Chaihu-Shugan-San ameliorates chronic atrophic gastritis by inhibiting nuclear factor-kappa B-mediated inflammation and apoptosis

Abstract

BACKGROUND

Chaihu-Shugan-San (CSS), a classic traditional Chinese medicine formula, has demonstrated significant efficacy in treating various gastrointestinal disorders.

AIM

To explore the therapeutic efficacy of CSS in alleviating chronic atrophic gastritis (CAG), and elucidate the underlying mechanisms of action.

METHODS

High performance liquid chromatography-mass spectrometry was used to identify the main active components of CSS. The therapeutic effects of CSS at doses of 925 mg/kg/day and 1850 mg/kg/day on N-methyl-N’-nitro-N-nitrosoguanidine (MNNG)-induced CAG were evaluated. Network pharmacology and molecular docking were used to predict the potential targets of CSS in CAG. The impact of CSS on the gut microbiota of rats was investigated by 16S rRNA sequencing.

RESULTS

The main active components of CSS were lipids and lipid-like molecules, phenylpropanoids and polyketides. In vivo experiments showed that CSS significantly ameliorated MNNG-induced CAG by inhibiting inflammation and apoptosis. The core target of CSS to alleviate CAG were tumor necrosis factor, interleukin (IL)-1β, IL-6, BAX, BCL2, caspase-3/caspase-9, and NFKBIA. Gene Ontology analysis of these core targets revealed their predominant association with the nuclear factor-kappa B (NF-κB) signaling complex and BAX apoptotic complex. Molecular docking demonstrated that six compounds in CSS, including baicalin, licoisoflavone B, licochalcone B, glabrone, glycyrrhiza flavonol A, and marmin exhibited strong binding affinities with NFKBIA. 16S rRNA sequencing indicated that CSS promoted beneficial changes in the colonic microbial community.

CONCLUSION

CSS alleviated CAG by inhibiting NF-κB-mediated inflammation and apoptosis, providing insights into its mechanism of action in protection against CAG.

Key Words: Chronic atrophic gastritis; Chaihu-Shugan-San; Inflammation; Apoptosis; Nuclear factor-kappa B

Core Tip: Our study demonstrated that Chaihu-Shugan-San (CSS) restored body weight, mitigated gastric tissue histology, and alleviated intestinal metaplasia in rats with chronic atrophic gastritis, reduced levels of inflammatory cytokines, and promoted antiapoptotic pathways. CSS was shown to downregulate the nuclear factor-kappa B signaling proteins, providing insights into its mechanism of action in protection against chronic atrophic gastritis development.

INTRODUCTION

Chronic atrophic gastritis (CAG) is a persistent gastrointestinal disorder characterized by the atrophy of the gastric mucosal epithelium and glands, leading to reduced quantity and function, thinning of the gastric mucosa, thickening of the mucosal basement layer, metaplasia of the pyloric and intestinal glands, or atypical hyperplasia[1,2]. It is considered a multifactorial disease or a precancerous condition, significantly contributing to the onset and progression of gastric cancer[3]. Timely management of CAG is crucial in preventing the development of gastric cancer[4]. Current medical practices for CAG typically involve nonspecific interventions, such as Helicobacter pylori eradication[5] and the use of mucosal protective agents[6], proton pump inhibitors, and other acid-suppressive medications[7,8]. Nevertheless, the condition often recurs and poses challenges for complete resolution. This not only impacts human health significantly but also imposes a substantial burden on healthcare systems, necessitating urgent exploration for novel and efficacious therapeutic interventions.

In accordance with the principles of traditional Chinese medicine (TCM), CAG can be categorized as gastric distention, discomfort in the upper abdomen, gastric irritation, and eructation[9]. The therapeutic strategies for CAG should place priority on the nourishment and tonification of the spleen, regulation of the liver, and harmonization of the stomach[10]. TCM interventions for CAG have shown beneficial effects in ameliorating cellular damage, inhibiting inflammatory response, boosting immune function, and augmenting antioxidant capacity[11-13]. Chaihu-Shugan-San (CSS), as described in the medical classic Jing Yue Quan Shu from the Ming Dynasty, is a TCM formula used for alleviating the manifestations associated with liver Qi stagnation. CSS is formulated from seven traditional Chinese herbal medicines, including Radix Bupleuri, Aurantii Fructus, Citrus Reticulata, Chuanxiong Rhizoma, Cyperi Rhizoma, Glycyrrhizae Radix et Rhizoma (licorice), and Paeoniae Radix Alba. This formula has been utilized in clinical and preclinical contexts to treat conditions such as depression and liver disorders, showing a positive safety profile and effectiveness[14-17]. Recent studies have shown that it also exerts protective effects against various gastrointestinal diseases. Specifically, CSS has been noted to improve gastric motility in rats with functional dyspepsia[18,19]. A meta-analysis of clinical trials has highlighted the therapeutic benefits of CSS in the management of reflux esophagitis[20]. However, further research is needed to clarify its precise treatment of CAG and the underlying mechanisms involved.

Network pharmacology involves identifying common targets shared by drugs and diseases to understand how drugs work therapeutically[21]. TCM compounds have complex compositions and extensive targets that are interconnected, forming a network system that regulates disease development[22,23]. Using network pharmacology to study the targets of TCM compounds reveals their mechanisms of action through biological regulatory networks, aligning with the overall regulatory theory and development trends of TCM[24]. This study analyzed the targets of treating CAG with CSS using network pharmacology and confirmed the therapeutic effects and mechanisms of CSS with in vivo investigations.

