Comparative retinal imaging in mouse models of proteinopathies identifies disease-specific biomarkers

Abstract

Purpose : In previous research we have shown that the App^NL-G-F mouse retina mimics the early, preclinical stages of Alzheimer’s disease, and that retinal imaging offers unique opportunities for research and drug discovery [1] . In this follow-up study, we have used in vivo and ex vivo retinal imaging in conjunction with immunohistochemistry and molecular assays, in a range of mouse models of neurodegenerative diseases characterized by protein aggregation, to identify the pathological correlates of imaging biomarkers.

Methods : Mouse models of alpha-synuclein, tau, and amyloid overexpression/aggregation underwent a battery of retinal examinations, including in vivo optical coherence tomography and electroretinography as well as ex vivo hyperspectral imaging (HSI; Snapscan, Imec). HSI spectra were analyzed as previously described [1]. Retinal samples were collected for immunohistochemistry and molecular studies of protein aggregation.

Results : Diverging patterns of retinal dysfunction and/or degeneration were observed in the different mouse models, illustrating the varied impacts of these protein aggregates on retinal structure and function. Moreover, while all protein aggregates led to HSI changes in the 450-600 nm range, distinct signatures were observed. Immunohistochemical and molecular studies revealed that the observed HSI changes correlated both with levels of the protein aggregates that are the hallmarks of these diseases, and with levels of soluble oligomers.

Conclusions : HSI can be used to quantify retinal amyloid beta, underscoring its potential as a biomarker for Alzheimer’s diagnosis and disease monitoring. Moreover, our research indicates that the HSI signatures of alpha-synuclein, tau, and amyloid aggregates are distinct. These findings add further evidence to suggest that retinal HSI may have utility in the diagnosis and stratification of proteinopathies associated with the most common forms of human neurodegenerative diseases.

[1] Vandenabeele et al. (2021) Acta Neuropathologica Communications [https://rdcu.be/cdbmS]

This is a 2021 ARVO Annual Meeting abstract.