Abstract
Purpose:
Advanced glycation end-products (AGEs) are thought to contribute to RPE barrier dysfunction eventually leading to retinal edema. These effects of AGEs appear to be mediated via the release of VEGF and VEGFR2 activation. We have recently shown that C-type natriuretic peptide (CNP) reverses the effect of AGEs on RPE permeability in vitro. Here we tested whether CNP can reverse the effect of AGEs on RPE barrier function in a diabetic eye disease model in vivo, and its effect on VEGF release induced by AGEs in RPE cultures in vitro.
Methods:
The integrity of RPE barrier function was evaluated by subretinal bleb reabsorption in rabbits. Glycated albumin (Glyc-alb, AGEs model) or non-glycated albumin was injected into the vitreous to a final concentration of 1 mg/mL, with or without CNP (220 ng). Two days later, subretinal blebs (1-5 µL) containing PBS were created and fluid reabsorption was followed for one hour using Spectralis OCT. Transepithelial electrical resistance (TEER) measurements assessed the permeability of monolayer-cultured ARPE-19 and human fetal RPE cells on transwell filters. Cells were treated with albumin or Glyc-alb (100 μg/mL), or VEGF (20 ng/mL), in the absence or presence of CNP (1nM-100 nM). The secretion of VEGF into the media was quantitated by ELISA assays.
Results:
The rate of subretinal fluid reabsorption was equal to 8.2±0.6 µl.cm-2.hr-1. In eyes receiving intravitreal Glyc-alb the rate of fluid reabsorption was significantly (p<0.05) decreased (2.2±0.4 µl.cm-2.hr-1). Co-administration of CNP reversed the effect of glyc-alb on fluid reabsorption. The administration of albumin did not significantly alter the rate of fluid reabsorption. In both ARPE-19 and human fetal RPE cultures, Glyc-alb induced 24±4% reduction in TEER, and a significant increase in VEGF secretion. Pretreatment with CNP blocked the effect of Glyc-alb on TEER; however it did not significantly alter the increase in VEGF secretion. In RPE cultures, the administration of VEGF caused a 22±4% drop in TEER, and this response was blocked by CNP.
Conclusions:
The administration of Gly-alb induced a decline in RPE barrier function both in vitro and in vivo. CNP reversed this effect both in models. As CNP did not affect the Glyc-alb-induced VEGF release, the action of CNP appears to be related to the suppression of VEGF receptor signaling. These data provide evidence that CNP would be efficacious in the treatment of diabetic macular edema.
Keywords: 666 pump/barrier function •
499 diabetic retinopathy •
701 retinal pigment epithelium