MATERIALS AND METHODS

Ultra-performance liquid chromatography-mass spectrometry

CSS granules were purchased from Beijing Tongrentang Chinese Medicine Co. Ltd. (Beijing, China). CSS granules (100 mg) were weighed and dissolved in 1 mL extraction solution (water-acetonitrile-isopropyl alcohol mixed solution, 1:1:1, v/v/v). The mixture was homogenized for 1 minute and extracted for 30 minutes by low temperature ultrasound. After that, it was centrifuged for 10 minutes at 12000 rpm at 4 °C and allowed to stand for 1 hour at -20 °C to precipitate protein. A further centrifugation step was carried out at 12000 rpm for 10 minutes at 4 °C. The supernatant was dried under vacuum. Thereafter, 0.2 mL 50% acetonitrile solution was added to the residue, which was homogenized. The homogeneous mixture was centrifuged at 14000 rpm for 15 minutes at 4 °C. The supernatant was carefully collected and used for ultra-performance liquid chromatography-mass spectrometry (UPLC-MS) analysis.

The instruments used included a Vanquish UPLC (Thermo, Waltham, MA, United States) and a Q Exactive HFX Hybrid Quadrupole Orbitrap MS (Thermo). Chromatographic separation was performed on a Waters HSS T3 (100 mm × 2.1 mm, 1.8 μm) at a low flow rate of 0.3 mL/minute. The column temperature was maintained at 40 °C and the injection volume was 2 μL each time. Mobile phase A was Milli-Q water (0.1% formic acid) and mobile phase B was acetonitrile (0.1% formic acid). The gradient elution conditions were set as follows: 0 minute, 100% solution A; 1 minute, 100% solution A; 12 minutes, 5% solution A; 13 minutes, 5% solution A; 13.1 minutes, 100% solution A; and 17 minutes, 100% solution A. The specific parameters of electrospray ionization source were set as follows: Sheath gas pressure, 40 arb; Auxiliary gas pressure, 10 arb; Spray voltage, + 3000 v/- 2800 v; Temperature, 350 °C; And ion transport tube temperature, 320 °C. The scanning mode was Full-ms-ddMS2 in positive and negative ion modes. The scanning range of the primary MS was (scan m/z range) 70-1050 Da, with a primary resolution of 70000 and secondary resolution of 17500. The data were analyzed with Progenesis QI software (Waters Corporation, Milford, MA, United States), and referred to the secondary MS database for Chinese herbal medicine built by San Shu Biotechnology for preliminary confirmation of components in CSS granules.

Animals and treatment

Male Sprague-Dawley rats (150-170 g) were adapted for at least 1 week before experiments and maintained in the specific pathogen-free conditions of temperature (22 °C ± 1 °C), relative humidity (50% ± 5%), alternating lighting (12 hours light: 12 hours dark cycle), and free access to sufficient food and water. The animals were randomly divided into four groups: (1) Control; (2) CAG model; (3) CAG model + CSS low dose (925 mg/kg); and (4) CAG model + CSS high dose (1850 mg/kg). The two doses of CSS (925 mg/kg/day and 1850 mg/kg/day) were selected based on clinical equivalent doses and previously published studies[25]. The control group had free access to drinking water and food. The CAG model was established by combining N-methyl-N’-nitro-N-nitrosoguanidine (MNNG) (200 mg/L) administration with an irregular diet regimen. The irregular diet involved alternating periods of feeding and fasting as follows: Food deprivation for 24 hours, followed by adlibitum feeding for 24 hours, then another 24 hours of food deprivation, followed by adlibitum feeding for 24 hours. This cycle was repeated throughout the 16-week modeling period. The rats were provided with MNNG for 10 weeks for model generation. After the model was established, the corresponding drugs were given by gavage for 6 weeks once daily. The body weight was recorded weekly. After the last administration, rats were killed. Also, blood samples and stomach tissues were collected. All animal experiments complied with ARRIVE guidelines. This study was approved by the Beijing Institute of Traditional Chinese Medicine Institutional Animal Care and Use Committee. with the ethical approval number No. BJTCM-R-2025-01-05.

Histopathological examinations

Stomach tissue samples were fixed and preserved in 4% paraformaldehyde. The samples were dehydrated in ethanol and xylene. After dehydration, the tissue samples were embedded in paraffin wax, and 5-μm thick serial sections were obtained by microtome for hematoxylin-eosin (HE) and alcian blue (AB) staining.

HE staining

HE staining was performed to visualize the morphological changes in the gastric mucosa. Five fields of gastric mucosa were taken from each section and the thickness of the glands in each field was measured with a micrometer. The average value of the five fields was calculated. The total number of gastric mucosa glands at 0.2-1.2 mm from the pyloric ring was counted and expressed as “unit/mm” at 100 × magnification. The mean numbers of gastric chief cells and parietal cells of the five fields in the gastric mucosa were counted from each section, starting 1.0 mm below the junction of the gastric antrum and the gastric body at 400 × magnification. After the whole gastric mucosa was observed under low magnification, 10 fields of view in the gastric antrum were selected to determine the presence or absence of inflammation. The degree of inflammatory cell infiltration was categorized into 7 grades from 0 to 3: (1) Grade 0: No inflammation; (2) Grade 0.5: Inflammation between 0 to 1 observed; (3) Grade 1: Multiple chronic inflammatory cell infiltrations seen in the pit of the stomach or the bottom of the inherent gland; (4) Grade 1.5: Inflammation between grade 1 to grade 2 observed; (5) Grade 2: More inflammatory cells in the gastric mucosa from the fovea to the myometrium; (6) Grade 2.5: Inflammation between grade 2 to grade 3 observed; and (7) Grade 3: Numerous inflammatory cells seen in the gastric mucosa.

