Advanced glycation end products (AGEs) are nonenzymatic glycosylated adducts of proteins that accumulate in vascular tissue during diabetes and aging. The aim of this work was to study the role of AGEs and of the oxidative mechanisms in diabetes-induced changes in vascular permeability. Intravital videomicroscopy was used to study albumin microvascular leakage in cremaster muscle. The extravasation of a fluorescent macromolecular tracer (fluorescein isothiocyanate-albumin) was measured for 1 h and, after computer-aided image analysis, was expressed as variations of normalized gray levels (arbitrary units). Extravasation of the macromolecular tracer was much higher in diabetic rats than in control rats (slope of extravasation versus time increased by >100%, P < 10(-4)). This increase was significantly inhibited when we blocked AGEs binding to their endothelial receptor by intravenous bolus of soluble recombinant receptor to AGEs (rR-RAGE) (slope of extravasation versus time decreased by 19, 30, and 40%, for 0.5, 2.5, and 5.15 mg/kg rR-RAGE, respectively) or by a 6 mg/kg intravenous bolus of antibody against RAGE (slope decreased by 53%). Systemic injection of probucol (an antioxidant) also significantly inhibited the increase in the extravasation of the macromolecular tracer occurring in experimental diabetes (slope decreased by 51%, P < 10(-4)). These results strongly suggest that in experimental diabetes the interaction of circulating AGEs and endothelial RAGE mediates albumin micro-vascular leakage, possibly via AGE-RAGE-dependent enhanced oxidant stress.

We studied in vivo interactions of nitric oxide (NO), oxidative stress, and prostanoids derived from the cyclooxygenase pathway in the arterioles studied by intravital microscopy in peripheral muscle. Topical administration of NO synthase (NOS) inhibitor Nomega-nitro-l-arginine (l-NNA) or cyclooxygenase inhibitor mefenamic acid (MA) alone leads to vasoconstriction. We found that l-NNA after MA induced an additional constriction, whereas MA after l-NNA induced a relative dilation. Therefore, an additional constriction was found when MA was administered after l-NNA in the presence of the thromboxane A2 synthase-PGH2 (TP) receptor antagonist SQ-29548. We also found a relative dilation when the TP receptor antagonist was administered after NOS inhibition by l-NNA. In the presence of superoxide dismutase and catalase, l-NNA-induced vasoconstriction is reduced, and the dilation observed after addition of MA in presence of the reactive oxygen species is no longer present. Taken together, these results showed that NO inhibition induced a shift in the synthesis or in the effects of cyclooxygenase products, in favor of constrictor prostanoids. This effect of NO inhibition disappears when reactive oxygen species are scavenged by superoxide dismutase and catalase.

Endothelial cells have been reported to generate reactive oxygen species such as the superoxide anion, hydrogen peroxide, and the hydroxyl radical. The aim of this work was to evaluate the role of reactive oxygen species in diabetes-induced changes in vascular permeability. Intravital videomicroscopy was used to study albumin microvascular leakage in the cremaster muscle. The extravasation of a fluorescent macromolecular tracer (FITC-albumin) was measured for 1 h and, after computer-aided image analysis, was expressed as variations of normalized gray levels (arbitrary units). The extravasation of the macromolecular tracer was greater in diabetic rats (5.28 ± 1.29 vs. 1.96 ± 0.41 AU at 1 h in diabetic and control rats, respectively). Administration of superoxide dismutase (SOD), which dismutates ·O^–₂ to H₂O₂, and of catalase which reacts with H₂O₂ to form H₂O and molecular oxygen failed to inhibit the increased extravasation of the macromolecular tracer when administered separately. However, a significant inhibition of diabetic increase in albumin extravasation was found when these 2 drugs were administered simultaneously, and in this case, the extravasation of the macromolecular tracer at 1 h was similar in diabetic rats (2.11 ± 0.61 AU) and normoglycemic rats (1.43 ± 0.48 AU). No difference was found in adherent leukocytes or in the leukocyte rolling flux between diabetic and normoglycemic rats. We conclude that reactive oxygen species are responsible for an increase in microvascular permeability likely via leukocyte-independent mechanisms.