Pulsatile versus oscillatory shear stress regulates NADPH oxidase subunit expression: implication for native LDL oxidation

Abstract

Shear stress regulates endothelial nitric oxide and superoxide (O2-*) production, implicating the role of NADPH oxidase activity. It is unknown whether shear stress regulates the sources of reactive species production, consequent low-density lipoprotein (LDL) modification, and initiation of inflammatory events. Bovine aortic endothelial cells (BAECs) in the presence of 50 microg/mL of native LDL were exposed to (1) pulsatile flow with a mean shear stress (tau(ave)) of 25 dyne/cm2 and (2) oscillating flow at tau(ave) of 0. After 4 hours, aliquots of culture medium were collected for high-performance liquid chromatography analyses of electronegative LDL species, described as LDL- and LDL2-. In response to oscillatory shear stress, gp91phox mRNA expression was upregulated by 2.9+/-0.3-fold, and its homologue, Nox4, by 3.9+/-0.9-fold (P<0.05, n=4), with a corresponding increase in O2-* production rate. The proportion of LDL- and LDL2- relative to static conditions increased by 67+/-17% and 30+/-7%, respectively, with the concomitant upregulation of monocyte chemoattractant protein-1 expression and increase in monocyte/BAEC binding (P<0.05, n=5). In contrast, pulsatile flow downregulated both gp91phox and Nox4 mRNA expression (by 1.8+/-0.2-fold and 3.0+/-0.12-fold, respectively), with an accompanying reduction in O2-* production, reduction in the extent of LDL modification (51+/-12% for LDL- and 30+/-7% for LDL2-), and monocyte/BAEC binding. The flow-dependent LDL oxidation is determined in part by the NADPH oxidase activity. The formation of modified LDL via O2-* production may also affect the regulation of monocyte chemoattractant protein-1 expression and monocyte/BAEC binding.

Figure 1

A, Fluorescent signals versus PCR cycles of gp91^phox versus GAPDH. B, Relative NADPH subunit gp91^phox mRNA expression normalized to GAPDH in response to pulsatile flow (PF) and oscillatory flow (OF) at 4 and 8 hours. C, Western blots of gp91^phox at 4 and 8 hours.

Figure 2

A, Fluorescence signal versus cycle number for Nox4 and gp91^phox and GAPDH. B, Bar graphs show relative mRNA expression for gp91^phox and Nox4 in response to pulsatile versus oscillatory flow conditions. Values are expressed as mean±SEM. #P<0.05 gp91^phox vs control; *P<0.05 Nox4 vs control.

Figure 3

Extracellular superoxide–measured production in BAECs by cytochrome c reduction assay. Rate of superoxide production remained unchanged under the static state (control). However, the rates in response to oscillatory versus pulsatile shear stress diverged starting at 2 hours of exposure.

Figure 4

Real-time intracellular superoxide production in response to flow measured by DHE. A, At time zero, column A illustrates BAECs under 3 conditions: pulsatile flow (PF), oscillatory flow (OF), and static state (control). B, Real-time merged images of phase and fluorescence at 4 hours demonstrate the localization of red fluorescence in the nuclei. In the presence of superoxide, DHE was converted to ethidium, which intercalated into the double-stranded DNA in the nuclei. C, Real-time fluorescent microscopy underscores the superoxide production as red fluorescence.

Figure 5

Flow regulation of native LDL oxidation. Pulsatile flow significantly reduced the ratios of oxidatively modified forms of LDL relative to static conditions by 51±12% for LDL⁻ and 30±7% for LDL²⁻, whereas oscillating flow increased LDL oxidation by 67±17% and 30±7%, respectively (*P<0.05, n=5).

Figure 6

Monocytes establishing solid adhesion to ECs were captured in response to pulsatile flow and oscillatory flow. A, Numbers of monocytes bound per high-power field in response to control (static condition), oscillatory, and pulsatile flow at 4 hours. B, Relative changes in MCP-1 mRNA expression.

Figure 7

Flow regulation of eNOS mRNA expression normalized with GAPDH in response to control, pulsatile flow, and oscillatory flow at 4 and 8 hours.

Figure 8

Rates of superoxide production in the presence of 2-DOG between static and oscillatory flow (OF) conditions. A, BAECs were incubated with 10 mmol/L of 2-DOG for 12 hours before SOD inhibited ferric cytochrome c reduction assay. The presence of 2-DOG significantly reduced the rates of superoxide formation between the control and 2-DOG samples, OF, and OF+2-DOG samples. The negative rates of superoxide production were attributable to the use of SOD. B, LDL oxidation in response to OF. The presence of 2-DOG significantly attenuated the extent of LDL oxidation, suggesting the role of NADPH as an important cofactor for superoxide production and the subsequent LDL oxidation.