Abstract
Purpose :
Using super-resolution microscopy, NIR autofluorescence (AF) signals of the RPE were attributed to melanin and melanolipofuscin granules in RPE cell bodies (Ach et al., ARVO 2020). In human RPE cell bodies, melanolipofuscin are more frequent than melanosomes (PMID: 32433758) which also impacts clinical NIR-AF image interpretation. Here, we report the NIR-AF signal of RPE cells including apical processes to further evaluate its contribution to clinical NIR-AF imaging.
Methods :
Retinal tissue cross-sections (10 µm thick) from five human donors (age range: 64-86 years) with normal maculas determined by ex vivo imaging by optical coherence tomography and confirmed by histology were imaged using a confocal microscope (white light laser source [excitation wavelengths in nm: 488, 635, 750, 780]; hybrid detector for optimized imaging in the NIR range; plan-apochromatic 40X 1.46 oil immersion objective; z-stacks in ~0.5µm steps). In addition, bright field images were taken to capture optically dense melanin pigment. RPE cells within areas of attached neurosensory retina at the fovea and perifovea were scanned. Finally, images were post-processed using FIJI.
Results :
Both 750 nm and 780 nm excitation, lead to NIR-AF emission from the RPE cell body, though very weak compared to short wavelength excitation from the same areas. In addition, individual spindle-shaped melanosomes are visible in the apical processes of the RPE cells, exhibiting a bright NIR-AF signal, but clearly lacking AF in the short and mid wavelength range (480 – 635 nm).
Conclusions :
There are probably two sources of RPE NIR-AF signals: spindle-shaped melanosomes in the apical processes as well as melanolipofuscin granules and more roundish melanosomes within the RPE cell bodies. The numerical predominance of melanolipofuscin granules within the cell bodies suggests that this signal is the leading source of NIR-AF imaging. These histological findings should be considered in the interpretation of future clinical NIR-AF imaging studies.
This is a 2021 ARVO Annual Meeting abstract.