Actions of Hydrogen Sulfide and ATP-Sensitive Potassium Channels on Colonic Hypermotility in a Rat Model of Chronic Stress

Objective To investigate the potential role of hydrogen sulphide (H2S) and ATP-sensitive potassium (KATP) channels in chronic stress-induced colonic hypermotility. Methods Male Wistar rats were submitted daily to 1 h of water avoidance stress (WAS) or sham WAS (SWAS) for 10 consecutive days. Organ bath recordings, H2S production, immunohistochemistry and western blotting were performed on rat colonic samples to investigate the role of endogenous H2S in repeated WAS-induced hypermotility. Organ bath recordings and western blotting were used to detect the role of KATP channels in repeated WAS. Results Repeated WAS increased the number of fecal pellets per hour and the area under the curve of the spontaneous contractions of colonic strips, and decreased the endogenous production of H2S and the expression of H2S-producing enzymes in the colon devoid of mucosa and submucosa. Inhibitors of H2S-producing enzymes increased the contractile activity of colonic strips in the SWAS rats. NaHS concentration-dependently inhibited the spontaneous contractions of the strips and the NaHS IC50 for the WAS rats was significantly lower than that for the SWAS rats. The inhibitory effect of NaHS was significantly reduced by glybenclamide. Repeated WAS treatment resulted in up-regulation of Kir6.1 and SUR2B of KATP channels in the colon devoid of mucosa and submucosa. Conclusion The colonic hypermotility induced by repeated WAS may be associated with the decreased production of endogenous H2S. The increased expression of the subunits of KATP channels in colonic smooth muscle cells may be a defensive response to repeated WAS. H2S donor may have potential clinical utility in treating chronic stress- induced colonic hypermotility.


Introduction
Different psychological and environmental stressors affect physiologic functions of the gastrointestinal tract and play important roles in the pathophysiology of gastrointestinal diseases [1]. Chronic stress causes colonic hypermotility [2,3,4,5,6] and precipitates or exacerbates the symptoms of two major motility disorders, irritable bowel syndrome and inammatory bowel disease [4,7]. The mechanisms that underline this increased colonic motility has received increased awareness in the past years. Experimental studies have revealed that some factors are involved, such as central nervous system,brain-gut axis, neurotransmitters, gastrointestinal hormones, and L-type Ca 2+ channels located in the colon [2,4,5,6,7,8].
Hydrogen sulfide (H 2 S) has recently been identified as a new ''gasotransmitter''. It is synthesized in many mammalian tissues and produces effects on various biological targets that have widespread consequences, ranging from cytotoxic to cytoprotec-tive [9]. Cystathionine b-synthase (CBS) and systathionine c-lyase (CSE) are two important enzymes for generation of endogenous H 2 S [9]. They have been shown to be expressed in the smooth muscle cells, enteric neurons, interstitial cells of Cajal, and epithelial cells of the gastrointestinal (GI) tract [10]. There is growing evidence that endogenous H 2 S might play an important role in several physiological processes including neurotransmission, pain, motility, and secretion [10,11,12]. Pharmacological studies show that exogenously applied NaHS, a H 2 S donor, inhibits gastric and intestinal motility, causing GI smooth muscle relaxation [13,14,15,16,17].
The mechanism through which H 2 S exerts its relaxant properties is related to the direct opening of ATP-sensitive potassium (K ATP ) channels located in the smooth muscle cells [9,10,14,15,18]. Other potential targets of action of H 2 S on GI smooth muscle include apamin-sensitive SK channels and delayed rectifier potassium channels [14,15]. K ATP channels are composed of at least two subunits: an inwardly rectifying K + channel six family (Kir6.x) that forms the ion conducting pore and a modulatory sulphonylurea receptor (SUR) that accounts for several pharmacological properties [19,20]. Both Kir and SUR subunits must be co-expressed, and combine in a 4:4 stoichiometry to generate a functional K ATP channel [19,20]. It is now well recognized that K ATP channels locate in GI smooth muscle cells, and Kir 6.1/SUR2B and Kir 6.2/SUR2B form the K ATP complex [21,22,23,24]. Differences exist in the functional and pharmacological properties of various K ATP channels in different tissues. In GI tract, the physiological role of K ATP channels may be related to the modulation of cell excitability [21]. Activation of K ATP channels leads to an increased hyperpolarization of membrane potential and results in the relaxation of GI smooth muscle [10].
Given the role of H 2 S and K ATP channels in GI motility, we investigated the possibility that H 2 S and/or K ATP channels contribute(s) to the colonic motility dysfunction in chronic stress. This involved an investigation of colonic H 2 S synthesis and the expression of two key enzymes for H 2 S synthesis over the course of repeated water avoidance stress (WAS). We also examined whether or not blocking H 2 S synthesis in sham stress could mimic the colonic hypermotility in chronic stress. Finally, we examined the potential role of exogenous H 2 S donor and K ATP channels in chronic WAS.

