Activation of NF-κB by Palmitate in Endothelial Cells
A Key Role for NADPH Oxidase-Derived Superoxide in Response to TLR4 Activation
Objective— We investigated whether NADPH oxidase–dependent production of superoxide contributes to activation of NF-κB in endothelial cells by the saturated free fatty acid palmitate.
Methods and Results— After incubation of human endothelial cells with palmitate at a concentration known to induce cellular inflammation (100 μmol/L), we measured superoxide levels by using electron spin resonance spectroscopy and the spin trap 1-hydroxy-3-methoxycarbonyl-2,2,5,5-tetramethylpyrrolidine (CMH). Palmitate exposure induced a >2-fold increase in superoxide levels, an effect associated with activation of NF-κB signaling as measured by phospho-IκBα, NF-κB activity, IL-6, and ICAM expression. Reduction in superoxide levels by each of 3 different interventions—pretreatment with superoxide dismutase (SOD), diphenylene iodinium (DPI), or knockdown of NADPH oxidase 4 (NOX4) by siRNA—attenuated palmitate-mediated NF-κB signaling. Inhibition of toll like receptor-4 (TLR4) signaling also suppressed palmitate-mediated superoxide production and associated inflammation, whereas palmitate-mediated superoxide production was not affected by overexpression of a phosphorylation mutant IκBα (NF-κB super repressor) that blocks cellular inflammation downstream of IKKβ/NF-κB. Finally, high-fat feeding increased expression of NOX4 and an upstream activator, bone morphogenic protein (BMP4), in thoracic aortic tissue from C57BL/6 mice, but not in TLR4−/− mice, compared to low-fat fed controls.
Conclusions— These results suggest that NADPH oxidase–dependent superoxide production links palmitate-stimulated TLR4 activation to NF-κB signaling in endothelial cells.
Saturated FFAs such as palmitate readily induce endothelial inflammation, including increased IKKβ-NF-κB signaling, via a mechanism that involves activation of Toll-like receptors (TLR) that are key components of the innate immune system. Among the consequences of TLR4-induced activation of NF-κB is impaired vascular insulin signaling and reduced nitric oxide production.1 Based on these and other observations, elevated circulating concentrations of saturated free fatty acids (FFA) are implicated in the mechanism underlying obesity-associated inflammation and insulin resistance in endothelial cells, but the mechanism underlying this link has yet to be established.
One potential mechanism whereby exposure to saturated FFA induces cellular inflammation is via reactive oxygen species (ROS) such as superoxide (O2·−)2 that can be generated by both mitochondrial electron transport and by cytosolic enzymes such as the NOX family of nicotinamide adenine dinucleotide phosphate (NADPH) oxidases. These enzymes transfer electrons from NADPH across cell membranes and are a major source of cytoplasmic ROS. The electron acceptor for this reaction is oxygen, producing superoxide radicals. Of 7 NOX homologues that have been identified in nonphagocytic cells, NOX4 is the major species expressed in endothelial cells, with NOX1, NOX2, and NOX5 being expressed at much lower levels. In vascular tissues of db/db mice, a genetic model of severe obesity and diabetes attributable to a mutation in the leptin receptor, expression of NOX1, NOX4, and p22phox (a smaller subunit which forms a stable heterodimer with NOX) are increased, as is NADPH oxidase–dependent superoxide generation, and these responses are implicated in the observed increase of inflammatory gene expression in the vasculature of these animals.3 Complementary studies using endothelial cell culture models report increased cellular NADPH oxidase levels after exposure to high levels of glucose, FFA, or insulin and are associated with increased superoxide production.4 These studies suggest that NADPH-mediated superoxide production has a central role in vascular inflammatory responses during nutritional excess.
Consistent with this hypothesis is recent evidence that NOX4 mediates lipopolysaccharide (LPS)-induced inflammatory responses in human endothelial cells.5 Because both LPS and palmitate activate IKKβ/NF-κB through TLR4 signaling, we sought to determine whether palmitate, like LPS, induces cellular inflammation via a mechanism involving NADPH oxidase–derived superoxide. In the current studies, we demonstrate that palmitate-activated TLR4 signaling results in NADPH oxidase–dependent superoxide production in endothelial cells, and that this sequence of events is required for the inflammatory response to palmitate in human endothelial cells. In addition, we report that high-fat (HF) feeding increases levels of both NOX4 mRNA and proinflammatory cytokines in vascular tissue from wild-type (WT) C57BL/6 mice compared to low-fat–fed controls and that these effects are absent in mice lacking TLR4. Collectively, these findings suggest that TRL4-dependent activation of NADPH oxidase is necessary for the proinflammatory effects of palmitate in vascular tissue, and that NADPH oxidase–generated superoxide plays a key role to couple TLR4 activation to cellular signaling via NF-κB.
