Comparative evaluation of three traditional herbal formulas on gastrointestinal motility in a mouse model of cold stress-induced dyspepsia

Abstract

BACKGROUND

Cold exposure has traditionally been considered a pathological factor that can easily impair gastrointestinal (GI) digestion. Shihosogan-tang (ST), Yijung-tang (YT), and Pyeongwi-san (PS) are well-known herbal formulas frequently used to treat GI disorders in East Asia.

AIM

To compare the effects of these herbal formulas on GI motility and investigate their mechanisms of action using a cold stress (CS)-induced dyspepsia mouse model.

METHODS

C57BL/6J mice were exposed to CS by immersion in cold water (10 ± 1 °C) while being restrained in conical tubes for 1 hour. This procedure was repeated six times over 2 weeks. Herbal formulas or mosapride (positive control) were administered orally five times per week over a 2-week period.

RESULTS

The pre-test results revealed that CS, rather than restraint stress, significantly delayed gut motility in mice. However, PS and ST notably improved gastric emptying and intestinal transit, surpassing YT. Additionally, PS and ST significantly reduced gastric potential of hydrogen and increased pepsin and lipase gene expression compared to CS. The observed mechanisms likely involved increased gastric acidity and enhanced levels of digestive enzymes, such as pepsin and lipase. Furthermore, PS administration elevated GI hormone levels and metabolites related to the gut microbiota (5-hydroxytryptamine and short-chain fatty acid) more effectively than ST and YT treatments.

CONCLUSION

PS more effectively alleviated CS-induced GI dysfunction than both YT and ST. These comparative findings offer valuable insights for clinical applications in the treatment of cold-related digestive disorders.

Key Words: Herbal formula; Cold stress; Dyspepsia; Gastrointestinal disorder; Glucagon-like peptide-1; Serotonin

Core Tip: This study systematically compared the therapeutic effects of three classical herbal formulas [Pyeongwi-san (PS), Shihosogan-tang, and Yijung-tang] in a mouse model of cold stress-induced dyspepsia. Among them, PS was the most effective in restoring gastrointestinal motility, enhancing gastric acid secretion and upregulating digestive enzymes gene expression, and modulating serotonin, short-chain fatty acids, bile acid receptors, and glucagon-like peptide-1 signaling. These findings elucidate the mechanisms of PS’s superior efficacy and strongly support its clinical application in treating cold-induced gastrointestinal dysfunction.

INTRODUCTION

Dyspepsia is an extremely common gastrointestinal (GI) dysfunction characterized by various symptoms including epigastric pain, burning sensation, feelings of fullness, early satiety, nausea, vomiting, belching, and bloating[1]. The global prevalence rates of functional dyspepsia are approximately 8.0% and 9.1% in developed and developing countries, respectively[2]. Dyspepsia is a global health concern[3]. Especially in Asia, 8%-30% of the population suffers from uninvestigated dyspepsia[4]. Dyspepsia is a chronic condition caused by various factors, such as Helicobacter pylori infection, medication side effects, lifestyle factors, and pregnancy[5]. These complexities make it challenging to achieve a permanent cure. Understanding the precise mechanisms underlying each cause is crucial for developing effective treatment approaches.

Cold stress (CS) is defined as a physiological condition in which exposure to low temperatures disrupts thermal homeostasis, although no universally accepted temperature threshold exists. In healthy individuals, CS has been identified as a significant environmental factor that can decrease the gastric emptying rate from 17% to 2% by reducing the number of gastric antral pressure waves[6]. Typically, cold exposure lowers the overall metabolic rate, which can decrease the production and activity of digestive enzymes in juvenile golden pompanos[7]. Gastric digestive enzymes function best under appropriate temperature and acidic conditions[8,9]. In addition, a decrease in the activity of digestive enzymes, such as proteases and lipases, in the GI tract has been observed in cold-exposed rodents[10]. Numerous animal studies have shown that CS significantly affects the composition of the gut microbiota and its metabolites, such as short-chain fatty acids (SCFA)[11,12]. Therefore, changes in the gut microbiota and their metabolites are crucial contributors to GI dysfunction[13].

Meanwhile, traditional Chinese medicine theory classifies cold as one of the six external pathogens, as described in the book Yellow Emperor’s Inner Canon, also known as Huang-Di-Nei-Jing[14]. When cold invades the stomach, it disrupts the normal flow of Qi, causing it to ascend and leading to symptoms of dyspepsia, such as belching[15]. Therefore, keeping warm is considered a preventive measure against various digestive disorders[16]. Numerous herbal medicines have been traditionally used to treat dyspepsia[17,18]. In the current study, we selected three typical digestive herbal formulas, Shihosogan-tang (ST) (Chai-Hu-Shu-Gan-Tang), Yijung-tang (YT) (Li-Zhong-Tang), and Pyeongwi-san (PS) (Ping-Wei-San), all of which were approved by the Korean National Health Insurance Service[19]. Although ST, YT, and PS have been widely used to treat dyspepsia in the Republic of Korea, China, and Japan, traditional medical theory attributes distinct benefits to each formula[20-22]. PS may alleviate stomach dampness and stagnation, ST addresses liver qi stagnation, and YT supports insufficient yang energy in the stomach. However, clear evidence to determine which of these formulas is the most effective for dyspepsia induced by CS is lacking. Furthermore, evidence to clarify the mechanism proposed by the traditional theory linking cold exposure to dyspepsia is insufficient.