Animals
Adult male Wistar rats weighting 200-220 g were obtained from the Experimental Animal Center of Wuhan University, Wuhan, Hubei Province, China. They were kept under conventional conditions in an environmentally controlled room (20-21uC, 60% humidity, 12:12 h light-dark cycle). All protocols were approved by the Institutional Animal Care and Use Committee of Wuhan University (Approval ID: WHU20110312) and adhered to the ethical guidelines of the International Association for the Study of Pain.

WAS Protocol
The WAS was conducted following the procedures modified from previous reports [2,3,8,25]. Briefly, the animals were placed on a block (86666cm), which was fixed in the center of a tank for a period of 1 h daily for 10 consecutive days. The tank was filled with fresh tap water up to 1 cm from of the top edge of the block. The sham WAS (SWAS) consisted of placing the rats similarly for 1 h daily for 10 days on the same platform in a waterless container. Stress sessions were performed between 10:00 AM and 13:00 PM to minimize diurnal variations in response. The total number of fecal pellets was counted at the end of each 1 h WAS or SWAS session.

Colonic Motility Tests in vitro
Rats were killed by cervical dislocation 30 min after the last stress session. The proximal colon was carefully removed and opened along the mesenteric border and pinned mucosa side up. The mucosa and submucosa were removed by sharp dissection under a dissection microscope. The circular muscle (CM) or longitudinal muscle (LM) strips (3610 mm) were cut along the circular or longitudinal axis. Each smooth muscle strip was suspended in an 6 ml organ bath which was filled with Tyrode's buffer containing(mM): NaCl 147.0, KCl 4.0, CaCl 2 2.0, NaH 2 PO 4 0.42, Na 2 HPO 4 2.0, MgCl 2 1.05, and glucose 5.5 (adjusted to pH 7.4 with NaOH). The organ bath was bubbled with a mixture of 95% O 2 plus 5% CO 2 and maintained at 37uC. One end of the strip was fixed to a hook on the bottom of the chamber and the other end was connected to a isometric force transducer (JZJOIH, Chengdu, CHN) at the top. Each muscle strip was placed under a resting preload of 1.0 g and equilibrated for 60 min, with Tyrode's buffer being changed every 20 min. To estimate the contractile activity of the strips, the area under the curve (AUC) of the spontaneous contractions from the baseline was measured and the result was expressed in g?min 21 . To normalize data, the value of AUC obtained before the treatment was considered 100% and the percentage of inhibition of the spontaneous contractions was estimated with the AUC obtained after the treatment.

