Pharmacological inhibition of hydrogen sulfide production mitigates constipation in a type 1 diabetes mouse model

Abstract

Constipation due to colonic contractility disorders is the predominant gastrointestinal symptom in diabetics. Hydrogen sulfide (H₂S) is an intestinal contractile agent at low concentrations and a relaxant at high concentrations. Cystathionine γ-lyase (CSE), cystathionine β-synthase (CBS), and sulfate-reducing bacteria (SRB) are among the factors that produce H₂S. This study investigated the effects of H₂S production inhibitors on colonic motility indices in mice with diabetic-induced constipation. Fifty-six mice were randomly allocated into four groups, including control, diabetic constipation (DC), disulfiram, and propargylglycine (PAG). Diabetes was induced using streptozotocin (STZ), followed by the administration of disulfiram and PAG. Blood and colon tissue samples were collected for analysis at the end of the study period. Measurements included body weight, blood glucose level, fecal parameters, and intestinal transit ratio (ITR). The gene expression levels of CBS, CSE, Bcl-2 antagonist/killer (BAK), and B-cell lymphoma 2 (BCL2) were measured, along with myosin light chain (MLC) protein expression. H₂S and gastrin levels, as well as the SRB content, were analyzed. Additionally, acetylcholinesterase (AChE) expression in colon tissue was evaluated. Histological assessment of the colon was also performed. Disulfiram and PAG administration improved the fecal pellet number and water content in mice with DC. H₂S inhibition decreased CBS and CSE gene expression, and improved SRB levels, ITR, and histological factors. The results of this study demonstrated that H₂S is an effective and key factor in regulating colonic motility in mice with DC. In the future, inhibitors of H₂S production may be used to manage the digestive complications associated with DC.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-38664-8.

Introduction

Diabetic constipation) DC) is the most common gastrointestinal disorder in diabetic patients, with a prevalence rate of 60%. The pathogenic mechanisms of constipation resulting from colonic motility disorders include autonomic neuropathy, endothelial dysfunction, enteric nervous system impairment, and alterations in gut microbiota¹. However, the primary cause of reduced colonic motility in diabetic patients has not yet been fully identified.

Hydrogen sulfide (H₂S) is a gaseous neurotransmitter produced by cystathionine γ-lyase (CSE) and cystathionine β-synthase (CBS). Studies have shown that H₂S plays a dual role in gastrointestinal motility: at low concentrations, it acts as a contractile agent and enhances intestinal motility, whereas at high concentrations, it exerts an inhibitory effect on gastrointestinal movements². A research study reported that H₂S levels and the expression of its producing enzymes are altered in diabetic patients and various diabetic models³. In addition, some studies indicate that H₂S plays a role in the development of diabetes and its complications^4,5. A recent study demonstrated that H₂S is involved in the pathophysiology of diabetic constipation, with increased levels observed in the colon of diabetic mice¹. However, the exact role of H₂S and its effect on colonic motility disorders caused by type 1 diabetes has not been extensively studied.

Contraction of the intestinal smooth muscle is a complex and interactive process involving both smooth muscle cells and gastrointestinal hormones. Myosin light chain (MLC) serves as a critical component of smooth muscle cells and is essential for regulating their contractile activity. Earlier studies have shown that variations in MLC expression are closely linked to changes in colonic smooth muscle contractility and intestinal transit in different mouse models of constipation^6,7. In the present study, we evaluated MLC and gastrin levels to examine colonic smooth muscle function and to explore potential alterations associated with reduced intestinal motility in DC mice.

In addition to enzymatic production, H₂S can be produced as a metabolic byproduct of sulfate-reducing bacteria (SRB)^8,9. Among the SRB families, bacterial species such as Desulfovibrio, Desulfomicrobium, Lawsonia, and Bilophila are found in the intestines of humans and animals. SRB are a diverse group of bacteria that metabolize organic compounds to acetate and CO₂. The species and abundance of SRB on the intestinal mucosal surface differed from those in the lumen. SRB are believed to contribute to weight loss and increased intestinal permeability. Furthermore, previous research has shown that SRB gavage increases H₂S production in mice and decreases colonic transit¹⁰.

Two types of pharmaceutical agents inhibit enzymatic H₂S production, including propargylglycine (PAG) and disulfiram. PAG is a commonly used inhibitor of H₂S synthesis. PAG is a selective and irreversible inhibitor that significantly reduces H₂S production by inhibiting CSE activity^11,12. On the other hand, disulfiram is a medication that aids in the treatment of chronic alcoholism by causing intense sensitivity to alcohol. This drug inhibits the hepatic enzyme aldehyde dehydrogenase. Studies have shown that disulfiram administration has a strong preventive effect on increasing blood glucose levels in diabetic mice. Moreover, disulfiram is recognized as a potential anticancer drug for various types of tumors, including breast cancer, and is considered a potential CBS inhibitor^13–15.

