Attenuated lipopolysaccharide-induced inflammatory bladder hypersensitivity in mice deficient of transient receptor potential ankilin1

Transient receptor potential ankyrin 1 (TRPA1) channel expressed by urothelial cells and bladder sensory nerve fibers might act as a bladder mechanosensor and nociceptive transducer. To disclose the role of TRPA1 in bladder function and inflammation-associated hypersensitivity, we evaluated in vitro and in vivo bladder function and inflammatory mechanosensory and nociceptive responses to intravesical lipopolysaccharide (LPS)-instillation in wild type (WT) and TRPA1-knock out (KO) mice. At baseline before treatment, no significant differences were observed in frequency volume variables, in vitro detrusor contractility, and cystometric parameters between the two groups in either sex. LPS-instillation significantly increased voiding frequency and decreased mean voided volume at 24–48 hours after instillation in WT but not in TRPA1-KO mice. LPS-instillation also significantly increased the number of pain-like behavior at 24 hours after instillation in WT mice, but not in TRPA1-KO mice. Cystometry 24 hours after LPS-instillation revealed shorter inter-contraction intervals in the WT mice compared with TRPA1-KO mice. In contrast, inflammatory cell infiltration in the bladder suburothelial layer was not significantly different between the two groups. These results indicate that TRPA1 channels are involved in bladder mechanosensory and nociceptive hypersensitivity accompanied with inflammation but not in physiological bladder function or development of bladder inflammation.

any of the cystometric parameters 16 . Another study demonstrated no functional changes in urethane-anesthetized cystometry in TRPA1-KO mice 17 .
In this study, we explored in vitro and in vivo bladder functional phenotypes of TRPA1-KO mice to understand the physiological role of TRPA1 in regulating bladder function. We also compared changes in bladder histology, mechanosensation, and nociception induced by intravesical instillation of lipopolysaccharide (LPS) between wild type (WT) and TRPA1-KO mice to reveal the role of TRPA1 in inflammation and mechanosensory and nociceptive bladder hypersensitivity due to inflammation.

Results
Phenotypic bladder function of TRPA1-KO mice in the absence of LPS instillation. Frequency/ volume measurements. There were no significant differences in body weight between WT and TRPA1-KO mice subjected to frequency/volume measurements for either sex (male, 24.4 ± 0.5 g vs. 23.3 ± 0.4 g, female, 19.4 ± 0.2 g vs. 19.0 ± 0.4 g, respectively; n = 10-22 per group). There were also no significant differences in any of the frequency/volume parameters between the two groups in either sex (Table 1). Mean flow rate was significantly higher in female mice in both WT and TRPA1-KO mice (Supplementary Table S1).
Decerebrated unanesthetized cystometry (CMG) measurements. There were no significant differences in any of the CMG parameters between WT and TRPA1-KO mice in either sex (n = 8-15 per group; Table 2). Inter-contraction interval in TRPA1-KO mice and mean voided volume in both WT and TRPA1-KO mice were significantly greater in female compared to male mice (Supplementary Table S1).
In vitro functional assessment of detrusor strips. There were no significant differences in the contractile responses to high K + (124 mM KCl) (n = 24-30 from n = 8-10 per group), or carbachol (CCh, 10 −8 -10 −3 M) (n = 8-10 from n = 8-10 per group) in either sex, or adenosine triphosphate (ATP,: 10 −6 -10 −2 M) in male mice (n = 8-10 from n = 8-10 per group) between WT and TRPA1-KO mice ( Fig. 1A-C). In contrast, the contractile responses to ATP in female TRPA1-KO mice were significantly weaker than those in female WT mice (n = 8 from n = 8 per group, Fig. 1C). There were no significant differences in the amplitude of contractions induced by electric field stimulation (EFS) at any frequency in either sex (n = 8-10 from n = 8-10 per group). Even after atropine administration, no significant differences were detected between the two groups in either sex.
In contrast, after repeated α, β-methylene-ATP (mATP) administration in female mice and tetrodotoxin (TTX) administration in male mice, the amplitude of contractions in TRPA1-KO mice were significantly smaller than that in WT mice in the same sex, although mATP and TTX administration markedly reduced contractility relative to baseline regardless of sex or genotype (Fig. 1D, Supplementary Fig. S1).
