He-Wei-Decoction Ameliorates Chronic Atrophic Gastritis via Modulation of the TLR4/NF-κB Signaling Pathway

Abstract

He-Wei-Decoction (HWD), a traditional Chinese medicine formula, emphasizes “strengthening the spleen and supplementing qi (Jianpi Yiqi),” and “resolving stasis and detoxifying (Huayu Jiedu).” This formula has been utilized in the treatment of chronic atrophic gastritis (CAG), however, its precise mechanisms of action remain to be elucidated. This study integrates network pharmacology with molecular dockings to identify the underlying mechanisms of HWD in the treatment of CAG. Then, immunohistochemistry assay analyzed the gastric mucosal lesions before and after treatment with HWD in CAG patients. The efficacy and mechanism of key active compounds in the treatment of CAG were validated using a 1-methyl-3-nitro-1-nitrosoguanidine (MNNG)-induced GES-1 cell in vitro model. A total of 165 active compounds from HWD were identified, along with 169 targets associated with CAG. Gene oncology (GO) analysis, Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis, and Component-Target-Pathway network analysis revealed that the Toll-like receptor (TLRs) signaling pathway may play a role in HWD’s regulation of inflammation in gastric mucosa. Molecular docking identified that luteolin, quercetin, and hederagenin potentially interact with key target proteins TLR4 and nuclear factor-kappaB (NF-κB). Experimental results validated that HWD significantly improved atrophic mucosa and inhibited the expression of TLR4, NF-κB, and cyclooxygenase 2 (COX2) proteins. The active compounds, luteolin, quercetin, and hederagenin, inhibit cell viability, migration and invasion, and reduce the expression of TLR4, NF-κB, and COX2. In summary, luteolin, quercetin, and hederagenin, the primary active components of HWD, can ameliorate the CAG by regulating the TLR4/NF-κB signaling pathway.

INTRODUCTION

Chronic atrophic gastritis (CAG) is a chronic inflammatory disease recognized as a precancerous lesion in gastric cancer (GC). CAG is characterized by mucosa atrophy, exposed blood vessels, and mucosal nodules.^1,2) Nevertheless, CAG is a reversible process that shows improvement upon treatment of the symptoms and inhibition of the inflammatory responses.^3–7) The inflammatory microenvironment of the gastric mucosa plays a crucial role in the progression of CAG, making anti-inflammatory treatment one of the important intervention strategies. Currently, traditional Chinese medicine (TCM) has been proven to inhibit inflammatory responses and effectively improve the inflammatory microenvironment of gastric mucosa. Shao and colleagues reported that Moluodan significantly reduced the expression of interleukin-6 (IL-6) and tumor necrosis factor α (TNF-α) in CAG rats, alleviating the inflammatory responses of the gastric mucosa.⁸⁾ Similarly, Qinhuayin significantly decreased the expression of cyclooxygenase 2 (COX-2) and TNF-α inflammatory factors in CAG rats.⁹⁾ Moreover, Mingyu and colleagues demonstrated that Weifuchun inhibited the inflammatory response (IL-1β, IL-6, and TNF-α) and the aggregation of macrophages in the gastric mucosa by suppressing the nuclear factor-kappaB (NF-κB) pathway.¹⁰⁾ Therefore, improving the inflammatory microenvironment may be key to the effective treatment of CAG with TCM, and screening for the anti-inflammatory molecules from TCM could provide new therapeutic options for CAG.^11–13)

He-Wei-Decoction (HWD) is a prescription for the treatment of CAG in TCM, which is based on an improved version of Si-Ni-San,¹⁴⁾ consisting of 12 herbs, namely, Radix Bupleuri (Bupleurum chinense DC., CH), Radix Paeoniae Alba (Paeonia lactiflora Pall., BS), Fructus Aurantii seu Ponciri (Citrus aurantium L., ZK), Radix Glycyrrhizae (Glycyrrhiza uralensis Fisch. GC), Astragalus chrysopterus bunge (Astragalus membranaceus Fisch., HQ), Cortex Magnoliae officinalis (Magnolia officinalis Rehd. et Wils. HP), Hericium erinaceus (Hericium coralloides Scop., HTG), Radix Notoginseng (Panax notoginseng Burk., SQ), Curcuma zedoaria (Curcuma aeruginosa Roxb., EZ), Rhizoma Sparganii (Sparganium stoloniferum Buch., SL), Herba Hedyotis diffusae (Hedyotis diffusa Willd., BHSSC), and Herba Scutellariae barbatae (Scutellaria barbata D.Don., BZL). HWD is developed based on the TCM theory of “strengthening the spleen and supplementing qi (Jianpi Yiqi), resolving stasis and detoxifying (Huayu Jiedu).” Clinical research indicates that HWD can effectively relieve clinical symptoms, enhance QOL, and reverse precancerous lesions. However, the complexity of HWD’s components and their synergistic interactions pose a challenge in elucidating its molecular mechanisms. Therefore, the aim of this study was to investigate the chemical and pharmacological basis of HWD in the treatment of CAG.

