Abstract
Purpose :
Diabetes impairs vision in the earliest stages of diabetic retinal disease (DRD) in humans, yet there is still no treatment available, in part due to a lack of understanding of the underlying mechanisms. We have developed the platform to recover and deeply characterize post-mortem human ocular samples in a standardized fashion and are analyzing them to improve our understanding of the molecular and cellular changes taking place during the onset and progression of diabetic retinal disease.
Methods :
Whole eyes and posterior poles are collected and processed from 7 different eye banks following established and standardized protocols. For each of the 19 donors, one eye was rapidly frozen while the other was fixed for 24 hours in 10% formalin before being transfered in saline until imaging. Each fixed sample was imaged using fundus camera and an OCX (LighTopTech) for extensive imaging of the neuroretinal and recording of all associated anatomical features. The fixed eyes were then processed and used for histological and immunofluorescence analysis. The contralateral fresh frozen eye was processed for isolation of all ocular tissues including vitreous and retinal regions (fovea, macula, mid- and far-peripheral retina) and used for validation of proteomic and transcriptomic changes identified in our previous cohort.
Results :
Fundus and optical coherence tomography (OCT) imaging revealed specific features including microaneurysms, vitreous hemorrhages, lipid exudates, disorganization of retinal inner layers (DRIL), and neovessels in 9 diabetic donors. The retinal region surrounding these features was further evaluated by OCT microscopy prior to being processed for immunofluorescence and assessment of specific changes involving Müller glial, microglia and endothelial cells in association with increased inflammation and complement activation. These findings were confirmed by immunoblot on tissues from the controlateral eyes.
Conclusions :
The current status of the biorepository demonstrates its potential as it shows that samples can be highly characterized and define, allowing concurrent analysis of specific features to identify their association with cellular and molecular changes. This approach is essential to continue to improve our understanding of DRD pathology and develop new therapies.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.