ORIGINAL RESEARCH published: 03 June 2020 doi: 10.3389/fphys.2020.00438 Mechanisms Associated to Nitroxyl (HNO)-Induced Relaxation in the Intestinal Smooth Muscle Mirko Gastreich-Seelig 1 , Marcel Jimenez 2 and Ervice Pouokam 1* 1 Institute for Veterinary Physiology and Biochemistry, Justus-Liebig-University Giessen, Giessen, Germany, 2 Department of Cell Biology, Physiology and Immunology and Neurosciences Institute, Universitat Autònoma de Barcelona, Barcelona, Spain The pharmacological properties of nitroxyl (HNO) donors in the gastrointestinal tract are unknown. We investigated the properties of this molecule in the regulation of gastrointestinal contractility focusing on its possible interaction with other gaseous signaling molecules such as NO and H2 S. Organ bath, Ca2+ imaging, and microelectrode recordings were performed on rat intestinal samples, using Angeli’s salt as HNO donor. Angeli’s salt caused a concentration-dependent relaxation of longitudinal or circular muscle strips of the ileum and the proximal colon. This relaxation was Edited by: strongly inhibited by the Rho-kinase inhibitor Y-27632 (10 µM), by the reducing agent Stephen J. Pandol, DTT or by the inhibitor of soluble guanylate cyclase (sGC) ODQ (10 µM) alone or in Cedars Sinai Medical Center, combination with the inhibitors of the endogenous synthesis of H2 S β-cyano-L-alanine United States (5 mM) and amino-oxyacetate (5 mM). Preventing endogenous synthesis of NO by the Reviewed by: Fabiola Zakia Mónica, NO synthase inhibitor L-NAME (200 µM) did not affect the relaxation induced by HNO. Campinas State University, Brazil HNO induced an increase in cytosolic Ca2+ concentration in colonic myocytes. It also Nazareno Paolocci, Johns Hopkins University, elicited myocyte membrane hyperpolarization that amounted to −10.6 ± 1.1 mV. ODQ United States (10 µM) and Apamin (1 µM), a selective inhibitor of small conductance Ca2+ -activated *Correspondence: K+ channels (SKca), strongly antagonized this effect. We conclude that HNO relaxes Ervice Pouokam the gastrointestinal tract musculature by hyperpolarizing myocytes via activation of the Ervice.Pouokam- Kamgne@vetmed.uni-giessen.de; sGC/cGMP pathway similarly to NO, not only inhibiting the RhoK and activating MLCP pouokamer@yahoo.de as do both NO and H2 S but also increasing cytosolic Ca2+ for activation of SKCa contributing to hyperpolarization. Specialty section: This article was submitted to Keywords: Ca2+ , gasotransmitter, membrane potential, motility, nitroxyl (HNO), small conductance Ca2+ - Gastrointestinal Sciences, activated K+ channels (SKca), soluble guanylate cyclase a section of the journal Frontiers in Physiology Received: 23 October 2019 INTRODUCTION Accepted: 08 April 2020 Published: 03 June 2020 Gastrointestinal motility is regulated by enteric neurotransmitters inducing contraction and Citation: relaxation of the smooth muscle cells. Excitatory neurotransmitters such as acetylcholine (ACh) Gastreich-Seelig M, Jimenez M elicit smooth muscle cell depolarization and contraction, whereas inhibitory neurotransmitters and Pouokam E (2020) Mechanisms cause smooth muscle hyperpolarization and relaxation. Inhibitory junction potentials are mainly Associated to Nitroxyl (HNO)-Induced Relaxation in the Intestinal Smooth Abbreviations: AOAA, amino-oxyacetate; CLA, β-cyano-L-alanine; DTT, 1,4-dithiothreitol; GI, gastrointestinal; L-NAME, Muscle. Front. Physiol. 11:438. Nω -nitro-L-arginine methylester hydrochloride; MLCP, myosin light chain phosphatase; ODQ, 1H-[1,2,4]oxadiazolo doi: 10.3389/fphys.2020.00438 [4,3-a]quinoxalin-1-one; RhoK, Rhokinase; RMP, resting membrane potential, sGC, soluble guanylate cyclase. Frontiers in Physiology | www.frontiersin.org 1 June 2020 | Volume 11 | Article 438 Gastreich-Seelig et al. HNO and Intestinal Motility due to ATP (or a related purine) and NO, causing a transient fast interplay between SNP and H2 S leads to HNO production inhibitory junction potential (IJPf) followed by a sustained IJP (Filipovic et al., 2013). Both the NO donor SNP and H2 S (IJPs), respectively. Hydrogen sulfide (H2 S) and nitric oxide (NO) [delivered by sodium sulfide (Na2 S)] undergo a fast chemical are two signaling molecules in the enteric nervous system with reaction at pH 7.4 and under aerobic conditions, forming the inhibitory effects on gastrointestinal smooth muscle (Schemann, intermediate [(CN)5 FeN(O)SH]3− . In a second reaction step, the 2005; Wood, 2008; Gallego et al., 2014; Jimenez et al., 2014). coordinated HSNO/SNO− is reduced by H2 S, which becomes These two signaling molecules exert their actions on oxidized to disulfide. The resulting disulfides may undergo smooth muscles via different mechanisms. In canine colonic further oxidations to polysulfides. The product of this second smooth muscle cells, exogenous NO caused accumulation step is the intermediate [(CN)5 Fe(HNO)]3− , which is converted of cytosolic cGMP (Shuttleworth et al., 1993). Endogenous to thiocyanate products after reaction with polysulfides and NO either depresses the release of ACh from interneurons elimination of nitroxyl (HNO) (Filipovic et al., 2013; Ivanovic- in descending enteric pathways or facilitates ACh release in Burmazovic and Filipovic, 2019). the ascending pathways; both mechanisms involve soluble Such a conversion of cyanide into thiocyanate being catalyzed guanylate cyclase (sGC) (Smith and McCarron, 1998). The by rhodaneses in vivo allowing therefore for the elimination of relative contribution of these pathways in the action of toxic cyanide (for review, see Mueller, 2006; Cipollone et al., NO may differ depending on either the species or the 2007). Thus, the combination NO/H2 S and therefore HNO segment of the gastrointestinal tract concerned. In contrast, appears therapeutically very interesting, as it is less toxic than H2 S does activate the opening of ATP-sensitive K+ (KATP ) the combination SNP/thiosulfate used for example in acute channels and the closure of voltage-dependent K+ channels hypertensive crises to regulate blood pressure (Butler Anthony (Kv ) in guinea pig gastric antrum myocytes (Zhao et al., and Feelisch, 2008; Filipovic et al., 2013). An interaction between 2009). In human and rat and mouse colon and jejunum, NO and H2 S in the presence of oxygen may lead to nitrous oxide relaxant effects of H2 S are dependent on apamin-sensitive (N2 O), sulfur, and water as described previously (Kurtenacker small conductance Ca2+ -activated K (SK) and on KATP and Löschner, 1938; Koppenol and Bounds, 2017) and shown channels (Gallego et al., 2008). In addition, activation of the in the following Reaction (1). The single-step formation of N2 O myosin light chain phosphatase (MLCP) may account for the being unlikely, the latter may result from dimerization of HNO relaxing properties of H2 S as shown in murine gastric fundus as shown by Reaction (2) (Kohout and Lampe, 1965; Bartberger (Dhaese and Lefebvre, 2009). et al., 2002). A third gaseous signaling molecule, nitroxyl (HNO), is getting [O2 ] more attention (Irvine et al., 2008; Pouokam et al., 2013; H2 S + 2NO −→ N2 O + S + H2 O (1) Eberhardt et al., 2014; Cowart et al., 2015). The most used HNO donor is Angeli’s salt, which decomposes spontaneously at 37◦ C in physiological solutions at neutral pH to produce equimolar 2HNO−→N2 O + H2 O (2) amounts of nitrite and HNO with a half-life of about 2.3 min. The formed HNO may in turn dimerize with elimination of water It has been recently confirmed that NO and H2 S cooperatively producing inert N2 O (Katrina et al., 2005). generate HNO in cells, as for example, a strong increase in the Its positive inotropic and lusitropic effects are important HNO-sensor CuBOT1 was observed in dorsal root ganglia (DRG) therapeutic actions in the heart. So far, HNO properties have neurons only when both gases were simultaneously applied been investigated mainly in the cardiovascular system (Irvine (Eberhardt et al., 2014). Similarly, this interplay was revealed with et al., 2008; Eberhardt et al., 2014). In the gastrointestinal tract, new generations of HNO-sensitive dyes such as TP-Rho-HNO upon repetitive application in Ussing chambers, the HNO-donor (Zhou et al., 2017) or HNOCL-1 (An et al., 2019) as endogenous Angeli’s salt induced a Ca2+ -dependent Cl− secretion without NO may be transformed to HNO by H2 S (Eberhardt et al., 2014), desensitization (Pouokam et al., 2013). The mechanism of action some effects of H2 S may be ascribed to HNO. clearly differed from that of the secretion induced by its sibling Since HNO presents interesting clinical properties, we aimed NO because it did not depend on the activity of the sGC, which to investigate the properties of this agent in the regulation is the prototypical site of action of NO. In contrast, HNO- of intestinal contractility, revealing its specific characters with evoked secretion was blocked by indomethacin, suggesting that regard to the well-known signaling molecules NO and H2 S. To cyclooxygenase metabolites such as prostaglandins mediate the our knowledge, this is the first report studying neuromuscular response, to which an activation of the basolateral Na+ -K+ - actions of HNO in the GI tract. ATPase, Ca2+ -dependent K+ , and ATP-sensitive K+ channels contributed (Pouokam et al., 2013). Recent studies showed a cross-talk of H2 S and NO signaling MATERIALS AND METHODS pathways (Yong et al., 2011; Filipovic et al., 2013; Miljkovic et al., 2013; Eberhardt et al., 2014; Zhou et al., 2017; An et al., Animals 2019). It is known that sodium nitroprusside (SNP) does release Female and male Wistar (160–220 g) or Sprague–Dawley NO in ambient light but not in the dark and that combining rats (12–18 weeks old) were used. The animals were bred equimolar SNP and H2 S does not allow NO release in ambient and housed at the Institute for Veterinary Physiology and light (Filipovic et al., 2013). It has been shown that a direct Biochemistry of the Justus-Liebig-University Giessen (Wistar Frontiers in Physiology | www.frontiersin.org 2 June 2020 | Volume 11 | Article 438 Gastreich-Seelig et al. HNO and Intestinal Motility rats) or at the Institute of Veterinary Physiology of the Isolation and Identification of Myocytes Universitat Autònoma de Barcelona (Sprague–Dawley rats) at Ca2+ imaging experiments were performed at isolated myocytes an ambient temperature of 22.5◦ C and air humidity of 50– from the proximal colon. The longitudinal muscle layer (devoid 55% on a 12 h/12 h light–dark cycle with free access to of mesenterium) from the colon was removed and cut in small water and food until the time of the experiment. The animals pieces of about 0.1 × 0.1 cm. These pieces were collected and were anesthetized with CO2 and killed by exsanguination. The transferred into the digestion solution consisting of collagenase Sprague–Dawley rats were used for microelectrode experiments type II (0.5 mg ml−1 ) and trypsin inhibitor (0.25 mg ml−1 ) and the Wistar rats for others. Experiments were approved by in Hank’s balanced salt solution (Ca2+ - and Mg2+ -free). If the named animal welfare officer of the Justus Liebig University cells from the circular muscle layer had to be prepared, after (administrative number 577_M) or by the Ethics Committee of stripping away the longitudinal muscle layer, the mucosa layer the Universitat Autònoma de Barcelona (administrative number was scrapped off using the edge of a glass slide; the remaining MJF-eut/01) and performed according to the German and tissue was minced, then proceeded as with the longitudinal European animal welfare law. layer. The tissues were then enzymatically digested for 30 min at 37◦ C then vortexed for 10 s and centrifuged 2,000 rpm for 3 min. The supernatant was discarded, and the digestion Isometric Force Measurements solution was added to the pellet. After shortly mixing, a For isometric force measurements, the muscle strips were second incubation period at 37◦ C for 30 min followed. The obtained as follows: the abdomen of euthanized animals was preparation was then vortexed for 5 s and centrifuged at opened, the ileum and colon were collected, and the lumen 1,500 rpm for 3 min. The supernatant was discarded, and was carefully cleaned. After a short period in ice-cold buffer the myocytes were resuspended in Dulbecco’s modified Eagle’s gassed with 5% (v/v) CO2 /95% (v/v) O2 , the tissues were cut medium (DMEM) F12 medium supplemented with fetal calf into 1.5 cm long pieces, which were then fixed in the organ serum until the beginning of measurements. At the end of bath. The chamber was filled with warm (37◦ C) and gassed (5% Ca2+ imaging experiments, tested myocytes were additionally CO2 /95% O2 , v/v) buffer solution. For longitudinal muscle strips, morphologically identified. After fixation for 10 min in 4% the pretension was set at 1.5 g, and after an equilibrium period (w/v) paraformaldehyde (PFA) at 37◦ C, their actin filaments of at least 15 min, the tension was lowered to 1 g. For circular were stained by fluorescein isothiocyanate (FITC)-conjugated muscle strips (1 × 0.5 cm), the tension was continuously set at phalloidin (800 nM) first at room temperature in the presence 1 g. The baseline was measured for 5 min before administration of 0.1% (v/v) Triton-X100, then overnight at 4◦ C. Nuclei were of any drug. As viability control, 10 µM carbachol and/or 30 mM counterstained with 40 ,6-diamidino-2-phenylindole (DAPI) for KCl were administered at the end of each experiment. 5 min. Staining was repeated at least three times with cells Isometric force was measured via a BioAmp-04/8 amplifier from different animals. Myocyte staining was analyzed using system and sampled via an A/D-converter with a sampling a fluorescence microscope (80i, Nikon, Düsseldorf, Germany). rate of 1 Hz (Föhr Medical Instruments, Seeheim, Germany). Digital images were taken with black and white camera (DS-2M For data analysis, the baseline just prior administration of B/Wc) using NS Elements 2.30 software. a drug was measured as mean over 1 min. To calculate the maximal relaxation induced by Angeli’s salt, the maximal reduction in muscle tone within 5 min after administration Microelectrode Recordings of the agonist was calculated and expressed as difference to For microelectrode recordings, the colon of Sprague–Dawley rats the baseline just prior administration of Angeli’s salt (1 Force; was removed and placed in carbogenated (95% O2 and 5% CO2 , see, e.g., Figure 4). As the action of Angeli’s salt on muscle v/v) Krebs solution, then opened along the mesenteric border. tone was only transient, in addition the changes in the area The mucosal and submucosal layers were gently removed and under the curve (AUC) over a 3-min interval (g·180 s) before 3 × 5 mm muscle strips were cut in a circular direction. and after administration of the HNO donor was calculated Electrophysiological experiments were performed with (see inset in Figure 3). To determine the frequency of phasic colonic strips pinned in a Sylgard-coated chamber with spontaneous contractions, first time derivatives (dg/dt) were the circular muscle layer facing upwards. The tissue was calculated. Only waves passing a threshold set at 0.1 g s−1 continuously perfused with carbogenated Krebs solution at were counted as contraction during a 1-min period before 37 ± 1◦ C and allowed to equilibrate for 1 h. Phentolamine, (control) and 3 min after administration of HNO. The calculated propranolol, and atropine (all at 1 µM) were added to block frequency was expressed as contractions/min (cpm). To prove α- and β-adrenoceptors and muscarinic receptors, respectively, whether Angeli’s salt effects were due to HNO and not nitrite, so that the action of Angeli’s salt could be measured under a control with decomposed Angeli’s salt was performed (see non-adrenergic non-cholinergic (NANC) conditions. To obtain Supplementary Figure S1) as previously done (Paolocci et al., stable microelectrode impalements, nifedipine 1 µM was added 2001; Fukuto et al., 2008) to prove that the decomposed to abolish mechanical activity. Circular smooth muscle cells form has different biological properties. The experiments were were impaled using glass microelectrodes filled with 3 M conducted in ambient light. For nitric oxide (NO) delivery, KCl (30–60 M of tip resistance). Membrane potential was SNP was used, as it is known to release NO in ambient light measured using a standard Duo 773 electrometer (WPI Inc., (Filipovic et al., 2013). Sarasota, FL, United States). Tracings were displayed on an Frontiers in Physiology | www.frontiersin.org 3 June 2020 | Volume 11 | Article 438 Gastreich-Seelig et al. HNO and Intestinal Motility oscilloscope (Racal-Dana Ltd., Windsor, United Kingdom) and Angeli’s salt (Cayman Chemical, Ann Arbor, MI, simultaneously digitalized (100 Hz) with a PowerLab 4/30 system United States) was dissolved in 0.01 N NaOH. Indomethacin and Chart 5 software for Windows (both from ADInstrument, and Nifedipine were dissolved in 96% (v/v) ethanol. Calyculin Castle Hill, NSW, Australia). IJPs were elicited by electrical field A (New England, Ipswitch, United States), glibenclamide, and stimulation (EFS) using two silver chloride plates placed 1.5 cm ODQ (Tocris, Bristol, United Kingdom) were dissolved in apart perpendicular to the longitudinal axis of the preparation. dimethyl sulfoxide (DMSO). Tetrodotoxin (TTX) (Alomone The protocol consisted of single pulse trains of EFS (0.4 ms Labs, Jerusalem, Israel) was dissolved in citrate buffer. Amino- pulse duration) at supramaximal voltage (30–40 V). The resting oxyacetate (AOAA), apamine (Alomone Labs, Jerusalem, Israel), membrane potential (RMP) and both the amplitude as well as atropine sulfate, carbachol, β-cyano-l-alanine (CLA), MRS2500 the duration of the IJP were measured and compared before and (Tocris, Bristol, United Kingdom), 1,4-dithiothreitol (DTT), after drug incubation. Nω -nitro-L-arginine methylester hydrochloride (L-NAME), phentolamine, propranolol, sodium hydrogen sulfide (NaHS), sodium nitroprusside (SNP; Enzo Life, Lausen, Switzerland), and Ca2+ Imaging Y-27632 (Tocris, Bristol, United Kingdom) were dissolved in The myocytes suspension was spun down at 1,500 rpm for 1 min. aqueous solutions. If not stated explicitly, drugs were purchased After discarding the supernatant, the cells were resuspended from Sigma Aldrich, St Louis, United States. in warm (37◦ C) Tyrode solution containing the Ca2+ -sensitive fluorescent dye fura-2 AM (6 µM; Life Technologies, Darmstadt, Germany) and pluronic acid (1.2 mg L−1 ; Life Technologies, Data Analysis and Statistics In general, results are given as means ± standard error of Darmstadt, Germany) and incubated for 1 h at 37◦ C. After the mean (SEM). The number of investigated tissues or cells incubation, 30 µl of the myocyte suspension was spread is indicated “n.” For imaging experiments (Ca2+ imaging and onto a poly-L-lysine-coated coverslip and incubated at room immunostaining), the same protocol had to be run at least three temperature for 15 min, before it was washed carefully. The times on different days and with different animals. When means coverslip was then mounted in the experimental chamber with of several groups had to be compared, an analysis of variance was a volume of about 3 ml. The preparation was superfused performed followed by Bonferroni or Tuckey post-hoc test. For hydrostatically with warm Tyrode solution (37◦ C). Perfusion the comparison of two groups, either a paired or an unpaired was stopped only if the drug applied had to be applied via a Student’s t-test or a Mann–Whitney U-test was applied. Statistical pipette. The perfusion rate was about 2 ml min−1 . Changes in analysis was performed with GraphPad Prism 6. Data were the cytosolic Ca2+ concentration were monitored as changes in considered significant when P < 0.05. the fura-2 ratio (R; emission at an excitation wave length of 340 nm divided by the emission at an excitation wave length of 380 nm). Experiments were carried out on an inverted RESULTS microscope (Olympus IX-50; Olympus, Hamburg, Germany), equipped with an