Shear-Induced Increase in Hydraulic Conductivity in Endothelial Cells Is Mediated by a Nitric Oxide–Dependent Mechanism

Abstract

This study addresses the role of nitric oxide (NO) and its downstream mechanism in mediating the shear-induced increase in hydraulic conductivity (L(p)) of bovine aortic endothelial cell monolayers grown on porous polycarbonate filters. Direct exposure of endothelial monolayers to 20-dyne/cm(2) shear stress induced a 4. 70+/-0.20-fold increase in L(p) at the end of 3 hours. Shear stress (20 dyne/cm(2)) also elicited a multiphasic NO production pattern in which a rapid initial production was followed by a less rapid, sustained production. In the absence of shear stress, an exogenous NO donor, S-nitroso-N-acetylpenicillamine, increased endothelial L(p) 2.23+/-0.14-fold (100 micromol/L) and 4.8+/-0.66-fold (500 micromol/L) at the end of 3 hours. In separate experiments, bovine aortic endothelial cells exposed to NO synthase inhibitors, N(G)-monomethyl-L-arginine and N(G)-nitro-L-arginine methyl ester, exhibited significant attenuation of shear-induced increase in L(p) in a dose-dependent manner. Inhibition of guanylate cyclase (GC) with LY-83,583 (1 micromol/L) or protein kinase G (PKG) with KT5823 (1 micromol/L) failed to attenuate the shear-induced increase in L(p). Furthermore, direct addition of a stable cGMP analogue, 8-bromo-cGMP, had no effect in altering baseline L(p), indicating that the GC/cGMP/PKG pathway is not involved in shear stress-NO-L(p) response. Incubation with iodoacetate (IAA), a putative inhibitor of glycolysis, dose-dependently increased L(p). Addition of IAA at levels that did not affect baseline L(p) greatly potentiated the response of L(p) to 20-dyne/cm(2) shear stress. Finally, both shear stress-induced and IAA-induced increases in L(p) could be reversed with the addition of dibutyryl cAMP. However, additional metabolic inhibitors, 2 deoxyglucose (10 mmol/L) and oligomycin (1 micromol/L), or reactive oxygen species scavengers, deferoxamine (1 mmol/L) and ascorbate (10 mmol/L), failed to alter shear-induced increases in L(p). Our results show that neither the NO/cGMP/PKG pathway nor a metabolic pathway mediates the shear stress-L(p) response. An alternate mechanism downstream from NO that is sensitive to IAA must mediate this response.