Role of NOX2 in Ceramide‐Induced Human Microvascular Endothelial Dysfunction

Abstract

Endothelial dysfunction, or the loss of nitric oxide (NO)‐mediated dilation to shear stress (flow‐induced dilation; FID), is observed in human adipose arterioles from patients with coronary artery disease (CAD). FID is maintained during CAD by release of the mitochondrial‐derived, pro‐inflammatory dilator, H2O2. Ceramide, a sphingolipid that when elevated in plasma is an independent risk factor for future cardiac events, also induces this transition in human arterioles (10µM, 16‐20hrs). The initial events leading to ceramide‐induced endothelial dysfunction are unknown. Here we hypothesize that activation of NADPH oxidase 2 (NOX2), an endothelial membrane enzyme activated by acute stress, occurs prior to the formation of mitochondrial H2O2 in arterioles exposed to chronic ceramide. Human arterioles (100‐250µm in diameter) were dissected from discarded surgical adipose tissues and prepared for videomicroscopy. Following preconstriction with endothelin‐1, internal diameters were measured in response to increased flow. To determine the minimum exposure time necessary to convert to H2O2‐dependent FID in arterioles exposed to ceramide, microvessels from healthy nonCAD adults were incubated with C2 ceramide for 30 min, 2 hrs, and 4 hrs. FID remained mediated by NO at 30min and 2hrs, however, following a 4hr incubation, FID was significantly impaired in the presence of PEG‐catalase (44.6% of maximal dilator capacity ±13.8 (SEM), n=8, p<0.05, one‐way ANOVA*) compared to vehicle control (83.5%±5.0, n=8). The source of H2O2 generated during FID in arterioles treated for 4hr with ceramide was not mitochondrial as increases in fluorescent intensity of mito peroxy yellow 1 (mitoPY1) were not observed during maximal flow. To test whether NOX2 is the initial source of H2O2 during FID in arterioles exposed to 4hr ceramide, nonCAD arterioles were first treated with the NOX2 inhibitor GSK2795039 (NOX2i, 10‐6 M) or underwent intraluminal administration of siRNA to decrease NOX2 expression prior to treatment with ceramide. FID was reduced in vessels treated with both ceramide and the NOX2 inhibitor in the presence of cPTIO (NO scavenger) (50.2%±30.1, n=4* vs. NOX2 inhibitor and ceramide alone 80.8%±6.4, n=7). L‐NAME also impaired FID in vessels with reduced expression of NOX2 and subsequently treated with ceramide (10.8%±38.9, n=3* compared to siNOX2 and ceramide alone 85.5%±5.8, n=3). This suggests that activation of NOX2 may be the initial event in the conversion of NO‐ to H2O2‐dependent FID due to ceramide and may serve as a potential therapeutic target in those at risk for developing cardiovascular disease due to increased plasma ceramide.