Therefore, these three herbal formulas were comparatively evaluated for their effects on ameliorating CS-induced GI dysfunction in a mouse model. In addition, we focused on digestive enzymes and gut microbiome-related metabolites, such as SCFA and serotonin, to explore the corresponding mechanisms. These findings may provide valuable evidence to enhance the scientific and clinical use of traditional herbal formulas for treating cold-induced dyspepsia.

MATERIALS AND METHODS

Herbal formula extraction

The Korean Pharmacopoeia-grade herbs used in the herbal formulas ST, YT, and PS were obtained from the Dongguk University Ilsan Medical Center (Goyang, Gyeonggi-do, Republic of Korea). The detailed components of the three herbal formulations and the proportions of each herb are shown in Figure 1A. Ethanol was used to extract all three herbal formulations. Briefly, each 100 g of powder was ground and then individually extracted with 1 L of 30% ethanol (Merck, Rahway, NJ, United States) by heating at 100 °C for 60 minutes. After centrifugation at 3000 × g for 10 minutes, the supernatants were filtered using filter paper (grade 4, Whatman, Kent, United Kingdom). The filtered extracts were evaporated by a rotary evaporator (EYELA N-1200A, Tokyo, Japan) and lyophilized at -70 °C, obtained yields of 24.8% for ST, 35.1% for YT, and 36.1% for PS, respectively (w/w, Figure 1B). The powders of herbal extract were stored at -20 °C for future animal experiments.

Figure 1 Components of herbal formulas, extraction procedure, and high-performance liquid chromatography fingerprinting. A: Components of the three herbal formulas; B: Overview of the herbal formula extraction process and yields; C: High-performance liquid chromatography-based fingerprinting of the three herbal formulas along with their standard compounds. YT: Yijung-tang; ST: Shihosogan-tang; PS: Pyeongwi-san; EtOH: Ethyl alcohol.

High-performance liquid chromatography fingerprinting analysis of herbal formula extraction

The three herbal formulas were fingerprinted using two-dimensional high-performance liquid chromatography (HPLC). ST, YT, and PS powders were dissolved in methanol to prepare 5 mg/mL solutions. Prior to analysis, the prepared solutions were sonicated for 30 minutes and filtered through a 0.2-μm syringe filter (Cytiva, Marlborough, MA, United States). Quantitative analysis was performed using a Dionex Ultimate 3000 Ultra-HPLC system equipped with a diode array detector (Thermo Fisher Scientific, Waltham, MA, United States) and Luna™ C18 column (250 mm × 4.6 mm, 5.0 μm) (Phenomenex, Torrance, CA, United States). The wavelength was set to 254 nm. The column was maintained at 30 °C, and injection volume was 10 μL. Paeoniflorin and liquiritin were used as reference standards for the three herbal formulas according to the Korean Herbal Pharmacopoeia (12^th edition). Stock solutions of paeoniflorin or liquiritin were diluted to concentrations ranging from 500 to 31.25 μg/mL to establish calibration curves.

The mobile phase, composed of water containing 0.1% formic acid (solvent A) and acetonitrile with 0.1% formic acid (solvent B), was delivered at a flow rate of 1.0 mL/minute using step gradient elution (Supplementary Table 1) for ST and paeoniflorin (retention time: 17.66 minute). Additionally, a mobile phase consisting of solvents A and B was applied at a flow rate of 1.0 mL/minute using the step gradient method (Supplementary Table 2) for YT, PS, and liquiritin (retention time: 13.81 minute) (Figure 1C).

Animals and experimental design

The animal experimental protocol was approved by the Institutional Animal Ethics Committee of Dongguk University (permit No. IACUC-2024-03260; approval date: 8 April 2024) and complied with the United States National Institutes of Health (NIH) guidelines for the humane care and use of laboratory animals. All sacrifices were conducted using an equal combination of Zoletil (tiletamine–zolazepam, Virbac, Carros, France) and Rompun (xylazine hydrochloride, Bayer, Leverkusen, Germany) for anesthesia, minimizing suffering as much as possible.

To determine whether restraint stress or CS had a more significant effect on GI motility, 20 male C57BL/6J mice (6 weeks old, Daehan Biolink, Eumsung, Republic of Korea) were randomly assigned to five groups of four mice each. All mice, except those in the normal group, were subjected to cold restraint stress under varying conditions: 10 °C for 30 minutes, 10 °C for 1 hour, 20 °C for 2 hours, and 20 °C for 4 hours (Supplementary Figure 1A). After 30 minutes of cold restraint stress, GI transit was assessed using the charcoal test.

