Introduction

Hydrogen sulfide (H2S), a recently characterized gasotransmitter, is endogenously produced by three enzymes, cystathionine β-synthase (CBS), cystathionine γ-lyase (CSE) and 3-mercaptopyruvate sulfurtransferase (MPST). CBS and CSE produce H2S from l-cysteine, while MPST produces H2S from 3-mercaptopyruvate, which is synthesized from l-cysteine by cysteine aminotransferase CAT (CAT/MPST pathway) [1, 2]. H2S can participate in a wide range of physiological processes, including neuromodulation, vasorelaxation, and cytoprotection [3,4,5,6]. In the bladder, previous reports showed exogenous H2S-induced contraction of the rat and guinea pig detrusor [7, 8] as well as relaxation of the pig bladder neck and human bladder tissues [9, 10], indicating that physiological roles of H2S in the bladder remain controversial. Recently, we reported that exogenous H2S induced relaxation of pre-contracted rat bladder dome (BL-D) and trigone (BL-T) tissues and intravesically instilled an H2S donor prolonged intercontraction intervals in rats [11]. We also detected that at least the CAT/MPST pathway-mediated endogenous H2S production was working in the rat BL-D and BL-T tissues [11]. These results indicate a possibility that H2S can function as an endogenous relaxation factor in the rat bladder.

Recently, bladder ischemia induced by a decrease in bladder blood flow (BBF) has been recognized as an etiologic factor of lower urinary tract symptoms (LUTS) including overactive bladder (OAB) and detrusor overactivity (DO) in human and animal models [12,13,14,15,16]. Hypertension can decrease pelvic arterial blood flow, resulting in dysfunction of the bladder [17,18,19]. The spontaneously hypertensive rat (SHR) develops DO [20, 21] and is a valuable tool for exploring the pathogenesis of DO [22]. Our and other groups reported that BBF was lower in SHRs compared with normotensive rats [18, 23, 24], and chronic treatment with an α1 adrenoceptor antagonist silodosin or an ATP-sensitive potassium (KATP) channel opener nicorandil prevented hypertension-related DO in SHRs via improving the decreased BBF [23, 24]. These lines of evidence indicate that hypertension is an important etiologic factor of LUTS via a decrease in BBF.

In SHRs, downregulation of the endogenous H2S system is reported and systemically administered H2S donors decreased blood pressure [25,26,27]. However, hypertension-related changes of the H2S system in the bladder have not clarified yet. In this study, therefore, we compared effects of H2S donors on the micturition reflex and on the bladder contractility, and the endogenous H2S system in the bladder between SHRs and normotensive Wistar rats.

Materials and methods

Ethics approval

Animal experimental procedures were approved by the Kochi University Institutional Animal Care and Use Committee (IACUC) (#J-34/K-48) and were conducted in accordance with NIH guidelines.

Animals

Eighteen-week old male Wistar rats (N = 43, Japan SLC, Hamamatsu, Japan) and male SHRs [N = 44, SHR/Izm, provided from the Disease Model Cooperative Research Association (Kyoto, Japan)] were used and for in vitro studies, tissues (BL-D, BL-T, liver, and cerebellum) were isolated after sacrifice by sodium pentobarbital (80 mg/kg, ip). At the day before tissue isolation or in vivo experiments, blood pressure and heart rate were measured with the tail cuff method, which included warming the whole animal body in the absence of anesthesia (BP-98A-L, Softron, Tokyo, Japan) [28]. Some parts of these tissues were quickly frozen in liquid nitrogen and stored at − 80 °C until use for Western blot and quantification of H2S content.