epifluorescence setup and an image analysis Nitroxyl Relaxes Gastrointestinal Muscle system (Till Photonics, Martinsried, Germany). Several regions The effects of Angeli’s salt were due to HNO, as the decomposed of interest, each with the size of about one cell, were selected. form of the salt was inefficient (see Supplementary Figure S1). The emission above 420 nm was measured from the regions The HNO donor Angeli’s salt induced a reduction in the of interest. Data were sampled at 0.2 Hz. The baseline in the basal tone, a decrease in spontaneous contractions leading fluorescence ratio of fura-2 was measured for several minutes to a reduction in the AUC of all three intestinal muscle before drugs were administered. preparations tested. In ileal longitudinal muscle, concentrations above 25 µM induced a transient fall in muscle tone, which Solutions and Reagents lasted for 3–5 min (Figure 1B). Concentrations lower than 25 The organ bath Parsons solution consisted of (in mM): NaCl µM were ineffective (data not shown). Muscle tone remained 107, KCl 4.5, NaHCO3 25, Na2 HPO4 1.8, NaH2 PO4 0.2, CaCl2 stable in time-dependent controls treated with the solvent only 1.25, MgSO4 1, and glucose 12. The solution was gassed with (Figure 1A). In addition, contractions induced by KCl (30 mM) carbogen (5% CO2 in 95% O2 , v/v); pH was 7.4. The composition or carbachol (10 µM) were unaffected after washout of the HNO of the Krebs solution was (in mM): glucose 10.1, NaCl 115.5, donor (Figures 1A,B). Concentration–response curves based NaHCO3 21.9, KCl 4.6, NaH2 PO4 1.1, CaCl2 2.5, and MgSO4 1.2 either on the amplitude of the maximal relaxation induced by bubbled with a carbogen (pH 7.4). The myocytes were dissociated the HNO donor (Figure 3A) or on the integrated change in in Ca2+ - and Mg2+ -free Hank’s balanced salt solution (HBSS, muscle tone over a 3-min period (Figure 3B) revealed a flat Life Technologies, Paisley, United Kingdom). For superfusion of concentration dependence of the effect of Angeli’s salt. the isolated myocytes during the imaging experiments, a Tyrode In addition, longitudinal muscle from the proximal solution was used (in mM): NaCl 140, KCl 5.4, CaCl2 1.25, colon responded with a concentration-dependent relaxation MgSO4 1, HEPES 10, and glucose 12.2; pH was 7.4. For additional (Figures 1D,E, 3C,D), which was not observed in parallelly identification of myocytes, FITC-conjugated phalloidin (Sigma- performed time-dependent control experiments (Figure 1C). Aldrich, St Louis, United States) and Roti-Mount Fluor DAPI In contrast to the ileum, the longitudinal muscle strips from (Carl Roth, Karlsruhe, Germany) were used. the colon responded to the HNO donor at much lower Frontiers in Physiology | www.frontiersin.org 4 June 2020 | Volume 11 | Article 438 Gastreich-Seelig et al. HNO and Intestinal Motility FIGURE 1 | Relaxation induced by different concentrations of Angeli’s salt (1–400 µM, arrows) on longitudinal segments of (B) rat ileum or (D,E) proximal colon and (A) or (C) time-dependent control experiments, where only the solvent of Angeli’s salt was administered. The inset (middle) shows an original tracing of an individual colonic longitudinal muscle strip with a transient relaxation induced by 1 µM Angeli’s salt, which is hard to recognize in the ensemble average depicted in (D) resulting from the averaging of eight muscle strips responding asynchronously with a short relaxation induced by the HNO donor. Line interruptions are caused by omitting washing periods of about 10 min, where the content of the organ bath was exchanged three times before the next concentration of Angeli’s salt was administered. KCl (30 mM) and or carbachol (CCh; 10 µM) were used to check tissue viability. Values are means of individual tracings (black lines) ± SEM (gray lines), n = 5–11. For statistics (B,D,E) are plotted in Figures 3A–D, respectively. concentrations. However, with higher concentrations of Angeli’s the HNO donor was concentration dependent and showed a salt, the duration of the induced relaxation was prolonged. maximum at a concentration of 50 µM with respect to the Contraction induced by carbachol (10 µM) was unaffected amplitude of relaxation (Figure 3E) and the decrease in the area (Figures 1D,E). under the curve over a 3-min period (Figure 3F). Angeli’s salt not A relaxing effect induced by Angeli’s salt was also observed in only reduced basal muscle tone but also suppressed the phasic circular muscle preparations from proximal colon (Figure 2B) contractions as shown when considering records from individual in comparison to a control series (Figure 2A). The effect of muscle preparations with an extended time scale (Figures 2C,D). Frontiers in Physiology | www.frontiersin.org 5 June 2020 | Volume 11 | Article 438 Gastreich-Seelig et al. HNO and Intestinal Motility FIGURE 2 | The HNO donor Angeli’s salt (25–100 µM, arrows) induces a relaxation of circular segments of (B) rat proximal colon compared to a (A) time-dependent control, where only the solvent of Angeli’s salt was administered. Line interruptions are caused by omitting washing periods of about 10 min, where the content of the organ bath was exchanged three times before the next concentration of Angeli’s salt was administered. KCl (30 mM) and carbachol (CCh; 10 µM) were used to check tissue viability. Data in (A,B) are means of individual tracings (black lines) ± SEM (gray lines), n = 5–6. (C,D) are higher magnifications of 5-min intervals of an individual muscle preparation contained in the respective ensemble averages in (A,B). For statistics (B), is plotted in Figures 3E,F. The frequency of phasic spontaneous contractions of of the HNO donor, 50 and 100 µM, were selected for further longitudinal muscle strips from the colon was significantly isometric measurements. (p < 0.05, Mann–Whitney U-test) reduced by Angeli’s salt. At 50 µM, the HNO donor reduced the frequency from 3.0 ± 0.6 (n = 7) cpm to 0.6 ± 0.0 cpm (n = 7). Similarly, they were reduced HNO Interferes With the Cholinergic from 1.8 ± 0.6 cpm (n = 8) to 0.0 ± 0.0 cpm (n = 8) with 100 Pathway for Relaxation µM of the donor. In order to investigate a possible interaction of HNO with All further experiments were performed with longitudinal MLCP, longitudinal muscle strips of the proximal colon were muscle strips from the proximal colon and two concentrations pre-contracted with 10 µM carbachol, and the amplitude of Frontiers in Physiology | www.frontiersin.org 6 June 2020 | Volume 11 | Article 438 Gastreich-Seelig et al. HNO and Intestinal Motility FIGURE 3 | Concentration-dependent relaxation induced by Angeli’s salt in (A,B) ileal longitudinal muscle, (C,D) colonic longitudinal, and (E,F) colonic circular muscle. The relaxing effect is either expressed as maximal reduction in muscle tone (1g; difference to baseline just prior to administration of the drug (A,C,E) or reduction in the area under the curve (AUC) over a 3-min period compared to the 3-min period just prior to administration of Angeli’s salt, as illustrated by the schematic inset where the arrow marks the administration of Angeli’s salt. Concentration–response curves in the colonic longitudinal muscle was constructed from two independent series of experiments (see Figure 1), in which the effect of 1–20 µM and 25–400 µM Angeli’s salt was tested, respectively. Values are means ± SEM, n = 6–11. relaxation caused by HNO was compared in the presence or Y-27632 (inhibitor of Rho-kinase). A preliminary test for the absence of one of the following smooth muscle contractile validity of Y-27632 (10 µM) – as inhibitor of the cholinergic apparatus desensitizers calyculin A (inhibitor of the MLCP) or pathway – was conducted. Indeed, this inhibitor significantly Frontiers in Physiology | www.frontiersin.org 7 June 2020 | Volume 11 | Article 438 Gastreich-Seelig et al. HNO and Intestinal Motility FIGURE 4 | Inhibition of Rho kinases by Y-27632 (10 µM) prevents the contraction induced by CCh (10 µM) of longitudinal muscle strips from (A) proximal colon and reduces the relaxation induced by HNO donor Angeli’s salt (AS). Data in (B) are expressed as difference between the muscle tone just prior to administration of Angeli’s salt and the muscle tone averaged over a 1-min period starting 1 min after administration of the HNO. Potentiating (C) the contractile action of the cholinergic system with calyculin A (100 nM) strengthens or weakens the (D) relaxing action of HNO. Values are means ± SEM, n = 7–12. *P < 0.05, ***P < 0.001 vs. control in the absence of the corresponding blocker (Mann–Whitney U-test). For statistics, see text. reduced the contraction induced by CCh (Figure 4A). In the Interplay Between NO, H2 S, and HNO presence of Y-27632, the relaxation induced by 50 µM Angeli’s The relaxing properties of NO depend on the activation of sGC. salt was not significantly reduced from −0.59 ± 0.14 g (n = 7) The potential interplay between NO, H2 S, and HNO may also to −0.28 ± 0.05 g (n = 11), whereas the relaxation induced involve this enzyme. To test this hypothesis, the relaxing action by 100 µM of HNO donor was reversed into a contraction of these three molecules was measured in the presence of the sGC of +0.24 ± 0.03 g (n = 11, measured 1 min after administration of inhibitor ODQ (10 µM). As expected, the inhibitor significantly the donor) in comparison to the untreated control, where HNO blocked (Figure 5A) the relaxation induced by the NO donor evoked a relaxation of 0.35 ± 0.03 g (n = 7, Figure 4B). sodium nitroprusside (SNP, 1 mM) from −1.02 ± 0.23 g (n = 6) A potentiation of the cholinergic-evoked contraction is to −0.32 ± 0.05 g (n = 9). Among the other donors, only 50 expected by inhibiting MLCP, for example with calyculin A. µM Angeli’s salt was significantly sensitive to the blocker as Indeed, the contractile response induced by CCh was slightly the relaxation was reduced from −0.37 ± 0.11 g (n = 7) to increased from 0.71 ± 0.28 g (n = 10) to 0.97 ± 0.15 g −0.03 ± 0.06 g (Figure 5C, n = 8). The relaxation induced by (n = 11) in the presence of 100 