Anti–phospho-IκBα, anti-IκBα, anti-TLR4 antibodies were obtained from Cell Signaling; anti-BMP4, anti-NOX4 antibodies were obtained from Abcam Inc; anti-GAPDH and anti-NOX2 antibodies from Santa Cruz Biotechnology; anti-ICAM antibody, Human IL-6 ELISA kits from R&D Systems; and TransAM NF-kB p65 kit, Active Motif, Carlsbad, Calif. The spin trap 1-hydroxy-3-methoxycarbonyl-2,2,5,5-tetramethylpyrrolidine (CMH) was purchased from Alexis Biochemical.
Superoxide dismutase (PEG-SOD), Diphenylene idonium (DPI), and LPS were obtained from Sigma-Aldrich. Dihydroethidium (DHE) was purchased from Molecular Probes/Invitrogen.
Palmitic (C 16:0) fatty acids were obtained from Alltech Associates Inc, and BSA (FFA-free) was purchased from Roche. FFA were dissolved in 0.1 mol/L NaOH at 70°C and then complexed with 10% BSA at 55°C for 10 minutes to achieve a final palmitate concentration of 100 μmol/L as described previously.6 Stock solutions of 5 mmol/L FFA with 10% BSA and 10%BSA control solutions were prepared 1 day before experiments. Palmitate preparations were assessed for LPS contamination using Amebocyte Lysate Test (Biowhittaker).
Human microvascular endothelial cells (HMECs) were purchased from Invitrogen-Cascade Biological and were cultured in RPMI 1640 supplemented with 10% fetal bovine serum (Hyclone Laboratories) and 12 μg/mL of bovine brain extract (Clonetics), L-glutamine (2 mmol/L), sodium pyruvate (1 mmol/L), and nonessential amino acids in the presence of penicillin (100 U/mL) and maintained at 37°C in 5% CO2. All Western bots were performed as described,7 using equal amounts of total protein for each condition and experiment. SDS gel electrophoresis was performed using a 4% by 20% gradient gel. Generation of stably-transfected HMECs with a dominant-negative myeloid differentiation factor-88 (DN-MyD88), interleukin-1 receptor-associated kinase (DN-IRAK), and phosphorylation-resistant mutant of IκBα that blocks NF-κB activation (NF-κB super repressor) was as previously described.8–10 The Silencer SiRNA Transfection Kit II (Ambion) was used to reduce expression of genes encoding TLR4, NOX4, NOX5. Silencer Negative Control siRNA was used as the scrambled control.
Total RNA Extraction, Quantitative RT-PCR
Thoracic aortic tissues from experiments described previously were used.1 Briefly both C57BL6 and TLR4−/− mice were fed either a low-fat or HF diet for 8 weeks, after which thoracic aorta tissue were collected. Thoracic aortic tissue total RNA was extracted using RNeasy Mini Kit (Qiagen) and Mouse NOX4, NOX2, NOX1, and BMP4 primer pairs were purchased from Applied Biosystems.
Endothelial superoxide radical was measured at room temperature by each of 2 different methods. The first of these used electron spin resonance spectroscopy (ESR) using the spin trap (CMH).11–13 HMECs were processed by washing once with ice-cold PBS and removed by scraping. After centrifugation the cells were resuspended in Krebs-HEPES buffer, and 0.1 mmol/L diethylenetriamine-penta-acetic acid (DTPA) was used to inhibit iron-catalyzed oxidation of the CMH trap. ESR studies were performed on a table-top x-band spectrometer Miniscope (Magnettech, Berlin, Germany). Recordings were made at room temperature using a small capillary tube. Instrument settings were biofield 3350, Sweep 60G, Microwave frequency 9.78 Ghz, microwave power 20 mW, and kinetic time of 10 minutes.