Inflammatory and nociceptive responses to LPS instillation. Because no phenotypic differences in bladder function were observed between WT and TRPA1-KO mice in either sex, inflammatory and nociceptive responses to LPS instillation were evaluated only in female mice. Frequency/volume measurements. There were no significant differences in voiding frequency or voided volume between the groups at baseline. In WT mice, LPS instillation led a significant increase in voiding frequency and a significant decrease in mean voided volume at 24-48 hours after instillation compared with the saline instillation (n = 6 per group). In contrast, in TRPA1-KO mice, there were no significant changes in either voiding frequency or mean voided volume after LPS instillation compared with the baseline (n = 7; Fig. 2A).
Bladder pain-like licking behavior. WT mice with LPS instillation exhibited a significant increase in the number of licking behavior at 24 hours after instillation compared with the baseline and the WT mice with saline instillation. In contrast, TRPA1-KO mice did not exhibit significant changes in the number of licking behavior compared with baseline at 24 hours after LPS instillation and significantly less behavior than the WT mice at 48 hours after LPS instillation (n = 6 per group; Fig. 2B).
Decerebrated unanesthetized CMG measurements. CMG measurements performed at 24 hours after LPS instillation revealed that interaction between LPS-instillation and genetically TRPA1-KO in inter-contraction interval and voided volume. Moreover, LPS-instilled WT mice exhibited significantly shorter inter-contraction intervals and lower mean voided volume than their baseline data. In contrast, LPS-instilled TRPA1-KO mice did not demonstrate any significant changes compared with their baseline data before LPS-instillation. TRPA1-KO mice showed significantly longer inter-contraction intervals and higher mean voided volumes than the LPS-instilled WT mice (n = 7 per group; Table 3 and Fig. 3).
Histological evaluation. Histological evaluation of the bladder specimens at 24 hours after saline or LPS instillation revealed that inflammatory cell infiltration and edema in the suburothelial layer were comparable in the bladders of both WT and TRPA1-KO mice received LPS instillation ( Fig. 4B,C,E,F). In contrast, neither inflammatory cell infiltration nor edema was observed in the bladder of saline-instilled WT mice (Fig. 4A,D). Compared to those of saline-instilled WT mice, the number of infiltrating inflammatory cells was significantly higher in the bladders of LPS-instilled WT and TRPA1-KO mice (p = 0.0009 and 0.034, respectively), which was not significantly different between the WT and TRPA1-KO mice receiving LPS instillation (p = 0.24; n = 7 per group; Fig. 4G).
mRNA expression of TRP channels in bladder and DRG. TRPA1 mRNA was not expressed in the bladders or the L6 DRGs of TRPA1-KO mice. Two-way ANOVA test revealed that there were no significant interaction effects between LPS instillation and genetic deletion of TRPA1 in the mRNA expression changes of any TRP channels. The expression of other TRP channels in the bladders or L6 DRGs were not significantly different between the WT and TRPA1-KO mice (n = 5 per group; Fig. 5A-C). LPS instillation significantly increased TRPM2 expression in the bladders compared to those with saline instillation (Fig. 5B), which was not observed in the L6 DRGs (Fig. 5C). LPS instillation did not affect the expressions of any of the other TRP channels significantly in the bladder or L6 DRGs.

Discussions
This study revealed that baseline bladder functions in TRPA1-KO mice were similar to those of WT mice and that TRPA1-KO mice exhibited less hypersensitive responses to LPS instillation, with histological inflammatory changes indistinguishable from those of WT mice. These results suggested that TRPA1 channels play minimal roles in physiological control of bladder function or development of LPS-induced bladder inflammation, while contributing to inflammation-induced bladder hypersensitivity in mice.
In vitro functional analysis of detrusor strips demonstrated that there were no significant differences in the contractile responses to high K + , CCh and EFS before drug administration and after atropine administration between WT and TRPA1-KO mice in either sex. These results suggested that intrinsic and cholinergic contractile property of detrusor smooth muscle were not different between WT and TRPA1-KO mice. Although female TRPA1-KO mice showed significantly lower contractile responses to ATP, this difference in the purinergic contractile property between WT and TRPA1-KO mice may not influence total contractile function.   A previous CMG study in conscious free-moving rats revealed no effect on any CMG parameters with application of a TRPA1 antagonist 16 , and another CMG study showed that none of the CMG parameters in urethane-anesthetized TRPA1-KO mice were different from those of the WT mice. We detected no significant differences in in vivo measurements of frequency/volume and CMG between the WT and TRPA1-KO mice in either sex. These findings indicate that TRPA1 channels are unlikely to contribute to the physiological control of the bladder.