For the main experiment, 72 male C57BL/6J mice, 6 weeks old, were purchased from Daehan Biolink (Eumsung, Republic of Korea). After a 1-week acclimation in a room with a temperature of 22 °C ± 2 °C, humidity of 40% ± 10%, an equal light and dark cycle, chow diet, and free access to water, the mice were randomly divided into six groups [normal, CS, mosapride (MO), YT, ST, PS] with 12 animals in each group. Four mice were housed per cage, and food intake was measured once a week. The total amount consumed was then divided by the number of days and the number of mice to calculate the average daily intake per mouse. Except for the normal group, all mice were restrained in 50 mL Falcon conical tubes and immersed in 10 °C ± 2 °C water for 1 hour (Figure 2A). CS was induced six times, thrice a week for 2 weeks. During this period, the herbal formula extracts (10 mL/1000 mg/kg) and MO (10 mL/2 mg/kg, positive control) were administered intragastrically once daily for 2 weeks (Figure 2A). The same volume of distilled water was administered via intragastric gavage to both the normal and CS groups, consistent with the treatment protocol used for the other groups. Since the primary objective of this study is to compare the effects, a single dosage was selected to ensure a consistent basis for evaluation. The dose of the herbal formula in mice was converted from the clinical dose based on body surface area[23]. After the CS exposure, the mice were dried immediately using sterilized tissue paper to prevent hypothermia. At the final CS exposure, rectal temperatures were measured using a digital thermometer (Testo 925; Lenzkirch, Germany) before and after 30 minutes of cold restraint stress exposure. Mice were sacrificed under anesthesia using Zoletil (Virbac, Carros, France) and Rompun (Bayer, Leverkusen, Germany). Serum was collected from the inferior vena cava. Stomach and small intestine tissues were removed and stored in deep freezer (-80 °C), 4% paraformaldehyde or RNA later stabilization solution (Invitrogen, Carlsbad, CA, United States). The potential of hydrogen (pH) of the stomach contents was determined using a pH indicator (pH range: 3.4-6.4, Toyo Roshi Kaisha, Tokyo, Japan).

Figure 2 Experimental schedule and basic characteristics of mice. A: Experimental schedule and schematic of the cold stress model; B: Comparison of mean bodyweight on the final experimental day (n = 12/group); C: Comparison of average daily food intake; D: The difference in rectal temperature between the start and 30 minutes after cold stress (n = 8 or 12). YT: Yijung-tang; ST: Shihosogan-tang; PS: Pyeongwi-san; CS: Cold stress; MO: Mosapride.

Phenol red assay for gastric emptying evaluation

After 20 hours of fasting, mice were subjected to the last 1 hour of CS treatment, excluding the normal group. The mice were allowed to recover 30 minutes at room temperature, and then 400 μL of 0.05% phenol red dissolved in 1.5% carboxymethyl cellulose sodium (CMC) was orally gavaged as a nonabsorbable dye. After 30 minutes, the mice were sacrificed, and their stomachs were removed under anesthesia to evaluate the gastric emptying rate through gross findings, gastric weight, and measurement of the forestomach and stomach areas. The gastric emptying rate was represented as a percentage of the forestomach and stomach areas.

Charcoal assay for determining intestinal transit rate

Similar to the gastric emptying experiment, the mice were orally administered 200 μL of 5% charcoal dissolved in 0.5% CMC after the final CS exposure to assess the intestinal transit rate. After 30 minutes, the mice were sacrificed, and their stomachs and intestines were removed to measure the distance traveled by the charcoal. The intestinal transit rate was calculated using the following formula: Intestinal transit rate (%) = (charcoal distance/gut length) × 100%. Herein, gut length was defined as the distance from the stomach to the cecum.

Histopathologic observation by hematoxylin-eosin staining and immunofluorescent staining

Stomach and jejunal tissues fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned into 4-μm-thick slices for histomorphological examination. After drying, tissue sections were deparaffinized, rehydrated, stained with hematoxylin-eosin, dehydrated, and mounted using Clarion™ mounting medium. For immunofluorescent staining, paraffinized jejunum tissues were sectioned at 4 μm. Following deparaffinization, rehydration, heat-induced antigen retrieval, and blocking of non-specific binding with 2% goat serum, the tissue sections were incubated with a 1:100 dilution of serotonin antibody [Serotonin Antibody (YC5/45) conjugated to Alexa Fluor^® 594] and MUC2 antibody (996/1 conjugated to DyLight 488). Finally, the slides were counterstained with 4′,6-diamidino-2-phenylindole and mounted in an anti-fade medium. The slides were photographed and examined under an optical microscope (BX61; Olympus, Tokyo, Japan). The percentage fluorescence density was estimated using the ImageJ software (NIH, Bethesda, MD, United States).