Superoxide measurement with dihydroethidium (DHE) was used to provide a second, complementary approach to measure endothelial superoxide production. HMECs were grown on coverslips coated with 2% gelatin grown to 80% confluence and treated with palmitate (100 μmol/L) for 2.5 hours in phenol-red-free EBM. After the cells were incubated with DHE (2 μmol/L) for 30 minutes, the following procedures were carried out in the dark: cells were washed twice with ice-cold PBS, fixed on ice with 2% paraformaldehyde (Electron Microscopy Sciences) for 3 minutes, and again washed twice with cold PBS. Cover slips were mounted onto slides with Gel Mount (Sigma). Superoxide production was measured using an Olympus fluorescence microscope using a TRITC filter (excitation 528 to 553 nmol/L, emission 600 to 660 nmol/L). Fluorescence intensity of these images was quantified using Image J software (NIH). The mean fluorescent intensity of approximately 80 to 100 cells was measured and expressed as fluorescence arbitrary units.
In all experiments, densitometry/ELISA measurements were normalized to controls incubated with vehicle and percent change relative to the control condition was calculated. Analysis of the results was performed using the STATA8 statistical package. Data are expressed as mean±SEM, and values of P<0.05 were considered statistically significant. A 2-tailed t test was used to compare mean values from studies involving 2 experimental groups. To compare responses after treatment with BSA or BSA-palmitate across time points, during the use of different inhibitors of superoxide or changes in TLR4 signaling, data were analyzed by 2-way analysis of variance using the Bonferoni-posthoc comparison test.
Effect of Palmitate on Superoxide Production and NF-κB Signaling in Endothelial Cells
Lipopolysaccharide is a known activator of NF-κB in most cell types, and previous investigators have demonstrated that LPS treatment increases superoxide generation as measured by the dye, 2,7 dichlorofluorescein-diacetate (DCF-DA).5 To validate the ESR method of superoxide detection, HMECs were treated with LPS (5 ng/mL) for 1 hour, followed by measurement of superoxide using the spin trap CMH. Consistent with previous results,5 LPS increased superoxide production by 2.2±0.4-fold (P<0.05) compared to BSA-treated HMECs.
To extend our previous observation that incubation of endothelial cells with palmitate-BSA for 3 hours increases phospho-IκBα protein levels (a marker of IKKβ-NF-κB signaling),14 we examined the effect of increasing palmitate-BSA concentrations (10 to 200 μmol/L) on production of both superoxide and the NF-κB–dependent cytokine, IL-6. Palmitate-BSA increased both IL-6 and superoxide levels dose-dependently, with concentrations of 100 μmol/L inducing a maximal >4-fold increase in IL-6 expression (Figure 1A) and a 2-fold increase in superoxide production (Figure 1G). Analysis of the time course of the effect of palmitate (100 μmol/L) on both cellular markers of NF-κB activation (IκBα phosphorylation, IκBα protein levels, NF-κB activity, IL-6 production, ICAM protein) and superoxide (O2·−) production in HMECs showed that the latter is significantly elevated before IκBα is phosphorylated (Figure 1B and 1H), consistent with a model in which superoxide mediates IKKβ activation. Both IL-6 and ICAM protein levels also increased in response to palmitate, with increased IL-6 expression preceding the change of ICAM protein (Figure 1E and 1F). The effect of palmitate to increase superoxide production was confirmed using an independent method based on the fluorescent dye DHE (Figure 1I).
Mechanisms Linking Palmitate-Induced Superoxide Production to Activation of Endothelial NF-κB Signaling
To determine whether superoxide is necessary for the effect of palmitate to increase phospho-IκBα and IL-6, HMECs were pretreated with 2 different inhibitors of superoxide production, superoxide dismutase (SOD) and diphenyleneiodonium (DPI) which inhibits flavin-containing oxidases such as NADPH oxidase. At doses between 100 to 300 U, SOD blocked the effect of palmitate to increase levels of both phospho-IκBα (Figure 2A) and IL-6 (Figure 2D). As expected, palmitate-induced superoxide production in HMECs was also inhibited by SOD (100 U) pretreatment (Figure 2G). Similarly, pretreatment with DPI (25 μmol/L) for 1 hour blocked the stimulatory effect of palmitate on both phospho-IκBα and IL-6 production (Figure 2B and 2E), while also preventing palmitate-mediated superoxide production as measured by ESR and the spin trap CMH (Figure 2H). These results collectively suggest that superoxide generation is required for palmitate-induced activation of IKKβ-NF-κB in endothelial cells.