Although studies have been conducted on DC, there is insufficient evidence regarding the role of H₂S and its relationships with fecal water content, MLC, gastrin, and SRB activity on colonic hypomotility. Accordingly, this study aimed to examine the role of H₂S and its pharmacological inhibitors in diabetes-induced constipation by integrating functional (fecal parameters and intestinal transit), biochemical (H₂S and gastrin levels), microbial (SRB abundance), histopathological (immunofluorescence (IF), hematoxylin and eosin (H&E), and transmission electron microscopy (TEM)) and molecular (CBS, CSE, apoptotic markers, and MLC expression) assessments to elucidate the mechanisms linking H₂S signaling to colonic motility dysfunction.

Materials and methods

Chemicals

STZ and PAG were purchased from Sigma-Aldrich (St. Louis, MO, USA). The RNA extraction kit, cDNA synthesis kit, SYBR Green Master Mix, housekeeping and target primers, and BCA kit were purchased from Parstous Co. and Amplicon Co. The gastrin ELISA kit was purchased from SunLong Biotech Co. Formalin was purchased from Alfasan Co. (Netherlands), and disulfiram tablets 500 mg were obtained under the brand Antabus from NOBEL iLAC Co.

Animals

Fifty-six male NMRI mice weighing 20–25 g (6–8 weeks) were obtained from the Laboratory Animal Reproduction Center. They were placed under standard thermal conditions and a 12-h light/dark cycle, ad libitum access to regular food and water. Animals were fasted for 8 h before the study, maintaining continuous access to water. Animal studies are reported in compliance with the ARRIVE guidelines and were approved by the Institutional Animal Care and Use Committee of Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran (Ethic code: IR.AJUMS.ABHC.REC.1402.001). All procedures were performed in accordance with the relevant guidelines and regulations for the care and use of laboratory animals.

Experimental group

Animals were placed in the laboratory 1 week before diabetes induction to habituate to the new environment. The determination of fecal parameters (number and wet weight) started 1 day before STZ injection and continued until the end of the study. The second criterion for entering the study was a significant decrease in the number and wet weight of fecal pellets. Therefore, animals with a significant decrease in fecal pellet number and wet weight were included in the study. Fifteen mice out of 42 met both the conditions. Finally, diabetic animals were assigned to three groups (5 in each group): diabetic control, disulfiram, and PAG. Animals in the disulfiram group received disulfiram (10 mg/kg, p.o. by gavage, dissolved in fresh physiological saline) for 3 consecutive days^14,16. Animals in the PAG group received PAG (10 mg/kg, intraperitoneally, dissolved in fresh physiological saline) for 3 consecutive days¹⁷. Additionally, the same number of intact mice were recruited for the control group (n = 5). Figure 1 provides an overview of the design and experimental timeline used in this study.

Fig. 1.

Graphical overview of the study design and experimental timeline.

Sample collection and preparation

Colon tissues were collected immediately after anesthesia induction and washed with normal saline. From each mouse, the entire colon tissue was harvested and systematically divided into predefined portions allocated for different downstream analyses, including gene expression analysis, protein evaluation, and H₂S measurement, all snap-frozen in liquid nitrogen; and approximately 2 cm of colon tissue for histopathological analyses. For protein extraction, RNA extraction, and H₂S determination, tissues were homogenized in appropriate buffers containing protease and phosphatase inhibitors, Samples were then centrifuged at 12,000 × g for 20 min at 4 °C, and the resulting supernatants were collected and stored at − 70 °C until further analysis. All tests were performed using colon tissues obtained from distinct mice in each group. The number of biological replicates (n) for each assay is indicated in the corresponding figure legends. Biological replicates represent independent samples derived from individual animals, with one sample per animal. Real-time PCR analyses were performed in duplicate for each sample.

Induction of diabetes

To induce type 1 diabetes, mice were fasted for 8 h before receiving a single intraperitoneal injection of STZ (160 mg/kg, dissolved in 0.1 M citrate buffer, pH = 4.5). Blood glucose levels were measured 3 days and 2 weeks after the STZ injection. Approximately 0.1 mL of blood was collected from the tail vein, and glucose concentration was determined using a glucometer (Diaban, SMM1000, Iran). Mice with fasting blood glucose levels > 200 mg/dL were considered diabetic¹⁸.

Measurement of stool parameters and intestinal transit ratio

Two weeks after diabetes induction, the animals were examined to confirm constipation, and the number and wet weight of fecal pellets were assessed both before and after the induction. If these indicators were reduced, the animals were considered constipated¹⁹. For each measurement, mice were housed individually in separate cages for 24 h (from 9:00 a.m. to 9:00 a.m. the next day), and fecal pellets were collected, counted, and weighed separately for each animal. Statistical analyses were performed using individual animal values (n = 5 animals per group). Counting fecal pellets continued throughout the study in all groups. Their numbers and weights were then determined. Next, to determine the dry weight, the samples were maintained at room temperature for 48 h and then weighed after complete drying. The water content of the feces was calculated using the following formula:⁷.