Network pharmacology is a promising approach for studying multi-component drugs. To address multi-component systems such as herbal formulas, the “network target” extends the concept of a drug target on a single molecule to its systematic effect on a biological network.^15–17) The TCM network pharmacology has been successfully utilized to analyze herbal formulas such as Si-Ni-San,¹⁸⁾ Wei-Fu-Chun,¹⁹⁾ Moluodan,⁸⁾ and Xiao-Ai Jie-Du decoction.²⁰⁾ This approach has provided a potential tool for analyzing the biological basis of herbal formulas and guiding the discovery of active compounds within herbs. Based on the results of network pharmacology, molecular docking can simulate the interactions between ligands and receptors, facilitating the screening of key active compounds and clarifying the molecular mechanisms of TCM.^21,22) These technologies are accurate and cost-effective, enabling us to identify the effective compounds in drugs, improve drug design, and elucidate biochemical pathways.

In the current study, we employed computational tools and resources to investigate the pharmacological network of HWD in relation to CAG, aiming to predict the active compounds, potential protein targets, and associated pathways. The primary components of TCMs were retrieved from the TCM Systems Pharmacology (TCMSP) database and analysis platform. These components were filtered based on their druggable properties, resulting in a total pf 165 candidate bioactive compounds. The STRING database was utilized to identify the targets corresponding to CAG. Subsequently, network pharmacology and molecular docking techniques were applied to identify the active compounds and their potential protein targets and pathways. Furthermore, in vitro experiments were conducted to validate the mechanism of the active compounds on CAG (Fig. 1). Interestingly, our findings indicated that luteolin, quercetin, and hederagenin, which are the active compounds of HWD, improve the inflammatory response of the gastric mucosa by inhibiting the TLR4/NF-kB pathway. Therefore, this study aims to provide basic data supporting the pharmacological effects and potential mechanism of HWD for CAG.

Fig. 1. Workflow of the Study

MATERIALS AND METHODS

Materials

Human gastric mucosa epithelial cell line GES-1 was obtained from the Guangzhou First’s People Hospital. Fetal bovine serum (FBS) (164210-50) was purchased from ProCell (Canton, MA, U.S.A.). RPMI Medium 1640 basic (C11875500BT) was purchased from GIBCO Life Technologies. 1-Methyl-3-nitro-1-nitrosoguanidine (MNNG, 70-25-7) was purchased from Selleck (Houston, TX, U.S.A.). Anti-TLR4 (GB15186-100), NF-κB p65 (GB11997-100), NF-κB pp65 (AF2006-50), and COX2 (GB11077-1-100) antibody were purchased from Servicebio (Hubei, China). MTT reagent (ST1537-5g) was purchased from Beyotime (Shanghai, China). Luteolin (HY-N0162), Quercetin (HY-18085), and Hederagenin (HY-N0256) were purchased from MedChemExpress (Monmouth Junction, NJ, U.S.A.).

Acquisition of the Active Compounds of HWD

The Traditional Chinese Medicine Systems Pharmacology database (TCMSP, Version: 2.3, https://old.tcmsp-e.com/tcmsp.php) was used to determine the active ingredients contained in CH, BS, ZK, GC, HQ, HP, SQ, SL, EZ, BZL, and BHSSC. The TCMSP database is a platform based on the systemic pharmacology of Chinese herbal medicine including 499 Chinese herbal medicines and more than 29000 ingredients. We utilized oral bioavailability (OB) ≥30% and drug-likeness (DL) ≥0.18 as the screening criteria to identify qualified active compounds. The HTG was investigated through the Shanghai Institute of Organic Chemistry of CAS. Chemistry Database[DB/OL] (https://organchem.csdb.cn.[1978-2024]). Then, the SMILES of HTG compounds was entered into the SwissADME (http://www.swissadme.ch/) to evaluate drug-likeness. Subsequently, the drug targets corresponding to these active compounds were queried using Swiss Target Prediction database (http://www.swisstargetprediction.ch/), and the canonical gene names of the drug targets were obtained from the Uniprot database (https://www.uniprot.org/).