AB staining

The sections were stained with AB to detect the types of intestinal metaplasia. Five photos of gastric mucosa were randomly obtained from each section using an Olympus BX53 microfilming system at a magnification of 400 ×. Each group was imaged under the same conditions. An Olympus DP73 image analysis system was used. The ratio of positively stained area to the total analyzed area was calculated for each photograph under the same conditions. The average ratio of five photographs per rat was taken and used for statistical comparisons among experimental groups.

Network pharmacology analysis

The diseases associated with seven herbs in CSS were downloaded from the HERB database (http://herb.ac.cn), and noncancerous digestive diseases and digestive cancers were selected. The chemical components of seven herbs in the CSS satisfying both oral bioavailability > 30% and drug-like properties > 0.18 were selected as candidate active ingredients according to the TCM Systems Pharmacology database and analysis platform (TCMSP) (https://old.tcmsp-e.com/). The potential targets in CSS using the TCMSP platform were identified. The keyword “chronic atrophic gastritis” was used to retrieve disease-related therapeutic targets from the GeneCards database (https://www.genecards.org/). Following elimination of duplicate target genes, the shared targets between the disease and drug were identified through systematic screening. The “drug and disease” intersection genes obtained by VEEN were imported into the STRING11.0 database for the generation of protein-protein interaction (PPI) analysis network diagram. The associated target genes were enriched into the Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways by DAVID Bioinformatics Resources (https://david.ncifcrf.gov/). Finally, the active ingredient-target-signaling pathway network was mapped.

Enzyme-linked immunosorbent assay

Serum gastrin level was detected using enzyme-linked immunosorbent assay (ELISA) kits (Fine Biotech, Wuhan, Hubei Province, China). Levels of tumor necrosis factor (TNF)-α, interleukin (IL)-1β, and IL-6 were detected using ELISA kits (Multisciences Biotech, Hangzhou, Zhejiang Province, China).

Western blotting

The total protein in the treated gastric mucosa was extracted with the radio immunoprecipitation assay lysate (Solarbio, Beijing, China) containing 10 μL phenylmethylsulfonyl fluoride (Solarbio), and the bicinchoninic acid protein assay kit (Solarbio) was used for quantification. Protein bands were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (Solarbio) and transferred to polyvinylidene fluoride membranes (Millipore, Billerica, MA, United States). The membranes were sealed in tris-buffered saline-tween-20 for 5 minutes at room temperature. The membranes containing the protein were incubated with anti-BAX, anti-BCL2, anti-phospho (p)-inhibitor of nuclear factor-kappa B (NF-κB) (IκB)-α Ser32/36, anti-IκBα, anti-p-p65Ser536, anti-p65, and anti-glyceraldehyde-3-phosphate dehydrogenase antibodies overnight at 4 °C. The membranes were incubated with horseradish-peroxidase-conjugated anti-mouse or anti-rabbit antibodies for 1 hour at 37 °C. Protein load was monitored by using an enhanced chemical luminescence reagent (Solarbio). The acquired image was analyzed with Tanon Image (Tanon, Shanghai, China). The catalog numbers and dilutions of all primary antibodies used in the Western blotting assay are provided in the Table 1.

Table 1 Antibodies.

Name	No.	Dilution ratio	Brands
BAX	R380709	1:500	Zenbio, China
BCL2	250413	1:500	Zenbio, China
p-IκBαSer32/36	340776	1:500	Zenbio, China
IκBα	R380682	1:500	Zenbio, China
p-p65Ser536	310013	1:500	Zenbio, China
p65	250060	1:500	Zenbio, China
GAPDH	60004-1-Ig	1:10000	Proteintech, China

p-IκB: Phospho-inhibitor of nuclear factor-kappa B; IκB: Inhibitor of nuclear factor-kappa B; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase.

Caspase-3/9 activity assay

The gastric mucosal tissue was lysed and homogenized on an ice bath with a glass homogenizer. The homogenate was transferred to a 1.5-mL centrifuge tube, lysed for a further 5 minutes on an ice bath, and centrifuged at 16000 g for 15 minutes at 4 °C, and the supernatant was extracted for protein content determination. The activities of caspase-3 and caspase-9 were determined (Beyotime, Shanghai, China).

Molecular docking

The three-dimensional protein structures of the genes of interest were obtained as pdb format files from the Protein Data Bank database (https://www.rcsb.org; last accessed on 3 March 2025). The structures of the core CSS chemical compounds as ligands were downloaded from the PubChem database (https://pubchem.ncbi.nlm.nih.gov; last accessed on 3 March 2025) in sdf format. Molecular docking was performed using the online CB-Dock2 database (https://cadd.labshare.cn/cb-dock2/) from the Yang Cao laboratory to visualize the results, as described previously (Liu et al[26]).

