Phenotypic Changes of Detrusor PDGFR Alpha Positive Cells in Spinal Cord Injury-Induced Detrusor Overactivity


 Volume accommodation occurs via a novel mechanism involving interstitial cells in detrusor muscles. The interstitial cells in the bladder are PDGFRα+ , and they restrain the excitability of smooth muscle at low levels and prevent the development of transient contractions (TCs). A common clinical manifestation of spinal cord injury (SCI)-induced bladder dysfunction is detrusor overactivity (DO). Although a myogenic origin of DO after SCI has been suggested, a mechanism for development of SCI-induced DO has not been determined. In this study we hypothesized that SCI-induced DO is related to loss of function in the regulatory mechanism provided by PDGFRα+ cells. Our results showed that transcriptional expression of Pdgfra and Kcnn3 was decreased after SCI. Proteins encoded by these genes also decreased after SCI, and a reduction in PDGFRα+ cell density was also documented. Loss of PDGFRα + cells was due to apoptosis. TCs in ex vivo bladders during filling increased dramatically after SCI, and this was related to the loss of regulation provided by SK channels, as we observed decreased sensitivity to apamin. These findings show that damage to the mechanism restraining muscle contraction during bladder filling that is provided by PDGFRα + cells is causative in the development of DO after SCI.


INTRODUCTION
As the bladder fills with urine the volume increases, but during much of the filling period, intravesical pressure remains low 1 . This accommodation occurs even though there is a natural tendency for the detrusor smooth muscle cells (SMCs) to contract in response to stretch 2,3 . In fact, during bladder filling non-voiding contractions (NVCs), detected as transient increases in intraluminal pressure, occur in cystometric records from all species including human. NVCs appear to correspond to localized contractions that are also observed in ex vivo bladder preparations and have been termed as 'spontaneous phasic contractions', 'micromotions' or 'transient contractions' [4][5][6][7][8] . Transient contractions (TCs) increase as bladder filling proceeds 4 . Prior experiments have shown that TCs are initiated by stretch-dependent non-selective cation channels expressed by detrusor SMC 2,3 . Inward currents generated by these channels depolarize SMCs and activate Ltype Ca 2+ channels, causing generation of Ca 2+ action potentials, Ca 2+ entry into SMCs and contraction. Action potentials propagate to other SMCs within the same muscle bundle, but do not spread to adjacent muscle bundles.
Several studies suggest that TCs initiate afferent nerve activity and provide a major source of the sensory information conveyed to the central nervous system during bladder filling [9][10][11][12] . A recent study has clearly demonstrated the link between TCs and sensory output from bladder 4 . Increased TCs may correspond to the sensory and mechanical behaviors associated with detrussor overactivity (DO). Normal bladders have the means to restrain development of TCs, but experiments have not been sufficiently rigorous to reveal the mechanisms responsible for restraining bladder excitability and the development of TCs during filling.
We discovered a novel mechanism involving interstitial cells in detrusor muscles.
Interstitial cells of the bladder were previously identified as c-Kit + cells and thought to provide excitatory input to the detrusor 13,14 , but more recent immunohistochemical evaluation showed few c-Kit + cells, other than mast cells, in the bladder of several species 15 . In fact, the interstitial cells in the bladder are PDGFRα + , and they provide inhibitory regulation of detrusor muscles [16][17][18][19][20] . Inhibitory regulation is enhanced by purines and by stretch, making it an ideal mechanism for regulating detrusor excitability and restraining TCs during bladder filling. Rigorous confirmation of the hypothesis that inhibitory regulation during bladder filling is provided by PDGFRα + cells would be to demonstrate that pathological conditions in which DO develops are associated with loss or remodeling of PDGFRα + cells. Therefore, we evaluated the status of PDGFRα + cells in spinal cord injury (SCI) animal models that are known to develop DO.
Clinical manifestations of SCI-induced bladder dysfunction involve a combination of storage and voiding problems. A myogenic origin of DO after SCI has also been suggested due to abnormal muscle reactivity ("the myogenic hypothesis") 21 without studying precise mechanisms for DO after SCI. Since obtaining human whole bladder due to SCI is extremely difficult, we investigated the mechanisms of DO in the acute phase of SCI using murine model.

