Activation of NF-kappaB by palmitate in endothelial cells: a key role for NADPH oxidase-derived superoxide in response to TLR4 activation

Abstract

Objective: We investigated whether NADPH oxidase-dependent production of superoxide contributes to activation of NF-kappaB 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 mumol/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-kappaB signaling as measured by phospho-IkappaBalpha, NF-kappaB 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-kappaB 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 IkappaBalpha (NF-kappaB super repressor) that blocks cellular inflammation downstream of IKKbeta/NF-kappaB. 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-kappaB signaling in endothelial cells.

Figure 1. Time course of palmitate-dependent activation of NF-κB signaling and superoxide prodution

A. IL-6 as determined by ELISA in HMEC treated with increasing concentrations of palmitate-BSA (0–200 μM) for 3 h. B–F. Measures of NF-κB signaling (phospho-IκBα, IκBα, NF-κB, IL-6, ICAM) in HMEC in response to 100 μM palmitate for time periods between 0–6 h. G–H. Superoxide levels as measured by ESR using the spin trap CMH in response to increasing concentrations of palmitate and for time period of 0–6 h using 100 μM of palmitate. I. DHE florescence in response to 3 h of palmitate-BSA (100 μM) or BSA control. * p<0.05 BSA vs palmitate-BSA, for each set of experiments n=(3–4).

Figure 2. Effect of superoxide inhibition on palmitate-dependent activation of endothelial cell NF-κB signaling

HMEC were pretreated with either of two inhibitors of superoxide -- SOD (100, 200, or 300 Units/ml) or DPI 25 μM -- or they were transduced with siRNA to NOX4, NOX5, or with a negative scrambled siRNA (Scb). HMEC were then treated with 100 μM palmitate for 3 h. A,B,C. Phospho-IκBα Western blot representative of 3 independent experiments (C-BSA control, F-palmitate, T-TNF-α). GAPDH protein levels were used as a loading control. D,E,F. IL-6 levels as measured by ELISA. *, p<0.05, n=3. G,H,I. Fold increase in superoxide production in response to palmitate as measured by ESR and spin trap CMH, * p<0.05, (n=3).

Figure 3. Effect of TLR4 signaling on palmitate-mediated superoxide production

A. HMEC were transduced with siRNA to TLR4 or a scrambled siRNA (con) and then treated with palmitate (100 μM) for 3h (FFA). IL-6 was measured (to assess NF-κB signaling) by ELISA and B. superoxide production was measured by ESR and the spin trap CMH. C. HMEC were transduced with retrovirus expressing GFP, DN-MyD88, DN-IRAK, or phosphorylation resistant IκBα (NF-κB super repressor). Following treatment with palmitate-BSA (F) or BSA alone (C), IL-6 and D. superoxide was measured and fold increase from control was calculated. * p<0.05, n=3.

Figure 4. Effect of palmitate or HF-feeding on vascular BMP4 and NOX4 levels

HMEC were treated with palmitate (100 μM) for time period between 0–6 h and 3 h in the presence or absence siRNA to TLR4. A. Western blot using anti-NOX4 antibody. Control (C), BSA alone, (F) palmitate-BSA. B. Same lysates were probed with anti-BMP4 antibody including addition of HMEC treated with NOX4 siRNA. Fold increase was normalized to GAPDH levels determined by Western blot (not shown) C. RNA was harvested from thoracic aortic from C57BL6 or TLR4 −/− mice fed either a low fat (10% saturated fat) or high fat (60% saturated fat) diet for 8 wk and mRNA content was assessed by quantitative PCR. Fold increase over LF group was calculated for BMP4, NOX4, NOX1 and NOX2. * p<0.05 (n=5 for each condition).