Endogenous Production of Hydrogen Sulphide
Production of H 2 S was measured based on previous reports [26,27]. Briey, the colonic tissue devoid of mucosa and submucosa was placed in a sealed vial containing physiological saline solution with L-cysteine (10 mM) and pyridoxal 59-phosphate (2 mM) which was connected to a second vial containing 0.5 mL of 1% (w v 21 ) zinc acetate. The first vial was bubbled with a gas mixture of 95% O 2 and 5% CO 2 at a rate of 1-4 mL?min 21 in order to minimize the degradation of H 2 S. The increase in pressure in the first vial forced the gases bubbled through the zinc acetate solution in the second vial and H 2 S was trapped as zinc sulphide. The incubation mixture was prepared on ice and the reaction was started by transferring the vials to a water bath at 37uC. The reaction was stopped at 30 min by injecting 0.5 mL of 50% (w v 21 ) trichloroacetic acid. Gas ow was allowed to continue for an additional 30 min to ensure complete trapping of H 2 S. H 2 S was measured by using a colorimetric method [26,28]. The content of the second vial was transferred to a test tube and 3.5 mL of distilled water, 0.4 mL of N,N-dimethyl-p-phenylenediamine sulphate (20 mM) in HCl (7.  Immunohistochemistry The streptavidin-peroxidase method was used for the immunohistological localisation of H 2 S-producing enzymes in proximal colon samples. The tissues were fixed overnight at room temperature in a pH-neutral, phosphatebuffered, 10% formalin solution and then embedded in paraffin. The sections were incubated at 100uC in 10 mM citrate buffer (pH 6.0) for 10 min to retrieve antigens, cooled for 20 min and then washed in phosphate-buffered saline. The sections were cut off, mounted on glass slides and de-paraffinized. Endogenous peroxidase activity was blocked with a 15-min incubation in 3% hydrogen peroxide. Mouse monoclonal anti-CSE and anti-CBS antibodies (1:250, 1:300, respectively) were applied to detect CSE and CBS. The sections were incubated in above primary antibodies overnight at 4uC. After washed in PBS, the sections were incubated in biotinylated anti-mouse or anti-goat secondary antibody and streptavidin-horseradish peroxidase. 3,39-Diaminobenzidine was used as a chromogen and hematoxylin was used for counterstaining.

Western Blotting
The proximal colon devoid of mucosa and submucosa was homogenized in ice-cold RIPA lysis buffer composed of 20 mM Tris-HCl, 0.1 mM PMSF, and 5ul/ml protease inhibitor cocktail. The protein concentration was determined by Bio-Rad protein assay (Bio-Rad Laboratories, Hercules, CA, USA). Then equal amounts of protein (50 ug) was loaded onto a 10% gel, subjected to SDS-PAGE, and electrotransferred onto polyvinylidene di-  (1:400) were used for the detection of Kir6.1, Kir6.2 and SUR2B respectively, with GAPDH (1: 10000) as the internal standard. Immune complexes were detected with horseradish peroxidase-conjugated secondary antibodies and enhanced with chemiluminescent substrate. Gradation of the protein bands was analyzed using Bandscan software (Glyko, Novato, USA). The gradation ratio of the target protein immunoreactive band and GAPDH from each lane was calculated as normalized blot optional density.

Data Analysis and Statistics
Data were expressed as mean 6 SEM. Significant difference between groups was evaluated using unpaired or paired Student's t-tests. IC 50 values (i.e., agonist concentrations that produced 50% inhibition of the observed maximum response, respectively) were analyzed using the 'sigmoidal dose-response (variable slope)' option for curve fitting. Statistical analysis was performed with Graph Pad Prism 5.0 (GraphPad Software, San Diego, CA, USA). A P value below 0.05 was considered statistically significant.

Repeated WAS Stimulated Colonic Transit
The colonic transits were measured by fecal pellet output. The fecal pellets per hour in the WAS rats at days 1, 3, 5, 7 and 10 were significantly higher than those in the SWAS rats (7. 12, respectively, n = 10/group, P,0.001), ever though it had decreased by ,28% from day 1 (Figure 1).

