Purchase this article with an account.
Sankarathi Balaiya, Wendi S Lambert, Bryan A Millis, David J Calkins, Edward M Levine; Development Of A Screening Platform To Identify Factors That Promote Retinal Regeneration In Mice. Invest. Ophthalmol. Vis. Sci. 2018;59(9):3109.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
We are developing an in vivo screening platform in mice to identify novel regeneration-promoting factors. Because an in vivo approach is low throughput and regeneration could be limited to a small number of cells, it is critical to rapidly visualize and quantify changes in Muller Glia/MG properties (proliferation, location, morphology, gene expression) in the whole retina and at cellular resolution. Here, we describe our efforts at achieving this following genetic stimulation of MG proliferation through MG-specific inactivation of the cell cycle regulator p27Kip1.
Adult p27Kip1(flox); Rlbp1:CreER mice received a single dose of tamoxifen (25µg/gbw). After 7 days, 5-ethynyl-2'-deoxyuridine/EdU (30µg/gbw) was given 3 times over 3 days. Retinas were subsequently harvested and fixed. EdU was detected with click-it chemistry and retinas were cleared by ScaleSQ. Flattened retinas were imaged using a multi-dimensional large image stitching approach with a fully automated widefield fluorescence microscope equipped and high speed large format sCMOS detector. 25 plane image-stacks were captured at 4 µm intervals to yield 100 µm depth throughout the entire tissue. Focus-stacking (Extended Depth of Focus/EDF) yielded a single plane of focused information per dataset. A custom script was validated and deployed to perform automated image segmentation and cell-number quantification.
EdU detection through the entire retina was achieved by extending the permeabilization and click-it reaction times. ScaleSQ restored the optical transparency of the retina within 2 hours, significantly reducing light scatter while preserving fluorescent signals. The superfast, high resolution imaging permitted the capture of as many as 3000 images per retina in less than one hour and EDF reduced the image collection to a single focused image suitable for quantification. Automated counting achieved a concordance of >85% when compared to manual counting.
We have developed a tissue and image processing pipeline that significantly streamlines the analysis of whole retinas, thereby increasing the feasibility of screening in an in vivo context. We are currently adapting this approach to work with antibody- and reporter-based detection methods.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.
This PDF is available to Subscribers Only