AOA hemihydrochloride | Inhibitor Screening

TLR2 and TLR4 interact with sulfide system in the modulation of mouse colonic motility

Laura Grasa1,2,3 Leticia Abecia4 Ainize Peña‐Cearra4,5 Sofia Robles1 Elena Layunta6 Eva Latorre7 José Emilio Mesonero1,2,3 Raquel Forcén1

Abstract

Background: H2S is a neuromodulator that may inhibit intestinal motility. H2S production in colon is yielded by cystathionine β‐synthase (CBS) and cystathionine γlyase (CSE) enzymes and sulfate‐reducing bacteria (SRB). Toll‐like receptors (TLRs) recognize intestinal microbiota. The aim of this work was to evaluate the influence of TLR2 and TLR4 on the endogenous and SRB‐mediated synthesis of H2S and its consequences on the colonic motility of mouse.
Methods: Muscle contractility studies were performed in colon from WT, Tlr2‐/‐, and Tlr4‐/‐ mice. The mRNA levels of TLR2, TLR4, CBS, CSE, and SRB were measured by real‐time PCR. Free sulfide levels in colon and feces were determined by colorimetric assays.
Results: NaHS and GYY4137, donors of H2S, reduced the contractility of colon. Aminooxyacetic acid (AOAA), inhibitor of CBS, and D‐L propargylglycine (PAG), inhibitor of CSE, increased the contractility of colon. In vivo treatment with NaHS or GYY4137 inhibited the spontaneous contractions and upregulated TLR2 expression. The in vivo activation of TLR4 with lipopolysaccharide increased the contractile response to PAG, mRNA levels of CSE, and the free sulfide levels of H2S in colon. In Tlr2‐/‐ and Tlr4‐/‐ mice, the contractions induced by AOAA and PAG and mRNA levels of CBS and CSE were lower with respect to WT mice. Deficiency of TLR2 or TLR4 provokes alterations in free sulfide levels and SRB of colon.
Conclusions and Inferences: Our study demonstrates interaction between TLR2 and TLR4 and the sulfide system in the regulation of colonic motility and contributes to the pathophysiology knowledge of intestinal motility disorders.

K E Y W O R D S
H2S, intestinal motility, microbiota, Toll‐like receptors

1 | INTRODUCTION

Colonic microbiota acts as a barrier against pathogens, drives the bacteria‐sensing receptors are TLR2 and TLR4, which recognize maturation of the immune system, and contributes to the main‐ the bacterial fragments lipoteichoic acid (LTA) and lipopolysacchatenance of gut homeostasis.1 Bacteria recognition is carried out ride (LPS), respectively. Expression of TLR2 and TLR4 has been by Toll‐like receptors (TLRs), a family of transmembrane proteins found in enteric neurons, glial cells, and smooth muscle cells of mouse colon, suggesting that TLR2 and TLR4 may play a chief role in the regulation of intestinal motility.4‐6 In fact, our group and other authors have described that TLR4 but not TLR2 is involved in the regulation of the normal spontaneous colonic motility in mice.7,8
Hydrogen sulfide (H2S) has emerged as an important gaseous signaling molecule playing numerous roles in health and disease. Previous studies have suggested that H2S is implicated in the regulation of several cellular processes in the gastrointestinal (GI) tract including inflammation, epithelial secretion, nociception, and motility.9 In fact, H2S has been proposed as a third gaseous neuromodulator, after nitric oxide and carbon monoxide (CO), that may inhibit GI motility.10
The endogenous production of H2S in the smooth muscle layer of rat and mouse colon is mainly through two enzymes: the cystathionine γ‐lyase (CSE) and the cystathionine β‐synthase (CBS).11,12 In fact, both CBS and CSE enzymes have been identified in enteric neurons, smooth muscle layers, and epithelial cells of colon.11‐13 Additionally, large amounts of H2S can be produced by the sulfatereducing bacteria (SRB) living in the colon.9 SRB are anaerobic microorganisms that conduct dissimilatory sulfate reduction to obtain energy, resulting in the release of a great quantity of sulfide in the intestine.
Understanding of H2S interactions with other signaling mediators is still a topic of active research. In fact, the cross talk between TLR and the H2S system in the intestine has not yet been elucidated in detail. As TLR2 and TLR4 have been found in the enteric nerves and smooth muscle, the same type of cells in which CBS and CSE are expressed, we hypothesize that TLR may modulate the synthesis of H2S and the motor responses induced by H2S in the intestine. Thus, the aim of the present study was to evaluate the influence of TLR2 and TLR4 on the endogenous and SRB‐mediated synthesis of H2S and its consequences on the colonic motility of mouse.

2 | MATERIALS AND METHODS

2.1 | Animals

All procedures were carried out under Project Licenses PI03/16 and PI13/17, approved by the in‐house Ethics Committee for Animal Experiments from the University of Zaragoza. Inbred C57BL/10 and mouse strains deficient for TLR2 (Tlr2‐/‐) and TLR4 (Tlr4‐/‐) were kindly provided by Ignacio Aguiló from the University of Zaragoza and bred at the Centro de Investigación Biomédica de Aragón (CIBA), Zaragoza, Spain. Their genotypes were periodically analyzed as described.14 Male Tlr2‐/‐ and Tlr4‐/‐ knockout and age‐matched (8‐12 weeks old) wild‐type (WT) mice were used in the experiments. All mice were housed under pathogen‐free conditions on a 12‐hour light/dark cycle with food and water ad libitum.The WT, Tlr2‐/‐, and Tlr4‐/‐ mice were treated with a single IP injection of NaHS or GYY4137 5 mg/kg for 3 or 24 hours, respectively. These treatments were previously shown to have effects on the digestive system.15‐18Other WT mice were treated with a single IP injection of lipoteichoic acid (LTA, 20 mg/kg for 3 hours) to activate TLR2 or ultrapure lipopolysaccharide (LPS, 5 mg/kg for 24 hours) to activate TLR4. These treatments have been used by other authors previously.19,20

