This study determined whether retinal degeneration during diabetes includes retinal neural cell apoptosis. Image analysis of retinal sections from streptozotocin (STZ) diabetic rats after 7.5 months of STZ diabetes identified 22% and 14% reductions in the thickness of the inner plexiform and inner nuclear layers, respectively (P < 0. 001). The number of surviving ganglion cells was also reduced by 10% compared to controls (P < 0.001). In situ end labeling of DNA terminal dUTP nick end labeling (TUNEL) identified a 10-fold increase in the frequency of retinal apoptosis in whole-mounted rat retinas after 1, 3, 6, and 12 months of diabetes (P < 0.001, P < 0. 001, P < 0.01, and P < 0.01, respectively). Most TUNEL-positive cells were not associated with blood vessels and did not colocalize with the endothelial cell-specific antigen, von Willebrand factor. Insulin implants significantly reduced the number of TUNEL-positive cells (P < 0.05). The number of TUNEL-positive cells was also increased in retinas from humans with diabetes. These data indicate that retinal neural cell death occurs early in diabetes. This is the first quantitative report of an increase in neural cell apoptosis in the retina during diabetes, and indicates that neurodegeneration is an important component of diabetic retinopathy.
Diabetic retinopathy remains a frightening prospect to patients and frustrates physicians. Destruction of damaged retina by photocoagulation remains the primary treatment nearly 50 years after its introduction. The diabetes pandemic requires new approaches to understand the pathophysiology and improve the detection, prevention, and treatment of retinopathy. This perspective considers how the unique anatomy and physiology of the retina may predispose it to the metabolic stresses of diabetes. The roles of neural retinal alterations and impaired retinal insulin action in the pathogenesis of early retinopathy and the mechanisms of vision loss are emphasized. Potential means to overcome limitations of current animal models and diagnostic testing are also presented with the goal of accelerating therapies to manage retinopathy in the face of ongoing diabetes.
Vascular endothelial growth factor (VEGF) may have a physiologic role in regulating vessel permeability and contributes to the pathophysiology of diabetic retinopathy as well as tumor development. We set out to ascertain the mechanism by which VEGF regulates paracellular permeability in rats. Intra-ocular injection of VEGF caused a post-translational modification of occludin as determined by a gel shift from 60 to 62 kDa. This event began by 15 min post-injection and was maximal by 45 min. Alkaline phosphatase treatment revealed this modification was caused by a change in occludin phosphorylation. In addition, the quantity of extracted occludin increased 2-fold in the same time frame. The phosphorylation and increased extraction of occludin was recapitulated in retinal endothelial cells in culture after VEGF stimulation. The data presented herein are the first demonstration of a change in the phosphorylation of this transmembrane protein under conditions of increased endothelial permeability. In addition, intra-ocular injection of VEGF also caused tyrosine phosphorylation of ZO-1 as early as 15 min and increased phosphorylation 4-fold after 90 min. In conclusion, VEGF rapidly increases occludin phosphorylation as well as the tyrosine phosphorylation of ZO-1. Phosphorylation of occludin and ZO-1 likely contribute to regulated endothelial paracellular permeability.
1 April 2003 in Progress in Neuro-Psychopharmacology and Biological Psychiatry
Blood-retinal barrier (BRB) breakdown is a hallmark of diabetic retinopathy, but the molecular changes that cause this pathology are unclear. Occludin is a transmembrane component of interendothelial tight junctions that may regulate permeability at the BRB. In this study, we examined the effects of vascular endothelial growth factor (VEGF) and diabetes on vascular occludin content and barrier function. Sprague-Dawley rats were made diabetic by intravenous streptozotocin injection, and age-matched animals served as controls. After 3 months, BRB permeability was quantified by intravenous injection of fluorescein isothiocyanate-bovine serum albumin (FITC-BSA), Mr 66 kDa, and 10-kDa rhodamine-dextran (R-D), followed by digital image analysis of retinal sections. Retinal fluorescence intensity for FITC-BSA increased 62% (P < or = 0.05), but R-D fluorescence did not change significantly. Occludin localization at interendothelial junctions was confirmed by immunofluorescence, and relative protein content was determined by immunoblotting of retinal homogenates. Retinal occludin content decreased approximately 35% (P < or = 0.03) in the diabetic versus the control animals, whereas the glucose transporter GLUT1 content was unchanged in rat retinas. Additionally, treatment of bovine retinal endothelial cells in culture with 0.12 nmol/l or 12 nmol/l VEGF for 6 h reduced occludin content 46 and 54%, respectively. These data show that diabetes selectively reduces retinal occludin protein expression and increases BRB permeability. Our findings suggest that the elevated VEGF in the vitreous of patients with diabetic retinopathy increases vascular permeability by downregulating occludin content. Decreased tight junction protein expression may be an important means by which diabetes causes increased vascular permeability and contributes to macular edema.