Purchase this article with an account.
A. Behzadian, A.B. El Remessy, M. Al Shabrawey, N. Ghaly, T. Franklin, R.B. Caldwell; Diabetes–induced Retinal Vascular Hyperpermeability Requires Urokinase Receptor (uPAR) . Invest. Ophthalmol. Vis. Sci. 2004;45(13):1100.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: Blood/retinal barrier (BRB) dysfunction and vascular injuries in diabetes have been attributed to elevated blood glucose and increased VEGF expression, yet the molecular mechanism of hyperglycemia induced (BRB) breakdown is not clear. In cultured retinal endothelial cells, we have shown that VEGF–induced paracellular permeability is mediated by expression of uPAR and increased extra–cellular proteolytic activity. Furthermore, high glucose medium was found to stimulate both VEGF and uPAR expression in these cells. The process involves the beta–catenin signaling pathway and is accompanied by redistribution of endothelial–cell junction proteins. Using transgenic uPAR knockout (ko) mice we now present direct evidence that uPAR plays a key role in diabetes–induced hyperpermeability in retina. Methods: uPAR ko mice were maintained by cross breeding on C57BL/6. Homozygous uPAR –/– mice showed no overt phenotypic abnormalities and no alterations in the retinal vascular pattern. Diabetes was induced in 8 weeks old uPAR –/– and their wild type littermates by streptozotocin injection. After 5 weeks, vascular permeability in retina was quantified by analyzing extravasation of fluorescein–conjugated albumin. Vitreous and retinal mRNA were examined for expression of proteases and VEGF mRNA respectively. Results: A 4–fold increase of vascular permeability was observed only in diabetic wild type as compared to control animals. The diabetic uPAR –/– mice showed no permeability increase over non–diabetic littermates irrespective of their genotype. There was no evidence of retinal morphologic or thickness changes among different groups. VEGF expression measured by real–time RT–PCR was the same in all diabetic animals irrespective of their genotype indicating that uPAR is required for increased vascular permeability in diabetes. All vitreous samples examined on gelatin zymograms were positive for MMP–2, whereas MMP–9 was present only in diabetic wild type but not in uPAR–/– animals. MMP–9 and pro–uPA are thought to interact to initiate a cascade of proteolysis on the cell surface. Conclusions: Our data indicate that the extracellular protease activity initiated by uPA/uPAR system is the critical step in diabetes–induced BRB breakdown.
This PDF is available to Subscribers Only