Key Points
• H2S is synthesized by two enzymes of the colon and sulfate‐reducing bacteria (SRB) of the intestinal lumen. Toll‐like receptors (TLRs) recognize intestinal bacteria. We evaluate the influence of TLRs on the sulfide system in the modulation of mouse colonic motility.
• Our study demonstrates the interaction between TLR2 and TLR4 and the sulfide system in the regulation of colonic motility.
• These findings contribute to the pathophysiology knowledge of intestinal motility disorders such as inflammatory bowel disease.

2.2 | Muscle contractility studies

Strips of proximal‐mid colon with intact mucosa were suspended in the direction of circular smooth muscle fibers in an organ bath and muscle contractility studies were performed as previously described.8To study in vitro the effect of the H2S donors in WT, Tlr2‐/‐, and Tlr4‐/‐ mice, NaHS (a rapid donor, 10‐1000 µmol/L) or GYY4137 (a slow donor, 0‐1000 µmol/L) was added every 15 minutes and the cumulative concentration‐response curves were performed.To study the inhibitory effect in vitro of the H2S synthesis enzymes in WT, Tlr2‐/‐, and Tlr4‐/‐ mice, aminooxyacetic acid (AOAA, 0.01‐10 mmol/L), an inhibitor of CBS, or D‐L propargylglycine (PAG, 0.01‐10 mmol/L), an inhibitor of CSE, was added every 15 minutes and the cumulative concentration‐response curves were performed.In mice treated in vivo with NaHS or GYY4137, only the spontaneous activity of the colon was recorded.
In mice treated with LTA or LPS, we added a single concentration of NaHS (1 mmol/L) or GYY4137 (1 mmol/L) or AOAA (10 mmol/L) or PAG (10 mmol/L) in each colonic strip.To estimate the responses to the drugs, the area under the curve (AUC) of spontaneous contractions from the minimum was measured before and after drug addition, and expressed as g/min. The inhibition or increase of the spontaneous contractions was quantified as the inhibition or increase in the AUC after the addition of the drugs over the baseline AUC, before the addition of the drugs.

2.3 | mRNA expression by real‐time PCR

The relative abundance of TLR2, TLR4, CBS, and CSE mRNA in the colon from mice was measured by real‐time PCR. Tissue samples for mRNA analysis were collected in RNAlater (Qiagen), and RNA extractions were carried out with the RNeasy mini kit (Qiagen) and the cDNA was synthesized using the NX M‐MuLV Reverse Transcriptase kit (Lucigen, Middleton, WI, USA) according to the supplier’s protocol. The cDNAs obtained were used to measure the mRNA levels by SYBR Green and using specific primers (Table 1). Reactions were run using the StepOnePlus Real‐Time PCR System (Life Technologies). The reaction mixture (10 µL) was comprised of 4.5 µL FastStart Universal SYBR Green Master (Roche, Mannheim, Germany), 0.5 µL of each primer 30 µmol/L, 2.5 µL of sterile distilled water, and 2 µL of cDNA template (200 ng). Each sample was run in triplicate, and the mean Ct was determined from the three runs. Thus, the relative mRNA expression was calculated as ΔΔCt = ΔCttreatment or type of mice – ΔCtcontrol or WT mice being ΔCt = Cttarget gene – Ctreference gene. GAPDH and actin were used as reference genes. Finally, the relative gene expression levels were converted and expressed as fold difference (=2−ΔΔCt).

2.4 | Determination of free sulfide levels in mouse colon

Sulfide levels in mouse colon were measured by using the zinc sulfide precipitation method with some modifications.21 Two samples of colonic tissue (120 mg) were harvested from each animal. The fresh samples were weighted out and immediately immersed in 1 mL of cold phosphate 50 mmol/L pH 6 or bicarbonate‐carbonate buffer 50 mmol/L pH 9.4. Samples were then homogenized on ice followed by centrifugation at 15 000 g for 10 minutes at 4°C. 500 µL of the clear supernatant from each sample was added to 400 µL of a pre‐mixed 1% w/v zinc acetate (350 µL) solution and 50 µL of NaOH 1.5 mol/L and incubated for 1 hour at RT. This was followed by centrifugation at 1200 g for 5 minutes to pellet the zinc sulfide formed. The supernatant was removed and the pellet washed with 1.5 mL of Milli‐Q water by vortexing thoroughly, followed by a centrifugation at 1200 g for 5 minutes. The supernatant was then removed and the pellet reconstituted with 480 µL of Milli‐Q water and mixed with 120 µL of pre‐mixed dye (60 µL of N,N‐dimethyl‐p‐phenylenediamine, NNDP, 20 mmol/L in HCl 7.2 mol/L, and 60 µL of FeCl3 30 mmol/L in HCl 1.2 mol/L). The solution was then incubated for 10 minutes to allow the color to develop and the absorbance was read on a spectrophotometer at 670 nm.
To determine free sulfide levels, one sample was prepared in a sodium phosphate buffer 50 mmol/L pH 6 and another sample in a bicarbonate‐carbonate buffer 50 mmol/L pH 9.4. At pH 6, >90% of free sulfides should exist as the neutral H2S species which are highly volatile and easily lost by dissipation into the atmosphere. At pH 9, >90% of free sulfides should exist as the anionic HS‐, which is retained in solution. Free tissue sulfide was then calculated by deducting the value of sulfides measured at pH 6 from pH 9.A standard curve of serially diluted NaHS was prepared in bicarbonate‐carbonate buffer 50 mmol/L pH 9.4. The concentrations of free sulfide were expressed as nmole per mg of tissue weight.