Immunofluorescence staining of NF-κB

The 5-μm sections were incubated with primary antibody of p-p65Ser536 (1:50 dilution) overnight at 4 °C, washed with phosphate-buffered saline (PBS), and incubated with secondary antibody of CY3 goat anti-rabbit IgG (1:200 dilution), and rewashed with PBS. Finally, 4’,6-diamidino-2-phenylindole was used for nuclear staining to detect the location of p-p65Ser536. Images of stained specimens were captured by scanning microscopy.

16S rDNA sequencing

Total DNA from rat fecal samples was extracted. Polymerase chain reaction (PCR) amplification was carried out on the V4 region of the bacterial 16S rDNA gene with the following barcode: 515 F-806 R. All PCRs were carried out with Phusion^® High-Fidelity PCR Master Mix (15 μL, New England Biolabs, Ipswich, MA, United States), primers (0.2 μM), and DNA template (10 ng). During thermal cycling, initial denaturation was performed at 98 °C for 1 minute, followed by 30 cycles at 98 °C (10 seconds), 50 °C (30 seconds), and 72 °C (30 seconds), and finally at 72 °C for 5 minutes. After the PCR products were purified, the sequencing libraries were prepared. The libraries that passed the quality check were subjected to paired-end sequencing.

Statistical analysis

All experiments were performed at least three times. Experimental values were given as mean ± SD. Statistical analysis was performed using GraphPad Prism 9. One-way analysis of variance was applied to analyze differences among the different groups, followed by Tukey’s significant post hoc test for pair-wise multiple comparisons. P < 0.05 was considered to be statistically significant.

RESULTS

Identification of the major chemical constituents in CSS granules

UPLC-MS was utilized to characterize the chemical constituents of CSS granules. The mass spectra of CSS positive and negative ion modes are presented in Figure 1A. UPLC-MS demonstrated that the identified compounds encompassed diverse chemical classes, predominantly comprising lipids and lipid-like molecules, phenylpropanoids and polyketides, organic oxygen compounds, organoheterocyclic compounds, benzenoids, as well as organic acids and derivatives (Figure 1B). Subsequent investigation focused on the two most abundant compound categories lipids and lipid-like molecules, along with phenylpropanoids and polyketides. The analysis revealed that prenol lipids, fatty acyls, and steroids/steroid derivatives constituted the predominant subclasses within lipids and lipid-like molecules. In addition, flavonoids, coumarins and their derivatives, and isoflavonoids represented the most abundant subclasses among phenylpropanoids and polyketides (Figure 1C). KEGG pathway analysis indicated that flavonoids derived from Radix Bupleuri, Cyperi Rhizoma, and licorice in CSS primarily modulate NF-κB/activator protein-1-mediated apoptosis, and digestion/absorption-related pathways (Figure 1D). Several herbal components of CSS (Radix Bupleuri, Cyperi Rhizoma, Citrus Reticulata, Paeoniae Radix Alba, Chuanxiong Rhizoma, and Aurantii Fructus) exhibited significant associations with various digestive system disorders, including noncancerous digestive diseases and digestive cancers (Figure 1E), with particularly strong correlations with CAG (Figure 1F).

Figure 1 The major chemical constituents in Chaihu-Shugan-San granules and their association with chronic atrophic gastritis. A: Mass spectra in positive and negative ion modes; B: Top six categories of active components in Chaihu-Shugan-San (CSS); C: Top three subclasses of lipids and lipid-like molecules, and phenylpropanoids and polyketides; D: Kyoto Encyclopedia of Genes and Genomes analysis of prenol lipids, and flavonoids; E: Number of digestive system diseases associated with the herbs in CSS; F: Relationship between the herbs in CSS and chronic atrophic gastritis. RT: Retention time; NF-κB: Nuclear factor-kappa B; AP-1: Activator protein-1; CAG: Chronic atrophic gastritis; CSS: Chaihu-Shugan-San.

CSS alleviated weight loss and gastric injury in rats with CAGs

To further investigate the therapeutic effects of CSS on CAG, CSS treatment was administered to rats with MNNG-induced CAG (Figure 2A). At week 10, the mean body weight of rats with CAG prior to CSS treatment was 250.8 ± 20.7 g. This was significantly lower than that of the rats in the control group, which had a mean body weight of 454.8 ± 17.9 g (Figure 2B) (P < 0.001). Upon completion of the CSS treatment, the mean body weights of rats treated with a high dose of CSS (394.7 ± 20.2 g, P < 0.05) for 6 weeks was higher than that of the CAG model group (323.5 ± 29.9 g). However, treatment with a low dose of CSS (371.2 ± 45.2 g, P > 0.05) did not result in a significant change in body weight in CAG rats (Figure 2B). MNNG-induced CAG rats showed decreased serum gastrin levels, while high-dose CSS treatment elevated gastrin levels (Figure 2C). Histopathological examination of the gastric tissue revealed that the color of the gastric tissue in the CAG model group was visibly paler and the gastric folds were lighter in comparison to those in the control group. Treatment with CSS led to a mitigation in gastric lesions (Figure 2D).