Cystometry

Cystometry was performed under urethane anesthesia (0.8 g/kg, ip) according to previous reports [11, 23]. Saline was intravesically instilled (2.4 mL/h) in all animals until 4–5 voidings were detected, and then either GYY4137 (a slow-releasing H2S donor [29]) (10−8, 10−7 and 10−6 M in saline) or vehicle (3.3 × 10−5 % N,N-dimethylformamide/saline) was instilled (2.4 mL/h). In the GYY4137-treated group, after detecting four–five voidings, the dose was increased in turn, and a total of 12–15 voidings were recorded after starting GYY4137 instillation. In the vehicle-treated group, vehicle instillation was continued until 15 voidings were recorded. Intercontraction intervals (ICI) and maximal voiding pressure (MVP) were evaluated from four to five voidings at each dose, and the relative values of these parameters were calculated as the ratio of averaged ICI and MVP measured at each dose to those measured during the first saline instillation. After cystometry, animals were euthanized by cervical dislocation under anesthesia. During the surgery and continuous cystometry, we monitored sufficient levels of anesthesia by confirming negative reflex responses to toe pinch every 30 min. If the level was insufficient, additional doses of urethane (0.05 g/kg/injection, ip) were administered.

In vitro organ bath experiments

Functional studies were performed according to previous reports [11, 30]. Strips of BL-D and BL-T were equilibrated and unstretched for 30 min in Krebs–Henseleit solution (10 mL) bubbled with 5% CO2/95% O2 (37 °C). A 1.0 g load was applied to each strip, and the load was readjusted to this level 30 min later. Following a 30-min period of equilibration, these strips were exposed to 100 mM KCl. After a 30 min-washout period, the relaxation response to NaHS (a rapid-releasing H2S donor [29]) (1 × 10−8–3 × 10−4 M) was measured cumulatively in bladder strips pre-contracted by carbachol (1 × 10−5 M). NaHS was administered after stable contraction was reached. NaHS solutions were quickly prepared before use.

Quantification of H2S

The content of H2S in bladder tissue samples was measured using the methylene blue method [11, 31]. Tissues were homogenized in 210 μL of ice-cold 50 mM Tris–HCl (pH 7.4). The homogenate (75 μL) was mixed with 425 μL of distilled water, 250 μL of 1% zinc acetate, 133 μL of 20 mM N,N-dimethyl p-phenylenediamine sulfate in 7.2 M HCl and 133 μL of 30 mM FeCl3 in 1.2 M HCl, then incubated for 30 min at room temperature. Next, 250 μL of 10% trichloroacetic acid was added, and the mixture was centrifuged (10,000 × g for 5 min). The absorbance of the supernatants (200 μL) was measured at 670 nm. All samples were assayed in duplicate, and the concentration of H2S was calculated based on a standard curve of NaHS (0–400 μM). The amount of H2S was normalized by protein content, determined using a commercially available kit (Protein Assay Rapid Kit, Wako, Osaka, Japan).

Western blot

Protein samples were prepared and Western blots were performed according to previous reports [11, 32]. Tissue samples (approximately 30 mg) were homogenized in 210 μL of ice-cold RIPA buffer (Wako) with a protease inhibitor cocktail (Roche Diagnosis, Basel, Switzerland). Protein (50 μg) was separated by SDS-PAGE and transferred to polyvinylidene difluoride membranes. After blocking with 5% skim milk, these membranes were incubated with primary antibodies as follows: rabbit polyclonal anti-CBS antibody (1:400 dilution, Proteintech, Rosemont, IL, USA), rabbit polyclonal anti-MPST antibody (1:400 dilution, Santa Cruz Biotechnology, Santa Cruz, CA, USA), rabbit polyclonal anti-CAT antibody (1:500 dilution, Abcam, Cambridge, UK), rabbit polyclonal anti-CSE antibody (1:200 dilution, Abcam), rabbit polyclonal anti-β-actin antibody (1:200 dilution, ANASPEC, Fremont, CA, USA) and mouse monoclonal anti-GAPDH antibody (1:3000 dilution, Acris Antibodies GmbH, Herford, Germany). The secondary antibodies were peroxidase-conjugated anti-rabbit IgG (1:3000) or anti-mouse IgG (1:2000) (GE Healthcare, Buckinghamshire, UK). The membranes were developed with Immobilon™ Western Chemiluminescent horseradish peroxidase substrate (Millipore, Billerica, MA, USA). Images were captured in a LAS4000mini (Fujifilm, Tokyo, Japan). Densitometric analysis was performed using MultiGauge software (Fujifilm). Target protein bands were normalized to β-actin or GAPDH.