We then examined whether TRPA1 channels were involved in the development of bladder inflammation and inflammation-associated hypersensitive responses induced by intravesical instillation of LPS. LPS instillation in WT mice significantly increased the number of bladder pain-like behavioral events at 24 hours after instillation. LPS instillation additionally led an increase in voiding frequency and a decrease in mean voided volume at 24-48 hours after instillation in WT mice. In contrast, TRPA1-KO mice showed weaker responses as determined by bladder pain-like behavior and frequent voiding provoked by LPS instillation at 24-48 hours after instillation. CMG measurements showed longer inter-contraction intervals and higher mean voided volumes in TRPA1-KO mice than those in WT mice, supporting the results of the frequency/volume measurements. Nevertheless, histological evaluation of inflammatory response to LPS instillation revealed that the infiltrating inflammatory cell numbers were similar between the WT and TRPA1-KO mice. Taken together, these results suggested that TRPA1 channels may not contribute to inflammation itself but were involved in increased nociception and mechanosensory responses induced by bladder inflammation. A previous study demonstrated that LPS directly activates TRPA1, which is involved in immediate nociceptive responses to LPS, independently of Tall -like receptor 4 18 . We cannot deny that the direct activation of TRPA1 by LPS contributed to the differences of the differences of nociceptive and micturition behaviors, even though our study evaluated the behavioral changes at 24 hours or later after LPS stimulation but not immediately after its instillation. Further studies evaluating the differences of immediate responses to LPS between WT and TRPA1-KO mice may be needed.
Previous rodent studies reported that TRPA1 antagonists attenuated bladder hyperalgesia and bladder hyperreflexia induced by cyclophosphamide or ifosfamide in inflammatory cystitis models 14,15,19 , which support the findings of the present study utilizing genetic deletion of TRPA1. As TRPA1-KO mice exhibited no detectable changes in baseline bladder functions, we speculated that there might be a compensatory increase in the expression of other TRP channels in TRPA1-KO mice. Previous studies, however, demonstrated that mRNA expression levels of TRPV1 and TRPM8 were not different between the TRPA1-KO and WT mice 20,21 . In the present study, mRNA expression levels of several TRP channels that could possibly be associated with bladder function, including TRPV1, TRPV4, TRPM2, TRPM7, and TRPM8, did not change in the bladder or the L6 DRGs in TRPA1-KO mice. Given that bladder hypersensitivity-associated behaviors were induced after the instillation, we expected an increase in expression of TRPA1channels in WT mice after LPS instillation. However, mRNA expression of TRPA1 did not change in the bladder or the L6 DRGs at least at 24 hours after LPS instillation. DeBerry et al. reported that the mRNA expression level of TRPA1 in L6-S1 DRGs of mice did not change in response to repeated cyclophosphamide administration 14 ; however, in the same cystitis model, they also observed an upregulation of TRPA1 mRNA expression by single-cell reverse transcription-polymerase chain reaction (RT-PCR) of neuronal cells in L5-S1 DRGs innervating the bladder 22 . A more detailed examination such as evaluation by single-cell RT-PCR of the neuronal cells in L6 DRGs innervating the bladder is needed to understand the precise roles of TRPA1 in DRG.
Interestingly, we found that the TRPM2 mRNA expression was upregulated at 24 hours after LPS instillation in both the WT and TRPA1-KO bladders. TRPM2 channels were reported to be expressed abundantly in immune cells including monocytes, macrophages, and neutrophils, contributing to inflammatory and neuropathic pain 23,24 . Thus, our observation regarding TRPM2 mRNA expression is consistent with those reports, with concomitant histological inflammatory changes in both mouse lines. The contribution of TRPM2 channel to physiological and pathophysiological bladder functions require further investigation.