Figure 2 Effects of Chaihu-Shugan-San on body weight and gastric injury in N-methyl-N’-nitro-N-nitrosoguanidine-treated rats. A: Schematic diagram; B: Body weight at different time points; C: Levels of gastrin in the serum of rats; D: Representative morphology of the stomach of rats. Upper, whole stomach. Lower, gastric mucosa after dissection. Data are shown as mean ± SD (n = 6). Differences were analyzed by one-way analysis of variance followed by Tukey’s post-hoc test. ^bP < 0.01. ^cP < 0.001. P values vs chronic atrophic gastritis model group. NS: No significance; H₂O: Water; MNNG: N-methyl-N’-nitro-N-nitrosoguanidine; CAG: Chronic atrophic gastritis; CSS: Chaihu-Shugan-San; Con: Control.

CSS mitigated gastric mucosal lesions in rats with CAG

HE staining revealed that the gastric mucosal glands in the control group were thick and abundant, with a large number of epithelial cells neatly arranged. In the CAG model group, the number of glands decreased, and their thickness was reduced. Inflammatory cell infiltration in the interstitium and mucosal congestion of the lamina propria were observed, with cells varying in size and shape and being disordered in arrangement. In MNNG-induced rats treated with low- and high-dose CSS, the pathological changes were attenuated, with remission in atrophy compared with the CAG model group (Figure 3A). The levels of gastric mucosal inflammation in the low-dose group (2.00 ± 0.34, P < 0.05) and high-dose group (1.03 ± 0.19, P < 0.001) were markedly lower than those in the CAG model group (2.49 ± 0.37) (Figure 3B). Following 6 weeks treatment with CSS, a notable reduction in both the quantity (25.7 ± 4.2 units/mm, P < 0.001) and thickness (205.1 ± 28.1 μm, P < 0.01) of the glands within the gastric mucosa of the CAG model group was detected when compared to the control group (Figure 3C). In contrast, the high-dose treatment group exhibited significantly elevated levels of glandular quantity (37 ± 4.5 units/mm, P < 0.01) and thickness (271 ± 40.9 μm, P < 0.05) in gastric tissue, with statistical significance relative to the CAG model group (Figure 3C). However, the low-dose treatment group did not display a significant increase in the number and thickness of gastric mucosal glands when compared to the CAG model group (Figure 3C, P > 0.05). There was a marked decrease in the number of chief cells in the CAG model group (131.2 ± 19.6 units/field, P < 0.05) relative to the control group. Conversely, the number of chief cells in the gastric mucosa of the high-dose group (176.9 ± 29.4 units/field, P < 0.05) was significantly higher than that in the CAG model group (Figure 3C). No significant difference was observed in the number of parietal cells within the gastric mucosa among the control group, low-dose group, and high-dose group when compared with the CAG group (Figure 3C, P > 0.05).

Figure 3 Effects of Chaihu-Shugan-San on histopathological changes of the gastric mucosa in methyl-N’-nitro-N-nitrosoguanidine-treated rats. A: Representative photomicrographs for hematoxylin and eosin staining of the gastric mucosa of rats. Upper: Scale bar = 200 μm. Lower: Scale bar = 50 μm; B: Quantification of inflammation score; C: Quantitation of the number of gastric glands, thickness of gastric glands, number of gastric chief cells, and number of gastric parietal cells; D: Representative photomicrographs for alcian blue staining of the gastric mucosa. Upper: Scale bar = 200 μm. Lower: Scale bar = 50 μm; E: Quantitative analysis of intestinal metaplasia of gastric mucosa. Data are shown as mean ± SD (n = 6). Differences were analyzed by one-way analysis of variance followed by Tukey’s post-hoc test. ^aP < 0.05. ^bP < 0.01. ^cP < 0.001. P values vs chronic atrophic gastritis model group. NS: No significance; CAG: Chronic atrophic gastritis; CSS: Chaihu-Shugan-San; Con: Control.

AB staining demonstrated the absence of blue staining in the deep mucosa lacking intestinal metaplasia in the control group. In the presence of intestinal metaplasia, the entire mucosa exhibited blue staining in the CAG model group. There was a small amount of blue staining, suggesting minimal intestinal metaplasia in the low-dose and high-dose groups (Figure 3D). The percentage of intestinal metaplasia in the CAG model group increased significantly (0.49% ± 0.12% and 8.31% ± 1.65% in the control group and CAG model group, respectively, P < 0.001). The areas of intestinal metaplasia in the low-dose (5.39% ± 1.34%, P < 0.01), and high-dose (2.65% ± 0.62%, P < 0.001) groups were significantly lower than in the CAG model group (Figure 3E).

CSS inhibited inflammation and apoptosis in rats with CAG

To obtain optimized results, the compounds obtained from TCMSP database mining were de-duplicated. A total of 139 active components were retrieved for the seven herbs constituting CSS. Active components lacking associated targets (16 components) were excluded from further analysis. The interaction relationships between the herbs and their respective active components were visualized (Supplementary Figure 1A). Two hundred and sixty putative targets associated with these 123 ingredients were predicted, and the ingredients-targets interaction network was constructed (Supplementary Figure 1B). The optimized enriched targets were imported into the DAVID database to conduct GO analysis (Supplementary Figure 1C) and KEGG analysis to visualize the relevant pathways based on the number of enriched genes (P < 0.05). The interaction relationships among active components, targets and KEGG pathways are summarized in Figure 4A. KEGG pathway results indicated that potential targets of CSS may participate in apoptosis, the IL-17 signaling pathway, and the TNF signaling pathway (Figure 4B).

