Investigating the integrity of the vasculature and structural relationships of pericytes, astrocytes and the endothelial glycocalyx in an ex vivo hyperglycaemic rat retinal model

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title
Background: Diabetes mellitus-induced retinopathy is the leading cause of adult-onset blindness. The retinopathy is characterised by the degeneration of the retinal microvascular and neural components that form the blood retinal barrier (BRB), and is due to diabetic metabolic derangements such as hyperglycaemia. The cellular components of the BRB include endothelial cells, pericytes and the glial cells (mainly astrocytes), which together maintain the integrity and barrier function of the retinal vasculature. However, there is limited data regarding the interconnecting biochemical pathways that lead to pericyte loss, astrocyte degeneration, and endothelial dysfunction as well as the thinning of the glycocalyx in the retinal vasculature of diabetics. As such, the underlying mechanism of how hyperglycaemia induces retinal damage is not fully understood, particularly at the early stages of the disease. Aim and Objectives: To establish and validate an ex vivo rat retinal model and use a multiimmunolabelled approach on the retina to study the effects of hyperglycaemia on the vasculature and cellular components of the BRB, including effect of time and the retinal preparation type. Methods: Eyes of euthanised Wistar and Sprague-Dawley adult rats (n = 68) were enucleated and the retinae were either dissected out (explant) or kept in situ after removal of the anterior segment, lens and vitreous (eyecup). Retinae were superfused continuously (for 1-, 2- or 3 hours) in a chamber with oxygenated Krebs-Henseleit buffer containing either physiological glucose (5.5 mmol/L; control group) or high glucose (25 mmol/L; hyperglycaemic group). For longer incubation (48 hours), explants were cultured in Dulbecco's Modified Eagle Medium containing the above glucose concentrations. Retinae were multi-immunolabelled to identify pericytes, astrocytes and the glycocalyx protein syndecan-1 with anti-neuron-glial 2 antigen (NG2), anti-glial fibrillary acidic protein (GFAP), and anti-syndecan-1 antibodies respectively. Retinal blood vessels were detected with FITC-conjugated lycopersicon esculentum (Tomato) lectin. Fluorescent signals were detected using confocal microscopy imaging where Z-stack images were randomly sampled from the mid-peripheral retina. Images were analysed using Image J software. Additionally, western blotting was used to determine the abundance of syndecan-1 and heat shock proteins in the rat retina. P value < 0.05 was considered statistically significant. Results: Retinal blood vessels, pericytes and astrocytes showed normal morphology in both explant and eyecup of normoglycaemia retinae, with no difference in fluorescence intensity between explant and eyecup preparations at each time point (1-3 hours). Similarly, there was no change in the vascular-associated syndecan-1 signal intensity over time. However, the syndecan-1 intensity was generally weak and localised mainly to the retinal arteries and their branches, but rarely in capillaries and was unevenly distributed throughout the entire retina in both retinal preparation types. Under hyperglycaemia, at the 3-hour time point of perfusion, the vessel width was reduced in the eyecup group, the neuron glial 2 (NG2) fluorescence intensity on pericytes was diminished with a noticeable reduction in number of pericyte nuclei bulge in the lower vascular bed in the explant and eyecup groups. Although the area coverage and fluorescence signal intensity of astrocytes did not change, retraction of astrocyte processes and signs of cellular disintegration were more evident in the hyperglycaemic eyecup group compared to that in the explant group. Syndecan-1 signal intensity was raised in hyperglycaemia compared to controls in the eyecup group. The preliminary investigation on retinae cultured for 48 hours showed signs of fragmentation and morphological deterioration of the BRB, which included narrowing of blood vessels, loss of pericytes and disintegration of astrocytes, especially in the hyperglycaemic group. Syndecan-1 signal intensity was also lower in the hyperglycaemic groups compared to the control groups cultured for 48 hours . Syndecan1 protein abundance in the 3-hour superfused retina using western blotting was significantly higher in the hyperglycaemic group than in control. Heat shock proteins showed higher expression in hyperglycaemic rat retina compared to the normoglycaemic group. Conclusion: This study validated the suitability of the ex vivo rat retinal model to study the components of the BRB. The study revealed that there is no significant difference between the retinal preparations (explants vs eyecup) and that the BRB is still structurally intact up to 3 hours in normoglycaemic conditions. The subtle morphological changes of the BRB (blood vessels, pericytes and astrocytes) induced by hyperglycaemia which were more apparent in the lower vascular layers,suggests that this ex vivo model is suitable to study the early pathogenesis of diabetic retinopathy. Furthermore, this study showed that syndecan-1 levels were affected by hyperglycaemia ex vivo, and that this occurred prior to morphological changes to the other BRB structures, thus supporting the notion that syndecan-1 plays a role in the pathophysiology of DR. Preliminary data suggest that heat shock proteins may play a role in response to hyperglycemia-induced oxidative stress.