fig. 6). In HUVEC transfected with scr siRNA or LXRα siRNA, T0901317 decreased the TNF-α-induced caspase 3/7 activity by 1.4-fold (P < 0.05 vs TNF-α) and 1.8-fold (P < 0.001 vs TNF-α), respectively. In contrast, in HUVEC transfected with LXRβ siRNA, the TNF-α-induced caspase 3/7 activity was not declined after T0901317 treatment (Fig. 9).
      Figure thumbnail gr9
      Fig. 9LXRβ is involved in the T0901317-mediated anti-apoptotic effects in HUVEC. Bar graphs represent the mean ± SEM of caspase 3/7 activity (n = 8/group) in untreated (open bar graphs) or TNF-α-treated (black bar graphs) HUVEC transfected with scrambled (scr), LXRα, or LXRβ siRNA, in the presence or absence of the LXR agonist T0901317 (T090), as indicated.

      4. Discussion

      The salient findings of the present study are that T0901317 LXR agonism improves directly TNF-α-induced endothelial dysfunction via its anti-apoptotic, anti-inflammatory, anti-oxidative properties and its capacity to improve NO bioavailability in the absence of its cholesterol-modulating effects, effects which also could be confirmed for the LXR agonist GW3965.
      The cardiovascular-protective effects of LXR ligands have particularly been explored in the field of atherosclerosis and are hereby mainly attributed to their dual role in lipid and lipoprotein homeostasis and their anti-inflammatory properties [
      • Blaschke F.
      • Takata Y.
      • Caglayan E.
      • et al.
      A nuclear receptor corepressor-dependent pathway mediates suppression of cytokine-induced C-reactive protein gene expression by liver X receptor.
      ,
      • Joseph S.B.
      • Castrillo A.
      • Laffitte B.A.
      • Mangelsdorf D.J.
      • Tontonoz P.
      Reciprocal regulation of inflammation and lipid metabolism by liver X receptors.
      ]. In detail, LXR ligands are known to inhibit the development of atherosclerosis via promoting reverse cholesterol transport through direct activation of genes involved in cellular cholesterol export, and via suppressing inflammation in macrophages [
      • Im S.S.
      • Osborne T.F.
      Liver X receptors in atherosclerosis and inflammation.
      ,
      • Joseph S.B.
      • McKilligin E.
      • Pei L.
      • et al.
      Synthetic LXR ligand inhibits the development of atherosclerosis in mice.
      ]. In contrast to the well-studied role of LXR signaling in macrophages, the impact of LXR action on other cells that directly may affect atherosclerosis, including endothelial cells and smooth muscle cells is less well investigated [
      • Im S.S.
      • Osborne T.F.
      Liver X receptors in atherosclerosis and inflammation.
      ]. Only recently, their capacity to improve vasomotor function in atherosclerotic apo E−/− mice has been demonstrated [
      • Chen J.
      • Zhao L.
      • Sun D.
      • et al.
      Liver X receptor activation attenuates plaque formation and improves vasomotor function of the aortic artery in atherosclerotic ApoE(−/−) mice.
      ] as well as their ability to induce eNOS phosphorylation in aortic endothelial cells of apo E−/− mice in vitro. Though, the role of LXR on endothelial function in the absence of its cholesterol-depending effects and a profound clarification of the impact of LXR on NO metabolism has not been investigated before.
      We demonstrated that LXRα, LXRβ, and RXRα are expressed in endothelial cells, with LXRβ being more abundantly expressed as LXRα. This is in agreement with previous findings showing that LXRα is more restricted and is mainly expressed in intestine, fat tissue, macrophage, kidney, lung, and in liver, whereas LXRβ is more ubiquitously expressed [
      • Im S.S.
      • Osborne T.F.
      Liver X receptors in atherosclerosis and inflammation.
      ]. Interestingly, we demonstrated that the protein expression of LXRβ, but not of LXRα, could be induced upon T0901317 supplementation. So far, it has been shown that the expression of LXRα can be upregulated by its own ligand in different human cells, due to the presence of a LXRE in the human LXRα promoter [
      • Li Y.
      • Bolten C.
      • Bhat B.G.