Measurement of GI hormones by enzyme-linked immunosorbent assay

Small intestine tissue (100 mg of jejunum) was homogenized in 1 mL of radioimmunoprecipitation assay buffer at 4 °C. After centrifugation at 5000 × g for 5 minutes, 100 μL of the supernatant was used to determine the concentrations of glucagon-like peptide-1 (GLP-1), (catalog No. CSB-E08118m), cholecystokinin (CCK), (catalog No. CSB-E08115m), and 5-hydroxytryptamine (5-HT) (5-HT/serotonin, catalog No. MBS723181) according to the user manual provided with the enzyme-linked immunosorbent assay kits (CUSABIO, Wuhan, Hubei Province, China and MyBioSource, San Diego, CA, United States).

Fecal SCFA quantification

The filtrates were transferred into 1.5-mL screw cap vials containing 100 μL inserts in preparation for HPLC analysis. The concentrations (mM/g of mouse feces) of lactate, formate, acetate, propionate, and butyrate were quantified using a Shimadzu HPLC system (Shimadzu, Japan) equipped with an LC-20AD pump, SIL-20A autosampler, and an SPD-20A UV/VIS detector. Analytes were separated using an Aminex HPX-87H column (300 mm × 7.8 mm, 9 μm particle size, Bio-Rad Laboratories, Hercules, CA, United States) at a constant flow rate of 0.6 mL/minute, with the column maintained at 50 °C. The HPLC system was operated in isocratic mode using 0.005 M sulfuric acid as the mobile phase, with detection at 210 nm. The injection volume was 10 μL, and the elution time for each run was 40 minutes. Standards for SCFAs were prepared at concentrations of 0.1-200 mmol/L, and calibration curves were generated for quantification, achieving an R² value > 0.999.

Real-time polymerase chain reaction analysis

Total RNA was extracted from frozen mouse stomachs and small intestines using the TRIzol reagent (Invitrogen, Carlsbad, CA, United States). Complementary DNA was synthesized using AccuPower RT PreMix, which included M-MLV reverse transcriptase, reaction buffer, and RNase inhibitor (BIONEER, Daejeon, Republic of Korea). Real-time polymerase chain reaction (PCR) analysis was conducted on 10 related genes including glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as a housekeeping gene, following previously described protocols for PCR amplification. Detailed information and primer sequences are presented in Supplementary Table 3. Gene expression was analyzed using the LightCycler^® 96 system (Roche, Basel, Switzerland). Quantification of relative gene expression was represented by 2^–ΔCt calculations and then normalized by GAPDH.

Statistical analysis

All data are presented as mean ± SD, with sample sizes of either 4 or 8 per group, as indicated in the figure legends. Statistical comparisons among groups were performed using one-way analysis of variance to assess overall group differences. When significant differences were detected, post hoc analyses were conducted using the least significant difference test to determine pairwise comparisons. A P value of less than 0.05 was considered statistically significant. All statistical analyses were performed using SPSS Statistics version 29.0 (IBM Corp., Armonk, NY, United States).

RESULTS

Quality control of three herbal formula extractions

A calibration curve was established by correlating the peak areas with the analyte concentrations. For paeoniflorin, the calibration equation was Y = 22.643 × X + 0.0487 with R² > 0.999 and the limit of detection (LOD) and limit of quantitation (LOQ) were 2.66 μg/mL and 8.06 μg/mL, respectively. Paeoniflorin content was 0.87% (w/w) in ST (Figure 1C). Similarly, the calibration curve for liquiritin was Y = 68.0810 × X + 0.0160 with R² > 0.999, and the LOD and LOQ were 0.13 μg/mL and 0.39 μg/mL, respectively. Liquiritin content was 0.10% (w/w) in PS and 0.15% (w/w) in YT (Figure 1C).

CS exerted a dominant suppression of GI motility compared to restraint stress

The pre-test showed the comparison results of CS and restraint stress. None of the four cold restraint stress conditions affected body weight compared to normal conditions (P > 0.05) (Supplementary Figure 1B). Under the 10 °C condition, 1-hour water immersion significantly lowered rectal temperature compared to 30-minute water immersion (P < 0.01) (Supplementary Figure 1C). However, under the 20 °C condition, no significant difference in rectal temperature was observed between the 2-hour and 4-hour durations (P > 0.05).

By contrast, none of the four cold restraint stress conditions affected gut length compared to normal conditions (P > 0.05) (Supplementary Figure 1D and E). However, only the 10 °C condition and 1-hour exposure significantly reduced the charcoal distance and intestinal transit compared to normal conditions (P < 0.05) (Supplementary Figure 1D, F, and G). Therefore, the condition of 10 °C for 1 hour was used in the subsequent experimental studies.

Herbal formulas did not reduce appetite, body mass, and rectal temperature

After CS stimulation and herbal formula treatment, neither body weight nor food intake showed significant changes compared to that in normal conditions (P > 0.05) (Figure 2B and C). None of the herbal formulas exacerbated the CS-induced reduction in rectal temperature (P > 0.05) (Figure 2D). PS treatment slightly ameliorated the reduction in rectal temperature compared to CS treatment, rather than MO, ST, and YT treatments, although no significant differences were observed.