1

To evaluate the ITR, animals received charcoal (1mL/100 g of B.W., P.O.). After 25 min, the mice were euthanized with a high dose of ketamine/xylazine mixture (100/10 mg/kg) via an intraperitoneal injection. Finally, the entire small intestine from the pylorus to the cecum was removed, and the whole length was measured. The distance traveled by charcoal and the total length of the small intestine were determined using the following formula:¹⁹.

2

Hydrogen sulfide measurement

Hydrogen Sulfide levels in tissue and plasma were measured using a H₂S Content Assay Kit (SunLong Biotech, catalog number: AK0617) according to the manufacturer’s instructions. H₂S reacts with N, N-dimethyl-p-phenylenediamine and ferric ammonium sulfate to form methylene blue, which has a maximum absorbance at 665 nm. Absorbance was measured with a spectrophotometer or microplate reader.

Sample preparation:

Tissue: 0.1 g of tissue was homogenized with 1 mL of Extraction I solution on ice and centrifuged (12,000 × g, 10 min, 4 °C). Then 0.15 mL of Extraction II solution was added to 0.8 mL of supernatant and centrifuged again. The final supernatant was used for measurement. Plasma: 1 mL of plasma was mixed with 1 mL of Extraction I solution and centrifuged, then 0.15 mL of Extraction II solution was added to 0.8 mL of supernatant and centrifuged again.

50 µL of sample was mixed with 75 µL of Reagent I and 75 µL of Reagent II and incubated for 10 min at room temperature. The absorbance was recorded at 665 nm; ΔA = AT − AB (test minus blank). The corresponding H₂S concentration (x, nmol/mL) was determined from a standard curve generated using the kit’s standard solutions. The value of x was then used to calculate H₂S content in tissue (nmol/g) or plasma (nmol/mL) according to the following formulas:²⁰.

3

4

where:

x: H₂S concentration from standard curve (nmol/mL).

V_SP: Supernatant volume in Extraction (0.8 mL).

V_EX2: Extraction solution II volume (0. 15 mL).

V_EX1: Extraction solution I volume (1 mL).

W: Sample weight (g).

V_L: Liquid sample volume (0.1 mL).

Quantification of gene expression in colon tissue using real-time PCR

RNA extraction was performed using the Total RNA Extraction kit according to the manufacturer’s instructions. Briefly, RNA was extracted, and the integrity and purity of the extracted RNA were determined by a Nanodrop at a wavelength of 260 nm/280nm. After that, cDNA was synthesized using the manufacturer’s instructions. Real-time PCR was used to evaluate the expression of the studied genes, including CBS, CSE, B-cell lymphoma 2 (BCL2), and Bcl-2 antagonist/killer (BAK).

The primers used for gene expression analysis are listed in Table 1. Primers were designed using Primer 3 software.

Table 1.

List of mRNAs and their corresponding forward and reverse Primers.

mRNA	Primer sequence
GAPDH	F-CATCACTGCCACCCAGAAGACTG	R-ATGCCAGTGAGCTTCCCGTTCAG
BCL2	F-CCTGTGGATGACTGAGTACCTG	R-AGCCAGGAGAAATCAAACAGAGG
BAK	F-GGAATGCCTACGAACTCTTCACC	R- CAAACCACGCTGGTAGACGTAC
CSE	F-GTGGGACAAGAGCCTGAGCAAT	R-GGATTTCCAGAGCGGCTGTATTC
CBS	F-GATCGCCAGAAAGCTGAAGGAG	R-CCACCTCATAGGCTGTTTGCTC

Real-time PCR was carried out using previously established protocols, and the results were reported according to the general formula 2^–ΔΔCt. GAPDH served as the housekeeping gene²¹.

Measuring the expression of myosin light chain protein by Western blot

For protein extraction, 100 mg of colon tissue was placed in a mixture of RIPA buffer containing protease inhibitors and homogenized for 60 s at 5000 rpm at 4 °C. The homogenized mixture was then placed on a shaker and incubated for 30 min. In the next step, the homogenized mixture was centrifuged at 12,000 rpm for 20 min at 4 °C, and the supernatant containing the extracted protein was stored at -70 °C. To measure MLC protein using the Western blot technique, the protein concentration in each sample was first determined using the Bradford method. For protein separation in sodium dodecyl sulfate (SDS)-PAGE, samples were boiled for 5 min in an adequate amount of SDS and 2-mercaptoethanol solution. In the next step, 15 µg of protein was electrophoresed on a 10% SDS-PAGE gel. The proteins were then transferred onto a nitrocellulose membrane, and a 5% skim milk solution was used to block the membrane pores. Subsequently, MLC protein was incubated overnight at 4 °C with a rabbit polyclonal antibody and a mouse monoclonal GAPDH antibody (SANTA CRUZ BIOTECHNOLOGY, INC.). Next, after washing three times with Tris-saline solution containing 0.1% Tween (TBST), the samples were incubated for two h with horseradish peroxidase-conjugated secondary antibodies. Finally, using the chemiluminescence method, MLC bands were visualized, and their density was quantified using FUSION FX software (Vilber, USA) and compared to the reference protein (GAPDH)²².

