Morphometric analysis of retinal structural components in СВА/С57 mice in experimental diabetes mellitus

Abstract

Purpose. To perform morphometric analysis of retinal structural components in СВА/С57 mice in experimental diabetes mellitus (DM) in an attempt to obtain measurable characteristics of possible retinal neurodegenerative changes required for comparison with those in the diabetic retina of other species.
Material and Methods. We retrospectively reviewed hematoxylin-and-eosin (H&E) stained specimens of eyes from eight СВА/С57 mice with diabetes duration of 6 months or less and healthy control animals. Total retinal thickness and thicknesses of the following retinal layers were measured: the photoreceptor layer (PRL), outer nuclear layer (ONL), outer plexiform layer (OPL), inner nuclear layer (INL), inner plexiform layer (IPL), ganglion cell layer (GCL), and nerve fiber layer (NFL). The numbers of neural-cell rows in both nuclear layers were calculated visually.
Results. The mean total thickness of the retina was 197.2 ± 6.32 µm for healthy controls and 227.8 ± 6.2 µm for diabetic animals. In controls and diabetic animals, the mean thicknesses for particular retinal layers were as follows: PRL, 54.4 ± 2.49 µm and 54.9 ± 1.69 µm, respectively; ONL, 47.1 ± 1.73 µm and 49.8 ± 1.69 µm, respectively; OPL, 15.7 ± 1.08 µm and 16.5 ± 0.75 µm, respectively; INL, 24.6 ± 1.094 µm and 32.1 ± 1.46 µm, respectively; IPL, 41.6 ± 1.79 µm and 52.9 ± 1.73 µm, respectively; and GCL plus NFL, 13.8 ± 0.92 µm and 21.6 ± 1.05 µm, respectively. In addition, the numbers of neural-cell rows in the ONL were 9.1 ± 0.32, and 10.0 ± 0.38, respectively, and in the INL, 3.6 ± 0.13 µm, and 3.8 ± 0.16, respectively. In diabetic mice, thicknesses of all inner retinal layers (INL, IPL, and GCL plus NFL) were increased due to edema (р ˂ 0.01 for all cases), resulting in a 30.6-µm increase in the total thickness of the retina. There was no microscopic evidence of neurodegeneration.
Conclusion: Microscopic images of the retina and measurements of thickness of individual retinal layers and numbers of neural-cell rows in retinal nuclear layers for mice with diabetes duration of 6 months or less indicated that a mouse model of streptozotocin-induced diabetes cannot be considered well suited for H&E staining assisted histological assessment of neurodegenerative changes.