      • et al.
      Induction of human liver X receptor alpha gene expression via an autoregulatory loop mechanism.
      ]. Our data suggest a specific regulation for LXRβ in endothelial cells.
      Since atherosclerosis develops in response to injury, with inflammation and oxidative stress-induced endothelial dysfunction being a hallmark at the early onset of the atherogenesis process [
      • Harrison D.G.
      Cellular and molecular mechanisms of endothelial cell dysfunction.
      ], we evaluated the effect of TNF-α in the presence or absence of T0901317 on endothelial cell apoptosis, inflammation, oxidative stress, and NO production. In agreement with the well-documented anti-inflammatory [
      • Blaschke F.
      • Takata Y.
      • Caglayan E.
      • et al.
      A nuclear receptor corepressor-dependent pathway mediates suppression of cytokine-induced C-reactive protein gene expression by liver X receptor.
      ] and recently demonstrated anti-apoptotic [
      • Lei P.
      • Baysa A.
      • Nebb H.I.
      • et al.
      Activation of Liver X receptors in the heart leads to accumulation of intracellular lipids and attenuation of ischemia-reperfusion injury.
      ] properties of LXR, T0901317 reduced TNF-α-induced VCAM-1 expression and subsequent MNC adhesion to endothelial cells, and decreased endothelial cell apoptosis. Both findings may partly be explained by the increased bioavailability of NO (see supra), which has besides vasorelaxing, also anti-inflammatory, and anti-apoptotic [
      • Van Linthout S.
      • Spillmann F.
      • Riad A.
      • et al.
      Human apolipoprotein A-I gene transfer reduces the development of experimental diabetic cardiomyopathy.
      ] properties. Furthermore, we demonstrated that, in line with the more abundant expression of LXRβ compared to LXRα, the T0901317-mediated anti-apoptotic effects in HUVEC were abrogated when LXRβ was knocked down by siRNA, suggesting that T0901317 mediates its endothelial-protective effects via the LXRβ.
      We provide several lines of evidence that T0901317 improves NO metabolism: 1) T0901317 improved the vasorelaxation in aortic rings stressed with TNF-α; 2) it upregulated the TNF-α-decreased [
      • Yan G.
      • You B.
      • Chen S.P.
      • Liao J.K.
      • Sun J.
      Tumor necrosis factor-alpha downregulates endothelial nitric oxide synthase mRNA stability via translation elongation factor 1-alpha 1.
      ] expression of eNOS and 3) raised the TNF-α-reduced pool of the substrate l-arginine as suggested by the T0901317-mediated decrease in TNF-α-upregulated mRNA expression of arginase II, the key isoform regulating arginase activity in endothelial cells [
      • Ryoo S.
      • Lemmon C.A.
      • Soucy K.G.
      • et al.
      Oxidized low-density lipoprotein-dependent endothelial arginase II activation contributes to impaired nitric oxide signaling.
      ], and the subsequent reduction in arginase activity, as well as the T0901317-mediated raise in TNF-α-downregulated mRNA expression of argininosuccinate synthase, which catalyzes the rate-limiting step in the arginine regeneration from l-citrulline. The importance of a functional citrulline-NO cycle via argininosuccinate synthase for basal and stimulated NO production, even in the excess of arginine, has only recently been demonstrated [
      • Goodwin B.L.
      • Pendleton L.C.
      • Levy M.M.
      • Solomonson L.P.
      • Eichler D.C.
      Tumor necrosis factor-alpha reduces argininosuccinate synthase expression and nitric oxide production in aortic endothelial cells.
      ,
      • Shen L.J.
      • Beloussow K.
      • Shen W.C.
      Accessibility of endothelial and inducible nitric oxide synthase to the intracellular citrulline-arginine regeneration pathway.
      ]. A direct LXR-mediated decrease in arginase II expression is supported by Marathe and coworkers, who discovered a functional LXR response element in the arginase II promoter [
      • Marathe C.
      • Bradley M.N.
      • Hong C.
      • et al.
      The arginase II gene is an anti-inflammatory target of liver X receptor in macrophages.
      ]. Though, in contrast to our findings, they found that LXR agonism induced arginase II expression in macrophages [
      • Marathe C.