Gastrin measurement

Gastrin levels in plasma were quantified using commercially available ELISA kits. Plasma was separated by centrifuging whole blood at 3,000 g for 10 min at 4 °C. The resulting supernatant was kept at − 70 °C for future analysis. Both standards and plasma samples were prepared according to the manufacturer’s guidelines for the gastrin assay. Briefly, 100 µL of each standard or plasma sample was pipetted into a 96-well ELISA plate that had been pre-coated with specific antibodies. The plates were incubated at 37 °C for 1 h and washed to remove any unbound substances. Next, 100 µL of an enzyme-linked detection antibody was added to each well, and incubation was carried out in accordance with the manufacturer’s instructions. After further washing, 100 µL of substrate solution was added, and the reaction was stopped by adding 50 µL of stop solution. Finally, absorbance was measured at 450 nm using a microplate reader²³.

Histology and immunofluorescence

After the experiment, the colon was separated and placed in a 10% formalin solution. The tissues were dehydrated using 70–100% ethanol using the passage method. After washing the tissues with glycol, the samples were saturated with paraffin, and a tissue block was prepared. The samples were cut with a microtome to a thickness of 5 μm, stained with H&E and alcian blue, and the pathological changes in the colon tissue were examined under a light microscope.

Imaging with an electron microscope to examine colon tissue

To prepare the samples for the scanning electron microscope, they were cut into 5 mm sections. This is because the stabilizing and drying solutions cannot quickly penetrate thicker samples. The samples were then placed in hydroxyl ethyl piperazine ethanesulfonic acid (HEPES) buffer (7.3, pH 0.1 M) containing 2.5% glutaraldehyde and 2% paraformaldehyde and fixed in distilled water containing 1% osmium tetroxide. Then, the water in the samples was removed using alcohol and acetone, and the drying material was completely separated from the samples using liquid carbon dioxide. The samples were then coated with layers of carbon and aluminum, followed by a layer of platinum. Each sample was then placed in a stable frame and examined under a scanning electron microscope²⁴.

Measurement of acetylcholinesterase (AChE) in colon tissue by immunofluorescence staining protocol

Tissue samples were fixed in 10% formalin for 2 weeks. Then, 5-micron tissue sections were prepared for IF staining. Antigen retrieval was performed using a sodium citrate buffer (pH 6) at 100 °C for 30 min. Endogenous peroxidase activity was blocked with 0.3% hydrogen peroxide. Sections were incubated with a blocking solution containing BSA to minimize non-specific binding.

Primary antibodies were diluted (1:100) in a blocking solution and incubated overnight at 4 °C. After washing with TBS, fluorophore-conjugated secondary antibodies (1:500) were applied for 2–3 h at room temperature in the dark. To visualize nuclei, DAPI staining was performed for 30 min. Finally, slides were mounted using an antifade mounting medium and analyzed with a fluorescence microscope (Olympus DP72)²⁵.

The absolute quantitative real-time PCR (qPCR) method for determining SRB count

Real-time PCR method and the Absolute Quantification model to determine the number of SRB bacteria (Desulfovibrio genus) in fecal samples. Real-time PCR was performed using specific primers for the target 16S rRNA gene. The forward primer DSV691 (5′- CCGTAGATATCTGGAGGAACATCAG-3′) and the reverse primer (5′-ACATCTAGCATCCATCGTTTACAGC-3′) were used for this gene. Absolute quantification was carried out using the pNovel TA Lethal plasmid containing a specific fragment of the 16 S rRNA gene. The bacterial count was calculated using a standard curve plotted against the primer copies. DNA extraction was performed using a stool DNA isolation kit, and quantification was performed using a 1000 NanoDrop ND spectrophotometer. DNA amplification and detection were carried out using the YTA qPCR Probe MasterMix kit. A standard curve was created using diluted pNovel TA Lethal plasmid DNA containing 16 S rRNA from the standard sample. The absolute plasmid quantity for each unknown fecal DNA sample was calculated using the standard curve. The amplification efficiency was determined using the slope of the standard curve²⁶.

Statistical analysis

The data were analyzed with GraphPad Prism software (version 8). Multiple comparison analysis was performed using one-way ANOVA followed by Tukey’s test. All data are presented as mean ± Standard deviation (SD). Differences were considered statistically significant at *p < 0.05, and more significant differences are reported as **p < 0.01, ***p < 0.001, ****p < 0.0001.

Results

Eeffect of disulfiram and PAG on the weight of mice in the studied groups

Induction of diabetes and subsequent constipation caused a substantial decrease in the weight of animals in the DC group compared to the control group (p < 0.001). However, PAG and disulfiram failed to revert their body weight to normal (Fig. 2A).

Fig. 2.

Effect of disulfiram (10 mg/kg) and propargylglycine (PAG) (10 mg/kg) on (A) body weight and (B) blood glucose levels in different groups. Data are expressed as mean ± SD. (n = 5) ***p < 0.001 ****p < 0.0001. DC represents the diabetic constipation group. PAG and Disulfiram, respectively, represent the diabetic-induced constipation animals that received PAG and Disulfiram for 3 consecutive days.

