The external urinary sphincter (EUS) acts cooperatively with the bladder to store and periodically eliminate urine. However, in neurogenic lower urinary tract dysfunction induced by spinal cord injury (SCI), this cooperative micturition control is disrupted, leading to detrusor overactivity (DO) and detrusor-sphincter dyssynergia (DSD), which causes inefficient voiding with high post-void residual volume. Previous studies showed that treatments targeting C-fiber bladder afferent pathways such as anti-NGF treatment or capsaicin-induced desensitization do not improve DSD [1], which led to a hypothesis that Aδ-fiber bladder afferents are responsible for the EUS hyperexcitability causing DSD.  We previously reported that the herpes simplex virus (HSV)vector-mediated treatment targeting Aδ-fiber bladder afferents with double mutant glycine (G2M) receptors with increased sensitivity to ivermectin (IVM), an FDA-approved anti-parasitic drug, can improve EUS activity, resulting in increased voiding efficiency (2019 ICS meeting). Activation of G2M receptors by IVM has been shown to induce a chloride ion influx and causes hyperpolarization to reduce excitability of neuronal cells [2]. In this study, we extended our previous study to examine the relationship between functional and molecular changes in SCI-induced voiding dysfunction with DSD and the electrical properties of bladder afferent neurons in SCI mice treated with G2M-encoding HSV vectors driven by neurofilament 200 promoter (NF200p) that can selectively target A-fiber afferent pathways.

All experiments were conducted in accordance with institutional guidelines and was approved by the institutional animal care and use committee. First, 8-9 weeks-old female C57BL/6N mice (n=30) were divided into 3 groups; (A) spinal intact (SI) mice, (B) SCI mice+NF200p-HSV vectors encoding wild-type glycine receptors (WT-GlyR) with IVM administration, (3) SCI mice+NF200p-HSV-G2M vectors with IVM administration. In SCI groups, the Th8/9 spinal cord was transected under isoflurane anesthesia. Thereafter, the bladder of SCI mice was emptied by perineal stimulation and bladder compression daily for 4 weeks post-SCI. Two weeks after SCI, NF200p-driven HSV vectors expressing G2M or WT-GlyR were inoculated into the bladder wall. One week after vector inoculation, IVM (50μl of 1μM solution) was intra-peritoneally injected daily for 7 days. Thereafter, SCI mice underwent single-filling cystometry (CMG) and EUS-electromyogram (EMG) recordings under an awake condition, followed by the removal of L6/S1 dorsal root ganglia (DRG) for RT-PCR studies of mechanosensitive channels such as ASIC 1-3 & Piezo2 and a C-fiber afferent marker, TRPV1. 
Next, another set of female mice (n=24) were used for patch-clamp recordings and divided into 2 groups; (1) control SCI mice and (2) SCI mice+NF200p-HSV vectors encoding G2M receptors. SCI mice in Groups 1 and 2 underwent inoculation of a total of 20μl of 2.5% fluoro-gold (FG) and 3x107 plaque-forming units of NF200p-driven HSV vectors expressing G2M receptors, respectively, into the bladder wall. At 4 weeks after SCI, L6-S1 DRG, which contain cell bodies of bladder afferent neurons carried through the pelvic nerve, were dissected under isoflurane anesthesia, and enzymatically dissociated into single neurons [3]. FG-labelled afferent neurons that innervate the bladder were identified using an inverted phase contrast microscope with fluorescent attachments, and cells with a diameter larger than 35μm were selected to assess the electrical properties of presumed Aδ-fiber bladder afferent neurons, which have a larger cell size than C-fiber afferent neurons. Whole-cell patch-clamp recordings were performed at room temperature on FG-labelled neurons within 24 hours after dissociation. We evaluated the characteristics of action potentials in bladder afferent neurons from 2 groups of SCI mice as well as spinal intact (SI) control mice. After evaluating action potential characteristics including the rheobase current that is the depolarizing current size required for action potential activation, IVM (10μM/50μL) was directly applied into the cell dish, and action potentials were again induced by depolarizing currents 60 seconds later to assess the rheobase current changed by IVM-induced G2M receptors activation in large-sized bladder afferent neurons.

Compared to the WT-GlyR and IVM group, treatment with HSV-G2M vectors and IVM application in SCI mice significantly increased the EMG activity reduction time, leading to improvement of voiding efficiency and residual volume without affecting non-voiding contractions during bladder filling (Fig. 1-A, B, C). RT-PCR showed reductions of ASIC 1-3 & Piezo2, but not TRPV1, in L6-S1 DRG (Fig. 1-B). Patch clamp recordings showed that FG-labelled, large-sized bladder afferent neurons show a tendency of lower rheobase current sizes for action potential activation in SCI mice (Group 1) compared to SI rats (52±8.4pA vs. 56±5.5pA, p=0.20).  Then, when IVM was applied to large-sized bladder afferent neurons, the rheobase current required for action potential activation was significantly increased from 46±4.0pA to 62±3.7pA (p<0.01) in SCI mice treated with HSV-NF200p-G2M vector (Group 2) whereas the rheobase current was not altered after IVM application in SI or SCI mice (Group 1) (Fig. 1-C).

The results of this study indicate that the improvement of inefficient voiding and DSD after IVM application in SCI mice treated with A-fiber-targeting HSV-NF200p-G2M vectors was associated with reductions in expression levels of mechanosensitive channels such as ASIC 1-3 and Piezo 2, but not TRPV1, in L6-S1 DRG that contain bladder afferent neurons. It has been shown that ASICs and Piezo2 act as mechanosensitive channels in afferent pathways and that ASIC1-3 receptors are expressed in TRPV1 expressing, unmyelinated C-fiber neurons as well as in mechanosensitive, myelinated A-fiber neurons, whereas Piezo2 are expressed in mechanosensitive, myelinated A-fiber neurons [1]. Also, in patch-clamp recordings, IVM application decreased the cell excitability of large-sized, presumed Aδ-fiber bladder afferent neurons obtained from SCI mice treated with HSV-NF200p-G2M vectors, as evidenced by the increased rheobase current for action potential activation after IVM application.  Taken together, it is plausible that activation of G2M receptors expressed in Aδ-fiber bladder afferent pathways by exogenous application of IVM suppresses Aδ-fiber bladder afferent activity in association with reduced expressions of mechanosensitive channels to improve inefficient voiding and DSD in SCI mice.

A druggable approach using exogenous application of specific ligands such as IVM can be achieved by gene delivery of designer receptors such as IVM-sensitive mutant glycine receptors (G2M) to Aδ-fiber bladder afferent pathways for the treatment of SCI-induced voiding dysfunction and DSD by using HSV vectors driven by an A-fiber-targeting promoter (NF200p).

Wada N et al. Therapeutic effects of inhibition of brain-derived neurotrophic factor (BDNF) on voiding dysfunction in mice with spinal cord injury.  Am Journal Physiol Renal Physiol (2019); 317: F1305-F1310
Lynagh T, et al. An improved ivermectin-activated chloride channel receptor for inhibiting electrical activity in defined neuronal populations. J Biol Chem (2010);285:14890-7
Shimizu T et al. Nerve growth factor-dependent hyperexcitability of capsaicin sensitive bladder afferent neurones in mice with spinal cord injury. Exp Physiol (2018); 103: 896-904