2.5 | Determination of free H2S levels in mouse feces

In order to evaluate the role of H2S produced in the colonic lumen by bacteria, free H2S levels were determined in mouse feces by a previously described method with some modifications.22,23
A sample of fresh mouse feces was collected into a pre‐weighed tube containing 500 µl of cold NaOH 1 mol/L. The tube with the feces was weighed again and the weight of the feces was calculated (30‐60 mg). Free sulfide was measured using the supernatant fraction following centrifugation (14 500 g, 5 minutes, 4°C) of NaOH slurries. The sulfide levels were determined using the methylene blue reaction by mixing 300 µL of fecal slurry with 150 µL of NNDP 0.17 mol/L in HCl 6 mol/L and adding immediately 150 µL of FeCl3 0.37 mol/L in HCl 6 mol/L. This was quickly sealed and allowed 20 minutes for full‐color development. Following centrifugation (12 000 g, 5 minutes), 100 µL of supernatant was diluted in 0.9 mL maleic acid buffer 0.86 mol/L pH 0.6, mixed and the absorbance was measured at 670 nm. A standard curve of NaHS was freshly prepared for each experiment and fecal sulfide was therefore calculated by extrapolation. The concentrations of sulfide were expressed as nmole per mg of feces.

2.6 | Microbial analysis

Freeze‐dried feces (20 mg) were homogenized in tubes containing zirconium and glass beads in a Precellys 24 tissue homogenizer (2 cycles of 5000 g for 30 seconds). Total DNA was isolated using FavorPrep Stool DNA Isolation Mini Kit (Favorgen Biotech Corp). Extracted DNA was eluted in 50 µL and quantified by NanoDrop‐100 Spectrophotometer (NanoDrop Technologies). Total bacteria gene copy number and four groups of bacteria were targeted by realtime PCR in QuantStudio 6 Flex Real‐Time PCR system (Thermo Fisher Scientific) with PerfeCTa SYBR Green SuperMix Low ROX (Quantabio). The analysis was performed with QuantStudio RT PCR software v1.3. The primers used in this study are in Table 2. Relative quantification to 16S rDNA gene was calculated using ΔΔCt method. The total bacteria 16S rRNA gene copy numbers were determined by standard curves made using serial dilutions of plasmid (containing 16S rRNA gene fragment) of known concentrations on a tenfold basis.

2.7 | Data analysis and statistics

Differences in the responses to the different concentrations of the drugs in mice were compared by one or two‐way analysis of variance (one or two‐way ANOVA) followed by Bonferroni’s post hoc test. IC50 values were calculated using a conventional concentration‐response curve with variable slope. Differences in bacteria levels, gene expression or sulfide levels were compared by Student´s unpaired t test.The results were expressed as the mean ± SEM with n denoting the number of animals used. The data were analyzed using the software GraphPad Prims version 5.00 (GraphPad Software), and the differences between P‐values <0.05 were considered to be statistically significant.

2.8 | Drugs and solutions

Lipoteichoic acid from Bacillus subtilis‐TLR2 ligand and ultrapure lipopolysaccharide from E coli O111:B4 strain‐TLR4 ligand were purchased from InvivoGen. NaHS, GYY4137, AOAA, and PAG were acquired from Sigma. AOAA, PAG and NaHS solutions were prepared in Milli‐Q water for in vitro studies. GYY4137 was dissolved in DMSO to obtain a stock solution of 3 mg/mL and the following solutions were made with Milli‐Q water or saline for in vitro or in vivo studies, respectively. NaHS, LPS, and LTA solutions were prepared in saline for in vivo studies.

3 | RESULTS

3.1 | In vitro effect of H2S on colonic spontaneous motility

To study the effect of H2S in colon motility, concentration‐response curves to rapid (NaHS, 10‐1000 µmol/L) and slow (GYY4137, 10‐1000 µmol/L) donors of H2S were performed in whole strips of mouse colon suspended in the circular direction in an organ bath. Both NaHS and GYY4137 reduced the contractility of the circular smooth muscle of the mouse colon in a concentration‐dependent manner (P < 0.001; Figure 1A‐D). The IC50 values were 465 µM for NaHS (95% confidence interval log IC50 = −3.33 ± 0.44; n = 7) and 192 µmol/L for GYY4137 (95% confidence interval log IC50 = −3.71 ± 0.32; n = 7).
To analyze the participation of H2S synthesis enzymes on the spontaneous contractions, concentration‐response curves to AOAA (0.01‐10 mmol/L), an inhibitor of CBS, and PAG (0.01‐10 mmol/L), an inhibitor of CSE, were performed on colonic strips in the same way in the organ bath. Both AOAA and PAG increased the contractility of the circular smooth muscle of the mouse colon in a concentration‐dependent manner (P < 0.01 and P < 0.001, respectively; Figure 1E‐H). The IC50 values were 40 µM for AOAA (95% confidence interval log IC50 = −4.38 ± 1; n = 4) and 700 µM for PAG (95% confidence interval log IC50 = −3.15 ± 0.48; n = 10).