Statistics

All values are expressed as mean ± SEM. Statistical differences were determined using an unpaired Student’s t test or Welch’s test. P values less than 0.05 were taken to indicate statistical significance.

Drugs and chemicals

GYY4137 [(p-methoxyphenyl)morpholino-phosphinodithioic acid] and NaHS were purchased from Cayman Chemical, Ann Arbor, MI, USA. All other chemicals (reagent grade) were commercially available.

Results

General features of the animals

Body weight, the bladder weight and the bladder body weight ratio (BBR) in SHRs were significantly lower than those in Wistar rats (Table 1). Systolic, mean and diastolic blood pressures in SHRs were significantly higher than those in Wistar rats (Table 1). There is no significant difference in heart rate between Wistar rats and SHRs (Table 1).

Table 1 The general features
Full size table

Effects of intravesically instilled GYY4137 on the micturition reflex

Results of continuous cystometry are shown in Figs. 1 and 2. The baseline values of ICI (sec) and MVP (cmH2O) just before intravesical instillation of vehicle or GYY4137 were 2152 ± 153 and 32.6 ± 0.8 in Wistar rats (N = 17) and 1362 ± 140 and 30.4 ± 0.5 in SHRs (N = 14). The baseline values of ICI values of SHRs were significantly shorter than those of Wistar rats (P < 0.05). In Wistar rats, GYY4137 significantly prolonged ICI at 10−6 M compared to the vehicle-treated group, while the GYY4137-induced prolongation was not detected in SHRs (Fig. 2). There was no significant difference in baseline values of MVP between Wistar rats and SHRs. GYY4137 showed no significant effect on MVP in Wistar rats or SHRs (data not shown).

Fig. 1
figure 1

Representative in vivo continuous cystometry traces in a Wistar rat (a) and a SHR (b). Vehicle (3.3 × 10−5 % DMF, upper traces) or GYY4137 (a slow-releasing H2S donor, 10−6 M, lower traces) was intravesically instilled (2.4 ml/h) following saline instillation. Arrows indicate the timing of each instillation. DMF N,N-dimethylformamide, SHR spontaneously hypertensive rat

Full size image
Fig. 2
figure 2

Effect of intravesically instilled GYY4137 (a slow-releasing H2S donor) on intercontraction intervals (ICI) in Wistar rats (a) and SHRs (b). Continuous cystometry (infusion rate, 2.4 ml/h) was performed under urethane anesthesia (0.8 g/kg, ip) in these rats. GYY4137 (10−8–10−6 M) or vehicle was intravesically instilled following saline instillation. Values are mean ± SEM. *P < 0.05, significantly different from the vehicle-treated group (unpaired Student’s t test). The number of samples per group is indicated in parentheses. SHR: spontaneously hypertensive rat

Full size image

In vitro organ bath experiments using pre-contracted rat bladder strips

Results of in vitro organ bath experiments are shown in Fig. 3 and Table 2. NaHS (1 × 10−8–3 × 10−4 M) dose-dependently induced the relaxation of BL-D and BL-T strips pre-contracted by carbachol in both Wistar rats and SHRs. In the BL-D, maximum relaxation in SHRs was significantly lower than that in Wistar rats, while there was no significant difference in the EC50 values for NaHS-induced relaxation between Wistar rats and SHRs (Table 2). In the BL-T, the EC50 values for NaHS-induced relaxation in SHRs were significantly higher than those in Wistar rats, while there was no significant difference in maximum relaxation between Wistar rats and SHRs (Table 2). We verified that NaHS or Krebs–Henseleit solution alone showed no contraction or relaxation on each tissue strip (data not shown).

Fig. 3
figure 3

Representative isometric force recordings in BL-D strips isolated from a Wistar rat (a) or a SHR (b). These strips were pre-contracted by carbachol (1 × 10−5 M) and after stable contraction was reached, NaHS (a rapid-releasing H2S donor, 1 × 10−8–3 × 10−4 M) was administered. BL-D bladder dome, C carbachol, SHR spontaneously hypertensive rat

Full size image
Table 2 Data from functional studies in the rat bladder tissues
Full size table

H2S content in the rat bladder

H2S was detectable in the bladder of Wistar rats and SHRs (Fig. 4). The level of H2S was significantly higher in SHRs than that in Wistar rats both in the BL-D and BL-T (Fig. 4).