Acquisition of Potential Therapeutic Target Genes of CAG

With “chronic atrophic gastritis” as keywords, the species “Homo sapiens” was selected for mining CAG-related target genes using the GeneCards database (https://www.genecards.org/). Subsequently, Jvenn (www.bioinformatics.com.cn) was employed to generated a Venn diagram, which facilitated the identification of intersection of the disease targets and drug targets obtained above. As a result, we identified a total of 169 common targets.

Protein–Protein Interaction (PPI) Network Construction

To elucidate the interaction of the therapeutic target genes and identify the hub genes, we entered the therapeutic genes into the Search Tool for the Retrieval of Interaction Genes/Proteins database (STRING, Version 12.0, http://string-db.org/), selected “Homo sapiens” as the species, and set the parameter at the highest confidence (0.700) level to obtain PPI data. The hub genes were identified by topology analysis. Visualization of the PPI network and topology analysis were performed via Cytoscape software.

Biological Function and Pathway Enrichment Analysis

Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were used to elucidate the mechanisms through biological processes (BP), cellular components (CC), molecular functions (MF), and key signaling pathways using the DAVID database (https://david.ncifcrf.gov). The species was limited to “Homo sapiens,” and the enrichment of pathway was considered significant when the modified Fisher’s exact false discovery rate (FDR) was <0.01. The GO and KEGG results were visually analyzed by an online bioinformatics platform (www.bioinformatics.com.cn).

Construction of the Component-Target-Pathway (C-T-P) Network

Based on common targets of compound-disease and inflammation-related pathways, a “C-T-P” regulatory network was constructed using the Cytoscape software. The data sources included the following: (1) the targets corresponding to these active compounds were queried using the Swiss Target Prediction database (http://www.swisstargetprediction.ch/), and the canonical gene names of the targets were obtained from the Uniprot database (https://www.uniprot.org/); (2) The targets corresponding to the key signaling pathways were obtained using the DAVID database (https://david.ncifcrf.gov). Then, the data sources were construed a node table of C-T-P network. The visualization of the C-T-P network and topology analysis were performed via the Cytoscape software. The proteins participated in the regulation of inflammation-related pathways were filtered by the Degree Centrality >10.

Molecular Docking

Molecular docking can predict the potential therapeutic effects of drug components by analyzing the binding potential between the main components and the targets. The specific methods were as follows: (i) The 2D structures of the main components were downloaded from PubChem databases (https://pubchem.ncbi.nlm.nih.gov/), then imported into the Chem3D software to convert and optimize them into 3D structures. (ii) The target proteins were obtained from the RCSB PDB (http://www.rcsb.org/), set species as “Homo sapiens,” and imported into the PyMOL software for processing. (iii) The main components and target proteins were transformed into PDBQT format by AutoDock Tools, and a docking grid box constructed for each target. The molecules with the lowest binding energy for each active compound in the docking conformation were allowed for semi-flexible docking by comparing with the original ligands and intermolecular interactions. (iv) The docking results were visualized with the PyMOL software.

Preparation of Gastric Mucosa Tissue

CAG specimens were collected from the Guangdong Provincial Second Hospital of Traditional Chinese Medicine. A total of 30 patients were either Helicobacter pylori (Hp)-negative or had eradicated Hp following treatment, and subsequently received HWD. Patients were administered HWD after meals, with a treatment duration of 6 months. We collected gastric mucosal biopsy tissues from patients both before and after the treatment. All sample collections were approved by the Ethics Committee of the hospital (IRB: K202307-001), and the study was conducted in accordance with the relevant guidelines.

Cell Culture

GES-1 cells were cultured in DMEM supplemented with 10% FBS, 100 U/mL penicillin, and 100 µg/mL streptomycin in 5% CO2 at 37°C. GES-1cells were passaged until 80% confluence, then sub-cultured and cryopreserved.

Model Preparation Method

To establish a model of gastric epithelial cells damage, GES-1 cells were treated with various concentrations of MNNG solutions (in the DMEM with 10%FBS) for 24 h to determine a suitable level for the cell model. Subsequently, the MTT assay was performed to determine the viability of GES-1 cells. Absorbance at 490 nm was measured using a microplate reader.

