A mouse model of corneal endothelial decompensation using cryoinjury

PURPOSE To develop a mouse model of bullous keratoplasty and evaluate the safety and efficacy of cryoinjury-induced corneal endothelial decompensation. METHODS Transcorneal freezing was performed on the right eye of each mouse. One cycle of cryoinjury was performed in 18 eyes (group A), and three cycles were performed in 17 eyes (group B). Pachymetry and intraocular pressure (IOP) measurements were done preoperatively, as well as at 1, 3, 7, 14, and 21 days after cryoinjury. At each post-cryoinjury time point, three mice from each group were euthanized, and the corneas underwent histology and electron microscopy. RESULTS In both groups, significant corneal edema was noted at post-cryoinjury day 1, which was maintained throughout the study period. IOP remained within normal range in group A, but increased significantly with time in group B (p=0.011 at day 1, 0.038 at day 3, 0.026 at day 14, and 0.008 at day 21). In group B, serious complications including hyphema (one case), severe iridocorneal adhesion (15 cases), and total cataract (three cases) were detected, while only mild iridocorneal adhesion (four cases) and cataract (three cases) were noted in group A. Live/dead cell assay, hematoxylin and eosin staining, and scanning electron microscopy revealed successful ablation of corneal endothelial cells and absence of regeneration in both groups. Hematoxylin and eosin staining and terminal deoxynucleotidyl transferase-mediated nick end labeling assay showed that apoptosis was mainly confined to the posterior stroma and endothelium in group A, while severe apoptosis was observed throughout all layers of the cornea in group B. CONCLUSIONS One cycle of cryoinjury was safer than three, while both were equally effective in inducing bullous keratopathy. This cryoinjury mouse model of bullous keratopathy was a consistently reproducible model that can be used for further studies on endothelial cell damage and rescue therapy.