Repeated WAS Increased Contractile Activity of Colonic Strips
The AUC of LM and CM strips of the WAS rats were both significantly higher than those of the SWAS rats (23.5963.23 g?min 21 vs. 14.3862.56 g?min 21 and 12.6262.13 g?min 21 vs. 7.9761.94 g?min 21 , n = 10/group, P,0.001, Figure 2A, 2C). To identify whether alteration in smooth muscle level contributes to the changes of colonic contractile response, colon segments were pretreated with tetrodotoxin (TTX) in organ bath for 30 min in order to block the inuence of the enteric nervous system on smooth muscle contractions. Figure 2B and

H 2 S Synthesis was Decreased Following Repeated WAS
The level of H 2 S produced by the colonic tissue devoid of mucosa and submucosa in the WAS rats was decreased, as compared with that in the SWAS rats (8.8561.63 nmol?min 21 ?g 21 vs. 13.2961.72 nmol?min 21 ?g 21 tissue, n = 10/group, P,0.001, Figure 3).

Repeated WAS Decreased Expression of H 2 S-producing Enzymes
In both SWAS and WAS rats colon, CSE was strongly expressed in the cytosols of the circular and longitudinal smooth muscle cells and the nucleus of the myenteric plexus neurons. CSE-immunoreactivity(IR) was diffuse in the mucosa and submucosa layers ( Figure 4A, 4B). A completely different pattern was found for CBS. It was primarily localized in the cytosols of myenteric plexus neurons and weakly localized in the epithelial cells. A diffuse pattern was also observed in the muscular layers ( Figure 4C, 4D). Western blotting analysis showed that WAS treatment resulted in down-regulation of the expression of CBS and CSE in the colonic tissue devoid of mucosa and submucosa (P,0.001, Figure 4E, 4F).

Inhibitors of H 2 S-producing Enzymes Increased Contractile Activity of Colonic Strips in SWAS Rats
Since our present experiments demonstrated that the two H 2 Sproducing enzymes were both down-regulated in the WAS rats, we determined whether alternatively blocking CSE and CBS using enzyme inhibitors, which mimiced a down regulation of the enzymes, produced an increase in the contractile activity in the SWAS rats.
PAG is an inhibitor of CSE and AOAA, an inhibitor of CBS. After 2 min of incubation with PAG (2 mM) or AOAA (2 mM), the AUC was measured. PAG significantly increased the AUC of LM and CM strips in the SWAS rats (Control: 15 21 , n = 10, P,0.001, Figure 5C, 5D).

Repeated WAS Increased Expression of Subunits of K ATP Channels
Western blotting analysis showed that there was a marked increase in the expression of Kir6.1 and SUR2B, whereas Kir6.2 had no change following WAS (P,0.05, Figure 8A, 8B).