3.2 | In vivo effect of H2S donors

To corroborate the in vitro inhibitory effect of the H2S donors on colonic motility, we examined the spontaneous motility of muscle strips isolated from mice treated in vivo with a single IP injection of NaHS or GYY4137 5 mg/kg for 3 or 24 hours, respectively. Both the rapid release of H2S through the donor NaHS and the slow release of H2S through the donor GYY4137 induced an inhibition of the spontaneous contractions of colon (Figure 2A, 2).
In order to establish a possible interaction between the H2S and the receptors TLR2 and TLR4 in the modulation of the intestinal motor function, we first studied mRNA expression of TLR2 and TLR4 in the colon of WT mice treated with NaHS or GYY4137. TLR2 expression was upregulated after the treatment with NaHS or GYY4137, being the expression of this receptor much higher (17fold) in the intestine of mice treated with GYY4137, the slow donor of H2S (Figure 2C). However, TLR4 mRNA level was not significantly modified by any of the donors of H2S (Figure 2D).

3.3 | TLR2 and TLR4 activation modulates H2S synthesis and H2S‐induced motor responses

The above results suggest that H2S could be a regulator of TLR expression. In the next step, we studied the effect of the TLR activation with specific ligands on the motor responses induced by H2S donors in colon. The in vitro addition of both NaHS 1 mmol/L and GYY4137 1 mmol/L reduced the contractility of the circular smooth muscle of mice treated with LTA (20 mg/kg for 3 hours), a specific ligand of TLR2, similar to those observed in mice treated with saline for 3 hours (Figure 3A, 3). In the same way, NaHS and GYY4137 reduced the contractility in the mice treated with LPS (5 mg/kg for 24 hours), a specific ligand of TLR4, similar to those observed in mice treated with saline for 24 hours (Figure 3A, 3).
The effects of the inhibition of CBS and CSE enzymes on the spontaneous motility of mice treated with LTA or LPS was also studied. AOAA (10 mmol/L) caused a contractile response in the circular smooth muscle of mice treated with LTA or LPS, similar to those observed in mice treated with saline (Figure 3C). PAG 10 mmol/L caused a contractile response in mice injected with LTA, similar to those observed in mice treated with saline (Figure 3D). However, the contractile response to PAG was significantly increased in mice treated with LPS (Figure 3D), suggesting that TLR4 could alter H2S synthesis mediated by CSE enzyme.
Thus, to establish a possible relation between the activation of TLR2 and TLR4 and the H2S synthesis in the modulation of the motor responses mediated by H2S, we studied mRNA expression of the H2S synthesis enzymes CBS and CSE and the levels of free sulfide in colon from mice treated with LTA or LPS.
The activation of TLR2 or TLR4 with LTA or LPS, respectively, did not modify the mRNA expression levels of CBS in colon (Figure 3E). Although the mRNA expression levels of CSE was not modified in the colon from mice treated with LTA, the mRNA levels of this enzyme were highly increased after the activation of TLR4 with LPS (Figure 3F).In addition, regarding sulfide levels, we observed that the activation of TLR2 with LTA did not modify the free sulfide levels in colon. However, the activation of TLR4 with LPS increased the free sulfide levels in colon (Figure 3G), suggesting that TLR4 activation increases the H2S synthesis mediated by CSE.

3.4 | H2S synthesis and H2S‐induced motor responses in Tlr2‐/‐ and Tlr4‐/‐ mice

To corroborate the previous results in which both TLR2 and TLR4 seem to show interaction with the H2S synthesis and the motor responses induced by H2S, experiments were performed on TLR2 or TLR4 deficient mice. NaHS (10‐1000 µmol/L) reduced the contractility in a concentration‐dependent manner in Tlr2‐/‐ (P < 0.001) and Tlr4‐/‐ (P < 0.05) mice (Figure 4A). The IC50 values of NaHS were 137 µmol/L for Tlr2‐/‐ mice (95% confidence interval log IC50 = −3.86 ± 0.24; n = 9) and 509 µmol/L for Tlr4‐/‐ mice (95% confidence interval log IC50 = −3.29 ± 0.66; n = 6).GYY4137 (10‐1000 µmol/L) reduced the contractility in a concentration‐dependent manner in Tlr2‐/‐ (P < 0.001) and Tlr4‐/‐ (P < 0.001) mice (Figure 4B). The IC50 values of GYY4137 were 110 µmol/L for Tlr2‐/‐ mice (95% confidence interval log IC50 = −3.95 ± 0.27; n = 6) and 175 µM for Tlr4‐/‐ mice (95% confidence interval log IC50 = −3.75 ± 0.40; n = 11).
No differences were found between the decreases in the contractility induced by NaHS or GYY4137 in Tlr2‐/‐ or Tlr4‐/‐ with respect to WT mice.Regarding enzyme inhibition, AOAA (0.01‐10 mmol/L) increased the contractility in a concentration‐dependent manner in Tlr2‐/‐ (P < 0.05) and Tlr4‐/‐ (P < 0.01) mice (Figure 4C). The IC50 values of AOAA were 20 µmol/L for Tlr2‐/‐‐ mice (95% confidence interval log IC50 = −4.57 ± 1.4; n = 9) and 580 µmol/L for Tlr4‐/‐ mice (95% confidence interval log IC50 = −2.23 ± 0.4; n = 7). PAG (0.01‐10 mmol/L) increased the contractility in a concentration‐dependent manner in Tlr2‐/‐ (P < 0.001) and Tlr4‐/‐ (P < 0.01) mice (Figure 4D). The IC50 values of PAG were 170 µmol/L for Tlr2‐/‐ mice (95% confidence interval log IC50 = −3.75 ± 0.55; n = 5) and 140 µM for Tlr4‐/‐ mice (95% confidence interval log IC50 = −3.83 ± 0.55; n = 5).
The contractions induced by AOAA or PAG were significantly lower in Tlr2‐/‐ and Tlr4‐/‐ with respect to WT mice (P < 0.001 and P < 0.05, respectively), which could be due to the presence of lower H2S levels in these TLR deficient mice.
For this reason, and in order to investigate the effects of the deficiency of TLR2 and TLR4 on the H2S synthesis, we studied the gene expression of CBS and CSE and the levels of free sulfide in colon from Tlr2‐/‐ and Tlr4‐/‐ mice.The expressions of CBS and CSE were significantly reduced in both Tlr2‐/‐ and Tlr4‐/‐ mice (Figure 4E, F). Surprisingly, the levels of free sulfide were incremented in Tlr4‐/‐ but unchanged in Tlr2‐/‐mice (Figure 4G).