nM calyculin A (Figure 4C). the H2 S donor sodium hydrogen sulfide (NaHS 100 µM) was MLCP blockade slightly potentiated the relaxing effect of not affected by ODQ; it amounted to −0.88 ± 0.24 g (n = 5) 50 µM Angeli’s salt and attenuated numerically the relaxing in the absence and −0.61 ± 0.11 g (Figure 5B) in the presence response to 100 µM Angeli’s salt; none of these effects, however, of the sGC blocker (Figure 5B). Increasing the concentration of reached statistical significance (Figure 4D). However, blocking AS to 100 µM partially overcomes the ODQ-induced inhibition, endogenous ACh effects on the basal tone with 1 µM TTX suggesting the presence of sGC-dependent and potentially sGC- and 1 µM atropine did not influence Angeli’s salt (50 µM)- independent inhibitory pathways (Figure 5D). induced relaxation as it amounted to −0.49 ± 0.06 g (n = 24) To assess the potential importance of endogenously produced and −0.42 ± 0.06 g (n = 26) in decrease values or to NO in HNO-induced relaxation, the relaxing activity of Angeli’s −120.05 ± 14.80 g·180 s (n = 24) and −105.05 ± 14.90 g·180 s salt was measured after a 30-min pre-incubation period of the (n = 26) for AUC values, in the absence and the presence of both tissue with the nitric oxide synthase (NOS) inhibitor L-NAME inhibitors, respectively. (200 µM). L-NAME did not affect HNO-induced relaxation Frontiers in Physiology | www.frontiersin.org 8 June 2020 | Volume 11 | Article 438 Gastreich-Seelig et al. HNO and Intestinal Motility FIGURE 5 | (A,C) The relaxing effect of the nitric oxide donor sodium nitroprusside (SNP; 1 mM) and 50 µM Angeli’s salt (AS) was significantly inhibited in the presence of the sGC inhibitor ODQ (10 µM). (D) When the concentration of Angeli’s salt was increased to 100 µM, this inhibition failed to reach significance. (B) Relaxation induced by the H2 S donor sodium hydrogen sulfide (NaHS; 100 µM) was unaffected by ODQ. (E,F) Inhibition of NOS with L-NAME (200 µM) did not affect HNO-induced relaxation. n = 5–9. *P < 0.05, vs. control in the absence ODQ (Mann–Whitney U-test). For statistics, see text. (Figures 5E,F) as the maximal relaxation was statistically enzymes. Thus, amino-oxyacetate (AAOA, 5 mM, cystathionine- unchanged from −1.29 ± 0.18 g (control, n = 8) to −1.32 ± 0.27 g β-synthase inhibitor) and β-cyano-L-alanine (CLA, 5 mM, (test, n = 7) for 50 µM Angeli’s salt and from −0.68 ± 0.05 g cystathionine-γ-lyase blocker) were used in the presence or (control, n = 8) to −0.75 ± 0.12 g (test, n = 7) for 100 µM absence of the sGC blocker ODQ (10 µM). Inhibition of Angeli’s salt. Calculating AUC led to the same conclusion. Similar endogenous synthesis of H2 S alone led to the reduction in the values of AUC were calculated for control and test (in the basal tone by −0.56 ± 0.08 g (AUC, −81.54 ± 13.22 g·180 s, presence of L-NAME): −163.3 ± 20.2 g·180 s (control, n = 8) vs. n = 6) vs. −0.10 ± 0.01 g for control (AUC, −3.85 ± 0.84 g·180 s, −148.1 ± 19.4 g·180 s (test, n = 7) for 50 µM Angeli’s salt and n = 6). This reduction lasted ∼5 min. Preventing H2 S synthesis, −91.6 ± 6.2 g·180 s (control, n = 8) vs. −77.8 ± 17.3 g·180 s (test, however, only slightly reduced the relaxation induced by 50 µM n = 7) for 100 µM Angeli’s salt. Angeli’s salt as shown in Table 1. A simultaneous blockade of The impact of endogenously produced H2 S on HNO endogenous synthesis of H2 S and sGC almost abolished the action was investigated using inhibitors of the H2 S-producing Angeli’s salt-evoked relaxation (Table 1). Frontiers in Physiology | www.frontiersin.org 9 June 2020 | Volume 11 | Article 438 Gastreich-Seelig et al. HNO and Intestinal Motility TABLE 1 | Impact of AOAA/CLA in the presence or not of ODQ on Angeli’s TABLE 2 | Impact of protection of thiol groups on Angeli’s salt- or SNP- and salt-evoked relaxation. NaHS-evoked relaxation. 1 Force (g) AUC (g·180 s) N 1 Force (g) AUC (g·180 s) N Angeli’s salt alone −0.70 ± 0.05 −96.4 ± 6.8 6 Angeli’s salt alone −1.49 ± 0.25 −186.8 ± 25.9 5 AOAA/CLA + Angeli’s salt −0.54 ± 0.07 −71.7 ± 10.0 6 DTT + Angeli’s salt −0.83 ± 0.07∗ −90.4 ± 10.9∗ 8 Angeli’s salt alone −0.97 ± 0.13 −133.0 ± 24.0 5 NaHS + SNP −0.73 ± 0.09 −103.3 ± 18.9 5 AOAA/CLA/ODQ + Angeli’s salt −0.31 ± 0.04∗∗ −1.8 ± 1.5∗∗ 5 DTT + NaHS + SNP −0.59 ± 0.07 −76.4 ± 10.6 8 Blockade of endogenous synthesis of H2 S with both AOAA and CLA (both The reducing agent DTT (500 µM) significantly inhibited the relaxation induced by 5 mM) slightly reduced the relaxation induced by Angeli’s salt (50 µM) in Angeli’s salt (50 µM) but only slightly reduced the relaxation induced by NaHS (H2 S longitudinal muscle from rat colon. Additional blockade of the sGC with ODQ donor; 100 µM) and SNP (NO donor; 1 mM). Longitudinal muscle from rat colon. (10 µM) significantly inhibited the response to Angeli’s salt. ∗∗ P < 0.01 vs. ∗ P < 0.05 vs. control in the absence of the corresponding blocker (Mann–Whitney control in the absence of the corresponding blockers (Mann–Whitney U-test) (see U-test) (see Supplementary Figure S3). Supplementary Figure S2). The enteric inhibitory neurotransmitters NO, ATP, or related Nitroxyl is highly thiophilic and may interact with thiol purines and H2 S, may induce GI myocyte relaxation via groups. In the presence of DTT (500 µM), known to maintain activation of sGC/cGMP, or P2Y1 purine receptors, or KATP , SH groups in reduced state or reduce protein disulfide bonds, respectively (He and Goyal, 1993; Smith and McCarron, 1998; the relaxing property of HNO donor was significantly impaired Gallego et al., 2008; Grasa et al., 2009). To identify the target (Table 2). Simultaneous administration of the H2 S donor (NaHS, candidates for HNO-induced relaxation, the hyperpolarization 100 µM) and NO donor (SNP, 1 mM) evoked a relaxation that induced by Angeli’s salt was challenged by ODQ (10 µM, was partially but not significantly sensitive to DTT (Table 2). inhibitor of the sGC), the bee venom apamin (1 µM, inhibitor of the SKca ), and glibenclamide (100 µM, inhibitor of KATP ). Mechanism of Action and Particularity As expected, the amplitude of the IJP was strongly reduced of HNO by apamin (1 µM) from −22.89 ± 2.08 mV (n = 12) to Calcium Ions Mobilization for Relaxation −6.39 ± 0.76 mV (n = 5) or any drug combination based on it H2 S induces the opening of KATP channels and the closure of (apamin + ODQ, from −22.89 ± 2.08 mV, n = 12 to −3.83 ± 1.59 voltage-dependent K+ channels in guinea pig gastric antrum mV, n = 5) (Figures 7A,C). In the presence of apamin, only myocytes (Wang, 2012) or activates small conductance calcium- the second component associated to the IJPs was recorded activated potassium channels (SKCa ) in mouse colon (Zhao et al., (Figure 7A). This second component was sensitive to ODQ 2009). HNO has been shown to stimulate sarco-endoplasmic (Figure 7A). Accordingly, both MRS2500 and apamin reduced reticulum Ca2+ -ATPase (SERCA) in cardiomyocytes (Irvine the fast component of the IJP (Figures 7A–C), whereas ODQ et al., 2008). To assess the role of Ca2+ in the HNO-induced (and L-NAME, not shown) reduced the second component relaxation, calcium imaging was performed on isolated myocytes of the IJP (Figures 7A,C). Apamin additionally elicits a slight of the proximal colon. depolarization as the resting membrane potential (RMP) is In myocytes from the longitudinal layer, Angeli’s salt (50 shifted upwards (Figure 7A). All these results confirm the µM) caused a significant increase in the fura-2 ratio signal from cotransmission process already described in several species 0.21 ± 0.08 (n = 22) to 0.44 ± 0.09 (n = 48) (P < 0.001, paired including the rat colon (Grasa et al., 2009) and the human small t test). In myocytes isolated from the circular layer, the response intestine (Gallego et al., 2014). was weaker, as the HNO donor increased the fura-2 ratio from 0.008 ± 0.001 (control; n = 13) to 0.057 ± 0.011 (test; n = 15) HNO Targets sGC and SKCa (P < 0.001; unpaired t-test). A representative result of 6 controls As shown in Figure 8, only glibenclamide did not modify the and 13 tests is shown in Figure 6. All the cells tested responded HNO donor-induced hyperpolarization, as the amplitude of with significant increase in the fura-2 signal to 30 mM KCl, hyperpolarization did not significantly change in the presence confirming the viability of the cells used (Figures 6B,C). of this blocker (-9.08 ± 1.42 mV, n = 5) compared to control (Angeli’s salt 25 µM, −10.64 ± 1.11 mV, n = 19). HNO Elicits Membrane Hyperpolarization Apamin and ODQ significantly reduced the Angeli’s salt- of Colonic Myocytes evoked hyperpolarization from −10.64 ± 1.11 mV (n = 19) Accumulation of cytosolic Ca2+ as observed in fura-2 to −4.13 ± 0.98 mV (n = 5) for the first inhibitor and from experiments (Figure 6) might be responsible for alternative −10.64 ± 1.11 mV (n = 19) to −1.73 ± 0.51 mV (n = 5) mechanisms of relaxation different from activation of sGC for the second one. In addition, combining ODQ with either pathways (Figures 1–3). Consequently, HNO inducing relaxation glibenclamide or apamin resulted in a very strong inhibition of indicates a more complex mechanism. Electrophysiological Angeli’s salt effect (Figure 8F). For the first combination, the experiments on myocytes unveiled that the HNO donor Angeli’s hyperpolarization was significantly reduced from −10.64 ± 1.11 salt (25 µM) induced cell membrane hyperpolarization by mV (n = 19) to −3.55 ± 0.90 mV (n = 9) and from −10.64 ± 1.11 −10.64 ± 1.11 mV (n = 19) from a resting membrane potential mV (n = 19) to −1.39 ± 0.53 mV (n = 6) for the second one. These of −40 mV (-38 to −45 mV, 95% confidence interval) (Figure 8). data lead to the conclusion that the hyperpolarization evoked by Frontiers in Physiology | www.frontiersin.org 10 June 2020 | Volume 11 | Article 438 Gastreich-Seelig et al. HNO and Intestinal Motility FIGURE 6 | (A–C) Changes in the fura-2 ratio evoked by Angeli’s salt (50 µM) in isolated rat colonic myocytes and (D) mechanisms underlying relaxation. (A) Photograph of isolated myocytes loaded with