Effect of disulfiram and PAG on the blood glucose of mice in the studied groups

STZ injection caused a notable increase in blood glucose levels in the DC group compared to the control group (p < 0.0001). Using PAG and disulfiram as H₂S-producing enzyme inhibitors did not reduce the blood glucose levels (Fig. 2B).

Effect of disulfiram and PAG on the number and water content of fecal pellets

The induction of diabetic constipation significantly decreased the number of fecal pellets (Fig. 3A) (p < 0.0001) and water content (Fig. 3B) (p < 0.001) in the DC group compared to those in the control group. The use of PAG and disulfiram improved the number of fecal pellets and the water content compared to those in the DC group (p < 0.05).

Fig. 3.

Effect of disulfiram (10 mg/kg) and propargylglycine (PAG) (10 mg/kg) on (A) fecal pellet number, and (B) fecal water content in different groups. Data are expressed as mean ± SD. (n = 5) *p < 0.05, ***p < 0.001 and ****p < 0.0001.

Effect of disulfiram and PAG on the intestinal transit ratio

Induction of diabetic constipation reduced ITR compared to the control group (p < 0.001). PAG and disulfiram significantly improved ITR (p < 0.05) (Fig. 4).

Fig. 4.

Effect of disulfiram (10 mg/kg) and propargylglycine (PAG) (10 mg/kg) on intestinal transit ratio (ITR). Data are expressed as mean ± SD. (n = 5) *P < 0.5. ***P < 0.001.

Effect of disulfiram and PAG on hydrogen sulfide

Induction of diabetic constipation increased H₂S levels in both plasma (p < 0.05) and colon tissue (p < 0.01) in the DC group compared to the control group. PAG and disulfiram administration, as inhibitors, significantly improved plasma and tissue H₂S levels compared to the DC group (p < 0.05 at both) (Fig. 5A and B).

Fig. 5.

Effect of disulfiram (10 mg/kg) and propargylglycine (PAG) (10 mg/kg) on (A) plasma H₂S levels, (B) tissue H₂S content, and (C) gastrin levels. Data are expressed as mean ± SD. (n = 5) * p < 0.05, ** p < 0.01.

Effect of disulfiram and PAG on plasma gastrin level

As indicated in Fig. 5C, the induction of diabetic constipation significantly increased the level of gastrin compared to the control group (p < 0.01). PAG and disulfiram administration did not affect gastrin hormone levels compared to the DC group.

mRNA expression analysis by real-time PCR

As shown in Fig. 6A and B, the induction of diabetic constipation significantly increased CSE and CBS mRNA expression in the DC group compared with the control group (p < 0.01 at both). PAG significantly reduced the CSE gene expression in colon tissue compared to the DC group (p < 0.01) but did not affect CBS gene expression. In contrast, disulfiram administration significantly reduced CBS gene expression (p < 0.05). However, it did not affect CSE gene expression in the colon tissue.

Fig. 6.

Effect of disulfiram (10 mg/kg) and propargylglycine (PAG) (10 mg/kg) on studied gene and protein expression levels. (A) Cystathionine γ-lyase (CSE) gene expression, (B) Cystathionine β-synthase (CBS) gene expression, (C) Bcl-2 antagonist/killer (BAK) gene expression, (D) B-cell lymphoma 2 (BCL2) gene expression, and (E) Sulfate-reducing bacteria (SRB) gene copy number. Data are expressed as mean ± SD (n = 5). * p < 0.05, ** p < 0.01, *** p < 0.001.

In addition, diabetic constipation significantly increased the expression of BAK (p < 0.001) and significantly decreased BCL2 gene expression (p < 0.05) in the DC group compared to the control group. Disulfiram reduced BAK gene expression in colon tissue (p < 0.05), whereas BCL2 gene expression was not affected compared to the DC group. In contrast, PAG administration did not significantly alter the expression of the BAK and BCL2 genes (Fig. 6C and D).

The expression of SRB significantly increased in the stool samples of the DC group compared to the Control group (p < 0.01). Administration of PAG and disulfiram reduced the abundance of these bacteria compared to the DC group (p < 0.05) (Fig. 6E).

Western blot analysis

The induction of diabetic constipation significantly increased the expression of MLC protein in the DC group compared to the Control group (p < 0.001). Disulfiram reduced MLC protein expression (p < 0.05). At the same time, PAG administration increased the MLC level compared to that in the DC group (p < 0.05) (Fig. 7).

Fig. 7.

Effect of disulfiram (10 mg/kg) and propargylglycine (PAG) (10 mg/kg) on myosin light chain (MLC) protein expression. Data are expressed as mean ± SD. (n = 4) *p < 0.05 ***p < 0.001. CO (Control), DIS (Disulfiram). Western blots were performed for all four samples in each group. One representative blot from each group is shown in Fig. 6, while the remaining replicate blots and full-length, uncropped images are provided in the Supplementary Information.