3.5 | Sulfate‐reducing bacteria and H2S levels in feces from Tlr2‐/‐ and Tlr4‐/‐ mice

The high levels of H2S in Tlr4‐/‐ mice could come from external sources to the intestinal tissue itself. Therefore, we decided to analyze by quantitative PCR some of the sulfate‐reducing bacteria (SRB) contained in the feces of Tlr2‐/‐ and Tlr4‐/‐ mice, as these bacteria may synthesize large amounts of H2S and influence in the total H2S levels present in the GI tract.
The determination of total bacteria DNA in the feces showed that the number of 16S rRNA gene copies was similar in Tlr2‐/‐ and Tlr4‐/‐ with respect to WT mice (Figure 5A).
Dissimilatory sulfite reductase (dsrA) gene copy number was quantified to determine the abundance of SRB in feces.24 As it is shown in Figure 5B, mice with a deficiency in TLR2 showed a high reduction, while mice with a deficiency in TLR4 showed an increase in the dsrA gene copy number.When we studied the expression of specific SRB in Tlr2‐/‐ mice, we found a decrease in Bilophila wadsworthia, a depletion of Desulfovibrio desulfuricans, but a high increase in Desulfovibrio piger bacteria (Figure 5C‐E). On the other hand, mice with a deficiency in TLR4 showed an increase in D desulfuricans and D piger, and no modification of B wadsworthia bacteria (Figure 5C‐E).Finally, we determined the free H2S levels in the feces of Tlr2‐/‐ and Tlr4‐/‐ mice. The H2S levels were reduced in Tlr2‐/‐ and unchanged in Tlr4‐/‐ with respect to WT mice (Figure 5F).

