Introduction & Objective: Diabetic polyneuropathy (DPN) is the most frequent complication of diabetes mellitus. Reduction of axon number and atrophy of dorsal root ganglion (DRG) neurons are causative pathological changes in DPN, while a decrease in the number of neurons has been inconclusive in rodent models. Because the pathogenesis of DPN differs between type 1 and type 2 diabetes, the pathological changes in DRGs may be different. In this study, we quantified DRGs in streptozotocin (STZ)-treated C57BL/6 and db/db mice in three dimensions.

Methods: STZ (160 mg/kg) was administered peritoneally to 8-week-old male C57BL/6 mice. At 8 and 16 weeks after administration, and in same-week-old db/db and db/m mice, lumbar L5 DRGs were dissected. Double immunofluorescent staining for the neuronal markers neurofilament (NF) and CGRP were applied to the tissue made transparent by the CUBIC method. Images were taken with a light sheet microscope. The captured 3D images were used to create training data, and a segmentation AI (DeepLab3+ modified) was trained. The trained AI processed the images and quantified them. The cells were divided into 4 groups as NF-low/high and CGRP+/- according to the median pixel value of each cell.

Results: Three-dimensional analysis showed that cell body volume and cell count decreased at both 8 and 16 weeks in the STZ group (p<0.05 vs. Control). Histograms of cell bodies showed a significant decrease in the range of 3000-6000 μm3 at both 8 and 16 weeks (p<0.05 vs. Control). When classified by NF/CGRP staining, the number and volume were reduced in the NF-low/CGRP- population (p<0.05). On the other hand, in db/db mice, although cell body volume was reduced (p<0.05 vs. Control), there was no significant difference in cell number. Also, there were no numerical changes in cell subpopulations.

Conclusions: Only DPN in type 1 diabetes showed a marked decrease in cell number of DRG neuron, which may lead to a new understanding of the pathogenesis of DPN.

Disclosure

H. Mizukami: None. T. Sasaki: None. H. Kushibiki: None. M. Ryuzaki: None. Z. Wang: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.