Abstract
Purpose :
The cadherins junction protein, alpha-catenin (ACat), is primarily located in the cell membrane of healthy retinal pigment epithelium (RPE) cells, but undergoes a redistribution to the cytosol in cells which exhibit damage-induced changes in morphology. Previously, efforts to quantify the correlation between cytosolic ACat levels and individual cell morphological metrics in situ were limited by small sampling areas, access to prohibitive computational power, and sufficient programming expertise. Here, we introduce two separate analysis protocols allowing for unbiased, computationally minimal, and easy-to-use quantification of ACat cellular distribution using the open-source Cell Profiler application and the commercially available IMARIS software suite.
Methods :
Damage to mouse RPE was induced by one of three methods (sub-retinal injection, light induced retinal degeneration, or systemic sodium iodate). Image sets of entire RPE sheets were acquired via confocal microscopy of fluorescently labeled ZO-1, ACat and nuclei. Maximum intensity projection images of each RPE sheet were processed using IMARIS 9.6 (Bitplane, Inc.) where individual cells were segmented, identified, and quantified morphologically. Incorrectly segmented cells and artifacts were manually rejected. ACat cytosolic intensities were derived using ZO-1 defined cell identification. For comparison, the same image sets were then processed using Cell Profiler without manual rejection of segmentation errors.
Results :
Entire RPE sheets from both control and damage groups were accurately and successfully segmented using both Cell Profiler and IMARIS analysis protocols on a single lab computer. Segmentation morphometrics of abnormal cells following damage matched previously reported (K. Donaldson, ARVO2017) results showing positive correlations between abnormally sized (area, perimeter) cells and increased ACat cytosolic expression. Once optimized, Cell Profiler and IMARIS protocols were able to be applied uniformly across image sets, enabling unbiased and repeatable segmentation and quantification.
Conclusions :
We present both open- and closed-source solutions using Cell Profiler or IMARIS for successful and efficient segmentation of RPE flatmounts. Additionally these protocols allow for simultaneous quantification of cytosolic ACat, a marker for abnormal RPE cells in damage models.
This is a 2021 ARVO Annual Meeting abstract.