PS alleviated the CS-induced inhibition of gastric motility, surpassing YT and ST

Cold exposure significantly increased stomach weight, size, and area compared to normal conditions owing to phenol red retention (P < 0.01) (Figure 3A-D). By contrast, PS treatment notably reduced stomach weight, area, and size compared to CS treatment (P < 0.05) (Figure 3A-D). Consequently, CS stimulation reduced gastric emptying rate by 28.8% compared to normal conditions (P < 0.01); however, the PS-treated group exhibited the highest gastric emptying rate among all the groups (P < 0.05) (Figure 3E).

Figure 3 Gastric emptying activity of herbal formulas under cold stress. A: Morphology of the stomach after phenol red intragastric administration; B: Stomach weight; C: Stomach area; D: Forestomach area; E: Gastric emptying rate were compared; F: Paraffin-embedded stomach tissue stained with hematoxylin-eosin; G: Morphology of the gastric mucosa and hemorrhagic lesions (indicated by orange arrows); H: Comparison of gastric potential of hydrogen; I: Comparison of gene expression levels of gastric digestive enzymes. ^aP < 0.05, compared with the normal group; ^bP < 0.01, compared with the normal group; ^cP < 0.05, compared with the normal group; ^dP < 0.01, compared with the cold stress group (n = 4 or 8). YT: Yijung-tang; ST: Shihosogan-tang; PS: Pyeongwi-san; CS: Cold stress; MO: Mosapride; GAPDH: Glyceraldehyde 3-phosphate dehydrogenase; HE: Hematoxylin-eosin; pH: Potential of hydrogen.

Histopathological observations of the stomach revealed that cold exposure significantly reduced the number of parietal cells compared to the normal condition. However, treatment with any of the herbal formulas or MO notably increased the number of parietal cells compared to CS treatment (Figure 3F). Morphological examinations of the gastric mucosa also showed that cold exposure visibly induced severe hemorrhagic lesions in the gastric mucosa compared with the normal condition. Nevertheless, PS treatment effectively ameliorated gastric mucosal lesions, surpassing ST, YT, and MO (Figure 3G). Furthermore, cold exposure significantly elevated gastric pH compared to normal conditions (P < 0.05) (Figure 3H). Only the PS and ST treatments markedly reduced the gastric pH compared to CS treatment (P < 0.05), whereas the YT and MO treatments did not show any significant difference (Figure 3H).

PS enhanced the gene expression of gastric digestive enzymes, outperforming both ST and YT

Cold exposure drastically reduced the expression of the following three gastric digestive enzyme genes: PGC, PGA5, and LIPF, compared to normal conditions (P < 0.01; Figure 3I). PS treatment dramatically increased the expression of the aforementioned gastric digestive enzyme genes compared to CS treatment (P < 0.05, P < 0.01) (Figure 3I). PS treatment was more effective than ST and YT treatments at regulating the expression of gastric digestive enzyme genes (Figure 3I).

PS improved the intestinal motility, surpassing ST and YT

After cold exposure and drug treatment, gut length did not show notable changes compared to normal conditions (P > 0.05) (Figure 4A and B). Cold exposure significantly reduced charcoal distance and intestinal transit compared to normal conditions (P < 0.05) (Figure 4C and D). Both the PS and ST treatments, but not the MO and YT treatments, showed a marked recovery in charcoal distance and intestinal transit, similar to normal conditions (P < 0.05, P < 0.01) (Figure 4C and D).

Figure 4 Intestinal transit activity of herbal formulas under cold stress. A: Morphology of the stomach and intestine after intragastric administration of charcoal; B: Gut length; C: Charcoal transit distance; D: Intestinal transit rate was compared; E: Paraffin-embedded jejunum tissue stained with hematoxylin-eosin; F-H: Villus height (F), villus width (G), and crypt depth (H) were compared. ^aP < 0.05, compared with the normal group; ^bP < 0.01, compared with the normal group; ^cP < 0.05, compared with the normal group; ^dP < 0.01, compared with the cold stress group (n = 4 or 8). YT: Yijung-tang; ST: Shihosogan-tang; PS: Pyeongwi-san; CS: Cold stress; MO: Mosapride.

Intriguingly, neither cold exposure nor drug treatment significantly affected jejunal villus height (Figure 4E and F). Histopathological observations of the jejunum showed that cold exposure significantly increased the villus width and crypt depth of the jejunum compared to normal conditions (P < 0.01) (Figure 4G and H). However, PS or ST treatment notably reduced the villus width of the jejunum compared to CS treatment, which was superior to the YT and MO treatments (P < 0.05, P < 0.01) (Figure 4G and H).