Fig. 4
figure 4

H2S content in the bladder of Wistar rats and SHRs. Values are mean ± SEM. *P < 0.05, significantly different from the Wistar group (Welch’s test for BL-D; unpaired Student’s t test for BL-T). The number of samples per group is indicated in parentheses. BL-D bladder dome, BL-T bladder trigone, SHR spontaneously hypertensive rat

Full size image

Expression levels of enzymes related to H2S biosynthesis in the rat bladder

Protein bands of CBS, MPST and CAT were detected in the BL-D and BL-T of both Wistar rats and SHRs (Fig. 5a–c). The expression level of MPST in the SHR BL-D was significantly higher than that in the Wistar rat BL-D, and that in the SHR BL-T showed a non-significant tendency (P = 0.09) to be higher than that in the Wistar rat BL-T (Fig. 5b). In contrast, there were no significant differences in the expression levels of CBS or CAT between Wistar rat and SHR bladder tissues (Fig. 5a, c). No protein bands for CSE were detected in the bladder of both Wistar rats and SHRs (data not shown).

Fig. 5
figure 5

Expression levels of enzymes related to H2S biosynthesis in the bladder of Wistar rats and SHRs. The amount of protein for CBS (a), MPST (b) or CAT (c) was normalized to the amount of protein for β-actin/GAPDH. Values are mean ± SEM. *P < 0.05, significantly different from the Wistar group (Welch’s test). The number of samples per group is indicated in parentheses. C and L indicate positive controls (cerebellum for CBS and MPST and liver for CAT). W and S indicate the Wistar and SHR groups. CAT cysteine aminotransferase, CBS cystathionine β-synthase, MPST 3-mercaptopyruvate sulfurtransferase, BL-D bladder dome, BL-T bladder trigone, SHR spontaneously hypertensive rat

Full size image

Discussion

In the present study, GYY4137 prolonged ICI in normotensive Wistar rats, but not in SHRs. NaHS-induced relaxation of pre-contracted bladder strips was attenuated in SHRs compared with Wistar rats. The H2S content in the bladder of SHRs was higher than that of Wistar rats. CBS, MPST, and CAT were detected in the bladder of both Wistar rats and SHRs, and the expression levels of MPST in the SHR bladder were significantly higher than those in the Wistar rat bladder.

It is reported that even at 6 weeks of age, SHRs show higher systolic blood pressure (180–200 mmHg) compared with normotensive controls (140–150 mmHg), and that the blood pressure of SHRs elevated with age [33]. On the other hand, hypertension-induced DO was detected at 17 and 21 weeks of age, but not at 12 weeks of age [20]. Based on the evidence and our previous reports [23, 24], in this study, we used SHRs and normotensive Wistar rats at 18 weeks of age. The SHRs showed obvious hypertension, and lower body weight, the bladder weight and the BBR compared with the Wistar rats. In addition, basal values of ICI in the SHRs were shorter than those in the Wistar rats, indicating that at 18 weeks of age, SHRs showed frequent urination due to DO. These differences between Wistar rats and SHRs were in line with previous reports [20, 23, 24]. At 18 weeks of age, intravesically instilled GYY4137, a slow-releasing H2S donor [29], prolonged ICI in Wistar rats and NaHS, a rapid-releasing H2S donor [29], relaxed pre-contracted bladder strips of Wistar rats, in line with our previous report [11]. These results suggest that intravesically administered GYY4137-induced inhibition of the micturition reflex in Wistar rats might be mediated at least by relaxation of the bladder smooth muscle. On the other hand, in SHRs, the GYY4137 showed no significant effect on ICI even at the same doses and SHR bladder strips showed NaHS-induced hypo-relaxation (BL-D) or lower potency in the NaHS-induced relaxation (BL-T) compared with the Wistar rat strips. These results suggest that intravesical H2S-induced suppression of the micturition reflex might be impaired in SHRs at least by attenuating H2S-induced relaxation of the bladder smooth muscle. In the SHR thoracic aorta, however, enhanced maximal vasorelaxation was observed in response to Na2S, another rapid-releasing H2S donor [29], compared with normotensive controls [34]. Therefore, hypertension-mediated changes in responsiveness to H2S might be different between the bladder and the vascular system.