Cell Viability Assay

GES-1 cells were digested with 0.25% trypsin and then cultured in a 96-well plate at a density of 8 × 10³/well at a volume of 100 µL per well for 24 h, with three replicate wells in each experimental group. Then the GES-1 cells were pretreated with MNNG solutions (20 µM, in the DMEM with 10% FBS) for 24 h. Following this, the model cells were treated with Luteolin, Quercetin, and Hederagenin in DMEM medium without serum for 2 h, and determined whether active compounds are gastroprotective after 24 h. The experimental cells were washed three times with PBS to remove MNNG and compounds. After that, MTT solution (0.5 mg/mL in DMEM) was added into each well and incubated for 4 h at 37°C. The resultant reduced MTT (formazan) was extracted with 150 µL DMSO and was transferred into a 96-well plate. Absorbance at 490 nm was measured using a microplate reader. The cell viability was reflected by comparing the difference in cellular activity between the drug treatment groups and the controls.

Transwell Assay

GES-1 cells were pretreated with MNNG (20 µM) for 24 h, and then were washed with PBS to remove MNNG. After that, the cells were treated with Luteolin, Quercetin, and Hederagenin in DMEM medium without serum for 2 h. Subsequently, the cells were harvested and Transwell assay performed.

The Transwell assay was used with 8-mm pore sized chambers. Transwell chambers were precoated with Matrigel for invasion assay. In brief, 2 × 10⁵ cells were suspended in 100 µL serum-free DMEM, and were then seeded in the top chamber, and 600 µL DMEM containing 10% FBS was added to the lower chamber. After 24 h incubation at 37°C in a 5% CO2 atmosphere, cells that remained in the upper surface of the membrane were removed by cotton swabs and the penetrating cells were fixed in methanol, and then stained with 0.1% crystal violet. Cell migration and invasion were quantified by counting cells on the lower surface using phase contrast microscopy.

Hematoxylin–Eosin (H&E) Staining

The gastric mucosae were collected and fixed with 4% paraformaldehyde, embedded in paraffin, sectioned into 5 µm slices, and subjected to standard H&E staining. The slices were visualized using a microscope (IX71, Olympus, Tokyo, Japan).

Immunohistochemistry (ICH)

The IHC of gastric mucosa was performed with the primary antibody and the EnVision™ Plus Kit (DAKO) according to the manufacturer’s instructions. After isolation and hydration, the antigen was first recovered with citrate buffer (pH 6.0), following which the endogenous peroxidase activity was quenched with 0.3% hydrogen peroxide solution. The slices were then incubated with primary antibody (anti-TLR4 (1:500 dilution), anti- NF-κB p65 (1:400 dilution), anti- NF-κB pp65 (1:400 dilution), and anti-COX2 (1:500 dilution)) at 4°C overnight. 3,3′- Diaminobenzidine (DAB) was used as a substrate to detect the binding of the antibody with the second antibody after incubation for 1 h at room temperature. Slices with normal IgG as a primary antibody were used as the negative control. The obtained images were analyzed using image J software.

Immunofluorescence

Cells were fixed in 4% paraformaldehyde at room temperature for 20 min and then rinsed with PBS three times. The cells were then blocked with 4% bovine serum albumin (BSA). The expression of TLR4, NF-κB p65, NF-κB pp65, and COX2 in GES-1 was analyzed. Cells were incubated at 4°C with anti-TLR4 (1:500 dilution), anti- NF-κB p65 (1:400 dilution), anti- NF-κB pp65 (1:400 dilution), and anti-COX2 (1:500 dilution), respectively, for 4 h and subsequently incubated with an FITC conjugated secondary antibody (1:500 dilution, Invitrogen) at 37°C for 1 h. Cells were counterstained with DAPI (1 µg/mL) and visualized on a fluorescence microscope (IX71, Olympus). The obtained fluorescence images were analyzed to count the number of fluorescence/DAPI-stained cells per 300 cells using the Image J software.

RT-PCR Assay

Model cells were treated with Luteolin, Quercetin, and Hederagenin in DMEM medium for 2 h. Subsequently, the cells were harvested at 1, 6, and 24 h to preform RT-PCR assay.