Our study has several limitations. First, potential compensatory changes in bladder function due to genetic deletion of TRPA1, which could mask the actual functional roles of TRPA1 cannot be excluded. Second, our study lacked the data of the inflammatory and nociceptive responses of TRP1-KO mice with saline instillation. TRP1-KO mice with saline instillation would be expected to show the similar results to WT mice with saline instillation, because there were little phenotypic differences in physiological condition between WT and TRPA1-KO mice. However, comparing 4 groups including TRP1-KO mice with saline instillation may be more suitable. Moreover, we evaluated changes by cystometry, histology, and RT-PCR only at 24 hours after LPS instillation. Evaluation at multiple time points is needed for precise determination of differences between WT and TRPA1-KO mice. Third, the degree of inflammatory changes in bladder tissue was quantified only by counting the number of infiltrating inflammatory cells by hematoxylin-eosin (H-E) staining and the types of inflammatory cells were not identified. Immunohistochemistry staining and digital image analysis could evaluate quantitative evaluation of each inflammatory cells in the bladder tissue more precisely, which may help to reveal the differences of inflammatory changes between WT and TRAP1-KO mice 13,25 . Finally, we did not conduct direct evaluations of bladder afferent nerve activity to assess bladder sensory function or of elevation of the oxdative stress markers such as hydrogen peroxide in the bladder tissue with LPS-induced inflammation. Further investigations including these evaluations are needed.
In conclusion, this is the first study demonstrating that TRPA1 channels contribute to inflammation-associated bladder mechanosensory and nociceptive hypersensitivity using TRPA1-KO mice. These findings suggest TRPA1 channels as promising pharmacological target for the treatment of inflammatory hypersensitive bladder disorders.

Animals.
A total of 19 male and 42 female TRPA1-KO mice and 18 male and 46 female WT littermates (10-12-weeks old) were used. Mice were maintained under standard laboratory conditions with a 12-h light (9:00 a.m. to 9:00 p.m.) and 12-h dark (9:00 p.m. to 9:00 a.m.) cycles with free access to food and water. Experimental protocols were approved by the animal ethics committee of the University of Tokyo Graduate School of Medicine in Tokyo, Japan, and were in line with the NIH guidelines for the care and use of experimental animals. TRPA1-KO mice were generated as reported previously 26 . The TRPA1-KO mouse line obtained from Jackson Laboratory (Bar Harbor, ME) was backcrossed with C57BL/6 J mice for at least ten generations to eliminate any background effects on the phenotypes.

Induction of cystitis.
Under isoflurane anesthesia, a polyethylene catheter (PE-10; Clay Adams, Parsippany, NJ, USA) was inserted transurethrally into the bladder. After draining urine, 100 µL saline containing 2.0 mg/ml LPS extracted from Escherichia coli O55:B5 was instilled intravesically for one hour. As a control, vehicle (saline) was instilled instead of LPS. During the 1-hour instillation period, LPS or saline was exchanged to a new solution at 30 minutes after instillation to avoid dilution with urine. After instillation, the solution was drained, and the mouse was allowed to recover from anesthesia. LPS dose was determined based on a previous study 27 and our pilot investigation.
Frequency/volume measurements. Each conscious mouse was separately placed in an MCM/ TOA-UF001-006 metabolic cage (Mitsubishi Chemical Medience, Tokyo, Japan) without any restraint, as reported previously [28][29][30] . After a 24-hour adaptation period, voided volume, voiding frequency, mean flow rate, and water intake volume were continuously recorded for 24 hours with a PowerLab (AD Instruments, Sydney, Australia) data acquisition system. All mice had free access to water and food during recordings. For evaluation of responses to LPS, after baseline measurements of voiding behavior for 24 hours, the mice received intravesical LPS or saline instillation and were placed back into the cage. After a 24-hour adaptation period, voiding behavior was monitored for another 48 hours.
Bladder pain-like nociceptive behavior. Assessment of bladder pain-like behavior was performed as described previously, with minor modifications 31,32 . Each mouse was placed under a lucent plastic box and habituated to the environment for one hour. Next, bladder pain-like behavior such as licking or biting of the skin of the lower abdomen close to the bladder was monitored on a video-camera (HDR-XR500V, Sony Corporation, Tokyo, Japan) for the next 30 minutes as baseline 33 . Thereafter, the mice received intravesical LPS or saline instillation, and bladder pain-like behavior was monitored again for 30 minutes at 24, 48, and 72 hours after instillation. Number of behaviors for each mouse was counted by two investigators in a blinded manner, and the values were averaged.