Figure 4 Effects of Chaihu-Shugan-San on inflammation and apoptosis in methyl-N’-nitro-N-nitrosoguanidine-treated rats. A: Chaihu-Shugan-San-active ingredient–target-signaling pathway network; B: Kyoto Encyclopedia of Genes and Genomes enrichment histogram; C: Levels of inflammatory cytokines tumor necrosis factor-α, interleukin (IL)-1β, and IL-6 in the serum of rats; D: Western blotting examined the apoptotic proteins BCL2 and BAX in the gastric mucosa of rats; E: Activity of caspase-3 and caspase-9. Data are shown as mean ± SD (n = 6). Differences were analyzed by one-way analysis of variance followed by Tukey’s post-hoc test. ^aP < 0.05. ^bP < 0.01. ^cP < 0.001. P values vs chronic atrophic gastritis model group. NS: No significance; KEGG: Kyoto Encyclopedia of Genes and Genomes; TNF: Tumor necrosis factor; IL: Interleukin; CAG: Chronic atrophic gastritis; CSS: Chaihu-Shugan-San; Con: Control; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase.

Based on the results of target prediction derived from network pharmacology analysis, our initial focus was on the therapeutic efficacy of CSS for CAG, through targets associated with inflammation and apoptosis. The serum TNF-α (105.0 ± 25.1), IL-1β (199.8 ± 42.3), and IL-6 (132.6 ± 33.2) levels in the CAG model group were significantly higher than those in the control group (P < 0.001). After administration of a high dose of CSS, the levels of TNF-α (57.9 ± 15.2, P < 0.01), IL-1β (103.7 ± 27.3, P < 0.001), and IL-6 (64.8 ± 13.9, P < 0.001) were significantly decreased compared with those in the CAG model group (Figure 4C). In the model group, both BCL2 expression and the BCL2/BAX ratio were significantly lower than those in the control group (P < 0.001). However, following CSS intervention, these parameters were markedly increased compared with the model group (P < 0.001). In contrast, BAX expression was significantly higher in the CAG model group than that in the control group (P < 0.001). Compared with the CAG model group, BAX expression was significantly reduced in both the low-dose group and the high-dose group (P < 0.001, Figure 4D). Activity of caspase-3 (84.3 ± 21.1) and caspase-9 (62.9 ± 14.7) was significantly higher in the CAG model group compared with the control group (Figure 4E, P < 0.001). Low dose (60.7 ± 13.4) and high dose (41.4 ± 10.2) of CSS decreased caspase-3 activity (Figure 4E). Similarly, caspase-9 activity was significantly lower in the high-dose group (33.9 ± 8.9) than in the control group (Figure 4E, P < 0.001).

CSS blocked activation of NF-κB in rats with CAG

A total of 452 therapeutic targets for CAG were identified from the GeneCards database. To investigate the pharmacological targets of CSS against CAG we performed a comparative analysis and visualized the 75 overlapping targets between CAG and CSS using a pie chart (Figure 5A), representing the candidate targets for CSS intervention in CAG. We constructed a PPI interaction network of these 75 crossover targets to elucidate the mechanisms underlying the potential targets. Network topology analysis identified several hub targets, including TNF, IL-1, IL-6, BAX, BCL2, caspase-3/caspase-9, and NFKBIA, which are hypothesized to play crucial roles in CSS-mediated therapeutic effects against CAG. GO analysis of these core targets revealed their predominant association with the NF-κB signaling complex and BAX apoptotic complex (Figure 5B). Further investigation into the interaction between CSS bioactive components and key targets in the NF-κB pathway revealed specific molecular interactions: Quercetin interacted with TNF, IL-1A, IL1B, and NFKBIA; Luteolin targeted TNF and NFKBIA; Kaempferol interacted with TNF; While and paeoniflorin modulated TNF and IKBKB. These findings provide mechanistic insights into the multicomponent, multitarget therapeutic strategy of CSS in CAG (Figure 5C). Six phenylpropanoids and polyketides in CSS, including baicalin, licoisoflavone B, licochalcone B, glabrone, glycyrrhiza flavonol A, and marmin, were individually subjected to molecular docking with IκBα. Each of these six ligands featured five potential active binding pockets for IκBα interaction, with all Vina scores consistently below 6 (Supplementary Table 1). For every ligand-IκBα complex, the active binding pocket demonstrating the highest Vina score was selected for subsequent structural visualization (Figure 5D).

Figure 5 Molecular docking analysis. A: Pie chart of Chaihu-Shugan-San (CSS)-chronic atrophic gastritis common targets; B: Protein-protein interaction network. Interaction score > 0.9; C: Relationship between active components in CSS and key targets of the nuclear factor-kappa B pathway; D: Visualization of molecular docking of inhibitor of nuclear factor-kappa B protein with baicalin, glabrone, licochalcone B, glycyrrhiza flavonol A, licoisoflavone B, and marmin. CAG: Chronic atrophic gastritis; CSS: Chaihu-Shugan-San; GO: Gene Ontology; NF-κB: Nuclear factor-kappa B; TNF: Tumor necrosis factor; IL: Interleukin; IκB: Inhibitor of nuclear factor-kappa B.