Histopathology

Microscopic examination of the control group revealed normal mucosal thickness, and the surface of the intestinal villi was long and covered with columnar cells. In addition, numerous goblet and mucous cells lined the walls of the Crypts of Lieberkühn. In the constipation group, the thickness of the intestinal mucosa was severely reduced, and the villi were destroyed. The cells covering the mucosa and the Crypts of Lieberkühn were destroyed, and a limited number of goblet cells were observed. In the disulfiram and PAG groups, the thickness of the intestine increased, and many goblet cells were observed compared to the constipation group (Fig. 8A). Alcian blue staining was used to confirm the nature of goblet cells. In this staining, secreted mucin is visible in blue.

Fig. 8.

Effect of disulfiram (10 mg/kg) and Propargylglycine (PAG) (10 mg/kg) on histopathological and molecular changes in colon tissue. (A) Hematoxylin and eosin (H&E) staining shows that the DC group exhibited severe mucosal thinning, villus destruction, and a reduced number of goblet cells. At the same time, the use of disulfiram and PAG improved intestinal thickness and goblet cell count. (B) Alcian blue staining confirms mucin secretion, with a significant reduction in blue-stained goblet cells in the diabetic constipation group, while disulfiram and PAG increase mucin production and crypt dilation. Goblet cells (G) and erythrocytes (E) are indicated by arrows. (C) Transmission electron microscopy (TEM) images reveal microvilli destruction and cytoplasmic vacuolation in the DC, with PAG improving cellular morphology. Villi (yellow arrows), nuclei (black arrows), and vacuoles (red arrows) are indicated in the image. (D) Immunofluorescence (IF) staining indicates increased acetylcholinesterase (AChE) expression in the diabetic constipation group, which was partially reduced by disulfiram and PAG. (E) Quantitative analysis of IF intensity demonstrating relative AChE expression levels among the experimental groups. Data are expressed as mean ± SD. (n = 3) * p < 0.05, *** p < 0.001 and **** p < 0.0001.

The number of blue goblet cells was significantly reduced in the constipated group. In the PAG and disulfiram groups, the number of these cells increased, and the ducts of the Crypts of the Lieberkühn were dilated and contained blue mucin, indicating significant repair and increased mucin production (Fig. 8B).

Transmission electron microscopy

Electron microscopy showed microvilli destruction, intracellular organelle destruction, cytoplasmic vacuolation, and nuclear destruction in mice with diabetic constipation. PAG administration improved cells’ morphological and functional parameters (Fig. 8C).

Immunofluorescence study

According to IF images, in mice with diabetic constipation, the expression of AChE was increased compared to that in the control group. Disulfiram and PAG partially inhibited this effect (Fig. 8D and E).

Discussion

Given the increasing prevalence of diabetes and its negative impact on global health, diabetic patients often experience various gastrointestinal symptoms, including chronic constipation, which significantly affects their quality of life. This study showed that a single intraperitoneal dose of STZ reduced fecal pellet count, fecal water content, and intestinal transit ratio 2 weeks after the onset of diabetes, consistent with the findings of Song et al.²⁷. The results of this study indicated that the body weight of mice in the DC group was significantly reduced compared to the control group, which is consistent with the findings of Quan’s study published in 2022¹. This similarity is due to the same method of inducing constipation in both studies, which was achieved by inducing diabetes with STZ, highlighting the key role of diabetes in weight loss. In contrast, our results differ from those of Ji Eun Kim’s 2020 study⁷, which created a constipation model by knocking out the leptin gene and inducing obesity, resulting in a significant increase in the body weight of mice. The difference in the constipation induction model can explain this discrepancy.

As previously mentioned, H₂S has a dual effect on gastrointestinal motility; at low concentrations, it acts as a contractile agent, whereas at high concentrations, it inhibits intestinal contractions. These dual effects are mediated through different molecular pathways, such as TRPV1 channels (for excitation) and ATP-sensitive potassium channels (for inhibition)^28,29. As shown in the current study, the elevated H₂S levels observed in diabetic constipated mice are likely to shift the balance toward inhibitory signaling, resulting in reduced propulsive movements. The decreased rate of propulsive contractions increases the time for water absorption along the colon, which in turn decreases the water content and fecal pellet number.

Our results also indicated a significant increase in the expression of CBS and CSE enzymes responsible for H₂S production and elevated plasma and tissue levels of H₂S in diabetic mice, consistent with the findings of Quan et al.¹. Our findings are consistent with the results of previous studies, which demonstrated that inhibition of CBS and CSE by reducing the synthesis of H₂S improved colonic and ileal motility in rats and mice, respectively^30,31.

The most important microbial producers of H₂S are sulfate-reducing bacteria 2S. SRB are categorized under the δ-Proteobacteria class and produce H₂S gas via the dissimilatory sulfite reductase enzyme complex³². The primary genus in this category is Desulfovibrio, which employs lactate and hydrogen as substrates. Studies have shown that approximately 50% of fecal H₂S is of bacterial origin, as only half of its synthesis is dependent on pyridoxal-5-phosphate. In germ-free mice, fecal H₂S levels are only 50% of those in mice with bacterial flora³³. This finding suggests an important role for bacteria in intestinal H₂S production.