      • Bradley M.N.
      • Hong C.
      • et al.
      The arginase II gene is an anti-inflammatory target of liver X receptor in macrophages.
      ], reducing hereby nitrite and excess NO production during immune responses. These findings underscore the modulating capacity of LXR agonists as it also follows from their anti-inflammatory properties [
      • Blaschke F.
      • Takata Y.
      • Caglayan E.
      • et al.
      A nuclear receptor corepressor-dependent pathway mediates suppression of cytokine-induced C-reactive protein gene expression by liver X receptor.
      ,
      • Joseph S.B.
      • Castrillo A.
      • Laffitte B.A.
      • Mangelsdorf D.J.
      • Tontonoz P.
      Reciprocal regulation of inflammation and lipid metabolism by liver X receptors.
      ] on the one hand, and their ability to enhance innate immunity to bacterial pathogens via preventing bacterial-induced macrophage apoptosis [
      • Valledor A.F.
      • Hsu L.C.
      • Ogawa S.
      • Sawka-Verhelle D.
      • Karin M.
      • Glass C.K.
      Activation of liver X receptors and retinoid X receptors prevents bacterial-induced macrophage apoptosis.
      ] on the other hand.
      TNF-α induces ROS [
      • Kataoka H.
      • Murakami R.
      • Numaguchi Y.
      • Okumura K.
      • Murohara T.
      Angiotensin II type 1 receptor blockers prevent tumor necrosis factor-alpha-mediated endothelial nitric oxide synthase reduction and superoxide production in human umbilical vein endothelial cells.
      ], which are important triggers of endothelial dysfunction via their induction of NOX [
      • Gao X.
      • Zhang H.
      • Belmadani S.
      • et al.
      Role of TNF-alpha-induced reactive oxygen species in endothelial dysfunction during reperfusion injury.
      ] and of eNOS uncoupling [
      • Pritchard Jr., K.A.
      • Groszek L.
      • Smalley D.M.
      • et al.
      Native low-density lipoprotein increases endothelial cell nitric oxide synthase generation of superoxide anion.
      ], i.e. when eNOS produces O2• instead of NO, a phenomena which we could demonstrate via the drop in TNF-α induced O2• production after l-NAME supplementation and the TNF-α induced decline in eNOS dimer/monomer ratio. Interestingly, eNOS uncoupling has been shown to require upstream activation of NOX [
      • Chalupsky K.
      • Cai H.
      Endothelial dihydrofolate reductase: critical for nitric oxide bioavailability and role in angiotensin II uncoupling of endothelial nitric oxide synthase.
      ]. ROS can oxidatively degrade NO and are able to induce arginase activity [
      • Chandra S.
      • Romero M.J.
      • Shatanawi A.
      • Alkilany A.M.
      • Caldwell R.B.
      • Caldwell R.W.
      Oxidative species increase arginase activity in endothelial cells through the RhoA/Rho kinase pathway.
      ]. Elevated arginase activity on its turn may limit the pool of l-arginine [
      • Li H.
      • Meininger C.J.
      • Hawker Jr., J.R.
      • et al.
      Regulatory role of arginase I and II in nitric oxide, polyamine, and proline syntheses in endothelial cells.
      ] to NOS and thus compromise NO production [
      • Chicoine L.G.
      • Paffett M.L.
      • Young T.L.
      • Nelin L.D.
      Arginase inhibition increases nitric oxide production in bovine pulmonary arterial endothelial cells.
      ] and increase O2• production by NOS uncoupling [
      • Kim J.H.
      • Bugaj L.J.
      • Oh Y.J.
      • et al.
      Arginase inhibition restores NOS coupling and reverses endothelial dysfunction and vascular stiffness in old rats.
      ]. A link between induced arginase activity and endothelial dysfunction has consistently been demonstrated [
      • Kim J.H.
      • Bugaj L.J.
      • Oh Y.J.
      • et al.
      Arginase inhibition restores NOS coupling and reverses endothelial dysfunction and vascular stiffness in old rats.