The effects of CSS on the NF-κB pathway were verified. IκBα was decreased notably at protein level in the gastric mucosa of the CAG model group (0.29 ± 0.06, P < 0.001) (Figure 6A). Conversely, p-IκBα (4.96 ± 1.55) and NF-κB p-p65 (4.13 ± 0.37) protein levels were significantly increased in the CAG model group compared with the control group (Figure 6A) (P < 0.001). Consequently, the nuclear location of NF-κB p-p65 by immunofluorescence staining was significantly increased in the CAG model compared with the control group (Figure 6B). Low- and high-dose CSS significantly upregulated IκBα protein levels (low dose: 0.46 ± 0.07; high dose: 0.59 ± 0.07) (both P < 0.001), and downregulated p-IκBα protein levels (low dose: 2.89 ± 1.07, P < 0.01; high dose: 1.63 ± 0.27, P < 0.001) and NF-κB p-p65 protein levels (low dose: 2.82 ± 0.42; high-dose: 1.67 ± 0.48) (both P < 0.001) in CAG rats. NF-κB p65 protein levels were not altered significantly in the control, low-dose, and high-dose groups compared with the CAG model group (Figure 6A). This formula decreased the nuclear location of p-p65 in the CAG model (Figure 6B).

Figure 6 Effects of Chaihu-Shugan-San on nuclear factor-kappa B signaling in methyl-N’-nitro-N-nitrosoguanidine-treated rats. A: Upper: Western blotting examined the nuclear factor-kappa B signaling proteins inhibitor of nuclear factor-kappa B (IκBα), phospho (p)-IκBα, p65, and p-p65. Lower: Densitometric quantitation of proteins; B: Immunofluorescence detection of expression and nuclear localization of p-p65. Data are shown as mean ± SD (n = 6). Differences were analyzed by one-way analysis of variance followed by Tukey’s post-hoc test. ^bP < 0.01. ^cP < 0.001. P values vs chronic atrophic gastritis model group. NS: No significance; CAG: Chronic atrophic gastritis; CSS: Chaihu-Shugan-San; Con: Control; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase; NF-κB: Nuclear factor-kappa B; IκB: Inhibitor of nuclear factor-kappa B; p: Phospho; DAPI: 4’,6-diamidino-2-phenylindole.

Effects of CSS on the diversity of colonic microbiota in rats with CAG

The microbial composition of rat feces was detected by 16S rDNA sequencing. The species accumulation curve initially showed an increasing trend and then tended to be smooth with the increase in the number of sequencing samples, which suggested that the sample size was sufficient to fully reflect the cecum microbial community (Figure 7A). The bacterial composition and relative abundance at the family level of the cecum microbiota were visualized using a community histogram, which showed that the following families were dominant in the rat cecum microbiota in the CAG and CAG + CSS groups: Ruminococcaceae, Lachnospiraceae, Muribaculaceae, and Prevotellaceae (Figure 7B). For α-diversity, the ACE (P = 0.14), Chao1 (P = 0.13), observed species (P = 0.16), Shannon (P = 0.65), and Simpson (P = 0.90) indexes did not significantly differ, but a significant difference in the phylogenetic diversity (PD) whole tree (P = 0.02) index was found between the CAG and CAG + CSS groups (Figure 7C). Principal coordinate analysis and the hierarchical cluster tree were drawn to display β-diversity between the CAG and CAG + CSS groups, which verified that the samples of different groups were separated (Figure 7D and E). Although α-diversity (within-sample diversity) was not significantly altered by CSS, the significant separation in β-diversity (between-sample diversity) indicated that CSS primarily modulated the compositional structure of the gut microbiota rather than its overall richness or evenness. By linear discriminant analysis effect size, 33 differential gut bacteria were identified between the CAG and the CAG + CSS groups (Figure 7F). CAG significantly reduced the abundance of genera with known short-chain fatty acid-producing capacity, such as Muribaculaceae. In contrast, CSS treatment significantly increased the abundance of genera linked to intestinal homeostasis, including Ruminococcaceae (Supplementary Figure 2).

Figure 7 Effects of Chaihu-Shugan-San on gut microbiome in methyl-N’-nitro-N-nitrosoguanidine-treated rats. A: Species accumulation curve estimates the rationality of sequencing sample quantity; B: Community histogram shows the compositional discrepancy at family level; C: Analysis of differences in the α-diversity indexes ACE, Chao1, phylogenetic diversity whole tree, observed species, Shannon and Simpson; D and E: Analysis of differences in the β-diversity index, as determined by principal coordinate analysis (D) and hierarchical cluster tree (E); F: Cladogram of linear discriminant analysis effect size (n = 3). PD: Phylogenetic diversity; CAG: Chronic atrophic gastritis; CSS: Chaihu-Shugan-San; PCoA: Principal coordinates analysis.

DISCUSSION

Chinese herbal compound therapy is a commonly used treatment for CAG[27], aiming at regulating the function of spleen and stomach and promoting the repair of gastric mucosa[28,29]. The frequently used Chinese herbals include Coptis rhizome, Scutellaria root, Ginseng, licorice, and Poria, harboring the abilities to eliminate heat and dry moisture, strengthen the spleen and qi, and relieve stomach discomfort[30,31]. Classic prescriptions, such as Banxia Xiexin decoction[32] and Liujunzi decoction[33], are frequently used in the clinical management of CAG. CSS is a formulation derived from TCM. Its primary application is to alleviate liver depression, regulate qi, and alleviate pain[34]. CSS could effectively regulate symptoms associated with liver qi discomfort, such as mood sadness, chest tightness, and hypochondria[35,36]. A meta-analysis also suggested that CSS has a distinct therapeutic impact on gastritis in clinical settings[37]. However, the precise mechanism and particular actions of this effect have not yet been elucidated. This study assessed the pharmacodynamic effects of CSS in a rat CAG model generated by MNNG. Additionally, the mechanism of CSS was studied using network pharmacology.