fura-2. Insets show (i) a fura-2 loaded myocyte that responded to Angeli’s salt and the same cell (ii) after staining of actin filament with phalloidin (green) and nucleus with 40 ,6-diamidino-2-phenylindole (DAPI) (blue). Scale bar, 50 µm. (C) Angeli’s salt induces an increase in the fura-2 ratio compared to (B) a time-dependent control. KCl (30 mM) was used for cell viability control. Values are given as means (symbols) ± SEM (parallel continuous lines), n = 6 for the time-dependent control and n = 13 for the test group with the HNO donor. For statistics, see text. (D) HNO may induce relaxation either directly by activating MLCP or indirectly inhibiting the RhoK. Additional increase in cytosolic Ca2+ concentration activates SKca for hyperpolarization and relaxation. Blocking RhoK with Y-27632 shunts HNO effects, inducing a shift toward Ca2+ -mediated responses, which also activates MLCK for contraction. The main mechanism for HNO-induced relaxation consists in hyperpolarizing myocytes via activation of sGC/cGMP. HNO is mediated by the activation of SKca channels and sGC, the HNO was not reduced, either in the presence of apamin or latter being the predominant pathway. MRS2500 (Table 3). This might be due to a major contribution of sGC pathway in inhibitory mechanisms. Accordingly, we HNO Is a Tone Regulator tested the effect of apamin in the presence of ODQ. However, Based on the electrophysiological measurements, the question as shown in Figure 5C, Angeli’s salt (50 µM) barely relaxes arising as if the relaxation elicited by HNO in organ bath was longitudinal muscle strips by −7.36 ± 4.00 g·180 s (n = 7) and only relying on the activation of sGC or whether SKca were −5.97 ± 2.37 g·180 s (n = 7), in the absence or presence of involved had to be addressed. However, the relaxing action of apamin; respectively. Frontiers in Physiology | www.frontiersin.org 11 June 2020 | Volume 11 | Article 438 Gastreich-Seelig et al. HNO and Intestinal Motility tone and the frequency of phasic contractions are reduced by the signaling molecule. The recovery used to occur after 5 min from the onset of the HNO donor (Figure 2D as example). This recovery profile might be due to the short half-life of the donor, which is ∼2–3 min at physiological pH and temperature (Switzer et al., 2009) or the pathways used to achieve the relaxation. In addition, NO and H2 S present similar periods of action depending on their concentration or the mode of administration or even the donor used. Neuromuscular transmission regulates smooth muscle contraction by either excitatory transmitters like ACh or inhibitory transmitters like NO. The excitatory neurotransmitter ACh binds to muscarinic G-protein-coupled receptors, initiating a physiological cascade through Gαq/11 leading step by step to activation of phospholipase Cβ (PLCβ) and synthesis of IP3 , which binds to IP3 R on sarcoplasmatic reticulum for intracellular release of Ca2+ . In addition to this pathway, receptor-operated (ROC) and stretch-activated (SAC) non-selective cation channels are activated and positively regulate voltage-dependent calcium channels (VDCCs), allowing for Ca2+ entry. Another parallel pathway is the G protein regulation of GDP–GTP exchange factor (Rho-GEF), RhoA, and activation of Rho-kinases (RhoK). Activation of RhoK and PKC by diacylglycerol (DAG) or by Ca2+ lead to phosphorylation and therefore inactivation of MLCP, the enzyme causing relaxation by dephosphorylating light chain of myosin. Accumulating Ca2+ activates the protein calmodulin initiating the activation of kinases such as PKC or myosin light chain kinase (MLCK). The latter phosphorylates light chain of myosin launching cross-bridge cycling (Taylor and Stull, 1988; FIGURE 7 | Identification of the fast inhibitory junction potential (IJPf) and Sanders, 2008). Relaxation is operated by hyperpolarization sustained IJP (IJPs) evoked by electrical field stimulation (EFS) under non-adrenergic non-cholinergic (NANC) conditions. (A) Apamin blocks the counteracting depolarization that resulted from excitation. This IJPf [first component in (A)], while ODQ blocks the IJPs [second component in can be achieved by activation of Ca2+ -dependent potassium (A)]. Apamin induces a shift of the resting membrane potential (RMP) to more channels like SKca in smooth muscle cells or alternatively in positive values. (A) The IJPf isolated after suppression of the IJPs by ODQ PDGFRα + cells that might transduce purinergic inputs to was also sensitive against MRS2500 (B). (C) The amplitude of the IJP was smooth muscle cells (Kurahashi et al., 2014). Relaxation can also strongly inhibited by apamin; this inhibition was not enhanced, when, in addition, ODQ was administered. In contrast, the duration of the IJP was be achieved by activating MLCP or by extrusion or restoring of significantly reduced by ODQ; this effect was significantly enhanced when, in cytosolic Ca2+ . addition, glibenclamide or apamin was present. Concentrations of drugs were In the present study, blocking the RhoK by the specific apamin 1 µM, glibenclamide 100 µM, and ODQ 10 µM. Values in (C) are inhibitor Y-27632 (Uehata et al., 1997; Narumiya et al., 2000) means ± SEM, n = 3–12. *P < 0.05, ***P < 0.001, ****P < 0.0001, vs. control blocked the response to CCh (Figure 4A). This confirms that in the absence of any drug (analysis of variance followed by Bonferroni test). For statistics, see text. the agonist activates RhoK during muscarinic contraction. In the presence of this inhibitor, the HNO donor-induced relaxation was strongly reduced (Figure 4B), revealing an inactivating property of HNO on the RhoK in the absence of the blocker. DISCUSSION The contraction observed with 100 µM (Figure 4B) may indicate a switch in the pathway depending on the concentration of Nitric oxide and hydrogen sulfide are both known to relax HNO, which, when it is higher, may affect a possible persistent intestinal smooth muscle cells (Shuttleworth and Sanders, 1996; blockade of RhoK or activity of MLCP. A possible reason for this Teague et al., 2002). In the present study, we show that also unexpected contraction might be the rise in the cytosolic Ca2+ HNO, released by the donor Angeli’s salt, elicits a relaxation concentration induced by HNO (Figure 6), which might be able of both circular and longitudinal muscle strips from the rat to elicit a contraction, when the RhoK is inhibited. A contrasting ileum or colon (Figures 1–3) as do NO and H2 S in different result was obtained in resistance arteries where NO relaxing species including rat (Tam and Hillier, 1992; Boeckxstaens et al., activity was potentiated with Y-27632 (Bolz et al., 2003). 1993; Stark et al., 1993; Teague et al., 2002). This relaxation On the other hand, inhibition of MLCP activity for example reaches a maximum within 3 min after application of the donor, by its specific blocker calyculin A (Ishihara et al., 1989; Bolz et al., which also matches with the profile of NO or H2 S in their 2003) should strengthen the contractile mechanism initiated relaxing property (Gallego et al., 2008, 2018). Both the basal for example by CCh. A partial effect was observed indeed Frontiers in Physiology | www.frontiersin.org 12 June 2020 | Volume 11 | Article 438 Gastreich-Seelig et al. HNO and Intestinal Motility FIGURE 8 | The hyperpolarization caused by the HNO donor Angeli’s salt (25 µM; A, n = 19) is insensitive to glibenclamide (B,F, n = 5), but sensitive to apamin (D,F, n = 5) or highly sensitive to ODQ alone (C,F, n = 5) or in combination with apamin (E,F, n = 6) or glibenclamide (F, n = 9). The discontinuous horizontal lines represent the resting membrane potential (RMP). Values are means ± SD, n values refer to (F), while (A–E) are representative original tracings. **P < 0.01, ****P < 0.0001, vs. control in the absence of any drug (analysis of variance followed by Bonferroni test). For statistics, see text. Frontiers in Physiology | www.frontiersin.org 13 June 2020 | Volume 11 | Article 438 Gastreich-Seelig et al. HNO and Intestinal Motility TABLE 3 | Amplitude of Angeli’s salt-evoked relaxation in the presence of This approach would allow for HNO production/effects either apamin and MRS2500. from direct interaction SNP/H2 S or secondary from NO/H2 S. 1 Force (g) AUC (g·180 s) N Using higher concentrations of H2 S could help improving HNO production/effects. How exactly the interplay between both gases Angeli’s salt alone −0.62 ± 0.11 −91.9 ± 19.0 8 leads to HNO formation, however, is still subject to debate Angeli’s salt in the presence −0.71 ± 0.12 −100.0 ± 16.8 11 (Ivanovic-Burmazovic and Filipovic, 2019). of apamin The strong inhibition of the NO-evoked relaxation by ODQ Angeli’s salt alone −0.62 ± 0.14 −83.4 ± 17.4 5 (Figure 5A) was expected since animals that lack GC have Angeli’s salt in the presence −0.69 ± 0.05 −100.4 ± 8.7 5 of MRS 2500 no response to NO (Groneberg et al., 2010). Although several interactions between NO and H2 S have been reported (Szabo, Blockade of SKca with apamin (1 µM) or of purine receptors P2Y1 with MRS2500 2017), we did not observe a significant reduction in NaHS (1 µM) does not inhibit the relaxation evoked by Angeli’s salt (50 µM) in longitudinal muscle from rat colon (see Supplementary Figure S4). response when tissue was incubated with ODQ (Figure 5B). Based on the data obtained with L-NAME, it seems that endogenously synthesized NO does not directly contribute to the (Figure 4C). A previous work on arteries also showed increasing relaxation induced by exogenous HNO. contractions upon administration of calyculin A (Ishihara et al., It is known that AAOA may increase spontaneous motility 1989). Thus, pre-incubation with this inhibitor leads to an (Gil et al., 2011). However, in the present experiments, enhanced pre-contraction status of the tissues. Under this we observed a transient reduction in the basal tone when condition, the HNO donor’s relaxing effect was potentiated endogenous synthesis of H2 S was hindered by both AOAA (Figure 4D) at 50 µM, what is not expected considering a and