4 | DISCUSSION

H2S has been considered as a gaseous signaling molecule regulating several functions in the GI tract, including the regulation of the intestinal motility.25 Here, we have shown that H2S and innate immune system are interconnected, showing that H2S increases TLR2 expression, TLR4 increases H2S synthesis mediated by CSE enzyme, and that TLR2 and TLR4 modify the populations of SRB in the colon.
Previous studies have described that NaHS relaxes the smooth muscle of guinea‐pig and rabbit ileum26,27 and human, rat and mouse colon.11,25 Our experiments in vitro and in vivo clearly demonstrate that NaHS, a rapid donor of H2S, and GYY4137, a slow donor of H2S, at physiologically relevant concentrations, inhibit the spontaneous motility of the circular smooth muscle in mouse colon, corroborating the inhibitory effect of H2S on the intestinal motility.
The endogenous production of H2S in the smooth muscle layer of rat and mouse colon is mainly through CBS and CSE enzymes.11,12 Both CBS and CSE enzymes have been identified in enteric neurons and smooth muscle layers of colon,11‐13 which are the main tissues involved in the intestinal motility. Our results demonstrate that the inhibition of CBS with AOAA or CSE with PAG increased the colonic motility, indicating that H2S might be a molecule that could regulate intestinal motility. Similarly, other authors have reported that AOAA and PAG are able to cause a smooth muscle depolarization and increase the spontaneous motility in rat and human colon.11,28
As far as we know, the cross talk between the H2S system and the bacteria‐sensing receptors (TLR) in the gastrointestinal tract has not yet been explored. In this work, we first demonstrate that the treatment with the H2S donors NaHS and GYY4137 increase the expression of TLR2 but not TLR4 in mouse colon. Interestingly, the release of H2S maintained in time through the administration of the slow donor GYY4137 induced a higher expression of TLR2 than in the case of the rapid release of H2S by NaHS. However, TLR2 activation with LTA did not modify the gene expression levels of CBS or CSE nor the free sulfide levels present in colon. These facts may explain that the responses induced by the CBS inhibitor AOAA and the CSE inhibitor PAG in the mice treated with LTA are similar to mice treated with saline.
Previous studies have shown interactions between H2S and TLR4 but not TLR2. In this sense, NaHS has an additive effect on osteoclast differentiation through activation of the TLR4 but not the TLR2 pathway, in rat periodontal tissue.29 H2S treatment decreased the expression of TLR4 and reduced inflammation in a rat model of renal ischemia‐reperfusion injury.30 In fact, H2S donors are able to reduce the infiltration of neutrophils and lymphocytes in several models and reduce the expression of many pro‐inflammatory cytokines, most likely related to its ability to suppress the activation of NF‐κB, a factor regulated by TLR2/TLR4 signaling.31 In addition, H2S supplementation inhibited TLR4 expression in a mouse model of mesangial cell overproliferation induced by high glucose.32 These evidences suggest that H2S is able to regulate TLR expression, although this regulation seems to be specific for the different tissues, species or models used.
Interestingly, and in relation to previous studies showing a relationship between H2S and TLR4, activation with LPS induces an increase in CSE expression and the free sulfide levels in colon, resulting in an increase of the motor response induced by PAG. Other studies have shown that the activation of TLR4 by LPS increases the biosynthesis of CSE and H2S in mouse macrophages through p38/ MAPKs and NF‐κB signaling.33 On the contrary, LPS decreased CSE expression in human endothelial cells and blocked H2S production in mouse aorta tissues.34 All these findings indicate that the activation of TLR4 by LPS may modulate the synthesis of H2S through the enzyme CSE. However, while a list of evidence of cellular regulation of H2S enzymes has been studied in other systems, the cellular source of the gene regulation could not be identified in the current study. In addition, TLR receptor activation of enzyme regulation may in fact take place in different cells types, like neurons and smooth muscle cells, but also in immune and epithelial cells of the mucosa and submucosa. In these sense, it is well‐known that mucosa‐submucosamuscle interactions might affect the muscle contractility.
On the other hand, the responses induced by the H2S donors NaHS and GYY4137 were not modified in mice with TLR2 or TLR4 activated or in mice with a deficiency in TLR2 or TLR4, indicating that TLR2 or TLR4 seem not to be involved in the direct relaxant effect of H2S on the intestinal smooth muscle. In fact, other authors have demonstrated that NaHS induces a smooth muscle hyperpolarization mediated by K channels, particularly apamin‐sensitive SK channels and glybenclamide‐sensitive K (ATP) channels in human and rat colon.25
In our study, the deficiency of TLR2 or TLR4 downregulates the gene expression levels of CBS and CSE present in colon. These facts may explain that the responses induced by the CBS inhibitor AOAA and the CSE inhibitor PAG in the Tlr2‐/‐ and Tlr4‐/‐ are also reduced in comparison with WT mice. Thus, our results indicate that the presence of TLR2 and TLR4 is necessary for maintaining the endogenous levels of H2S synthesized by CBS and CSE that modulate the intestinal motility. However, although TLR receptors may be required to maintain endogenous H2S enzymes, we cannot determine exactly whether this regulation occurs at the cellular level or through multicellular signaling pathways, where different cell types can interact to obtain the final biological effect desired.
Surprisingly, the free sulfide levels found in the colon from Tlr2‐/‐ or Tlr4‐/‐ mice are not reduced, as it could be expected because of the downregulation of the CBS and CSE enzymes found in these mice. Several putative origins of H2S in the large intestine have been described: enteric neurons and smooth muscle layers expressing CBS and CSE enzymes,11‐13 blood or vascular tissues35 and finally, the luminal bacteria.36 H2S produced by luminal bacteria has the potential to modify GI function and participates in motility disorders when intestinal microbiota is altered.37
In the colon, H2S can be free or bound to luminal contents. Fecal components have a large capacity for binding and catabolizing H2S, thereby rendering it inactive. Therefore, total concentrations are not representative of the active luminal sulfide; free levels of H2S have been reported between 0.2‐1.1 nanomol/mg in mouse feces.38 Then, we measured the free H2S levels and the sulfate‐reducing bacteria present in the feces of Tlr2‐/‐ or Tlr4‐/‐ mice, detecting levels of free H2S about 0.6 nanomol/mg in feces of WT mice and these levels were reduced in Tlr2‐/‐ but not modified in Tlr4‐/‐ mice.
Sulfate‐reducing bacteria (SRB) can perform anaerobic respiration utilizing sulfate (SO42−) as terminal electron acceptor, reducing it to H2S. In our study, the amount of total bacteria was not modified in Tlr2‐/‐ or Tlr4‐/‐ compared to WT mice. In fact, it has been described that the impact of TLR deficiency on the composition of the intestinal microbiota is minimal under homeostatic conditions.39 However, SRB abundance was strongly reduced in Tlr2‐/‐ and increased in Tlr4‐/‐ mice, indicating that TLR may be able to modulate the populations of SRB.
SRB include a large amount of bacteria of different genera.40 In our work, we have focused in studying the levels of the sulfate‐reducing bacteria Desulfovibrio desulfuricans and Desulfovibrio piger, as they are considered the most abundant SRB in human feces, and the sulfite‐reducing bacterium Bilophila wadsworthia, which has also been isolated in these samples.41‐43 In our study, we demonstrate that effectively TLR2 and TLR4 modulate the levels of D piger, B wadsworthia, and D desulfuricans. In this way, alterations in the amount of bacteria of genus Desulfovibrio present in the cecal contents of TLR2 deficient mice, but not in TLR4 deficient mice have been reported previously.39
Several authors have described alterations in fecal sulfide concentrations and activity of SRB in chronic intestinal motility disorders as inflammatory bowel disease (IBD).38 Within this context, recent studies have described in IBD different alterations of TLRs showing the critical importance of these innate immunity components in sensing bacteria to maintain intestinal homeostasis.44 Our study demonstrates the interaction between the TLR and the sulfide system in the regulation of the colonic motility, an intestinal function altered in IBD patients. Thus, our study contributes to the understanding of the pathophysiology of the motility disorders and suggests the modulation of the sulfide system as a possible therapeutic target.
In conclusion, our study demonstrates the interaction between TLR2 and TLR4 and the sulfide system in the regulation of colonic motility and contributes to the pathophysiology knowledge of motility disorders such as IBD.

