Abstract
Purpose:
Characteristics of autofluorescent granules (AG) in retinal pigment epithelium (RPE) cells like number of AG in single cells, distribution across the fundus, AG size and position within RPE cells are still not known. We have recently shown that Structured illumination microscopy (SIM) enables three-dimensional high resolution imaging of AG in RPE cells. Here we present the self-written software AGES (Analysis of Granules Established by SIM) for automated analysis of 3D-SIM-image data to characterize AG within RPE.
Methods:
Highly resolved SIM images were processed with the newly developed software AGES for differentiation and analysis of intracellular AG. As AG are heterogeneous in brightness, an algorithm applying a rising threshold is used. AGES enables breaking down clusters of densely-packed AG into their composing granules. 3D-SIM images from macular and temporal RPE cells were examined (two donors, 32 years and 91 years) using three different laser excitation wavelengths (488, 568 and 647 nm). We analyzed number, size, position and fluorescence intensity of AG in 10-20 cells per ocular region of each donor.
Results:
AGES is able to detect AG separately, even in complex structures and AG being next to each other. The mean number of AG per macular cell was 77,9 ± 37,8 (32-year-old donor) and 112,0 ± 55,0 (91-year-old donor) and per temporal cell 44,4 ± 18,7 (32-year-old donor) and 67,5 ± 29,3 (91-year-old donor). The granula diameter (1,00 ± 0,23 µm) and the granula volume (0,61 ± 0,44 µm 3) at all sites were in average similar for both donors. AG were distributed towards the cell edge.
Conclusions:
3D-high-resolution-SIM and the self-written software AGES allows detailed examination and automated analysis of AG in RPE cells. For the first time characteristics of AG like exact number and localization within single RPE cells are determinable.
Keywords: 701 retinal pigment epithelium •
551 imaging/image analysis: non-clinical •
695 retinal degenerations: cell biology