PS promoted the secretion of intestinal serotonin, but not CCK

Double immunofluorescence staining showed that cold exposure reduced the fluorescence intensity of 5-HT (serotonin) and MUC2 antibodies in the jejunum compared to normal conditions (Figure 5A). By contrast, PS treatment significantly increased the fluorescence intensity of both antibodies compared to CS treatment, whereas MO treatment had no effect (Figure 5A). Interestingly, ST treatment increased the fluorescence intensity of the 5-HT antibody only, whereas YT treatment increased the fluorescence intensity of the MUC2 antibody only (Figure 5A).

Figure 5 Changes in intestinal serotonin, cholecystokinin, and bile acid receptors. A: Immunofluorescence staining of the jejunum for 5-hydroxytryptamine (5-HT) and MUC2; B: Comparison of 5-HT concentration in intestinal tissue; C and D: Comparison of gene expression levels of 5-HT receptors; E: Comparison of cholecystokinin concentration in intestinal tissue; F and G: Comparison of gene expression levels of bile acid receptors. ^aP < 0.05, compared with the normal group; ^bP < 0.01, compared with the normal group; ^cP < 0.05, compared with the normal group; ^dP < 0.01, compared with the cold stress group (n = 4 or 8). YT: Yijung-tang; ST: Shihosogan-tang; PS: Pyeongwi-san; CS: Cold stress; MO: Mosapride; 5-HT: 5-hydroxytryptamine; DAPI: 4’,6-diamidino-2-phenylindole; 5HT3R: 5-hydroxytryptamine type 3 receptor; 5HT4R: 5-hydroxytryptamine type 4 receptor; CCK: Cholecystokinin; FXR: Farnesoid X receptor; TGR5: Takeda G-protein-coupled receptor 5; GAPDH: Glyceraldehyde 3-phosphate dehydrogenase.

Consistent with the immunofluorescence findings, cold exposure significantly reduced 5-HT concentration in the intestinal tissue compared to normal conditions (P < 0.05) (Figure 5B). However, only PS treatment noticeably increased the intestinal 5-HT concentration (P < 0.01) (Figure 5B). Although cold exposures significantly upregulated the 5-hydroxytryptamine type 3 receptor (5HT3R) and 5-hydroxytryptamine type 4 receptor (5HT4R) gene expressions, only PS and ST treatments significantly downregulated 5HT3R expression (P < 0.05) (Figure 5C and D). However, drug treatments did not increase 5HT4R gene expression.

Cold exposure significantly reduced the intestinal CCK concentration compared to normal conditions (P < 0.01) (Figure 5E). However, although YT and MO treatments significantly enhanced the intestinal CCK concentration (P < 0.05), neither PS nor ST treatment caused a significant increase compared to CS treatment (P > 0.05) (Figure 5E).

PS suppressed the gene expression of bile acid receptors, similar to YT and ST

Farnesoid X receptor (FXR) and Takeda G-protein-coupled receptor 5 (TGR5) are the nuclear and membrane bile acid receptors, respectively. Cold exposure markedly upregulated the expression of intestinal FXR and TGR5 compared to normal conditions (P < 0.05) (Figure 5F and G). All herbal formulas, as well as MO, drastically downregulated FXR and TGR5 expressions compared to CS (P < 0.05 and P < 0.01) (Figure 5F and G).

PS restored the fecal SCFA, surpassing YT and ST

Cold exposure significantly reduced fecal propionic acid and butyric acid levels, but not acetic acid levels, compared to normal conditions (P < 0.05, P < 0.01) (Figure 6A). PS treatment dramatically increased fecal acetic acid, propionic acid, and butyric acid levels compared to CS treatment (P < 0.05 or P < 0.01) (Figure 6A). However, YT and ST treatments showed a relatively modest increase in fecal propionic acid levels compared to PS treatment (P < 0.05) (Figure 6A).

Figure 6 Alterations in short-chain fatty acids, and related mechanisms. A: Fecal levels of short-chain fatty acids (SCFAs), including acetic acid, propionic acid, and butyric acid, were measured by high-performance liquid chromatography; B and C: Messenger RNA (mRNA) expression of SCFA receptors G protein-coupled receptor 41 (B) and G protein-coupled receptor 43 (C) was determined by quantitative polymerase chain reaction; D and E: Intestinal glucagon-like peptide-1 (GLP-1) levels (D), GLP-1 mRNA expression (E) were compared among groups; F and G: Gene expression of inflammatory cytokines tumor necrosis factor-α (F) and interleukin-1β (G) was assessed. ^aP < 0.05, compared with the normal group; ^bP < 0.01, compared with the normal group; ^cP < 0.05, compared with the normal group; ^dP < 0.01, compared with the cold stress group (n = 4 or 8). YT: Yijung-tang; ST: Shihosogan-tang; PS: Pyeongwi-san; CS: Cold stress; MO: Mosapride; GAPDH: Glyceraldehyde 3-phosphate dehydrogenase; GPR41: G protein-coupled receptor 41; GPR43: G protein-coupled receptor 43; GLP-1: Glucagon-like peptide-1; TNF-α: Tumor necrosis factor-alpha; IL-1β: Interleukin-1 beta.