REFERENCES

1. Tojo R, Suárez A, Clemente MG, et al. Intestinal microbiota in health and disease: role of bifidobacteria in gut homeostasis. World J Gastroenterol. 2014;20(41):15163‐15176.
2. Akira S, Takeda K. Toll‐like receptor signalling. Nat Rev Immunol. 2004;4(7):499‐511.
3. Abreu MT. Toll‐like receptor signalling in the intestinal epithelium: how bacterial recognition shapes intestinal function. Nat Rev Immunol. 2010;10(2):131‐144.
4. Barajon I, Serrao G, Arnaboldi F, et al. Toll‐like receptors 3, 4, and 7 are expressed in the enteric nervous system and dorsal root ganglia. J Histochem Cytochem. 2009;57(11):1013‐1023.
5. Brun P, Giron MC, Qesari M, et al. Toll‐like receptor 2 regulates intestinal inflammation by controlling integrity of the enteric nervous system. Gastroenterology. 2013;145(6):1323‐1333.
6. Wang Y, Devkota S, Musch MW, et al. Regional mucosa‐associated microbiota determine physiological expression of TLR2 and TLR4 in murine colon. PLoS ONE. 2010;5(10):e13607.
7. Anitha M, Vijay‐Kumar M, Sitaraman SV, Gewirtz AT, Srinivasan S. Gut microbial products regulate murine gastrointestinal motility via Toll‐like receptor 4 signaling. Gastroenterology. 2012;143(4):1006‐160000.
8. Forcen R, Latorre E, Pardo J, Alcalde AI, Murillo MD, Grasa L. Toll‐like receptors 2 and 4 exert opposite effects on the contractile response induced by serotonin in mouse colon: role of serotonin receptors. Exp Physiol. 2016;101(8):1064‐1074.
9. Pimentel M, Mathur R, Chang C. Gas and the microbiome. Curr Gastroenterol Rep. 2013;15(12):356.
10. Farrugia G, Szurszewski JH. Carbon monoxide, hydrogen sulfide, and nitric oxide as signaling molecules in the gastrointestinal tract. Gastroenterology. 2014;147(2):303‐313.
11. Gil V, Gallego D, Jimenez M. Effects of inhibitors of hydrogen sulphide synthesis on rat colonic motility. Brit J Pharmacol. 2011;164(2B):485‐498.
12. Linden DR, Sha L, Mazzone A, et al. Production of the gaseous signal molecule hydrogen sulfide in mouse tissues. J Neurochem. 2008;106(4):1577‐1585.
13. Schicho R, Krueger D, Zeller F, et al. Hydrogen sulfide is a novel prosecretory neuromodulator in the guinea‐pig and human colon. Gastroenterology. 2006;131(5):1542‐1552.
14. Pardo J, Wallich R, Martin P, et al. Granzyme B‐induced cell death exerted by ex vivo CTL: discriminating requirements for cell death and some of its signs. Cell Death Differ. 2008;15(3):567‐579.
15. Fiorucci S, Antonelli E, Distrutti E, et al. Inhibition of hydrogen sulfide generation contributes to gastric injury caused by anti‐inflammatory nonsteroidal drugs. Gastroenterology. 2005;129(4):1210‐1224.
16. Li L, Bhatia M, Zhu YZ, et al. Hydrogen sulfide is a novel mediator of lipopolysaccharide‐induced inflammation in the mouse. FASEB J. 2005;19(9):1196–1198.
17. Zhao HY, Yan R, Zhou XG, Ji F, Zhang B. Hydrogen sulfide improves colonic barrier integrity in DSS‐induced inflammation in Caco‐2 cells and mice. Int Immunopharmacol. 2016;39:121‐127.
18. Chen S, Bu D, Ma Y, et al. GYY4137 ameliorates intestinal barrier injury in a mouse model of endotoxemia. Biochem Pharmacol. 2016;118:59‐67.
19. Buchholz BM, Chanthaphavong RS, Bauer A. Nonhemopoietic cell TLR4 signaling is critical in causing early lipopolysaccharide‐induced ileus. J Immunol. 2009;183(10):6744‐6753.
20. Mayerhofer R, Fröhlich EE, Reichmann F, et al. Diverse action of lipoteichoic acid and lipopolysaccharide on neuroinflammation, blood‐brain barrier disruption, and anxiety in mice. Brain Behav Immun. 2017;60:174‐187.
21. Ang AD, Konigstorfer A, Giles GI, Bhatia M. Measuring free tissue sulfide. Adv Biol Chem. 2012;2:360‐365.
22. Florin TH. Hydrogen sulphide and total acid‐volatile sulphide in faeces, determined with a direct spectrophotometric method. Clin Chim Acta. 1991;196(2–3):127‐134.
23. Moore J, Babidge W, Millard S, Roediger W. Colonic luminal hydrogen sulfide is not elevated in ulcerative colitis. Dig Dis Sci. 1998;43(1):162‐165.
24. Beaumont M, Andriamihaja M, Lan A, et al. Detrimental effects for colonocytes of an increased exposure to luminal hydrogen sulfide: The adaptive response. Free Radical Bio Med. 2016;93:155‐164.
25. Gallego D, Clave P, Donovan J, et al. The gaseous mediator, hydrogen sulphide, inhibits in vitro motor patterns in the human, rat and mouse colon and jejunum. Neurogastroent Motil. 2008;20(12):1306‐1316.
26. Hosoki R, Matsuki N, Kimura H. The possible role of hydrogen sulfide as an endogenous smooth muscle relaxant in synergy with nitric oxide. Biochem Biophys Res Commun. 1997;237(3):527‐531.
