Spinal cord injury (SCI), besides resulting in paralysis, can induce lower urinary tract (LUT) dysfunctions that are also highly debilitating and increase morbidity in afflicted individuals. Neurogenic LUT dysfunction includes bladder overactivity, detrusor-sphincter dyssynergia (DSD), urinary retention, frequent urinary tract infections and potential damage to the upper urinary tract. Alterations in neurotrophin signalling has been shown to play a major role in the development of LUT dysfunction. Previous studies have indicated a time dependent downregulation in brain derived neurotrophic factor (BDNF) dependent signalling that results in further deterioration of LUT function [1]. Compensating for the decrease in BDNF may represent a treatment for preventing degeneration of LUT function. Thus, our aim was to elucidate whether LM22B-10, a selective small molecule agonist of TrkB/C receptors [2], affects the development of bladder overactivity and DSD in a spinal cord contused mouse model.

Spinal cord contusion (SCC) surgery and functional assessments: Adult female C57Bl/6 mice (8-12 weeks old) were anesthetized using 2% isoflurane, a laminectomy performed, and the spinal cord exposed between T9-T10 vertebrae. The exposed cord was subjected to severe contusion injury (75 kDy force; Infinite Horizon Impactor, Precision Instrument). Sham controls underwent laminectomy surgeries without contusion (N=4). The area partially devoid of column segments was packed with haemostatic sponge and the muscle and skin sutured. After surgery, the animals had their bladders expressed twice daily by gentle abdominal compression and were given daily prophylactic antibiotics and analgesics for up to one week. Mice were evaluated for hindlimb locomotion recovery using the Basso mouse scale [3] at one, three, seven, 14, 28 and 42-days following injury. LM22B-10 was administered by subcutaneous implanted osmotic pumps that delivered 5 mg/kg/day of drug (N=8) or vehicle (N=9) over a four-week period (Alzet model 1004, vehicle consisted of 50% DMSO with sterile saline). Osmotic pumps were implanted at the time of SCC surgeries without priming which delays drug release by 24-48 hours. At 42 days post SCC, mice were subjected to decerebrate cystometrogram and external urethral sphincter electromyogram (CMG-EUS-EMG) recordings to examine bladder and EUS activities, respectively.

Histology: Following cystometric measurements, the urinary bladder was dissected out and the mouse transcardially perfused with 1 x tris-buffered saline (1xTBS) followed by 4% paraformaldehyde for perfusion fixation of the spinal cord and bladder. Bladders were weighed then further fixed flat in a dissection dish with 4% PFA. Spinal cord and bladder tissues were stored overnight in 30% sucrose solution then embedded in optimal cutting temperature compound for cryosectioning. Spinal cord and bladder tissue sections were processed for immunofluorescence to image glial fibrillary acidic protein (GFAP) and TrkB receptors and/or to Trichome staining for visualization of tissue collagen content. Slides were imaged using brightfield montage microscopy (Olympus Fluoview3000) and analyzed using HCImage software (Hamamatsu Photonics). Quantitative data were expressed as mean ± SEM. Unpaired Student’s t-test determined differences between contused vs. sham controls and parameters with and without treatment. One-way ANOVA multiple comparison was performed to determine between group differences followed by Tukey’s multiple comparisons test.

Histological evaluation of spinal cord injury sites showed reduced scarring and decreased collagen deposition (Fig. 1A, white circles indicate necrotic core) and by void area bound by GFAP and TrkB expressing cells in the spinal cords (Fig. 1B) of LM22B-10 treated mice compared to vehicle controls. Furthermore, LM22B-10 treated mice showed improvement in hindlimb locomotion recovery at 28- and 42-days post injury compared to vehicle treated group (Fig. 1C). Bladders of LM22B-10 treated SCC mice also showed decreased collagen deposition and bladder hypertrophy (Fig. 1E and 1F) compared to controls. Continuous filling CMG recordings in chronic SCC mice demonstrated bladder overactivity and non-voiding contractions that were not present in sham controls (Fig. 2B vs. 2A). EUS-EMG recordings from sham mice show a guarding reflex as bladder pressure increased, and decreased EUS tonic activity accompanied by bursting as the bladder emptied; characteristic of normal rodent voiding (Fig. 2A inset, tonic EUS activity denoted by red arrows). Conversely, vehicle treated SCC mice showed high baseline pressures, consistent non-voiding contractions with increased EUS tonic activity during bladder contraction which is the hallmark of DSD. (Fig 2B inset). SCC mice that received LM22B-10 treatment exhibited reduced numbers of non-voiding contractions that correlated decreased tonic activity (Fig. 2C) when compared to vehicle controls.

These data demonstrate that activation of TrkB/C signalling by LM22B-10 reduced bladder overactivity and DSD in the contused mouse model. Our data showed that contusion injury resulted in the formation of large scar regions and secondary cysts within the spinal cord that can prevent neural regeneration. This scar region was smaller in LM22B-10 treated mice which suggests that the drug may be eliciting its positive effects by reducing deposition of inhibitory extracellular proteins around the glial scar and/or promoting neural growth.

LM22B-10 treatment improved hindlimb function and reduced the development of bladder overactivity and DSD in chronic SCC mice that correlated with decreased spinal cord scar volume and secondary injuries. Targeting neurotrophin signalling pathways could hold significant potential as a treatment for neurogenic LUT dysfunction.

Frias B et al., J Neurosci.;35(5):2146-60, 2015
Matyas J et al., J Neursci., 37(14):3956-3971, 2017
Basso et al., J Neurotrauma, 23(5):635-59, 2006