By contrast, cold exposure markedly upregulated the expression of G protein-coupled receptor 41 (GPR41) (P < 0.05) (Figure 6B), but not G protein-coupled receptor 43 (GPR43) (Figure 6C). PS and ST treatments notably downregulated GPR41 expression, whereas YT and MO treatments did not (P < 0.05) (Figure 6B). However, none of these drugs affected GPR43 expression in our study.

PS and ST suppressed the GI hormone GLP-1, surpassing YT and MO

Cold exposure markedly elevated intestinal GLP-1 Levels compared to normal conditions (P < 0.05) (Figure 6D). Additionally, cold exposure significantly upregulated the GLP-1 gene expression in the intestinal tissue compared to normal conditions (P < 0.05) (Figure 6E). All drug treatments notably reduced GLP-1 Levels compared to CS treatment (P < 0.05) (Figure 6D). A similar pattern was observed for intestinal GLP-1 expression, although no significant changes were observed with MO and YT treatments.

Regarding intestinal inflammation, cold exposure notably upregulated the tumor necrosis factor-alpha (TNF-α) and interleukin-1 beta (IL-1β) gene expression levels in intestinal tissue compared to normal conditions (P < 0.05 or P < 0.01) (Figure 6F and G). However, MO treatment more effectively inhibited TNF-α and IL-1β gene expressions than herbal formula treatments. Among the three herbal formulas, ST exhibited the most potent inhibition of the inflammatory genes.

DISCUSSION

In traditional medical theory, cold is considered a critical pathological factor that directly contributes to GI disorders[15]. Cold restraint stress has been frequently used in animal models to investigate GI dysfunction because of its simplicity[24]. Numerous studies have demonstrated that both cold-water immersion and restraint stress can result in GI dysfunction, often leading to conditions such as dyspepsia[25,26]. A previous animal study has proved that gut motility in mice is temperature-dependent[27]. However, the dominant effects of cold and restraint stress on GI dysfunction remain unknown. Therefore, in preliminary experiments, we compared the relative sensitivity of gut motility to variations in temperature or restraint duration. We observed that 1-hour cold water-immersion apparently induced more sensitivity-induced intestinal transit than 4-hour restraint stress in normal-temperature water immersion (Supplementary Figure 1). Thus, CS was more effective than restraint stress in inhibiting intestinal motility.

Although cold exposure often leads to increased food intake and body weight as a compensatory mechanism to maintain core temperature and energy balance[28], our studies did not observe any evident changes in body mass, body temperature decline, and food intake following intermittent cold exposure (1 hour, 10 °C ± 2 °C, six times over 2 weeks) or co-treatment with any herbal formulas. Thus, alterations in basic metabolism and appetite can be ruled out as contributing mechanisms in the current study.

Previous studies have indicated that CS can delay gastric emptying[29], with one human study reporting a dramatic decrease in total gastric emptying from 17% to 2% during CS[6]. To compare the effects of the herbal formulas on improving CS-induced decline in gastric motility, we evaluated the gastric emptying rate by gavage with phenol red dye. As expected, cold exposure clearly delayed stomach emptying in our study, as evidenced by enlargement of the stomach area and mass. After treatment with the herbal formulas, PS demonstrated the most significant enhancement in stomach motility compared to the other formulas and the positive control drug, MO. In general, short-term cold exposure stimulates gastric acid secretion via central gamma-aminobutyric acid mechanisms and activation of the vagus nerve[6,30]. Intriguingly, in the current study, we discovered that long-term intermittent cold exposure slows gastric acid secretion, as evidenced by an increase in the pH of the gastric contents and partial loss of gastric parietal cells. Severe hemorrhagic lesions in the gastric mucosa were also observed in the cold exposure group. Many crucial gastric enzymes, such as gastric pepsin and lipase, function efficiently in acidic environments[31,32]. Thus, PS treatment maintains a gastric acidic environment, contributing to the activity of these gastric enzymes. In addition, the upregulation of PGC, PGA5, and LIPF gene expression in the stomach tissues suggests that PS may enhance the production of pepsin and gastric lipase in the stomach, which are involved in digestion and gastric motility. Overall, PS treatment ameliorated intermittent CS-induced gastric motility impairment and gastric mucosal injury, outperforming YT and ST treatments. A potential mechanism underlying this improvement in gastric motility is PS treatment effectively preventing the decline in gastric acid secretion, thereby enhancing the quantity and activity of gastric digestive enzymes.