Discussion
The repetitive WAS rat model stimulates the desire to survive as well as fear of the surrounding environment, and is an acceptable animal model for studying the possible mechanisms involved in altered colonic motility and visceral hyperalgesia induced by chronic stress [2,3]. The present study found that WAS increased the fecal pellet number, a validated measure of colonic motility [2,3], and the spontaneous contractile activity of colonic strips in the presence or absence of TTX. These data indicate that WAS induces colonic hypermotility, and the increased contractile activity induced by WAS might be directly due to myogenic changes in the colon, even though the potential changes of enteric nervous system and brain-gut axis may have contribution as well.
H 2 S is produced enzymatically by GI tissues and causes welldefined physiologic effects including inhibition of motility and causing smooth muscle relaxation [10,11,12]. However, little is known about the link between the hypermotility induced by repeated stress and the endogenous synthesis of H 2 S. We, therefore, examined the possibility that hypermotility induced by repeated stress may have been due to the decrease in H 2 S synthesis. First, we observed the endogenous H 2 S produced by the colonic tissue devoid of mucosa and submucosa was significantly decreased in the WAS rats. Second, the expression of CSE and CBS in the colon devoid of mucosa and submucosa were both down-regulated in the WAS rats. Third, alternative blocking H 2 S- producing enzymes using PAG or AOAA, which mimiced a down ''regulation'' of the enzymes, produced a increase in the contractile activity in the SWAS rats. Because of the 'nitric oxide (NO)-like effects' [26], we did not test hydroxylamine (HA), the other inhibitor of CBS on the contractile activity in the SWAS rats. These results suggest an important role for the decreased H 2 S production in chronic WAS-induced colonic hypermotility.
In the present study, we indentified the localisation of H 2 Sproducing enzymes. In both SWAS and WAS colon, CSE was strongly expressed in smooth muscle cells of the lamina propia whereas CBS-IR was less intense but still positive in muscle layers, which is consistent with the previous reports [26,29]. In addition, we found the two H 2 S-producing enzymes were present in neurons of the myenteric plexus, and a similar staining was described in rat jejunum [30]. CSE-IR was intense in the nucleus of the myenteric neurons, whereas CBS-IR was weak in the cytosols. CSE appears to be the major source of H 2 S synthesis in both SWAS and WAS colon and CBS makes a minor contribution as well. Note that major differences might exist in rat colon. For example, CBS-IR was apparent in lamina propria but CSE-IR was quite diffuse [31,32]. CSE was expressed in neurons of the enteric nervous system but CBS was not [26]. Perhaps the different technical approaches account for these major differences in the distribution of the enzymes.
Myenteric neurons, located between the longitudinal and circular muscle layers in the gastrointestinal wall, mainly regulate gastrointestinal motility [33]. CSE-IR or CBS-IR neurons, which can produce H 2 S, could be regarded as inhibitory motor neurons and therefore, H 2 S of neural origin can be considered a 'neurotransmitter'. However, the release of H 2 S after neuronal stimulation was not demonstrated in enteric neurons [13] and H 2 S does not participate in neurally mediated relaxation [26]. Perhaps the appropriate parameters of stimulation have not been used to elicit H 2 S release [26]. Further studies are needed to evaluate the role of H 2 S of neural origin. Endogenous H 2 S of non-neural origin can be produced by smooth muscle cells. It may regulate motility through direct interaction with some molecular targets in the smooth muscle cells and might be an intracellular signaling molecule in this sense. The present study showed that CSE and CBS expression in the colon devoid of mucosa and submucosa were both down-regulated in the WAS rats. Theoretically, the decreased H 2 S production from either a muscular or a neuronal source might contribute to colonic motility disorder.   Several plausible mechanisms for colonic hypermotility were considered in chronic stress animal models. Central structures involved in stress-induced hypermotility include the increased mRNA expression of arginine vasopressin and corticotropinreleasing factor in the hypothalamus [5,6]. We reported recently that the plasma hormones levels of substance P, thyrotropinreleasing hormone, motilin, and cholecystokinin in the WAS rats were increased, whereas peptide YY was decreased. These gastrointestinal hormones and the brain-gut peptides influenced colonic motility by directly decreasing the IKv and IBK Ca of proximal colonic smooth muscle cells [2]. Moreover, chronic stress can remodel the gastrointestinal smooth muscle cells to cause motility dysfunction. For example, 9-day heterotypic chronic stress