Because NOX4 is the most abundant isoform of NADPH oxidase in vascular tissue, we hypothesized that palmitate-mediated activation of NF-κB and superoxide production are both NOX4-dependent. To test this hypothesis, we selectively reduced expression of either NOX4 or NOX5 (as a control, since NOX5 is not highly expressed in endothelial cells) in HMECs using siRNA. As expected, treatment with NOX4 siRNA, but not NOX5 siRNA, reduced NOX4 protein levels (Figure 2C). Moreover, palmitate-mediated increases of phospho-IκBα, IL-6, and superoxide production in HMECs were reduced after NOX4 siRNA treatment, whereas neither siRNA targeting NOX5 nor scrambled control siRNA had measurable effects (Figure 2C, 2F, and 2H). Reduction of endothelial NOX2 protein by siRNA had no significant effect on palmitate-mediated increases in phospho-IκBα protein or on palmitate-mediated superoxide production as measured by ESR and the spin trap CMH (see supplementary Figure I, available online at http://atvb.ahajournals.org). These results demonstrate that endothelial NOX4 is necessary for palmitate-mediated superoxide generation and subsequent activation of inflammatory signaling in HMECs.
Role of TLR4/MyD88/IRAK Signaling in Palmitate-Mediated Superoxide Production
We have previously shown that in endothelial cells, palmitate-mediated activation of IKKβ is dependent on TLR4.1 To determine whether TLR4 also lies upstream of superoxide generation in response to palmitate, we determined whether the latter response is blocked by inhibition of TLR4 signaling. This was accomplished using siRNA to reduce TLR4 expression or with dominant negative constructs of MyD88 (DN-MyD88) and IRAK (DN-IRAK), previously shown to inhibit palmitate-dependent increases in NF-κB signaling.1 We found that reduction of TLR4 protein expression using siRNA prevented palmitate-mediated increases of both IL-6 production and superoxide production (Figure 3A and 3B). Similar results were obtained when HMECs were transfected with viral constructs expressing (DN-IRAK) or (DN-MyD88), which inhibit signal transduction downstream of TLR4 (Figure 3C and 3D), whereas transducing HMECs with NF-κB super-repressor to inhibit NF-κB activation1 (which inhibits palmitate-mediated increases of IL-6) had no effect on the increase of superoxide production induced by palmitate. Taken together, these findings suggest that in response to palmitate, superoxide generation occurs downstream of TLR4/MyD88/IRAK signaling but upstream of NF-κB signaling.
Role of TLR4 Signaling in the Induction of BMP4 and NOX4 by Palmitate or HF Feeding
Bone morphogenetic protein 4 (BMP4) is reported to induce and activate NADPH oxidase in endothelial cells, and administration of BMP4 to mice induces hypertension and endothelial dysfunction via a mechanism that is NADPH oxidase–dependent.15 We therefore investigated in HMECs (1) whether incubation in palmitate increases both BMP4 and NOX4 protein levels, and (2) whether this effect is dependent on TLR4. Both NOX4 and BMP4 expression increased within 1 hour of treatment with 100 μmol/L palmitate, and inhibition of TLR4 signaling by siRNA attenuated both responses (Figure 4A and 4B). By comparison, reduction of NOX4 protein levels by NOX4 siRNA had no effect on the palmitate-induced increase of BMP4 protein levels (Figure 4B), suggesting that TLR4 is upstream of both BMP4 and NOX4.
We next asked whether obesity induced by a HF diet in C57BL6 mice is associated with increased vascular expression of either BMP4 or NOX4 and whether this increase is dependent on TLR4 signaling. We examined stored thoracic mRNA samples from a previously described study1 in which C57BL6 (WT) and TLR4−/− mice were fed either a HF or a LF diet for 8 weeks. In response to HF feeding, both WT and TLR4−/− mice displayed similar increases of body weight, body fat content, and serum insulin and free fatty acids (FFA) levels compared to mice fed a LF diet, and levels of phospho-IκBα, ICAM, and IL-6 (markers of vascular NF-κB signaling) were each increased in thoracic aorta extracts from WT mice but not from mice lacking TLR4 on the same diet.1 In WT mice fed a HF diet, levels of both BMP4 and NOX4 mRNA were also increased in thoracic aorta (as measured by quantitative PCR) compared to low-fat–fed WT mice. By comparison, HF feeding did not increase NOX4 or BMP4 expression in TLR4−/− mice, suggesting that TLR4 is necessary for induction of NOX4 in vascular tissue in this setting (Figure 4C). In contrast, HF feeding was not associated with increased NOX1 or NOX2 expression in either WT or TLR4−/− mice (Figure 4C), suggesting that the effect of HF feeding on the vasculature is specific to NOX4 expression in comparison to other NOX isoforms. HF feeding was also associated with increased aortic NOX4 protein expression compared to the LF-fed group (see supplemental Figure II).
