Quantifying Molecular Disease Mechanisms in Intact Tissue using Automatic and Adaptive Refractive Index Compensation for Light-Sheet Fluorescence Microscopy

Abstract

Optical tissue clearing has revolutionized researchers’ ability to perform fluorescent measurements within intact tissue. One common complication to all optically cleared tissue is a spatially heterogeneous refractive index (RI), leading to light scattering and first-order defocus. We designed C-DSLM (cleared tissue digital scanned light-sheet microscopy) as an adaptive light-sheet fluorescence microscopy (LSFM) method intended to automatically generate in-focus images of cleared tissue.1 C-DSLM uses electrotunable lenses and computational autofocusing to de-couple volumetric imaging from sample movement. C-DSLM automatically corrects for first-order defocus and maintains co-planarity between the exciting light-sheet and focus of the detection arm. Using C-DSLM and tissue clearing (PACT), we investigated multiple animal models difficult to image using standard methods. Here we present our results from two translational animal models of disease. By intentionally under-clearing mouse brain tissue with an endogenous fluorescent reporter of myelin, we created heterogeneous RI tissue that retained both lipid-rich myelin and myelin progenitor cells. Because C-DSLM can automatically account for this heterogeneous RI in large samples, we discovered regional differences in brain-wide myelin progenitor cell distribution linked to changes in circuit function and behavior due to the deletion of proteolipid protein.1 Because C-DSLM imaging is inertia-free, it can accommodate fragile samples such as intact retinas. We discovered that the primary plexus vasculature over-compensates for deficient vertical sprouts and secondary plexus in a rat model of retina development exposed to severe hyperoxic injury.2 As optical tissue clearing becomes more widely adopted, simple and affordable high-throughput imaging methodologies are required. C-DSLM is one such methodology that has already produced multiple advancements in quantifying translational animal models. 1) Ryan, D.P. et al. Nature Communications 2017 2) Singh, J.N. et al. Journal of Biomedical Optics 2017