MNNG is a nitroso chemical with strong genotoxicity and carcinogenicity[38]. It functions as an alkylating agent, directly interacting with DNA molecules and causing alkylation of bases, leading to structural damage to DNA[39]. MNNG is commonly used in research as a mutagen and carcinogen[40]. Several studies have demonstrated that MNNG is effective in causing CAG and precancerous lesions in animal models[41,42]. Consequently, animals fed with MNNG are widely recognized as a routinely used model for CAG. In this study, we treated rats with MNNG for 16 weeks to induce CAG. Consistent with previous studies[43], the MNNG-induced CAG rats had reduced body weight, stomach atrophy, and significant damage to the gastric mucosa[44]. CSS alleviated weight loss and gastric injury, and mitigated gastric mucosal lesions in rats induced by MNNG, indicating its efficacy in the treatment of CAG.

Network pharmacology is a methodology that uses bioinformatics and network analysis to investigate the chemical interactions and impacts on biological networks[22]. In CAG, many researchers focus on the signaling pathways related to gastric mucosal inflammation and damage repair to explore how the existing drugs or natural products could alleviate CAG using network pharmacology[12,45]. In this study, we also used network pharmacology to examine the therapeutic targets of CSS in CAG. We found that inflammatory cytokines (such as TNF-α, IL-6, and IL-1β), and apoptosis-related proteins (BCL2 and caspase-3) formed a core network. Through experimental analysis, CSS was found to inhibit the levels of TNF-α, IL-6, and IL-1β. Additionally, CSS treatment effectively upregulates the BCL2/BAX ratio, indicating that its protective effects on the gastric mucosa and the amelioration of atrophic lesions may be partially mediated by modulating the balance of BCL2 family proteins and suppressing abnormal apoptosis of gastric mucosal epithelial cells. We conducted pathway enrichment analysis using these targets and discovered that they were mostly linked to apoptosis and inflammatory pathways, including TNF and IL-17 signaling. Upon reviewing the pathway-specific regulatory molecules on the KEGG website, we identified NF-κB as a potential core protein among the pathways.

NF-κB is a central hub downstream of multiple inflammatory pathways, including the IL-17 signaling pathway and TNF signaling pathway predicted by network pharmacology[46,47]. Therefore, validating the NF-κB pathway allows us to capture a key convergent mechanism of CSS action. NF-κB is a crucial transcription factor whose stability is regulated by IκBα, an essential inhibitory protein[48]. Under resting conditions, IκBα binds to the NF-κB heterodime, sequestering it in the cytoplasm in an inactive state. When a cell undergoes stimulation, such as an inflammatory signal, IκBα is phosphorylated by the IκB kinase complex. This complex is ubiquitinated and degraded and releases NF-κB; then NF-κB can enter the nucleus and activate the transcription of target genes, triggering physiological processes such as inflammation and immune response[49,50]. When gastric mucosa is damaged, such as by Helicobacter pylori infection, the use of nonsteroidal anti-inflammatory drugs, or other physicochemical factors, the degradation of IκBα, and the activation of NF-κB are abnormally enhanced[51,52]. This leads to an increased expression of inflammatory mediators, including TNF-α, IL-1β, and IL-6, which exacerbate the inflammatory state of gastric mucosa and contribute to the development of gastritis[53]. Inhibition of the NF-κB activity may serve as a treatment approach for gastritis. Studies have demonstrated that some drugs or natural products can play an anti-inflammatory and protective role in CAG by blocking the NF-κB signaling pathway[54-56]. Consistently in this study, our study demonstrates that CSS treatment suppresses IκBα phosphorylation and enhances its stability, thereby inhibiting NF-κB activation induced by CAG.

The gut flora plays an important role in the maintenance and function of the gastrointestinal barrier[57]. Hence, the investigation of CAG has increasingly focused on the examination of gut flora diversity using 16S rDNA sequencing[58]. Prior research has demonstrated that the diversity of intestinal flora in CAG patients is significantly lower than that in the normal population[59]. Additionally, investigations have indicated a decline in the abundance of intestinal microorganisms in CAG rats that were induced by MNNG[60]. Intriguingly, our study showed that CSS can increase the diversity of intestinal flora in CAG rats, indicating that the ameliorating effect of CSS on CAG may be associated with the regulation of intestinal flora homeostasis. Muribaculaceae is mainly associated with short-chain fatty acid[61]. The increase in the relative abundance of Muribaculaceae may contribute to treat CAG by improving the balance of intestinal flora[62]. previous literatures reported that Ruminococcaceae attenuates inflammation[63], and its elevated abundance may also contribute to the protection of the intestinal barrier[64]. We suggest that CSS-induced the beneficial structural changes in the gut microbiota may reduce systemic low-grade inflammation, thereby creating a favorable environment for gastric mucosal repair.

CONCLUSION

Our exploration of the pharmacological activity and therapeutic mechanism of CSS reveals that it alleviates CAG symptoms by regulating the NF-κB signaling pathway. These findings may provide new insights into the clinical application of CSS and are expected to facilitate further research and development in the field of CAG therapy.