27. Teague B, Asiedu S, Moore PK. The smooth muscle relaxant effect of hydrogen sulphide in vitro: evidence for a physiological role to control intestinal contractility. Br J Pharmacol. 2002;137(2):139‐145.
28. Martinez‐Cutillas M, Gil V, Mañé N, et al. Potential role of the gaseous mediator hydrogen sulphide (H2S) in inhibition of human colonic contractility. Pharmacol Res. 2015;93:52‐63.
29. Irie K, Ekuni D, Tomofuji T, et al. Combined effects of hydrogen sulfide and lipopolysaccharide on osteoclast differentiation in rats. J Periodontol. 2012;83(4):522‐527.
30. Lobb I, Zhu J, Liu W, Haig A, Lan Z, Sener A. Hydrogen sulfide treatment ameliorates long‐term renal dysfunction resulting from prolonged warm renal ischemia‐reperfusion injury. Can Urol Assoc J. 2014;8(5–6):E413‐418.
31. Wallace JL, Ferraz JG, Muscara MN. Hydrogen sulfide: an endogenous mediator of resolution of inflammation and injury. Antioxid Redox Signal. 2012;17(1):58‐67.
32. Ding T, Chen W, Li J, Ding J, Mei X, Hu H. High glucose induces mouse mesangial cell overproliferation via inhibition of hydrogen sulfide synthesis in a TLR‐4‐dependent manner. Cell Physiol Biochem. 2017;41(3):1035‐1043.
33. Zheng YJ, Luo NX, Mu DZ, et al. Lipopolysaccharide regulates biosynthesis of cystathionine gamma‐lyase and hydrogen sulfide through toll‐like receptor‐4/p38 and toll‐like receptor‐4/ NF‐kappa B pathways in macrophages. In Vitro Cell Dev Biol Anim. 2013;49(9):679‐688.
34. Bourque C, Zhang Y, Fu M, et al. H2S protects lipopolysaccharide‐induced inflammation by blocking NFkappaB transactivation in endothelial cells. Toxicol Appl Pharmacol. 2018;338:20‐29.
35. Fiorucci S, Distrutti E, Cirino G, Wallace JL. The emerging roles of hydrogen sulfide in the gastrointestinal tract and liver. Gastroenterology. 2006;131(1):259‐271.
36. Blachier F, Davila A‐M, Mimoun S, et al. Luminal sulfide and large intestine mucosa: friend or foe? Amino Acids. 2010;39(2):335‐347.
37. Jimenez M, Gil V, Martinez‐Cutillas M, Mane N, Gallego D. Hydrogen sulphide as a signalling molecule regulating physiopathological processes in gastrointestinal motility. Brit J Pharmacol. 2017;174(17):2805‐2817.
38. Medani M, Collins D, Docherty NG, Baird AW, O'Connell PR, Winter DC. Emerging role of hydrogen sulfide in colonic physiology and pathophysiology. Inflamm Bowel Dis. 2011;17(7):1620‐1625.
39. Ubeda C, Lipuma L, Gobourne A, et al. Familial transmission rather than defective innate immunity shapes the distinct intestinal microbiota of TLR‐deficient mice. J Exp Med. 2012;209(8):1445‐1456.
40. Linden DR. Hydrogen sulfide signaling in the gastrointestinal tract. Antioxid Redox Sign. 2014;20(5):818‐830.
41. Loubinoux J, Bronowicki JP, Pereira IA, Mougenel JL, Faou AE. Sulfate‐reducing bacteria in human feces and their association with inflammatory bowel diseases. FEMS Microbiol Ecol. 2002;40(2):107‐112.
42. Rinttila T, Kassinen A, Malinen E, Krogius L, Palva A. Development of an AOA hemihydrochloride extensive set of 16S rDNA‐targeted primers for quantification of pathogenic and indigenous bacteria in faecal samples by realtime PCR. J Appl Microbiol. 2004;97(6):1166‐1177.
43. Jia W, Whitehead RN, Griffiths L, et al. Diversity and distribution of sulphate‐reducing bacteria in human faeces from healthy subjects and patients with inflammatory bowel disease. FEMS Immunol Med Microbiol. 2012;65(1):55‐68.
44. Cario E. Toll‐like receptors in inflammatory bowel diseases: a decade later. Inflamm Bowel Dis. 2010;16(9):1583‐1597.
45. Grasa L, Abecia L, Forcén R, et al. Antibiotic‐induced depletion of murine microbiota induces mild inflammation and changes in toll‐like receptor patterns and intestinal motility. Microb Ecol. 2015;70(3):835‐848.
46. Hirata I, Naito Y, Takagi T, et al. Endogenous hydrogen sulfide is an anti‐inflammatory molecule in dextran sodium sulfate‐induced colitis in mice. Dig Dis Sci. 2011;56(5):1379‐1386.
47. Yang G, Wu L, Bryan S, Khaper N, Mani S, Wang R. Cystathionine gamma‐lyase deficiency and overproliferation of smooth muscle cells. Cardiovasc Res. 2010;86(3):487‐495.
48. Guo X, Xia X, Tang R, Zhou J, Zhao H, Wang K. Development of a real‐time PCR method for Firmicutes and Bacteroidetes in faeces and its application to quantify intestinal population of obese and lean pigs. Lett Appl Microbiol. 2008;47(5):367‐373.
49. Nee LM. An Exploration of the Gut Bacteria in a Mouse Model of Autism Spectrum Disorder on A Ketogenic Diet. [Senior Theses]. Hartford, CT: Trinity College; 2017.
50. Schneeberger M, Everard A, Gómez‐Valadés AG, et al. Akkermansia muciniphila inversely correlates with the onset of inflammation, altered adipose tissue metabolism and metabolic disorders during obesity in mice. Sci Rep. 2015;5:16643.