      ]. We demonstrated that treatment with T0901317 decreased TNF-α-induced NOX 4 and p22phox mRNA expression and NOX activity in HUVEC. In addition, we showed a T0901317-mediated induction in TNF-α-impaired NO production, paralleled by a reduction in TNF-α upregulated O2• production. Furthermore, T0901317 decreased the production of the marker of eNOS uncoupling, nitrotyrosine, which is generated by the conversion of tyrosine into nitrotyrosine by the cytotoxic peroxynitrite, which is produced by NO and O2•. However, analysis of the eNOS dimer to monomer ratio indicated that T0901317 did not convert the TNF-α-decreased eNOS dimer to monomer ratio. This implies that T0901317 could not decline the TNF-α-induced eNOS uncoupling, but leads to a direct reduction in oxidative stress, including a decrease in NOX activity and O2• production, and a direct induction of NO.
      Furthermore, we demonstrated that LXR agonism reduces endothelial AT1R expression. The expression levels of AT1R define the biological efficacy of Ang II, which is an important mediator of endothelial dysfunction [
      • Oak J.H.
      • Cai H.
      Attenuation of angiotensin II signaling recouples eNOS and inhibits nonendothelial NOX activity in diabetic mice.
      ]. In line with the T0901317-mediated reduction in AT1R, we further showed that T0901317 reduced the Ang II responsiveness. In detail, we demonstrated that less MNCs adhered after Ang II stimulation to HUVEC pretreated with TNF-α and T0901317 compared to HUVEC pretreated with TNF-α alone. Therefore, we suggest that LXR agonism besides its direct improvement in TNF-α-induced endothelial dysfunction may further impede the atherogenesis process via the suppression of Ang II-induced (inflammatory) processes.
      In conclusion, LXR agonism improves TNF-α-impaired endothelial function via its anti-apoptotic, anti-inflammatory, anti-oxidative properties and its capacity to restore TNF-α-impaired NO production independent of its cholesterol-modulating effects.

      Funding

      This study was supported by the Berlin-Brandenburg Center for Regenerative Therapies – BCRT (Bundesministerium für Bildung und Forschung – 0313911) to CT and SVL, by DFG funding through the Berlin-Brandenburg School for Regenerative Therapies.

      Conflict of interest statement

      None.

      Acknowledgments

      We thank Thomas Düsterhöft, Ulrike Fritz, Annika Koschel, and Gwendolin Matz (in alphabetical order) for excellent technical assistance.

      References

        • Deanfield J.E.
        • Halcox J.P.
        • Rabelink T.J.
        Endothelial function and dysfunction: testing and clinical relevance.
        Circulation. 2007; 115: 1285-1295
        • Gao X.
        • Xu X.
        • Belmadani S.
        • et al.
        TNF-alpha contributes to endothelial dysfunction by upregulating arginase in ischemia/reperfusion injury.
        Arterioscler Thromb Vasc Biol. 2007; 27: 1269-1275
        • Goodwin B.L.
        • Solomonson L.P.
        • Eichler D.C.
        Argininosuccinate synthase expression is required to maintain nitric oxide production and cell viability in aortic endothelial cells.
        J Biol Chem. 2004; 279: 18353-18360
        • Riad A.
        • Westermann D.
        • Van Linthout S.
        • et al.
        Enhancement of endothelial nitric oxide synthase production reverses vascular dysfunction and inflammation in the hindlimbs of a rat model of diabetes.
        Diabetologia. 2008; 51: 2325-2332
        • Yan G.
        • You B.
        • Chen S.P.
        • Liao J.K.
        • Sun J.
        Tumor necrosis factor-alpha downregulates endothelial nitric oxide synthase mRNA stability via translation elongation factor 1-alpha 1.
        Circ Res. 2008; 103: 591-597
        • Goodwin B.L.
        • Pendleton L.C.
        • Levy M.M.
        • Solomonson L.P.
        • Eichler D.C.
        Tumor necrosis factor-alpha reduces argininosuccinate synthase expression and nitric oxide production in aortic endothelial cells.
        Am J Physiol Heart Circ Physiol. 2007; 293: H1115-H1121
        • Kataoka H.
        • Murakami R.
        • Numaguchi Y.