Next, we compared endogenous H2S system between SHR and Wistar rat bladder tissues. Higher H2S content and an increase in MPST protein expression were detected in the SHR bladder compared with the Wistar rat bladder. These results indicate that endogenous H2S system might be upregulated to compensate the attenuated relaxation response to H2S in the SHR bladder. Enhancement of the bladder afferent activity, a cause of DO [35], is reported in SHRs [36]. Since it is reported that H2S promotes the release of sensory neuropeptides from the bladder afferents to mediate inhibitory neurotransmission to the pig bladder neck [9], in SHRs, H2S system might be upregulated to augment the inhibitory signal, thereby counteracting the enhanced bladder afferent activity. On the other hand, several groups reported that in SHRs, plasma H2S levels were reduced and CSE mRNA expression and activity of CSE were decreased in the thoracic aorta compared with normotensive controls [26, 37, 38]. Since in the SHR thoracic aorta, Na2S-induced vasorelaxation was enhanced [34], unlike the bladder, to compensate the decreased H2S production in the vascular system, H2S-induced vasorelaxation might be enhanced in SHRs compared with the control. In addition, responses to H2S in the thoracic aorta are different between young prehypertensive and adult SHRs. At 4 weeks of age, higher potency of the vasorelaxant response to Na2S was detected in the SHR thoracic aorta, while Na2S-induced hyper-relaxation was observed in SHRs at 16 weeks of age compared with each control [34]. Since hypertension-induced DO is detected after hypertension is developed (at least after 12 weeks of age) [20], responses to H2S and levels of the endogenous H2S system in the bladder might be different among prehypertensive and pre-detrusor overactive SHRs, hypertensive and pre-detrusor overactive SHRs, and hypertensive and detrusor overactive SHRs. Further studies are needed to investigate “time-dependency” of hypertension-mediated changes of responses to H2S and the endogenous H2S system in the SHR bladder.

Although mechanisms for H2S-induced bladder smooth muscle relaxation remain controversial, one of the targets for H2S is KATP channels, because glibenclamide, a KATP channel blocker, reduced GYY4137-induced relaxation of the pre-contracted pig bladder strips [9]. In this study, H2S-induced bladder relaxation was attenuated in SHRs, indicating a possibility that KATP channels-mediated relaxation might be impaired in the SHR bladder. However, it is reported that mRNA levels of KATP channel subunits are increased in the SHR bladder compared with normotensive controls [24], indicating that KATP channels-mediated responses can be enhanced. These findings suggest the involvement of other mechanisms rather than KATP channels in the impairment of H2S-induced bladder relaxation in SHRs. Another possible mechanism for the impairment might be related to nitric oxide (NO), which is reported to show synergistic action with H2S on vasorelaxation [39]. In general, NO plays an important role in the relaxation of the urethra sphincter smooth muscle, while the detrusor is less sensitive to the effects of NO [40, 41]. However, NO is reported to depress sensitized bladder afferent nerve activity in rats with cyclophosphamide-induced cystitis [42, 43], indicating that NO might modulate bladder hyperactivity induced by pathological conditions and H2S has a possibility to enhance the NO-induced modulation synergistically. In SHRs, mRNA expression levels of neuronal NO synthase were lower in the bladder compared with the normotensive rat bladder [18]. Therefore, impairment of the synergistic action of NO and H2S might be a cause of the attenuation of H2S-induced bladder relaxation in SHRs. As it is unclear whether H2S and NO modulate bladder smooth muscle relaxation and bladder afferent nerve activity synergistically, further studies are necessary to clarify the synergistic action and the attenuation mechanisms in the SHR bladder focusing on the interaction between H2S and NO.

Conclusion

H2S-induced bladder relaxation in SHRs is impaired, which might be a cause of hypertension-mediated induction of LUTD. To compensate the less responsiveness, endogenous H2S levels might be increased in the SHR bladder.