Changes in protein expression in SCI-induced detrusor muscles
Three approaches were used to characterize changes in protein expression. Firstly, immunohistochemistry was used to examine the expression and distribution of PDGFRα and SK3 immunoreacitvity in SCI. PDFGRα and SK3-like immunoreactivity decreased in detrusor muscles after SCI in a time dependent manner (Fig. 2). We also used PDFGRα/eGFP mice to examine the density of green nuclei in control and SCI. The density of eGFP were decreased in SCI (3 day and 7day) compared to control. Since SCI can show bladder distension, we normalized the eGFP expression by calculated the surface area (e.g. 63 mm 2 in control vs 111 mm 2 in SCI, n=3, respectively). PDGFRα + cells also decreased to 53% (3 day) and 35% (7 day) after SCI in PDGFRα/eGFP mice, ( Fig. 3A-C). We confirmed these findings by Western analysis and verified reduction in PDGFRα protein in detrusor muscles of SCI mice (n= 4, Fig. 3D & E). These findings were consistent with the transcriptional changes observed after SCI (see Fig. 1), and suggest overall reduction in PDGFRα + cells and reduced expression of SK3 that would negatively impact the regulation of excitability provided by PDGFRα + cells in the bladder.

Apoptosis of PDGFRα + cells in SCI
We examined changes in the expression of apoptosis pathways to better understand the fate of PDGFRα + cells after SCI. RNA-seq of whole muscle samples showed geneset scores computed for the apoptosis-related KEGG and GOBP terms, respectively, using the FAIME algorithm 22 . The apoptosis-related geneset score was significantly increased (t-test: P <0.05) in detrusor muscles after SCI (Fig. 4A). Apoptosis-related genes, including Apaf1, Capns1, and Casp3, were significantly upregulated (FC>1.5 and FDR <0.05) in detrusor muscles 3 days after SCI (Fig. 4B). Time dependent increases in expression of Apaf1 (Fig. 4C) and Caspase3 (Fig, 4D) in detrusor muscles after SCI (as compared to sham; n=4 for each period) were confirmed by qPCR.

Ex vivo preparation to confirm the role of PDGFRα + cells in SCI
Ex vivo bladder preparations were used to characterize the relationship between intravesical volumes and pressures in bladders after SCI. Pressure-volume studies done ex vivo exclude extrinsic neural reflexes during filling. SK channels are highly expressed in detrusor PDGFRα + cells, and antagonists of these channels increase TCs during filling 17 . We examined the effects of the SK channel antagonist, Apamin (300nM) on bladders from control and SCI (up to 1 month).
Apamin (300 nM added to the bathing solution) increased the amplitude (6.9±1.2 cmH2O, P<0.01) and frequency (65±12 events, P<0.01, n=11) of TCs during bladder filling, as compared to control (2.1±0.5 cmH2O in amplitude and 22±9.1 events in frequency; Fig.   5A, Table 1). At 1, 3, 7, 14 and 30 days after SCI, the amplitude and frequency of TCs during filling were increased ( Fig. 5B-F, Table 1). Enhanced TCs during filling persisted for at least 30 days after SCI, and apamin failed to induce significant changes of amplitude (from 7 days) and frequency (from 1 day) of TCs after SCI ( Fig. 5B-F, Table 1) indicating that development of DO after SCI progressed for up to 1 month after SCI. Bladder capacity and filling time to reach 30cmH2O were also increased in all of SCI groups compared to sham ( Table 2). The enlarged bladder capacity due to lack of voluntary voiding and prolonged infusion time was prominent in 30 days.