Our results showed increased SRB expression in the feces of diabetic constipated mice compared to the control group, which may indicate a higher presence of these bacteria in the gut and, consequently, more H₂S production, leading to increased hypomotility. The presence of SRB in fecal samples from groups treated with disulfiram and PAG significantly decreased, which may be based on the hypothesis that reduced H₂S production in these groups and increased intestinal contractions subsequently improved motility, reducing the retention of feces in the colon, which in turn decreased SRB presence.

Ritz et al.¹⁰ demonstrated that the instillation of SRB by gavage in mice reduced the intestinal transit ratio while increasing H₂S production in the colon. In addition, Chassard et al. reported in 2012 that in patients with constipation-predominant irritable bowel syndrome, the number of SRB was significantly higher than that in the control group³⁴.

Our research showed that SRB levels in diabetic constipated mice significantly changed, and the H₂S produced by these bacteria may contribute to the progression of DC. On the other hand, Rowan et al.³⁵ showed that SRB play a role in the pathogenesis of ulcerative colitis by overproducing H₂S, disrupting mucosal integrity, and inhibiting butyrate utilization. These findings confirm the harmful role of high levels of SRB and overproduction of H₂S in digestive disorders, including constipation³⁵. In a 2014 human study, Wang et al. showed that oral consumption of the probiotic Lactobacillus plantarum P-8 for 4 weeks reduced the Desulfovibrio species, which is the predominant form of SRB³⁶. Sawin et al. showed that the consumption of the prebiotic glycomacropeptide significantly reduced the predominant form of SRB in fecal samples³⁷. Therefore, it can be suggested that the consumption of probiotics and prebiotics can play a role in modulating the SRB population and, as a complementary therapeutic approach, can improve the gastrointestinal complications caused by increased SRB levels and reduce their negative effects. Despite these findings, the exact contribution of bacterial and enzymatic sources of H₂S in regulating colonic transit in diabetic constipation remains unclear and requires further investigation.

BAK plays an essential role in normal cells, whereas BCL2 promotes cell survival by inhibiting apoptosis. Our study showed that BAK expression increased in the colonic tissue of DC mice, while BCL2 expression decreased. This could be due to increased enzymatic and bacterial H₂S levels in the colonic tissue of DC mice. These findings are consistent with recent studies that showed increased BAK expression in the colonic tissue of loperamide-induced constipation mice³⁸ and with previous studies in diabetic patients³⁹. The studied inhibitors, PAG and disulfiram, did not have a significant effect on improving apoptotic pathways. This could be due to various factors, such as the lack of bacterial inhibitors of H₂S synthesis and the simultaneous impact of diabetes on the apoptotic pathway in our study. However, disulfiram slightly improved BAK expression in the colon tissue of DC mice, which may be related to its protective and antioxidant properties, as reported in previous studies.

The regulation of gastrointestinal motility involves multiple factors, including H₂S, smooth muscle contractility (assessed via MLC as a key contractile marker of smooth muscle), and hormones such as gastrin, an essential gastrointestinal hormone involved in intestinal motility that links these to H₂S-mediated hypomotility. As anticipated, DC mice showed elevated H₂S and MLC protein expression and plasma gastrin levels, yet reduced fecal output, water content, and ITR, indicating that H₂S-dominant hypomotility overrides compensatory increases in MLC and gastrin. H₂S inhibition by PAG/disulfiram improved motility indices and histology without altering gastrin levels, underscoring H₂S as the primary driver linking elevated levels to delayed transit, water absorption, and constipation.

Our results demonstrated an increase in MLC expression in the colonic tissue of the DC group, which may represent a compensatory response aimed at enhancing contractile activity and, consequently, improving colonic motility. However, this upregulation of MLC expression failed to restore colonic motility in the DC group, potentially due to reduced MLC phosphorylation, dysfunction of myosin light chain kinase (MLCK) and myosin light chain phosphatase (MLCP) enzymes, or alterations in calcium levels. Supporting this hypothesis, previous studies have reported increased MLC expression accompanied by reduced phosphorylation in constipation and diabetic models^7,40. Furthermore, studies have shown that high concentrations of H₂S predominantly inhibit cytochrome c oxidase (Complex IV) in the electron transport chain, leading to decreased electron flow, reduced oxygen consumption, and impaired mitochondrial function, ultimately resulting in decreased ATP production. ATP reduction levels can adversely affect the activities of MLCK and MLCP enzymes, leading to reduced phosphorylation of MLC^41,42.

