Non-invasive Detection of Reactive Oxygen Species in ex vivo Retinal Explants Models Using Autofluorescence Multispectral Imaging

Abstract

Purpose : Glycolytic and mitochondrial stressors are known to produce reactive oxygen species (ROS). These stressors have been implicated in the pathogenesis of neurodegenerative disorders. The purpose of the study was to detect ROS induced by these stressors using a novel autofluorescence multispectral imaging (AFMI) in an ex vivo organotypic retinal explant model

Methods : Eyes from Wister rats were enucleated after post-mortem and retinal explants were prepared. Explants were treated with iodoacetic acid (IAA) and rotenone to inhibit the glycolytic and mitochondrial stress, respectively. Explants underwent 24 hour treatment with stressors. Tissue viability was assessed using fluorescent live/dead cell assay to determine the working concentration. ROS production was determined with dihydroethidium stain. Further, explants were imaged using novel AFMI microscope to determine the ROS levels non-invasively without exogenous fluorophores. AFMI consists of 33 spectral channels covering specific excitation and emission wavelengths. Spectral features were extracted from acquired images. Channel analysis was performed, and uncorrelated channels were selected. The principle component analysis (PCA) was used to reduce the dimensions and PCA scores were used to construct the classifier

Results : A dose-dependent tissue death was observed with IAA after 24 hours of exposure: IAA (50µM) produced significant levels of ROS compared to control group. Similarly, tissue death was increased proportionally with the dose of rotenone and 500 µM produced signigicant ROS. Further, AFMI spectral channels showed significant differences in spectral signals between control and tissues treated with IAA and rotenone. The accuracy of these changes was supported by receiver operating characteristics curves with areas under the curves (AUC) of 0.97 and 1.00, respectively. The potential clinical application of this technology was evaluated using UV-free channels through a similar analysis. Similar results noted in UV free channels with IAA and rotenone, AUC of 0.88 and 0.98 respectively.

Conclusions : ROS can be detected using novel AFMI method in explants models. AFMI has potential as novel imaging tool in ophthalmic practice to diagnose and monitor neurodegenerative diseases like glaucoma and age-related macular degeneration. Further, research will explore in disease models such as optic nerve crush model

This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.