was found to increases the synthesis and release of norepinephrine in plasma, which enhanced the transcription of the pore-forming a 1C subunit of Ca v 1.2 (L-type) channels in colonic circular smooth muscle cells [4]. And the cellular mechanisms by which norepinephrine enhances expression of the a 1C -subunit is related to the phosphatidylinositol 3-kinase (PI3K)/Akt/GSK-3bsignaling pathway [7]. Indeed, H 2 S is intermingled with the synthetic and release pathway for some neural and or hormonal factors. For example, H 2 S can regulate the release of corticotrophin-releasing hormone and NO [28,34], and enhance the ability of NO to relax smooth muscle [35]. On the other hand, NO can increas the expression and activity of CSE [28,36]. Future studies should address the possible mechanisms behind chronic stress-induced H 2 S regulation in the colonic motility, and the link between H 2 S and neural and or hormonal factors in these changes.
NaHS, the H 2 S donor, is an important pharmacological tool for investigating the effects of H 2 S in vitro. Our data showed exogenously applied NaHS concentration-dependently inhibited the spontaneous motility of colonic muscle strips from both SWAS and WAS rats, which is consistent with the previous studies [13,14,15,16,17]. The inhibitory effect of NaHS on smooth muscle cells is largely through an direct action on K ATP channels [9,10,14,15,18]. The result that glybenclamide, a K ATP channel blocker, significantly reduced the inhibitory effect induced by NaHS raised the possibility that NaHS might act via K ATP channel. The molecular mechanisms underlying the effect of H 2 S on K ATP channels are still largely unknown. Because H 2 S is a reductant, it is possible that H 2 S directly interacts with K ATP channel proteins by reducing the cysteine residues [37]. Taken together, less H 2 S production through ''less activation'' of K ATP channels may be one of the reasons why repeated WAS induced colonic hypermotility.
It is well recognized that Kir 6.1/SUR2B and Kir 6.2/SUR2B form the K ATP complex located in the GI smooth muscle cells [21,22,23,24]. Heterologously expressed Kir 6.1 demonstrates a conductance of ,35 pS, whereas Kir 6.2 has a conductance of ,80 pS [21]. The data that the single-channel conductance of K ATP channel is ,42 pS and Kir 6.1 is strongly expressed on the plasma membrane, whereas Kir 6.2 appeared to be restricted to the cytosol, suggest that Kir 6.1 is the major pore-forming isoform of K ATP channel [21]. Despite the observation of a marked decrease in the capacity of WAS colon to synthesize H 2 S, the pharmacological evidence in vitro showed that the NaHS IC 50 for the WAS rats was significantly lower than that for the controls. This indicates either an increase in the number of K ATP channels, or an increase in the frequency of openings in this model of WAS. We next investigated the expression of the isoforms of K ATP channels. After WAS, the protein expression of the pore-forming Kir 6.1, except for Kir 6.2, and SUR2B, the major binding site for the K + channel opener was markedly increased, indicating that the number of K ATP channels was increased. We do not know the detailed mechanisms by which WAS enhances expression of Kir 6.1/SUR2B. As activation of K ATP channels leads to the relaxation of GI smooth muscle [10], it may be an adaptive responses following exposure to a repeated psychological stressor in rats, aiming at improving disturbed gut function and ameliorating symptoms.
There is growing evidence that H 2 S might have important therapeutic potentia. For example, H 2 S maintains the gastric mucosal integrity and administration of H 2 S donor or L-cysteine, a precursor of endogenous H 2 S synthesis, results in enhanced gastric ulcer healing [28]. Moreover, H 2 S is an anti-inammatory mediator and promotes resolution of colitis in rats [32,38]. Our pharmacological experiments in vitro showed that the IC 50 values of NaHS for the WAS rats were lower than those for the controls, partly due to the increased expression of Kir 6.1 and SUR2B in smooth muscle cells. It is thus noteworthy that the H 2 S donor has potential contribution to reverse stress-induced colonic hypermotility. However, the putative beneficial vs toxic effects of H 2 S has been controversial for many years [26,32]. Further studies are needed to evaluate its physiological, pharmacological and toxicological effects in GI tract.
In summary, this study provides evidence that the decreased production of endogenous H 2 S may be one of the reasons for chronic WAS-induced colonic hypermotility, and the increased expression of Kir 6.1 and SUR2B in colonic smooth muscle cells may be a defensive response to chronic stress. H 2 S donor can markedly inhibited the colonic contractions of the WAS rats. Thus, enhancement of endogenous H 2 S synthesis or delivery of appropriate concentrations of exdogenous H 2 S may have potential clinical utility in treating chronic stress-induced colonic hypermotility.