A New Quantitative Metric for Precise Classification of Diabetic Podocyte Injury Using Scanning Electron Microscopy

Podocytes are highly interdigitated epithelial cells in the glomerulus that maintain the kidney's filtration barrier, and their injury underlies the progression of diabetic kidney disease. Early podocyte damage is challenging to detect using light or transmission electron microscopy; new techniques like super-resolution microscopy remain limited in capturing the three-dimensional topography of podocytes. Scanning electron microscopy (SEM) offers superior spatial resolution and surface detail; however, standardized quantitative methods to analyze podocyte ultrastructure are lacking. In this study, we developed and compared three analytical approaches to quantify podocyte injury from SEM images in a streptozotocin-induced diabetic mouse model. Using ImageJ software, we measured the slit diaphragm (SD) fraction via (1) thresholding, (2) ridge detection, and (3) foot process plot profiling, comparing diabetic and nondiabetic podocytes. The ridge detection method showed the best diagnostic accuracy (88% sensitivity and 93% specificity), successfully distinguishing diabetic from healthy podocytes. Furthermore, SD fraction measurements correlated negatively with biomarkers of podocyte dysfunction and diabetic stress. This work establishes the first reliable, quantitative pipeline for detecting subtle early podocyte injury in diabetic kidney disease using SEM, providing a valuable tool for future mechanistic and therapeutic studies.