In this study we investigated responses of the murine bladder and the status of
PDGFRα + cells after SCI. PDGFRα + cells and the functions provided by these cells in regulating bladder contractions during filling were greatly decreased after SCI. TCs increased dramatically during bladder filling, and accompanying this change in function, Pdgfra and Kcnn3 expression decreased. Expression of PDGFRα and SK3 protein also decreased. Detrusor muscles displayed apoptotic loss of PDGFRα + cells. The increase in TCs was due to decreased sensitivity to apamin during bladder filling, which is consistent with the reduction in expression of SK3 and loss of PDGFRα + cells that express SK3 channels. These findings demonstrate a novel mechanism for development of DO after SCI that is linked to loss of the inhibitory regulation provided by PDGFRα + cells during bladder filling.
PDGFRα + cells regulate detrusor excitability during bladder filling 23 . SK channel antagonists potentiate the amplitude of spontaneous contractions in murine 24 , guinea pig 25 and human 26 detrusor muscles. The effects of apamin are likely due to blocking SK channels in PDGFRα + cells which have a high expression of SK3 channels, because the current density from SK channels is minimal in SMCs and not even resolvable in murine SMCs at physiological potentials 27 . Thus, PDGFRα + cells express a powerful mechanism to suppress excitability and allow the bladder to fill with minimal activation of TCs and sensory discharge 23  IHC showed that the distribution of PDGFRα + cells decreases in detrusor muscles after SCI. The immunohistochemistry findings were confirmed using a reporter strain of mice with expression of eGFP in nuclei of PDGFRα + cells, as a relatively low density of eGFP + cells was found in bladders of PDGFRα/eGFP mice after SCI. However, IHC is not reliable method to quantify loss of protein, so protein expression in extracts of detrusor muscles was measured by Western analyses. Western blots showed the downregulation of PDGFRα expression and SK3 expression which was mirrored by transcriptional expression.

Loss of PDGFRα + cells caused by SCI were due to apoptotic changes in detrusor
PDGFRα + cells although the mechanism inducing apoptosis in detrusor PDGFRα + cells by SCI has not been elucidated. Apaf1 as well as caspase3 had a trend to be increased after SCI compare to sham indicating that cell death occurred as a result of SCI. We speculate that apoptosis of PDGFRα + cells may be caused by the reduced expression and function of neurotropins 28,29 . The mechanism causing damage to the regulatory function provided by PDGFRα + cells after SCI is an important topic for future research.
Ex vivo preparations were used in the present study to isolate bladders and exclude connections from central and spinal reflexes. Release of mediators from the urothelium may also influence detrusor excitability, but substances and receptors involved in such a mechanism have not been identified. Ex vivo bladder developed increased TCs after SCI as compared to sham bladders (see Table 1), and enhanced TCs persisted for at least 30 days after SCI. Effects of apamin on responses to bladder filling remained for up to 72 hr after SCI, but the sensitivity to apamin decreased after SCI due to partial loss of PDGFRα + cells and downregulation of SK3 channels. These findings present in ex vivo bladders suggest that myogenic mechanisms are sufficient to generate DO.
SCI patients have less opportunity to see a urologist due to other complications that often need to be treated ahead of bladder dysfunction. This prevents patients from early and appropriate examination and treatment of lower urinary tract dysfunction in the early phase of SCI. Although there are many reports of DO after recovery from spinal shock [30][31][32][33][34][35] , only a few reports confirm development of DO in the acute phase of SCI (i.e. 3 -40 days after SCI) 36,37 . Early treatment to avoid higher intravesical pressure with lower bladder compliance followed by vesicoureteral reflux associated with DO has been suggested to keep patients' renal and bladder function serving as a 'low pressure tank' without waiting for the irreversible complications of SCI 36,37 . Given the importance of an early intervention for a treatment especially focusing on a myogenic aspect, preventing the phenotypic change of PDGFRα + cells and rescuing the function of SK channels in bladder PDGFRα + cells might be a promising target to avoid development of DO after SCI.