Decerebrated unanesthetized CMG measurements. CMG measurements were performed in decerebrate, unanesthetized mice according to the previously reported methods (20). In brief, under isoflurane anesthesia, a PE-50 catheter (Clay Adams) was implanted into the bladder through the dome. Skin from the scalp was removed, and skull was opened, and decerebration was performed at pre-collicular position. To eliminate the potential influence of anesthesia, more than two hours after surgery, the intravesical catheter was connected via a three-way stopcock to a pressure transducer (DX-100, Nihon Kohden, Tokyo, Japan) and a syringe pump (KDS 200, Muromachi Kikai, Tokyo, Japan) for saline instillation at 10 µl/min. Intravesical pressure and voided volume were recorded continuously using a data acquisition program (PowerLab). After stable and reproducible recordings were obtained, micturition cycles during a 1-hour period were averaged, and the following cystometric parameters were analyzed: basal pressure (minimum bladder pressure), threshold pressure (bladder pressure at onset of micturition), maximum pressure (maximum bladder pressure during micturition), voided volume, residual volume (infusion volume-voided volume) and inter-contraction interval (duration between two micturitions) 34 .
In vitro functional studies of detrusor strips. In vitro assessment of detrusor muscle function was performed as described in our previous reports 28,30,35 . After mice were sacrificed by pentobarbital sodium overdose, bladder bodies were harvested, and longitudinal, full-thickness detrusor strips were transferred to 5 ml organ baths. After a 2-hour equilibration period with a 10 mN stable tension, functional assessment was initiated. Briefly, the strip was first exposed to a high K + -containing (124 mM KCl) Krebs solution. Then, contractile responses to the following stimuli were examined: CCh (10 −8 -10 −3 M), ATP: 10 −6 -10 −2 M, electric field stimulation (EFS; pulse width, 0.8 milliseconds, 50 V; pulse duration, 10 seconds; stimulation interval, 2 minutes at 2, 4, 8, 16, 32, and 48 Hz). After baseline EFS measurement, contractions were evoked by EFS again after 10 −6 M atropine administration, purinoceptor desensitization by repeated administrations of 10 −5 M mATP, and finally 10 −6 M TTX.
Histological examination. After mice were sacrificed by pentobarbital sodium overdose, bladders were harvested and opened at the dorsal side of the bladder neck along the midline toward the dome. Isolated bladder specimens were fixed in 4% paraformaldehyde-phosphate buffered saline, embedded in paraffin, and cut into 3-μm sections. Histological evaluation was performed with hematoxylin-eosin (H-E) staining, and three sections of submucosal layer with maximum infiltration of inflammatory cells per bladder were chosen by a double-blinded evaluator. Thereafter, number of infiltrating inflammatory cells was counted using a counting grid at 400× magnification by two investigators, and the values were averaged.
Real-time RT-PCR. After mice were sacrificed with pentobarbital sodium overdose, bladders and bilateral L6 DRGs were dissected and immediately placed in RNAlater RNA stabilization reagent (QIAGEN, Venlo, Netherland). Total RNA extracted from tissues using miRNeasy Mini kit (QIAGEN) was reverse-transcribed to cDNA using SuperScript VILO Master Mix (Life Technologies, Carlsbad, CA, USA) according to the manufacturer's protocol. For relative quantification of mRNA expression, real-time RT-PCR was performed using Power SYBR Green PCR Master Mix and gene specific primers with the StepOnePlus Real-Time PCR System (Life Technologies). Primer sequences are provided in Supplementary Table S2. Drugs. CCh, ATP, atropine, and TTX were purchased from Wako Chemical (Tokyo, Japan). LPS, and mATP were purchased from Sigma-Aldrich (St. Louis, MO, USA). Krebs solution was composed of 118 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl 2 , 25.0 mM NaHCO 3 , 1.2 mM KH 2 PO 4 , and 11 mM glucose (pH 7.4).
Statistical analysis. All data were expressed as means ± standard error of means. Dose responses for CCh and ATP, and frequency responses for EFS were analyzed by f-test of nonlinear regression, and other results were analyzed using unpaired Student's t test, Tukey's test, the Mann-Whitney U test, repeated measures analysis of variance (ANOVA) followed by Dunnett's post hoc analysis, or two-way ANOVA test. P < 0.05 was considered statistically significant.