        • Okumura K.
        • Murohara T.
        Angiotensin II type 1 receptor blockers prevent tumor necrosis factor-alpha-mediated endothelial nitric oxide synthase reduction and superoxide production in human umbilical vein endothelial cells.
        Eur J Pharmacol. 2010; 636: 36-41
        • Gao X.
        • Zhang H.
        • Belmadani S.
        • et al.
        Role of TNF-alpha-induced reactive oxygen species in endothelial dysfunction during reperfusion injury.
        Am J Physiol Heart Circ Physiol. 2008; 295: H2242-H2249
        • Pritchard Jr., K.A.
        • Groszek L.
        • Smalley D.M.
        • et al.
        Native low-density lipoprotein increases endothelial cell nitric oxide synthase generation of superoxide anion.
        Circ Res. 1995; 77: 510-518
        • Katz S.D.
        • Rao R.
        • Berman J.W.
        • et al.
        Pathophysiological correlates of increased serum tumor necrosis factor in patients with congestive heart failure. Relation to nitric oxide-dependent vasodilation in the forearm circulation.
        Circulation. 1994; 90: 12-16
        • Im S.S.
        • Osborne T.F.
        Liver X receptors in atherosclerosis and inflammation.
        Circ Res. 2011; 108: 996-1001
        • Miao B.
        • Zondlo S.
        • Gibbs S.
        • et al.
        Raising HDL cholesterol without inducing hepatic steatosis and hypertriglyceridemia by a selective LXR modulator.
        J Lipid Res. 2004; 45: 1410-1417
        • Van Linthout S.
        • Spillmann F.
        • Lorenz M.
        • et al.
        Vascular-protective effects of high-density lipoprotein include the downregulation of the angiotensin II type 1 receptor.
        Hypertension. 2009; 53: 682-687
        • Van Linthout S.
        • Spillmann F.
        • Graiani G.
        • et al.
        Down-regulation of endothelial TLR4 signalling after apo A-I gene transfer contributes to improved survival in an experimental model of lipopolysaccharide-induced inflammation.
        J Mol Med (Berl). 2011; 89: 151-160
        • Van Linthout S.
        • Spillmann F.
        • Riad A.
        • et al.
        Human apolipoprotein A-I gene transfer reduces the development of experimental diabetic cardiomyopathy.
        Circulation. 2008; 117: 1563-1573
        • Blaschke F.
        • Takata Y.
        • Caglayan E.
        • et al.
        A nuclear receptor corepressor-dependent pathway mediates suppression of cytokine-induced C-reactive protein gene expression by liver X receptor.
        Circ Res. 2006; 99: e88-99
        • Imayama I.
        • Ichiki T.
        • Patton D.
        • et al.
        Liver X receptor activator downregulates angiotensin II type 1 receptor expression through dephosphorylation of Sp1.
        Hypertension. 2008; 51: 1631-1636
        • Lei P.
        • Baysa A.
        • Nebb H.I.
        • et al.
        Activation of Liver X receptors in the heart leads to accumulation of intracellular lipids and attenuation of ischemia-reperfusion injury.
        Basic Res Cardiol. 2013; 108: 323
        • Tontonoz P.
        • Mangelsdorf D.J.
        Liver X receptor signaling pathways in cardiovascular disease.
        Mol Endocrinol. 2003; 17: 985-993
        • Chen J.
        • Zhao L.
        • Sun D.
        • et al.
        Liver X receptor activation attenuates plaque formation and improves vasomotor function of the aortic artery in atherosclerotic ApoE(−/−) mice.
        Inflamm Res. 2012; 61: 1299-1307
        • Stangl V.
        • Lorenz M.
        • Meiners S.
        • et al.
        Long-term up-regulation of eNOS and improvement of endothelial function by inhibition of the ubiquitin-proteasome pathway.
        FASEB J. 2004; 18: 272-279
        • Cardellini M.
        • Menghini R.
        • Martelli E.
        • et al.
        TIMP3 is reduced in atherosclerotic plaques from subjects with type 2 diabetes and increased by SirT1.
        Diabetes. 2009; 58: 2396-2401
        • Hsu J.J.