Spinal cord injury (SCI) animal model
All experimental procedures were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and the animal use protocol, reviewed and approved by the Institutional Animal Use and Care Committee at the University of Nevada. C57BL/6 (male mice, 8-12 wks old), Pdgfra tm11(EGFP)Sor /J (PDGFRα/eGFP, Jackson lab) and smMHC/Cre/eGFP (SMC/eGFP) from Jackson Lab used for SCI operations and age-matched control. Laminectomies were performed under isoflurane anesthesia (3 -4 % with a balance of oxygen for induction followed by 2 % for maintenance), and the spinal cord (T11-T13) were exposed without any damage or compression to the surrounding dura. Dumont #5 forceps were positioned in the middle of the exposed spinal cord segment to perform complete spinal cord transection. Complete spinal cord transection was done at T12 confirmed by retraction of rostal and caudal cut ends of spinal cord under surgical microscope, which had a space approximately at 2 mm. Control animals received sham operations with exposing the vertebrae at same level as SCI without damaging any spinal cord and dura. Enrofloxacin (5 mg/kg) was applied subcutaneously for three days after SCI followed by twice a week after SCI surgery until ex vivo or molecular evaluation was done. The bladder was manually squeezed to eliminate the residual urine of bladder once daily. Bladders were collected for experiments at 1, 3, 7, 14 and 30 days after SCI surgery and in sham control.

Molecular preparation
Dissection of detrusor smooth muscles and RNA isolation in sham and SCI were identical as previously described 18 .
For quantitative analysis of transcripts, PDGFRα + cells and SMC/eGFP were purified by fluorescence-activated cell sorting and detrusor muscles for molecular tests. Total RNA was isolated from detrusor muscles, Transcriptomes profiled by mRNA-seq (Novogene Co Ltd) were investigated to identify the genes and pathways potentially involved in the regulation of excitability in PDGFRα + cells upon SCI treatment. Total RNA was obtained from detrusor muscles in sham and SCI. Using the SAM tool 38 , the genes with false discovery rate (FDR) < 5% and fold change (FC) > 1.5% were deemed to be differentially expressed. The FAIME algorithm 22 was applied to assign gene expression-based geneset scores for the "apoptosis" related genes defined by both the Kyoto Encyclopedia of Genes and Genomes (KEGG) 39 and Gene Ontology (GO) 40 databases. The FAIME method generates geneset scores using the rank-weighted gene expression of individual samples, which determines whether an a priori defined set of genes shows statistically significant, concordant expression differences between two biological states (e.g. sham vs. SCI), and provides a mechanistic interpretation of the deregulated genes.

Automated Capillary Electrophoresis and Chemiluminescent Western Blotting
Muscles were snap-frozen in liquid N2, and stored at −80°C. For analysis, the methods for homogenization, centrifugation and collection of the supernatants were identical as previously reported 41,42

Drugs
All reagents were purchased from Sigma-Aldrich (St Louis, MO, USA) and apamin (Tocris, UK) solubilized in the bath solution for ex vivo recordings.

Statistical analyses
All data are expressed as means ± SEM. "n" represents the number of experiments.
All statistical analyses were performed using GraphPad Prism. A paired and unpaired Student's t test was used to compare groups of data and differences were considered to be significant at P < 0.05. This study was carried out in compliance with the ARRIVE guidelines.  Far fewer eGFP + nuclei (reporter for PDGFRα + cells) were found in detrusor muscles after SCI. D. Representative Wes full length gel image of PDGFRα expression in murine detrusor muscles following SCI. 100,000 x g pellet, 1µg/lane, γ-action was used for normalization. E. Normalized signal intensities of PDGFRα by g-actin following SCI periods.      PDGFRα immunoreactivity of detrusor muscle layer in control and SCI. Immunoreactivity of PDGFRα (green) and SK3 (red) in control (sham). SCI (24hr, 48hr and 72 hr).

Figure 3
Density of PDFGRα/eGFP cells and Microcapillary electrophoresis and immunodetection of PDGFRα by WES in control and SCI detrusor muscles. A-C. Far fewer eGFP+ nuclei (reporter for PDGFRα+ cells) were found in detrusor muscles after SCI. D. Representative Wes full length gel image of PDGFRα expression in murine detrusor muscles following SCI. 100,000 x g pellet, 1μg/lane, γ-action was used for normalization. E. Normalized signal intensities of PDGFRα by g-actin following SCI periods.  The effects of apamin on transient contractions (TCs) of sham, 1,3,7,14 and 30day after SCI using ex vivo preparation. A-F. Ex vivo pressure-response curve for control and apamin application. Aa,b-Fa,b. Expanded time scales with adjustment of baseline under control (a) and apamin (b) from above panels.