Computational Identification and 3D Morphological Characterization of Renal Glomeruli in Optically Cleared Murine Kidneys

The aim of this study was to establish an accessible methodology for the objective identification and 3D morphological characterization of renal glomeruli in mice. 3D imaging of the renal cortex was performed by light sheet microscopy on iDISCO+ optical cleared kidneys of six C57BL/6J mice after labelling of the capillary endothelium by lectin injection. 3D images were processed with the open source software ImageJ, and statistical analysis done with GraphPad Prism. Non-visual delimitation of the external surface of the glomeruli was ensured by greyscale-based thresholding, the value of which was determined from the statistical analysis of the voxel frequency distribution. Exclusion of false-positive identification was done by successive volume- and shape-based segmentation. Renal glomeruli were characterized by their number, surface area, volume, and compactness. Average data were expressed as mean ± SD. The number of glomeruli was equal to 283 ± 35 per mm(3) of renal tissue, representing 1.78 ± 0.49% of the tissue volume. The surface area, volume and compactness were equal to 20,830 ± 6200 µm², 62,280 ± 14,000 µm(3) and 0.068 ± 0.026, respectively. The proposed standardized methodology allows the identification of the renal glomeruli and their 3D morphological characterization, and is easily accessible for biologists.