In both cultured endothelial cells and vascular tissue in vivo, nutritional excess rapidly induces cellular inflammation at the level of NF-κB signaling. Growing evidence implicates vascular inflammation as a key mechanism linking cardiovascular disease to obesity and related metabolic disorders,16–18 and we have recently shown that during HF feeding, inflammation and insulin resistance develop in the vasculature of mice well before these changes are observed in liver, skeletal muscle, or fat.19 The steady increase in prevalence of obesity and its metabolic sequelae20 heightens the need for an improved understanding of how nutrient excess affects the vasculature. In light of evidence of a key role for TLR4 signaling in the stimulation of inflammatory signaling in palmitate-treated endothelial cells,1 the present studies were undertaken to delineate the pathways linking TLR4 activation to the activation of IKKβ/NF-κB signaling in vascular tissue. We demonstrate in cultured human endothelial cells that superoxide production is required for palmitate to induce NF-κB signaling, and further that palmitate increases superoxide production by activating NADPH oxidase through the TLR4/MyD88/IRAK pathway. Finally, in a mouse model of DIO and insulin resistance, expression of both BMP4 (which induces NADPH oxidase) and NOX4 (the predominant NADPH oxidase isoform in endothelial cells) increases in response to 8 weeks of HF feeding in thoracic aorta of C57BL6 mice, but not TLR4−/− mice, despite no differences in weight or adiposity. These results collectively suggest that the mechanism whereby palmitate-induced TLR4 signaling activates NF-κB in endothelial cells is dependent on NADPH oxidase–generated superoxide.
Palmitic, oleic, and linoleic acids constitute up to 70% of circulating FFA in humans, with each typically being present in concentrations between 10 to 50 μmol/L.21,22 Although the current cell culture studies relied on concentrations of palmitic acid (100 μmol/L) higher than those usually found in humans, lower concentrations of palmitic/BSA (50 μmol/L) also increase endothelial levels of IL-6. Thus, the responses investigated in this article can be elicited by palmitate at levels present in the circulation of normal humans, which reinforces the clinical relevance of these studies. Because long-term exposure to higher concentrations of palmitate (eg, 100 to 300 μmol/L for 24 hour) can trigger apoptosis in human endothelial cells,23 we chose conditions that do not induce this effect.
Superoxide is an important participant in redox cell signaling and, when present in excess, is implicated in the development of vascular diseases such as hypertension and atherosclerosis. Superoxide, along with other ROS, is implicated in the activation of redox-sensitive transcription factors such as NF-κB,24 and also regulates inflammatory responses such as leukocyte adhesion25,26 and induction of monocyte chemoattractant protein (MCP-1).27 Because both pharmacological and genetic interventions that reduce ROS also attenuate TNF-α–mediated activation of endothelial inflammatory signaling, these mediators appear to contribute to vascular inflammation in at least some conditions.
Interest in the role of ROS in cellular inflammation has grown with accumulating evidence that this mechanism may contribute to the effect of nutrient excess to cause insulin resistance in muscle, liver, fat and vascular tissue.1,28 In a recent study, San Martin et al reported that NOX1, NOX4, p22phox mRNA levels and vascular superoxide production (measured by dihydroethidium fluorescence) are each increased in the aorta of db/db mice, a genetic model of obesity and diabetes. These findings extend our previous work in a murine DIO model showing that vascular inflammatory markers such as ICAM, VCAM, SOCS3, and IL-6 also increase in response to HF feeding.1,19 Combined with the present findings of increased vascular expression of NOX4 and BMP4 in the aorta of DIO mice, these results support the hypothesis of a cause-and-effect relationship between ROS signaling and vascular inflammation at the level of NF-κB signaling. As TLR4−/− mice fed a HF diet are both protected from the development of vascular inflammation and fail to induce vascular NOX4/BMP4, we propose that HF feeding triggers vascular inflammation via a sequence of events that includes TLR4 activation and superoxide generation.