Total RNA was extracted using TRIzol (Invitrogen, Waltham, MA, U.S.A.). Extracted RNA was then reversely transcribed to cDNA via the PrimeScript RT Master Mix (Perfect Real Time, Bioscience). The resulting cDNA was then subjected to the RT-PCR assay using the TB Green® Premix Ex Taq GC (Perfect Real Time, Bioscience). The RT-PCR assay was performed on an ABI 7500 System (Applied Biosystems, Waltham, MA, U.S.A.), with ACTB serving as the internal control. The mRNA levels of TLR4 and NF-κB were normalized to the ACTB gene. Relative mRNA levels of gene expression were calculated using a classical 2^−ΔΔCT method. Primers for the PCR assay used are listed in Supplementary Table S1.

Statistical Analysis

Descriptive statistics including mean and standard deviation (S.D.) were conducted with GraphPad Prism. t-Tests were performed using SPSS, and p < 0.05 was considered significant (*p < 0.05, **p < 0.01). To compare the difference between before treatment group and after treatment group, we used the Wilcoxon Signed Rank Test. The test was preformed using SPSS, and the corresponding Z-value and p-value were calculated to determine whether the observed differences were statistically significant.

RESULTS

Identification of Bioactive Compounds in HWD

To discover the potential bioactive compounds in HWD, a total of 165 candidate bioactive compounds were screened out under the screening condition of OB ≥ 30% and DL ≥ 0.18, based on TCMSP, Chemistry Database, and SwissADME. Among the herbs in HWD, GC, HQ, HTG, and BZL contained a higher number of bioactive compounds. Specifically, there are 77 kinds of GC, 22 kinds of BZL, 18 kinds of HQ, and 12 kinds of HTG (Supplementary Table S2). According to the TCM theory, components of GC, HQ, and HTG are utilized for strengthening the spleen and supplementing qi, and those of BZL are employed for detoxification. GC, HQ, HTG, and BZL are considered as the primary herbs in HWD for the treatment of CAG.

Prediction of the Potential Targets and Pathways of HWD Treatment of CAG

We retrieved 5667 protein targets corresponding to 165 candidate bioactive compounds from the SWISS Target Prediction database. Following normalization through the UniProt database and the removal of duplicates, a total of 769 unique protein targets were obtained for further analysis. At the same time, using “chronic atrophic gastritis” as the keyword, we obtained 891 CAG-related targets from the GeneCards database by deleting duplicate entries. Then, the 769 drug targets and 891 disease targets were intersected, resulting in 169 potential targets for HWD in the treatment of CAG (Fig. 2A); 169 potential targets were input into the STRING database to obtain protein–protein interaction (PPI) information and make them a PPI network with 114 nodes and 246 edges. Topological analysis revealed that nodes with the degree top 6 are AKT1, PIK3CA, SRC, TP53, MAPK1, and MAPK8 (Fig. 2B), identifying as the hub genes for bioactive compounds in the treatment of CAG.

Fig. 2. Prediction of the Potential Targets and Pathways of HWD Treatment of CAG

(A) Venn diagram of intersection targets of active components of HWD and CAG. (B) PPI network of core targets. GO function (C) and KEGG pathway (D) enrichment analysis of HWD against CAG.

Subsequently, we performed GO analysis for the 169 potential targets using the DAVID 6.9 database. A total of 517 GO terms were significantly enriched in the data, with the counts for CC, MF, and BP being 87, 207, and 906, respectively (p < 0.05). According to the number of associated genes, the top 12 GO terms were selected (Fig. 2C). The CC terms dominantly relate to the cytoplasm, plasma membrane, nucleus, and extracellular region. The MF terms are protein binding, ATP binding, metal ion binding, enzyme binding, protein homodimerization activity, and protein kinase activity. The BP terms are involved in transcription by RNA polymerase II, apoptotic process, positive regulation of cell population proliferation, positive regulation of DNA-templated transcription, chromatin remodeling, and inflammatory response. These results indicated that active compounds may exert their effects through biological processes such as apoptosis, cell proliferation, inflammatory responses, protein interactions, and kinase activation.

Additionally, we performed KEGG enrichment analysis, which revealed 175 signaling pathways (p < 0.05). As shown in Fig. 2D, these signaling pathways mainly involve cancer-related and inflammation-related processes, including the EGFR tyrosine kinase inhibitor resistance pathway, PD-L1 expression and PD-1 checkpoint pathway, ErbB signaling pathway, C-type lectin receptor signaling pathway, IL-17 signaling pathway, TNF signaling pathway, and Toll-like receptor (TLR) signaling pathway. Previous studies report that CAG, as a chronic inflammatory disease, can be significantly reversed through the inhibition of inflammation-related pathways, thereby reducing the incidence of GC.⁵⁾ In the following, we focused on the regulation of inflammation-related signaling pathways by bioactive compounds.