CLA, indicating a contribution of endogenous H2 S to the potential inhibitory property of this agent on RhoK as mentioned maintenance of basal muscle tone of rat colonic musculature. The above. Such a property would have slightly weakened the relaxation induced by exogenous HNO may be partly potentiated MLCP inhibition by calyculin A, as MLCP is physiologically by endogenous H2 S, as the HNO-evoked maximal relaxation was inactivated by RhoK, but the potentiation observed would have slightly reduced by the combination AOAA/CLA (Table 1). This not occurred. Apparently, a pre-contraction may change the indicates a potential switch in the predominant pathway(s) for priority of HNO for target selection. This can be observed with relaxation depending on the concentration and the nature of the a higher concentration of the HNO donor (100 µM, Figure 4D) gasotransmitter(s) present in situ. Only additional administration where the expected reduction in relaxing action of HNO occurs, of ODQ in the presence of AOAA/CLA abolished AS-evoked indicating indeed a direct activation of MLCP or an indirect relaxation. If a cooperative interplay between NO and H2 S could activation through inhibition of RhoK by HNO. This property lead to endogenous synthesis of HNO, release of HNO could be activating MLCP is also known for NO (Bolz et al., 2003; indirectly sensed by its interaction with thiol groups knowing Gallego et al., 2016) and H2 S (Dhaese and Lefebvre, 2009). its high thiophilicity. This thiophilic property was revealed with Figure 6D summarizes the mechanisms of HNO actions on DTT (Table 2). The relaxation obtained by the simultaneously the contractile apparatus. Preventing the effects of endogenous administration of the NO donor SNP and the H2 S donor NaHS ACh with TTX and atropine could not influence Angeli’s salt- was partly sensitive to DTT, indicating indeed a partial thiophilic induced relaxation, indicating indeed that HNO may act in characteristic under both gaseous signaling molecules. Thus, non-pre-contracted tissues – that is under normal basal tone – exogenous NO and H2 S may partially lead to an HNO-like by preferentially selecting sGC/cGMP pathway and secondary effect. Whether endogenously produced NO cooperates with activating SKca channels. endogenous H2 S for HNO-like effects still has to be investigated It has been shown that NO released by a pure NO donor in more detail, as we used in the present study only one NOS like diethylamine-NON-Oate (DEA NON-Oate) interacts with inhibitor (L-NAME) at a single concentration. H2 S for HNO production or effects (Eberhardt et al., 2014; In contrast to previous findings showing an activation of Wild et al., 2015; Teicher et al., 2017). This reveals the interplay the SERCAs in cardiomyocytes (Cheong et al., 2005; Tocchetti between both gases NO and H2 S, contrasting with a potential et al., 2007; Sivakumaran et al., 2013), cytosolic accumulation direct interaction between SNP and H2 S as shown previously of Ca2+ upon administration of the HNO donor also makes (Filipovic et al., 2013). Combining NO from DEA NON-Oate sense, as this could activate SKca in colonic myocytes, leading to and H2 S from Na2 S synergistically induces release of calcitonin hyperpolarization and therefore to relaxation. The corresponding gene-related peptide (CGRP), a substance known to be released hyperpolarization was observed under non-cholinergic non- during HNO effects (Paolocci et al., 2001; Eberhardt et al., 2014; adrenergic conditions to isolate inhibitory responses (Figure 8). Wild et al., 2015; Teicher et al., 2017). In trigeminal neurons, Under these pharmacological conditions, the KATP channels Na2 S/H2 S is known to be the limiting factor for HNO-like known as H2 S target (Gallego et al., 2008; Zhao et al., 2009) effects (Wild et al., 2015). In the present study, we used a did not play a role in the response to HNO as shown by 10:1 concentration ratio for SNP and NaHS in order to release the missing sensitivity against glibenclamide, the prototypical excess NO in the ambient light. Under such conditions, a rapid blocker of this class of ion channels (Figure 8). On the contrary, interaction between SNP and H2 S should lead to little amounts the HNO-evoked hyperpolarization was dependent both on of HNO. The excessive release of NO from SNP is also expected sGC and SKca as the hyperpolarization was strongly reduced to interact – as does NO from DEA NON-Oate – with H2 S. by the corresponding inhibitors ODQ and apamin, respectively Frontiers in Physiology | www.frontiersin.org 14 June 2020 | Volume 11 | Article 438 Gastreich-Seelig et al. HNO and Intestinal Motility (Figure 8). One possibility is that HNO might be acting on enteric bonds, and the amplitude of its response may be dependent inhibitory neurons causing the release of purines acting on P2Y1 on the balance –S–S–/SH at its targets. In our previous study receptors that activate SKCa channels. However, MRS2500 did focusing on colonic epithelial ion transport (Pouokam et al., not modify the mechanical relaxation induced by HNO (Table 3), 2013), interaction of HNO with thiols was also observed, as and therefore, it is possible that the effect of HNO is not on the prosecretory property of HNO was sensitive to L-cysteine. enteric inhibitory neurons but a direct effect on post-junctional This property was Ca2+ dependent and mediated by activation cells. These results are consistent with calcium measurements in of the secretory machinery consisting among others of the colonic myocytes. Whether these mechanisms are located only in basolateral Na+ -K+ -ATPase, the Ca2+ -dependent K+ and the smooth muscle cells or other post-junctional cells such as ICC or KATP channels. The present work shows an involvement of Ca2+ PDGFRα + cells is currently unknown. As PDGFRα + cells may in HNO-induced relaxation. In contrast to the dependence of the express more SKca than smooth muscle cells (Koh et al., 1997; prosecretory property on eicosanoids revealed by the sensitivity Kurahashi et al., 2011; Baker et al., 2015), the involvement of of the HNO-induced secretion to 1 µM indomethacin (Pouokam such cells when considering the muscle strips and not isolated et al., 2013), in the present study, the same concentration of myocytes cannot be excluded. indomethacin does not reduce HNO-induced relaxation. Indeed, All these results indicate a clear activation of SKca and similar values of force or AUC were recorded in the absence sGC/cGMP pathway by HNO in colonic myocytes. Acute and presence of 1 µM indomethacin: 0.74 ± 012 g (n = 11) or analgesic effects of the HNO donor Angeli’s salt are prevented −101.20 ± 18.89 g·180 s (n = 11) for control and 0.72 ± 0.13 g by treatment with ODQ, glibenclamide, or KT5823 (inhibitor (n = 12) or −103.15 ± 20.45 g·180 s (n = 12) for test in of protein kinase G), suggesting a mechanism of action via the presence of the inhibitor. These observations indicate a activation of the cGMP/PKG/KATP pathway (Longhi-Balbinot predominant role of eicosanoids in the secretory process but et al., 2016). A further pathway that can be activated by HNO their secondary or minor role in mediating HNO effects on after interaction with sGC is the release of CGRP and activation the motility. of KATP channels as shown during vasodilatation (Favaloro and HNO alone may act as a hybrid signaling molecule between Kemp-Harper, 2007). Activation of sGC by HNO, as it is the NO and H2 S with more regulatory properties and some case with NO, may occur via coordination to the ferrous enzyme particularities. Consequently, HNO appears to be an excellent as shown on purified bovine lung sGC (Miller et al., 2009). It candidate for substituting NO and or H2 S in the therapy of is also known that the oxidized ferric enzyme is not influenced diverse GI tract disorders. Nitroxyl-prodrugs (temposil, dipsan) by HNO (Miller et al., 2009). Due to its high thiophilicity, have been used so far as alcohol deterrent agents or undergo HNO modifies regulatory thiols on sGC, leading to inhibition clinical trials (CXL 1427) (Cowart et al., 2015). Investigating on of the enzyme activity (Miller et al., 2009). Low concentrations the appropriate in situ concentration and the ideal therapeutic of Angeli’s salt up to 10 µM activate sGC from bovine lung, window as it is known for each drug (Paolocci et al., 2007) could whereas high concentrations from 100 µM began inhibiting be a great benefit for the GI tract. enzyme activity (Miller et al., 2009). It has been shown that HNO relaxes the gastrointestinal tract musculature by ODQ inhibits sGC by oxidation of the ferrous to the ferric hyperpolarization of myocytes via activation of the sGC/cGMP heme (Zhao et al., 2000). In Figure 5C, ODQ significantly pathway similarly to NO, inhibiting the RhoK and activating reduced Angeli’s salt effect, revealing the capacity of HNO to MLCP as do both NO and H2 S. HNO also increases cytosolic activate sGC more likely at the ferrous heme. At 100 µM Ca2+ for activation of SKCa , contributing to hyperpolarization. of the HNO donor (Figure 5D), ODQ failed to inhibit the Using two different rat strains led to same conclusions. The relaxation. Here, we could speculate that Angeli’s salt still can effects of HNO imply contribution of thiols and do not activate the heme counteracting ODQ action. HNO induces depend primarily on eicosanoids. To our knowledge, this intradisulfide formation in PKGIα, therefore activation of the study is the first one showing the effects of HNO on the kinase leading to vasorelaxation in mesenteric arteries in vitro intestinal motility. and arteriolar dilatation in vivo in mice. This activation of the Although the pharmacological tools and the techniques used kinase is similar to binding of cGMP (Donzelli et al., 2017). isometric force measurements, electrophysiology, Ca2+ imaging, NO activates the sGC/cGMP pathway, but other types of K+ and immunostaining help understanding the mechanistic of channels, namely, the KNO1, KNO2, and the BK channels HNO actions, there are some limitations