By contrast, the charcoal test is a straightforward and reliable method for assessing changes in intestinal motility, particularly in rodent models[33]. Our charcoal test results clearly demonstrate that cold exposure impairs intestinal motility. Both PS and ST effectively ameliorated this dysfunction, whereas YT had no effect. Moreover, intestinal histopathological findings showed that cold exposure predominantly induced intestinal villus width and crypt depth enlargement rather than length extension. Cold exposure generally reduced blood flow to peripheral tissues, including the intestines, thereby limiting the oxygen and nutrients needed for energy-intensive processes, such as villus elongation[34]. Meanwhile, increased crypt depth and villus width may represent a more adaptive response to CS, aimed at maintaining the structural integrity of the gut rather than solely enhancing nutrient absorption[35]. MUC2, the main mucin glycoprotein produced by intestinal epithelial cells, is mainly present in the mucus layer and is essential for the formation and protection of the intestinal barrier[36]. The decrease in the intestinal MUC2 levels supports the finding that cold exposure impairs the intestinal barrier. Consequently, the pathological findings suggest that the intestine may undergo compensatory epithelial remodeling following cold exposure. The increases in villus width and crypt depth may be part of a remodeling process that strengthens the intestine’s ability to adapt to CS. Thus, the inhibition of villus width and crypt depth expansion implied that PS and ST effectively alleviated CS-induced damage to intestinal motility.

Serotonin is an essential neurotransmitter, with approximately 90%-95% located in the gut, where it is principally produced by enterochromaffin cells and can be regulated by tryptophan metabolism through gut microbiota[37]. Peripheral serotonin plays a key role in the regulation of gut motility by acting as a signaling molecule that triggers peristaltic and secretory reflexes[38]. Several studies have shown that CS reduces serotonin levels and alters the structure of gut microbiota[35,38,39]. Our study also confirmed a reduction in intestinal serotonin levels following cold exposure. Interestingly, the gene expression of the two main serotonin receptors was simultaneously upregulated by CS. This inconsistency suggests a compensatory mechanism in the response to CS. Only PS treatment enhanced intestinal serotonin levels and reduced intestinal 5HT3R gene expression. Therefore, the direct or indirect reduction of intestinal serotonin may be one of the mechanisms by which PS treatment improves intestinal motility.

Many studies have reported that alterations in the gut microbiota are closely associated with the production of SCFAs and changes in the bile acid profile[40,41]. SCFAs are crucial for gut health, serving as an energy source for enterocytes and promoting gut motility by modulating intestinal muscle contraction, fluid absorption, and gut barrier function[42]. They also help regulate inflammation and support the overall health of the gut microbiome, thereby maintaining a balanced and functional digestive system[43]. Consistent with previous results, cold exposure visibly reduced SCFA production in the gut, particularly in the presence of butyric and propionic acids[44]. Meanwhile, GPR41 and GPR43, which are G protein-coupled receptors that are activated by SCFAs, are involved in immune and inflammatory responses[45]. In our study, low SCFA levels led to the compensatory upregulation of GPR41 and GPR43, thereby amplifying the inflammation process. Consequently, elevation of SCFA production and reduction of GPR41 and TNF-α levels present another potential mechanism illuminating how PS treatment improves intestinal function.

CS can upregulate the bile acid synthesis enzyme CYP7B1, increasing the liver and fecal bile acids[46]. In addition, both FXR (a nuclear receptor for bile acids) and TGR5 (a membrane receptor for bile acids) are important regulators of intestinal motility[47]. FXR primarily influences gut homeostasis and inflammation, whereas TGR5 directly promotes intestinal motility through bile acid signaling and the release of gut hormones, such as CCK and GLP-1[48,49]. Although bile acid levels were not determined in the present study, the gene expression of intestinal FXR and TGR5 suggested that the downregulation of bile acid receptors by herbal formulas is a plausible mechanism for regulating intestinal motility.

Finally, this study has several limitations that should be mentioned. First, only male mice were used in the experiments, which may limit the generalizability of the findings to both sexes, considering the known influence of sex hormones on metabolic and neurochemical responses. Second, the sample size was relatively small. Third, only a single dose of each herbal formula was tested, which does not allow for the evaluation of dose-dependent effects. Future studies should include both sexes, larger sample sizes, and multiple dosage levels to validate and expand upon the current findings.

CONCLUSION

PS treatment was significantly more effective than either YT or ST treatment in alleviating GI dysfunction caused by intermittent CS. The observed benefits of PS may be closely associated with its ability to modulate key physiological pathways including enhancement of serotonin and SCFA levels, reduction of GLP-1 activity, and downregulation of intestinal bile acid receptors. This molecular and metabolic regulation may play a critical role in mitigating the adverse effects of CS on GI function. Furthermore, this comparative analysis not only highlights the distinct therapeutic advantages of PS over alternative treatments, but also elucidates the underlying mechanisms contributing to its efficacy. Such insights are particularly valuable for developing targeted clinical interventions for cold-related digestive disorders and offer a foundation for future research and potential translational applications in therapeutic practice.

ACKNOWLEDGEMENTS

We sincerely thank Dr. Hwang SJ from the Institute of Bioscience and Integrative Medicine, Daejeon University, for his invaluable technical support in the evaluation of GI motility.