A potential link between LPS signaling, which like palmitate activates TLR4, and NADPH oxidase has been described in cultured phagocytic cells, in which LPS treatment leads to an NADPH-dependent oxidative burst.29 LPS also increases NADPH-dependent superoxide production, as measured by DCF fluorescence, in human endothelial5 and immune cells.30 Our current work extends these findings by demonstrating that TLR4 signaling is necessary for palmitate-mediated superoxide production in endothelial cells. Although Park and et al5 demonstrated a direct interaction between TLR4 receptor and NADPH oxidase in endothelial cells, recent studies in immune cells suggest that IRAK4 may mediate the effect of TLR4 to activate NADPH oxidase.15 Thus, although growing evidence indicates that TLR4 activation induces NADPH oxidase in both endothelial and immune cells, the mechanisms underlying this effect remain incompletely defined and may differ between cell types.
Whereas LPS treatment leads to a rapid (<1 hour) activation of TLR4/MyD88/IRAK signaling, our studies show that palmitate-dependent activation of IκBα signaling develops more slowly (3 hours), suggesting that inflammation induced by LPS and palmitate may involve distinct mechanisms. One potential explanation for the slower onset of palmitate-mediated effects is that unlike LPS, palmitate does not bind the TLR4 receptor, but rather induces changes in the composition of lipid rafts. Lipid rafts are dynamic assemblies of sphingolipids and cholesterol that are implicated as key components of cell surface signaling through their effect to physically concentrate receptors, downstream kinases, and adaptor proteins involved in signaling pathways. In macrophages and other immune cells, assembly of TLR4 and its accessory proteins (CD14, MD2) in lipid rafts is essential for LPS-mediated activation of NF-κB signaling.31,32 Whether palmitate affects lipid raft formation or composition in endothelial cells, and in this way induces vascular inflammation, remains to be investigated.
The source of vascular superoxide generation during diabetes also remains incompletely understood. Several studies suggest that hyperglycemia increases ROS production via stimulation of mitochondrial respiration.33 Because, by comparison, specific inhibition of NADPH oxidase attenuated palmitate-induced superoxide generation in HMECs, our findings suggest a critical role for NADPH oxidase in this response. The present studies, however, do not exclude the mitochondria as a source of superoxide production in the presence of excess palmitate.
In summary, we conclude that endothelial inflammation induced by the saturated FFA palmitate is dependent on NADPH oxidase–generated superoxide and that the TLR4 signaling pathway connects palmitate to NADPH oxidase activation. Further in vivo studies are necessary to determine whether vascular dysfunction in DIO and other nutrient excess states is similarly dependent on endothelial NADPH oxidase activity.
Sources of Funding
This study was supported by NIH grants DK073878 (to F.K.), U54 CA116847 (to F.K., M.W.S., I.S., and D.M.H.), and DK52989 and DK68384 (to M.W.S.).
Received November 13, 2008; revision accepted June 4, 2009.
Kim F, Pham M, Luttrell I, Bannerman DD, Tupper J, Thaler J, Hawn TR, Raines EW, Schwartz MW. Toll-like receptor-4 mediates vascular inflammation and insulin resistance in diet-induced obesity. Circ Res. 2007; 100: 1589–1596.
Bedard K, Krause KH. The NOX family of ROS-generating NADPH oxidases: physiology and pathophysiology. Physiol Rev. 2007; 87: 245–313.
San Martin A, Du P, Dikalova A, Lassegue B, Aleman M, Gongora MC, Brown K, Joseph G, Harrison DG, Taylor WR, Jo H, Griendling KK. Reactive oxygen species-selective regulation of aortic inflammatory gene expression in Type 2 diabetes. Am J Physiol Heart Circ Physiol. 2007; 292: H2073–H2082.
Inoguchi T, Li P, Umeda F, Yu HY, Kakimoto M, Imamura M, Aoki T, Etoh T, Hashimoto T, Naruse M, Sano H, Utsumi H, Nawata H. 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; 49: 1939–1945.