Furthermore, our study revealed that DC could elevate plasma gastrin levels, but the studied inhibitors failed to alter this level. This finding did not agree with Kim et al.⁷, who reported a decrease in gastrin levels in a leptin gene knockout model of constipation. This discrepancy may be attributed to differences in the induction methods of constipation models. In line with previous research, Migdalis⁴³ demonstrated an increase in gastrin levels in diabetic patients, and a 2017 study confirmed elevated gastrin levels in diabetic mice^43,44. Our findings demonstrated elevated gastrin levels and MLC expression in DC mice. However, these increases failed to counteract colonic hypomotility, highlighting the critical role of H₂S in colonic motility disorders in DC mice.

Mucin, the primary constituent of mucus, plays a key role in determining the viscosity of gastrointestinal secretions. It facilitates the lubrication of luminal contents, including undigested food particles, thereby promoting their movement and eventual elimination with ease. In individuals with chronic constipation, a decrease in mucin secretion has been noted, which may impair lubrication and disrupt normal gastrointestinal function. This reduction in mucin production may slow the transit of intestinal contents and worsen constipation symptoms, highlighting a potential link between mucin secretion and the motility disturbances seen in constipation⁴⁵. Histological examination in this study revealed that diabetic constipation led to morphological and functional impairments in intestinal epithelial cells, with reduced mucin secretion, as observed by H&E and alcian blue staining. Furthermore, electron microscopy images showed damage to microvilli, as well as intracellular and nuclear organelles, in the colon of diabetic mice. These findings suggested significant tissue damage to the intestinal mucosa in this model.

Our findings suggest that excessive H₂S levels in diabetic constipation disrupt colonic mucosal integrity and reduce mucin secretion. In contrast, Wallace et al. reported that H₂S under physiological conditions protects gastrointestinal )GI( mucosal by activating cyclooxygenase-2 )COX-2(-mediated prostaglandin production⁴⁶. Together, these findings suggest H₂S at physiological doses protects the GI mucosa and promotes contractions by increasing mucin secretion rate while high concentration has the opposite effects.

AChE plays a critical role in modulating acetylcholine levels at the synaptic cleft. Its increased activity can lead to reduced acetylcholine availability, resulting in decreased intestinal motility and contributing to constipation. Similarly, a study conducted by Bharucha et al. has shown that elevated AChE activity is associated with impaired gastrointestinal motility, as reduced acetylcholine can decrease the contraction of smooth muscle in the intestines⁴⁷.

Immunofluorescence results from our study showed elevated AChE expression in the colon of constipated diabetic mice. Given the contractile effect of acetylcholine on intestinal muscles, this finding supports a reduction in intestinal motility in diabetic constipation. These results align with the study by Ji Eun Kim et al.⁷, which focused on a constipation model and decreased colonic motility.

Administration of these inhibitors improved the morphological and functional parameters of the intestinal mucosa, including increased mucosal thickness, improved shape and number of epithelial cells, and restoration of Lieberkühn crypts. Moreover, the number of goblet cells responsible for mucin secretion significantly increased in these groups. Alcian blue staining revealed increased goblet cells and blue mucin secretion, indicating enhanced secretory activity in the intestinal mucosa.

Although this study provides valuable insights into the role of H₂S in diabetic constipation, several aspects warrant further investigation in future studies. First, an in vitro study comparing colonic muscle strips from healthy and diabetic mice under varying H₂S concentrations could provide additional mechanistic insights into contractility and frequency of contraction. Second, electrophysiological studies examining the effects of H₂S on ionic channels and currents may further clarify the cellular mechanisms underlying H₂S-mediated modulation of smooth muscle activity. Third, the assessment of key molecular markers involved in H₂S biosynthesis, such as heme oxygenase-1 (HO-1), could enhance our understanding of the enzymatic regulation of diabetic constipation. Incorporating these approaches in future investigations would complement the present findings and provide a more detailed understanding of the cellular, molecular, and electrophysiological mechanisms through which H₂S modulates colonic motility.

Conclusion

This study showed that diabetes leads to and exacerbates constipation through enzymatic and bacterial pathways. Although we have proven that disulfiram and PAG improve intestinal motility and diabetic constipation by inhibiting hydrogen sulfide production, owing to study limitations, including limited financial resources, clinical trials with large and randomized populations are needed to generalize the medicinal use of these compounds in humans. Considering the positive effects observed, it is necessary to investigate the safety and efficacy of these compounds in humans.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

All authors contributed to the study conception and design. Material preparation was performed by SA.M. and M.B. Data collection and analysis were performed by R.K. and F.S. M.S and A.A contributed to the practical work. A.R contributed to histological analysis and manuscript writing. The first draft of the manuscript was written by R.K. and SA.M, and all authors collaborated on the final writing and editing of the manuscript. All authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

This paper is a part of Razieh Kazemzadeh’s Ph.D. thesis. This work is financially supported by the Persian Gulf Physiology Research Center of Ahvaz Jundishapur University of Medical Sciences (Grant No. APRC-0203).

Data availability

The datasets used and/or analyzed during the current study available from the first author, Razieh Kazemzadeh, on reasonable request. Gmail: kazemzadehrazieh@gmail.com.

Declarations

Competing interests

The authors declare no competing interests.