        • Lu J.
        • Huang M.S.
        • et al.
        T0901317, an LXR agonist, augments PKA-induced vascular cell calcification.
        FEBS Lett. 2009; 583: 1344-1348
        • Griendling K.K.
        • Minieri C.A.
        • Ollerenshaw J.D.
        • Alexander R.W.
        Angiotensin II stimulates NADH and NADPH oxidase activity in cultured vascular smooth muscle cells.
        Circ Res. 1994; 74: 1141-1148
        • Chandra S.
        • Romero M.J.
        • Shatanawi A.
        • Alkilany A.M.
        • Caldwell R.B.
        • Caldwell R.W.
        Oxidative species increase arginase activity in endothelial cells through the RhoA/Rho kinase pathway.
        Br J Pharmacol. 2012; 165: 506-519
        • Joseph S.B.
        • Castrillo A.
        • Laffitte B.A.
        • Mangelsdorf D.J.
        • Tontonoz P.
        Reciprocal regulation of inflammation and lipid metabolism by liver X receptors.
        Nat Med. 2003; 9: 213-219
        • Joseph S.B.
        • McKilligin E.
        • Pei L.
        • et al.
        Synthetic LXR ligand inhibits the development of atherosclerosis in mice.
        Proc Natl Acad Sci U S A. 2002; 99: 7604-7609
        • Li Y.
        • Bolten C.
        • Bhat B.G.
        • et al.
        Induction of human liver X receptor alpha gene expression via an autoregulatory loop mechanism.
        Mol Endocrinol. 2002; 16: 506-514
        • Harrison D.G.
        Cellular and molecular mechanisms of endothelial cell dysfunction.
        J Clin Invest. 1997; 100: 2153-2157
        • Ryoo S.
        • Lemmon C.A.
        • Soucy K.G.
        • et al.
        Oxidized low-density lipoprotein-dependent endothelial arginase II activation contributes to impaired nitric oxide signaling.
        Circ Res. 2006; 99: 951-960
        • Shen L.J.
        • Beloussow K.
        • Shen W.C.
        Accessibility of endothelial and inducible nitric oxide synthase to the intracellular citrulline-arginine regeneration pathway.
        Biochem Pharmacol. 2005; 69: 97-104
        • Marathe C.
        • Bradley M.N.
        • Hong C.
        • et al.
        The arginase II gene is an anti-inflammatory target of liver X receptor in macrophages.
        J Biol Chem. 2006; 281: 32197-32206
        • Valledor A.F.
        • Hsu L.C.
        • Ogawa S.
        • Sawka-Verhelle D.
        • Karin M.
        • Glass C.K.
        Activation of liver X receptors and retinoid X receptors prevents bacterial-induced macrophage apoptosis.
        Proc Natl Acad Sci U S A. 2004; 101: 17813-17818
        • Chalupsky K.
        • Cai H.
        Endothelial dihydrofolate reductase: critical for nitric oxide bioavailability and role in angiotensin II uncoupling of endothelial nitric oxide synthase.
        Proc Natl Acad Sci U S A. 2005; 102: 9056-9061
        • Li H.
        • Meininger C.J.
        • Hawker Jr., J.R.
        • et al.
        Regulatory role of arginase I and II in nitric oxide, polyamine, and proline syntheses in endothelial cells.
        Am J Physiol Endocrinol Metab. 2001; 280: E75-E82
        • Chicoine L.G.
        • Paffett M.L.
        • Young T.L.
        • Nelin L.D.
        Arginase inhibition increases nitric oxide production in bovine pulmonary arterial endothelial cells.
        Am J Physiol Lung Cell Mol Physiol. 2004; 287: L60-L68
        • Kim J.H.
        • Bugaj L.J.
        • Oh Y.J.
        • et al.
        Arginase inhibition restores NOS coupling and reverses endothelial dysfunction and vascular stiffness in old rats.
        J Appl Physiol. 2009; 107: 1249-1257
        • Oak J.H.
        • Cai H.
        Attenuation of angiotensin II signaling recouples eNOS and inhibits nonendothelial NOX activity in diabetic mice.
        Diabetes. 2007; 56: 118-126