Park HS, Chun JN, Jung HY, Choi C, Bae YS. Role of NADPH oxidase 4 in lipopolysaccharide-induced proinflammatory responses by human aortic endothelial cells. Cardiovasc Res. 2006; 72: 447–455.
Gallis B, Corthals GL, Goodlett DR, Ueba H, Kim F, Presnell SR, Figeys D, Harrison DG, Berk BC, Aebersold R, Corson MA. Identification of flow-dependent endothelial nitric-oxide synthase phosphorylation sites by mass spectrometry and regulation of phosphorylation and nitric oxide production by the phosphatidylinositol 3-kinase inhibitor LY294002. J Biol Chem. 1999; 274: 30101–30108.
Bannerman DD, Tupper JC, Erwert RD, Winn RK, Harlan JM. Divergence of bacterial lipopolysaccharide pro-apoptotic signaling downstream of IRAK-1. J Biol Chem. 2002; 277: 8048–8053.
Ferri N, Garton KJ, Raines EW. An NF-kappaB-dependent transcriptional program is required for collagen remodeling by human smooth muscle cells. J Biol Chem. 2003; 278: 19757–19764.
Landmesser U, Cai H, Dikalov S, McCann L, Hwang J, Jo H, Holland SM, Harrison DG. Role of p47(phox) in vascular oxidative stress and hypertension caused by angiotensin II. Hypertension. 2002; 40: 511–515.
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.
Kim F, Tysseling KA, Rice J, Pham M, Haji L, Gallis BM, Baas AS, Paramsothy P, Giachelli CM, Corson MA, Raines EW. Free fatty acid impairment of nitric oxide production in endothelial cells is mediated by IKKbeta. Arterioscler Thromb Vasc Biol. 2005; 25: 989–994.
Miriyala S, Gongora Nieto MC, Mingone C, Smith D, Dikalov S, Harrison DG, Jo H. Bone morphogenic protein-4 induces hypertension in mice: role of noggin, vascular NADPH oxidases, and impaired vasorelaxation. Circulation. 2006; 113: 2818–2825.
Kim F, Pham M, Maloney E, Rizzo NO, Morton GJ, Wisse BE, Kirk EA, Chait A, Schwartz MW. Vascular inflammation, insulin resistance, and reduced nitric oxide production precede the onset of peripheral insulin resistance. Arterioscler Thromb Vasc Biol. 2008; 28: 1982–1988.
Knopp RH, Retzlaff B, Walden C, Fish B, Buck B, McCann B. One-year effects of increasingly fat-restricted, carbohydrate-enriched diets on lipoprotein levels in free-living subjects. Proc Soc Exp Biol Med. 2000; 225: 191–199.
Artwohl M, Roden M, Waldhausl W, Freudenthaler A, Baumgartner-Parzer SM. Free fatty acids trigger apoptosis and inhibit cell cycle progression in human vascular endothelial cells. Faseb J. 2004; 18: 146–148.
Lum H, Roebuck KA. Oxidant stress and endothelial cell dysfunction. Am J Physiol Cell Physiol. 2001; 280: C719–C741.
Chen XL, Zhang Q, Zhao R, Ding X, Tummala PE, Medford RM. Rac1 and superoxide are required for the expression of cell adhesion molecules induced by tumor necrosis factor-alpha in endothelial cells. J Pharmacol Exp Ther. 2003; 305: 573–580.
Chen XL, Zhang Q, Zhao R, Medford RM. Superoxide, H2O2, and iron are required for TNF-alpha-induced MCP-1 gene expression in endothelial cells: role of Rac1 and NADPH oxidase. Am J Physiol Heart Circ Physiol. 2004; 286: H1001–H1007.
Ryan KA, Smith MF Jr, Sanders MK, Ernst PB. Reactive oxygen and nitrogen species differentially regulate Toll-like receptor 4-mediated activation of NF-kappa B and interleukin-8 expression. Infect Immun. 2004; 72: 2123–2130.
Yvan-Charvet L, Welch C, Pagler TA, Ranalletta M, Lamkanfi M, Han S, Ishibashi M, Li R, Wang N, Tall AR. Increased inflammatory gene expression in ABC transporter-deficient macrophages: free cholesterol accumulation, increased signaling via toll-like receptors, and neutrophil infiltration of atherosclerotic lesions. Circulation. 2008; 118: 1837–1847.