that we will need (Koh et al., 1995). to tackle in further studies. Indeed, there is a need to We can conclude that HNO presents similarities with NO better understand the mechanism of action of HNO on the and H2 S when it comes to mechanisms of action or pathways contractile apparatus, with focus on the single components of this regulated. All three gaseous signaling molecules cause GI machinery. Sarcoplasmatic/endoplasmatic reticulum calcium tract relaxation. HNO causes relaxation via activation of the ATPase (SERCA) 2 gene encodes two spliced variants SERCA2a sGC/cGMP pathway leading to hyperpolarization (like NO), and SERCA2b. The SERCA2a is specific for heart tissue (Khan activation of SKCa channels (like H2 S), activation of MLCP et al., 1990; Spencer et al., 1991; Lytton et al., 1992) but can be (like both NO and H2 S), and inhibition of RhoK. It also expressed in low amounts in smooth muscle (Vandecaetsbeek corrects impaired pathways and induces a fine tuning between et al., 2009), whereas SERCA2b is mostly expressed by smooth motility-regulating properties of NO and H2 S. HNO may also muscle cells and non-muscle cells (Spencer et al., 1991; Lytton present a motility “sensor” property. It interferes with disulfide et al., 1992) SERCA2b apparently presents a twofold higher Frontiers in Physiology | www.frontiersin.org 15 June 2020 | Volume 11 | Article 438 Gastreich-Seelig et al. HNO and Intestinal Motility affinity for cytosolic Ca2+ ions and a lower maximal turnover number 577_M) and the Ethics Committee of the Universitat rate (Vandecaetsbeek et al., 2009). Other studies, however, point Autònoma de Barcelona. out a contradictory property, a rather lower turnover rate for Ca2+ transport and ATP hydrolysis (Lytton et al., 1992). The lower expression of SERCAa in smooth muscle cells may AUTHOR CONTRIBUTIONS account for potential difference in Ca2+ signaling compared to cardiomyocytes. However, we cannot confirm this since we did EP designed the experiments. MG-S, MJ, and EP performed not look for the expression of SERCA2 in colonic myocytes used the in vitro experiments, analyzed the recordings, and in our study for calcium imaging. We used isolated myocytes wrote the manuscript. deprived from vascular tissue. In addition, how HNO activates sGC or modulates cGMP still has to be investigated. Endogenous production of HNO remains a striking issue, as very small FUNDING amounts if produced are difficult to detect and so a potential This work was supported by Deutsche Forschungsgemeinschaft, interplay of endogenous NO and H2 S remains questionable. Grant PO 2143/1-1 and the ERASMUS + Program. Developing more HNO donors with longer half-life is needed to explore the multitude of HNO effects. ACKNOWLEDGMENTS DATA AVAILABILITY STATEMENT The diligent care of Mrs. B. Buß, E. Haas, B. Schmitt, A. Stockinger, and Emma Martínez is a pleasure to acknowledge. We The datasets generated for this study are available on request to thank Prof. Martin Diener for helpful discussion. the corresponding author. SUPPLEMENTARY MATERIAL ETHICS STATEMENT The Supplementary Material for this article can be found The animal study was reviewed and approved by Animal online at: https://www.frontiersin.org/articles/10.3389/fphys. welfare officer of the Justus Liebig University (administrative 2020.00438/full#supplementary-material REFERENCES Cowart, D., Venuti, R., Guptill, J., Noveck, R., and Foo, S. (2015). A phase 1 study of the safety and pharmacokinetics of the intravenous nitroxyl prodrug, An, W., Ryan, L. S., Reeves, A. G., Bruemmer, K. J., Mouhaffel, L., Gerberich, CXL-1427. J. Am. Coll. Cardiol. 65(10 Suppl.), A876. J. L., et al. (2019). A Chemiluminescent Probe for HNO quantification and Dhaese, I., and Lefebvre, R. A. (2009). Myosin light chain phosphatase activation real-time monitoring in living cells. Angew. Chem. Int. Ed. 58, 1361–1365. is involved in the hydrogen sulfide-induced relaxation in mouse gastric fundus. doi: 10.1002/anie.201811257 Eur. J. Pharmacol. 606, 180–186. doi: 10.1016/j.ejphar.2009.01.011 Baker, S. A., Hennig, G. W., Ward, S. M., and Sanders, K. M. (2015). Temporal Donzelli, S., Goetz, M., Schmidt, K., Wolters, M., Stathopoulou, K., Diering, S., sequence of activation of cells involved in purinergic neurotransmission in the et al. (2017). Oxidant sensor in the cGMP-binding pocket of PKGIα regulates colon. J. Physiol. 593, 1945–1963. doi: 10.1113/jphysiol.2014.287599 nitroxyl-mediated kinase activity. Sci. Rep. 7:9938. Bartberger, M. D., Liu, W., Ford, E., Miranda, K. M., Switzer, C., Fukuto, J. M., Eberhardt, M., Dux, M., Namer, B., Miljkovic, J., Cordasic, N., Will, C., et al. et al. (2002). The reduction potential of nitric oxide (NO) and its importance (2014). H2S and NO cooperatively regulate vascular tone by activating to NO biochemistry. Proc. Natl. Acad. Sci. U.S.A. 99:10958. doi: 10.1073/pnas. a neuroendocrine HNO–TRPA1–CGRP signalling pathway. Nat. Commun. 162095599 5:4381. Boeckxstaens, G. E., Pelckmans, P. A., Herman, A. G., and Van Maercke, Y. M. Favaloro, J. L., and Kemp-Harper, B. K. (2007). The nitroxyl anion (HNO) is a (1993). Involvement of nitric oxide in the inhibitory innervation of the human potent dilator of rat coronary vasculature. Cardiovasc. Res. 73, 587–596. doi: isolated colon. Gastroenterology 104, 690–697. doi: 10.1016/0016-5085(93) 10.1016/j.cardiores.2006.11.018 91003-z Filipovic, M. R., Eberhardt, M., Prokopovic, V., Mijuskovic, A., Orescanin-Dusic, Bolz, S. S., Vogel, L., Sollinger, D., Derwand, R., de Wit, C., Loirand, G., et al. Z., Reeh, P., et al. (2013). Beyond H2S and NO interplay: hydrogen sulfide and (2003). Nitric oxide-induced decrease in calcium sensitivity of resistance nitroprusside react directly to give Nitroxyl (HNO). A new pharmacological arteries is attributable to activation of the myosin light chain phosphatase and source of HNO. J. Med. Chem. 56, 1499–1508. doi: 10.1021/jm3012036 antagonized by the RhoA/Rho kinase pathway. Circulation 107, 3081–3087. Fukuto, J. M., Jackson, M. I., Kaludercic, N., and Paolocci, N. (2008). “Examining doi: 10.1161/01.Cir.0000074202.19612.8c nitroxyl in biological systems,” in Methods in Enzymology. Nitric Oxide, Part Butler Anthony, R., and Feelisch, M. (2008). Therapeutic uses of inorganic nitrite F, eds E. Cadenas, and L. Packer (San Diego, CA: Academic Press), 411–431. and nitrate. Circulation 117, 2151–2159. doi: 10.1161/CIRCULATIONAHA. doi: 10.1016/s0076-6879(07)00826-9 107.753814 Gallego, D., Clave, P., Donovan, J., Rahmati, R., Grundy, D., Jimenez, M., et al. Cheong, E., Tumbev, V., Abramson, J., Salama, G., and Stoyanovsky, D. A. (2005). (2008). The gaseous mediator, hydrogen sulphide, inhibits in vitro motor Nitroxyl triggers Ca2+ release from skeletal and cardiac sarcoplasmic reticulum patterns in the human, rat and mouse colon and jejunum. Neurogastroenterol. by oxidizing ryanodine receptors. Cell Calcium 37, 87–96. doi: 10.1016/j.ceca. Motil. 20, 1306–1316. doi: 10.1111/j.1365-2982.2008.01201.x 2004.07.001 Gallego, D., Malagelada, C., Accarino, A., De Giorgio, R., Malagelada, J. R., Cipollone, R., Ascenzi, P., and Visca, P. (2007). Common themes and Azpiroz, F., et al. (2014). Nitrergic and purinergic mechanisms evoke inhibitory variations in the rhodanese superfamily. IUBMB Life 59, 51–59. doi: 10.1080/ neuromuscular transmission in the human small intestine. Neurogastroenterol. 15216540701206859 Motil. 26, 419–429. doi: 10.1111/nmo.12293 Frontiers in Physiology | www.frontiersin.org 16 June 2020 | Volume 11 | Article 438 Gastreich-Seelig et al. HNO and Intestinal Motility Gallego, D., Malagelada, C., Accarino, A., Gori, A., Malagelada, J. R., Azpiroz, Miljkovic, J. L., Kenkel, I., Ivanović-Burmazović, I., and Filipovic, M. R. (2013). F., et al. (2018). Functional neuromuscular impairment in severe intestinal Generation of HNO and HSNO from Nitrite by Heme-Iron-Catalyzed dysmotility. Neurogastroenterol. Motil. 30:e13458. doi: 10.1111/nmo.13458 Metabolism with H2S. Angew. Chem. Int. Ed. 52, 12061–12064. doi: 10.1002/ Gallego, D., Mane, N., Gil, V., Martinez-Cutillas, M., and Jimenez, M. (2016). anie.201305669 Mechanisms responsible for neuromuscular relaxation in the gastrointestinal Miller, T. W., Cherney, M. M., Lee, A. J., Francoleon, N. E., Farmer, P. J., King, tract. Rev. Esp. Enferm. Dig. 108, 721–731. doi: 10.17235/reed.2016.4058/2015 S. B., et al. (2009). The effects of nitroxyl (HNO) on soluble guanylate cyclase Gil, V., Gallego, D., and Jimenez, M. (2011). Effects of inhibitors of hydrogen activity: interactions at ferrous heme and cysteine thiols. J. Biol. Chem. 284, sulphide synthesis on rat colonic motility. Br. J. Pharmacol. 164, 485–498. 21788–21796. doi: 10.1074/jbc.M109.014282 doi: 10.1111/j.1476-5381.2011.01431.x Mueller, E. G. (2006). Trafficking in persulfides: delivering sulfur in biosynthetic Grasa, L., Gil, V., Gallego, D., Martín, M. T., and Jiménez, M. (2009). P2Y1 pathways. Nat. Chem. Biol. 2, 185–194. doi: 10.1038/nchembio779 receptors mediate inhibitory neuromuscular transmission in the rat colon. Br. Narumiya, S., Ishizaki, T., and Uehata, M. (2000). Use and properties of ROCK- J. Pharmacol. 158, 1641–1652. doi: 10.1111/j.1476-5381.2009.00454.x specific inhibitor Y-27632. Methods Enzymol. 325, 273–284. doi: 10.1016/ Groneberg, D., Konig, P., Wirth, A., Offermanns, S., Koesling, D., and Friebe, s0076-6879(00)25449-9 A. (2010). Smooth muscle-specific deletion of nitric oxide-sensitive guanylyl Paolocci, N., Jackson, M. I., Lopez, B. E., Miranda, K., Tocchetti, C. G., Wink, cyclase is sufficient to induce hypertension in mice. Circulation 121, 401–409. D. A., et al. (2007). The pharmacology of nitroxyl (HNO) and its therapeutic doi: 10.1161/circulationaha.109.890962 potential: not just the Janus face of NO. Pharmacol. Ther. 113, 442–458. doi: He, X. D., and Goyal, R. K. (1993). Nitric oxide involvement in the peptide VIP- 10.1016/j.pharmthera.2006.11.002 associated inhibitory junction potential in the guinea-pig ileum. J. Physiol. 