Hypothesis / aims of study
Diabetes is a growing epidemic in the United States, with projections that 1 in 3 Americans will be affected by 2050. As a chronic disease, diabetes is associated with high costs and various complications leading to end organ dysfunction. The most common of these complications is diabetic bladder dysfunction (DBD), with a prevalence as high as 87%. Recently, it has been established that inflammation in specific tissues underlies the vast majority of these complications. This inflammation is triggered by a multimeric complex known as the NLRP3 inflammasome. Accordingly, our laboratory has shown that NLRP3 plays a critical role in the symptoms of DBD. In this original study, we have evaluated whether NLRP3 mediates changes in specific nerve cell populations in the bladder that could explain functional deficits in diabetic patients, specifically diminished sensation (fullness) and overactivity. Sensation of bladder fullness is transferred to the CNS via Aδ-fibers, large myelinated afferent nerves in the bladder wall, while overactivity is thought to be triggered through C-fibers, small non-myelinated afferent nerves in the urothelium [1,2]. For these, and related, studies our lab has crossed a genetically diabetic mouse strain (the Akita mouse) with a NLPR3-/- strain to create a diabetic mouse lacking NLRP3 expression. We then utilize this novel model to assess the effect of NLRP3 on nerve sub-types during DBD.
Study design, materials and methods
Four groups of 15 week old mice were used: 1) control non-diabetic mice, 2) diabetic mice (Akita), 3) NLRP3-/- (null) non-diabetic mice, 4) NLRP3-/- (null) diabetic (Akita) mice. Previous studies from our lab have documented these diabetic mice (Akita) develop DBD symptoms at this time point and that deletion of NLRP3 had no effect on blood glucose in the nondiabetic or diabetic mouse. Blood glucose was assessed with a standard glucometer following lancing of the submandibular vein. For evaluation of neuronal changes in number and density, bladders were fixed in neutral buffer formalin, embedded in paraffin and sectioned (5 µm). Transverse sections from the lower third of the bladder were stained with a rabbit anti-NF-200 antibody (for Aδ-fibers) or a rabbit anti-calcitonin gene-related peptide (CGRP) antibody (for C-fibers) and visualized with a goat anti-rabbit IgG antibody conjugated to Alexa Fluor 488 using standard methods and citrate antigen retrieval (pH 6.0). The entire cross sections were imaged (20X) with Zen software, using tiling micrographs stitched into a continuous image by the software, and exported as TIFF files. Files were then imported into NIS Elements software and calibrated. A hue spectrum or manual extraction was used to create a region of interest (ROI) composed of the bladder wall for Aδ-fibers and the urothelium for C-fibers. The area of the ROIs was then provided by the software in µm2 and individual neurons marked and counted. Aδ-fibers were defined as fluorescent areas >50 um2 that stained positive with a nuclear co-stain (DAPI), which was used to exclude blood vessels containing auto-fluorescent red blood cells. C-fibers were defined as continuous fluorescent fibers >1 µm. Data was reported as the mean ± standard error (SEM). Statistical analyses were performed using GraphPad InStat software (GraphPad Software, inc., La Jolla, CA) and a one-way ANOVA with a Turkey’s post-hoc test. Results were considered significant at P < 0.05.
Results
As shown in Figure 1A, there was a marked decrease in the number of Aδ-fibers of diabetic mice compared to the nondiabetic controls [(n=6) (p<0.001)]. No differences were seen in the size of the bladder wall (Figure 1B). This decrease in nerve count, correlated with no change in bladder wall size, results in a dramatic decrease in Aδ density (Figure 1C) (p<0.001) in the diabetics. Importantly, diabetes had no effect on either the number of Aδ fibers (Figure 1A) or their density (Figure 1C) in the NLRP3-/- mice. Diabetes caused a significant increase in C-fiber number within the urothelial layer (Figure 2A) with no change in the size of the urothelium (Figure 2B) (p<0.05) and, thus, an increase in C-fiber density (Figure 2C) (p<0.001). Similar to Aδ-fibers, the diabetic change in C-fibers was completely absent in the NLRP3-/- mice (Figures 2A-C).
Interpretation of results
The decrease in Aδ-fiber number and density in the diabetic mice suggests that a lack of sufficient signaling through these fibers may be a possible mechanism by which diabetics experience reduced bladder sensation. Early DBD is also associated with bladder activity, which we have previously demonstrated in 15 week old diabetic mice. The increase in C-fiber number and density in the diabetic mice suggest a possible mechanism by which early DBD causes bladder overactivity. Ultimately, the absence of these pathologic changes in the diabetic NLRP3-/- mice suggests that the NLRP3 inflammasome may play an important role in these neuropathic changes and subsequent development of symptoms of early DBD.