461, Paolocci, N., Saavedra, W. F., Miranda, K. M., Martignani, C., Isoda, T., Hare, J. M., 485–499. doi: 10.1113/jphysiol.1993.sp019524 et al. (2001). Nitroxyl anion exerts redox-sensitive positive cardiac inotropy Irvine, J. C., Ritchie, R. H., Favaloro, J. L., Andrews, K. L., Widdop, R. E., and Kemp- in vivo by calcitonin gene-related peptide signaling. Proc. Natl. Acad. Sci. U.S.A. Harper, B. K. (2008). Nitroxyl (HNO): the Cinderella of the nitric oxide story. 98, 10463–10468. doi: 10.1073/pnas.181191198 Trends Pharmacol. Sci. 29, 601–608. doi: 10.1016/j.tips.2008.08.005 Pouokam, E., Bell, A., and Diener, M. (2013). Actions of Angeli’s salt, a nitroxyl Ishihara, H., Ozaki, H., Sato, K., Hori, M., Karaki, H., Watabe, S., et al. (1989). (HNO) donor, on ion transport across mucosa-submucosa preparations from Calcium-independent activation of contractile apparatus in smooth muscle by rat distal colon. Eur. J. Pharmacol. 715, 133–141. doi: 10.1016/j.ejphar. calyculin-A. J. Pharmacol. Exp. Ther. 250, 388–396. 2013.05.031 Ivanovic-Burmazovic, I., and Filipovic, M. R. (2019). Saying NO to H2S: A Story Sanders, K. M. (2008). Regulation of smooth muscle excitation and contraction. of HNO, HSNO, and SSNO. Inorg. Chem. 58, 4039–4051. doi: 10.1021/acs. Neurogastroenterol. Motil. 20(Suppl. 1), 39–53. doi: 10.1111/j.1365-2982.2008. inorgchem.8b02592 01108.x Jimenez, M., Clave, P., Accarino, A., and Gallego, D. (2014). Purinergic Schemann, M. (2005). Control of gastrointestinal motility by the "gut brain"– neuromuscular transmission in the gastrointestinal tract; functional basis for the enteric nervous system. J. Pediatr. Gastroenterol. Nutr. 41(Suppl. 1), future clinical and pharmacological studies. Br. J. Pharmacol. 171, 4360–4375. S4–S6. doi: 10.1111/bph.12802 Shuttleworth, C. W., and Sanders, K. M. (1996). Involvement of nitric oxide in Katrina, M. M., Herbert, T. N., and John, P. T. (2005). Donors of HNO. Curr. Top. neuromuscular transmission in canine proximal colon. Proc. Soc. Exp. Biol. Med. Chem. 5, 649–664. doi: 10.2174/1568026054679290 Med. 211, 16–23. doi: 10.3181/00379727-211-43950c Khan, I., Spencer, G. G., Samson, S. E., Crine, P., Boileau, G., and Grover, Shuttleworth, C. W., Xue, C., Ward, S. M., de Vente, J., and Sanders, K. M. (1993). A. K. (1990). Abundance of sarcoplasmic reticulum calcium pump isoforms in Immunohistochemical localization of 3’,5’-cyclic guanosine monophosphate in stomach and cardiac muscles. Biochem. J. 268, 415–419. doi: 10.1042/bj2680415 the canine proximal colon: responses to nitric oxide and electrical stimulation Koh, S. D., Campbell, J. D., Carl, A., and Sanders, K. M. (1995). Nitric oxide of enteric inhibitory neurons. Neuroscience 56, 513–522. doi: 10.1016/0306- activates multiple potassium channels in canine colonic smooth muscle. J 4522(93)90350-o Physiol 489(Pt 3), 735–743. doi: 10.1113/jphysiol.1995.sp021087 Sivakumaran, V., Stanley, B. A., Tocchetti, C. G., Ballin, J. D., Caceres, V., Zhou, L., Koh, S. D., Dick, G. M., and Sanders, K. M. (1997). Small-conductance Ca2+- et al. (2013). HNO enhances SERCA2a activity and cardiomyocyte function by dependent K+ channels activated by ATP in murine colonic smooth muscle. promoting redox-dependent phospholamban oligomerization. Antioxid. Redox Am. J. Physiol. Cell Physiol. 273, C2010–C2021. doi: 10.1152/ajpcell.1997.273.6. Signal. 19, 1185–1197. doi: 10.1089/ars.2012.5057 C2010 Smith, T. K., and McCarron, S. L. (1998). Nitric oxide modulates cholinergic Kohout, F. C., and Lampe, F. W. (1965). On the role of the nitroxyl molecule in the reflex pathways to the longitudinal and circular muscle in the isolated guinea- reaction of hydrogen atoms with nitric oxide. J. Am. Chem. Soc. 87, 5795–5796. pig distal colon. J. Physiol. 512(Pt 3), 893–906. doi: 10.1111/j.1469-7793.1998. doi: 10.1021/ja00952a052 893bd.x Koppenol, W. H., and Bounds, P. L. (2017). Signaling by sulfur-containing Spencer, G. G., Yu, X., Khan, I., and Grover, A. K. (1991). Expression of isoforms molecules, Quantitative aspects. Arch. Biochem. Biophys. 617, 3–8. doi: 10.1016/ of internal Ca2+ pump in cardiac, smooth muscle and non-muscle tissues. j.abb.2016.09.012 Biochim. Biophys. Acta 1063, 15–20. doi: 10.1016/0005-2736(91)90347-B Kurahashi, M., Mutafova-Yambolieva, V., Koh, S. D., and Sanders, K. M. Stark, M. E., Bauer, A. J., Sarr, M. G., and Szurszewski, J. H. (1993). Nitric (2014). Platelet-derived growth factor receptor-alpha-positive cells and not oxide mediates inhibitory nerve input in human and canine jejunum. smooth muscle cells mediate purinergic hyperpolarization in murine colonic Gastroenterology 104, 398–409. doi: 10.1016/0016-5085(93)90407-4 muscles. Am. J. Physiol. Cell Physiol. 307, C561–C570. doi: 10.1152/ajpcell. Switzer, C. H., Flores-Santana, W., Mancardi, D., Donzelli, S., Basudhar, D., 00080.2014 Ridnour, L. A., et al. (2009). The emergence of nitroxyl (HNO) as a Kurahashi, M., Zheng, H., Dwyer, L., Ward, S. M., Koh, S. D., and Sanders, K. M. pharmacological agent. Biochim. Biophys. Acta 1787, 835–840. doi: 10.1016/j. (2011). A functional role for the ’fibroblast-like cells’ in gastrointestinal smooth bbabio.2009.04.015 muscles. J. Physiol. 589(Pt 3), 697–710. doi: 10.1113/jphysiol.2010.201129 Szabo, C. (2017). Hydrogen sulfide, an enhancer of vascular nitric oxide signaling: Kurtenacker, A., and Löschner, H. (1938). Über die Einwirkung von Stickoxyd auf mechanisms and implications. Am. J. Physiol. Cell Physiol. 312, C3–C15. doi: Thiosulfat und Sulfid. Z. Anorg. Allg. Chem. 238, 335–349. doi: 10.1002/zaac. 10.1152/ajpcell.00282.2016 19382380404 Tam, F. S., and Hillier, K. (1992). The role of nitric oxide in mediating non- Longhi-Balbinot, D. T., Rossaneis, A. C., Pinho-Ribeiro, F. A., Bertozzi, M. M., adrenergic non-cholinergic relaxation in longitudinal muscle of human taenia Cunha, F. Q., Alves-Filho, J. C., et al. (2016). The nitroxyl donor, Angeli’s coli. Life Sci. 51, 1277–1284. doi: 10.1016/0024-3205(92)90017-j salt, reduces chronic constriction injury-induced neuropathic pain. Chem. Biol. Taylor, D. A., and Stull, J. T. (1988). Calcium dependence of myosin Interact. 256, 1–8. doi: 10.1016/j.cbi.2016.06.009 light chain phosphorylation in smooth muscle cells. J. Biol. Chem. 263, Lytton, J., Westlin, M., Burk, S. E., Shull, G. E., and MacLennan, D. H. (1992). 14456–14462. Functional comparisons between isoforms of the sarcoplasmic or endoplasmic Teague, B., Asiedu, S., and Moore, P. K. (2002). The smooth muscle relaxant reticulum family of calcium pumps. J. Biol. Chem. 267, 14483–14489. effect of hydrogen sulphide in vitro: evidence for a physiological role to control Frontiers in Physiology | www.frontiersin.org 17 June 2020 | Volume 11 | Article 438 Gastreich-Seelig et al. HNO and Intestinal Motility intestinal contractility. Br. J. Pharmacol. 137, 139–145. doi: 10.1038/sj.bjp. Yong, Q. C., Cheong, J. L., Hua, F., Deng, L. W., Khoo, Y. M., Lee, H. S., 0704858 et al. (2011). Regulation of heart function by endogenous gaseous mediators- Teicher, C., De Col, R., and Messlinger, K. (2017). Hydrogen sulfide mediating crosstalk between nitric oxide and hydrogen sulfide. Antioxid. Redox Signal. 14, both excitatory and inhibitory effects in a rat model of meningeal nociception 2081–2091. doi: 10.1089/ars.2010.3572 and headache generation. Front. Neurol. 8:336. doi: 10.3389/fneur.2017. Zhao, P., Huang, X., Wang, Z. Y., Qiu, Z. X., Han, Y. F., Lu, H. L., et al. (2009). 00336 Dual effect of exogenous hydrogen sulfide on the spontaneous contraction of Tocchetti, C. G., Wang, W., Froehlich, J. P., Huke, S., Aon, M. A., Wilson, G. M., gastric smooth muscle in guinea-pig. Eur. J. Pharmacol. 616, 223–228. doi: et al. (2007). Nitroxyl improves cellular heart function by directly enhancing 10.1016/j.ejphar.2009.05.014 cardiac sarcoplasmic reticulum Ca2+ cycling. Circ. Res. 100, 96–104. doi: 10. Zhao, Y., Brandish, P. E., Di Valentin, M., Schelvis, J. P., Babcock, G. T., and 1161/01.RES.0000253904.53601.c9 Marletta, M. A. (2000). Inhibition of soluble guanylate cyclase by ODQ. Uehata, M., Ishizaki, T., Satoh, H., Ono, T., Kawahara, T., Morishita, T., Biochemistry 39, 10848–10854. doi: 10.1021/bi9929296 et al. (1997). Calcium sensitization of smooth muscle mediated by a Rho- Zhou, Y., Zhang, X., Yang, S., Li, Y., Qing, Z., Zheng, J., et al. (2017). Ratiometric associated protein kinase in hypertension. Nature 389, 990–994. doi: 10.1038/ visualization of NO/H2S cross-talk in living cells and tissues using a nitroxyl- 40187 responsive two-photon fluorescence probe. Anal. Chem. 89, 4587–4594. doi: Vandecaetsbeek, I., Raeymaekers, L., Wuytack, F., and Vangheluwe, P. (2009). 10.1021/acs.analchem.7b00073 Factors controlling the activity of the SERCA2a pump in the normal and failing heart. Biofactors 35, 484–499. doi: 10.1002/biof.63 Conflict of Interest: The authors declare that the research was conducted in the Wang, R. (2012). Physiological implications of hydrogen sulfide: a whiff absence of any commercial or financial relationships that could be construed as a exploration that blossomed. Physiol. Rev. 92, 791–896. doi: 10.1152/physrev. potential conflict of interest. 00017.2011 Wild, V., Messlinger, K., and Fischer, M. J. M. (2015). Hydrogen sulfide determines Copyright © 2020 Gastreich-Seelig, Jimenez and Pouokam. This is an open-access HNO-induced stimulation of trigeminal afferents. Neurosci. Lett. 602, 104–109. article distributed under the terms of the Creative Commons Attribution License doi: 10.1016/j.neulet.2015.06.056 (CC BY). The use, distribution or reproduction in other forums is permitted, provided Wood, J. D. (2008). Enteric nervous system: reflexes, pattern generators the original author(s) and the copyright owner(s) are credited and that the original and motility. Curr. Opin. Gastroenterol. 24, 149–158. doi: 10.1097/MOG. publication in this journal is cited, in accordance with accepted academic practice. No 0b013e3282f56125 use, distribution or reproduction is permitted which does not comply with these terms. Frontiers in Physiology | www.frontiersin.org 18 June 2020 | Volume 11 | Article 438