Visualizing Compound-Target-Pathway Network and Molecular Docking of HWD

To identify the anti-inflammatory molecules, we constructed C-T-P networks using the Cytoscape software. As shown in Fig. 3A, bioactive compounds derived from GC, HQ, HTG, and BZL are associated with inflammation-related pathways, including the IL-17 signaling pathway, TNF signaling pathway, TLR signaling pathway, T cell receptor signaling pathway, and NOD-like receptor signaling pathway. The C-T-P network showed that six proteins participated in the regulation of inflammation-related pathways with a high frequency, indicating their significant role in the enrichment pathways. The six proteins are MAPK14, AKT1, PIK3CA, MAPK1, MAPK3, and MAPK8. Consistent with the previous results of PPI analysis, these genes occupy core positions and are associated with the TNF signaling pathway and TLR signaling pathway. As previous reported, the TLR signaling pathway regulates the expression of cytokines and chemokines such as TNF-α, IL-2, IL-6, IL-12, and COX2, by activating signaling pathways like NF-κB and MAPK, thereby playing an important role in modulating of the inflammatory microenvironment of gastric mucosa.^5,23–26) Subsequently, we focus on the candidate active components that regulate the TLR signaling pathway.

Fig. 3. Visualizing the Interaction Network and Molecular Docking of Active Components of HWD

(A) The interaction network of active components of HWD to the target of inflammation related pathways. (B) 2D molecular structures and relative molecular weight of candidate active components. (C) Molecular docking mode of active components of HWD with core targets of TLR signaling pathway.

Using AutoDock Tools, the core targets were docked with the candidate molecules one by one to obtain the binding energy. TLR4 and NF-κB proteins served as core targets, while the active compounds of GC, HQ, HTG, and BZL were used as candidate molecules. If the binding energy is less than 0 kcal/mol, it indicates that the component and the target can spontaneously combine; if the binding energy is less than −4.25 kcal/mol, it represents good docking affinity; if the binding energy is less than −7 kcal/mol, it is considered to have very strong binding affinity. The molecular docking results showed that luteolin (MOL000006), quercetin (MOL000098), and hederagenin (MOL000296) exhibited strong binding activity (binding energy<−8.00 kcal/mol) with the receptors TLR4 and NF-κB (Table 1). The docking pattern is shown in Fig. 3B.

Table 1. The Molecular Docking Results

Compound	Binding energy(kcal/mol)
	TLR4	NF-κB
Luteolin	−8	−9.2
Quercetin	−8.5	−9.1
Hederagenin	−9.3	−8.8

HWD Effected on the Anti-inflammatory in Gastric Mucosa

We observed a significant improvement in gastric mucosal lesions before and after treatment with HWD in CAG patients. Specifically, the administration of HWD resulted in significant improvements in atrophic mucosa (Table 2). The results indicated a pathological efficacy rate of 96.7% for HWD treatment. Additionally, H&E staining confirmed the beneficial effects of HWD on the pathology of the gastric mucosa in CAG disease (Fig. 4A). Prior to HWD treatment, the gastric mucosa exhibited an increased nuclear-to-cytoplasmic ratio and a slight increase in cell density. By contrast, following HWD treatment, nuclear morphology gradually returned to normal, and cell density decreased. Therefore, HWD prevents the progression of CAG and maintains effective clinical outcomes for this condition.

Table 2. Comparison of Pathological Changes of Gastric Mucosa in Patients

Case	Gender	Age	Atrophic mucosa		Intestinal metaplasia		Atypical hyperplasia
			Before	After	Before	After	Before	After
1	M	47	3*	2	3	2	0	0
2	M	41	2	0	0	0	0	0
3	M	53	1	0	1	0	0	0
4	W	53	2	1	0	1	0	0
5	W	69	2	1	1	1	0	0
6	W	40	1	0	1	2	0	0
7	M	45	1	0	0	1	0	0
8	M	63	1	0	2	0	0	0
9	M	40	2	0	0	0	1	0
10	W	50	2	1	1	1	0	0
11	M	51	1	0	1	0	0	0
12	W	56	1	0	0	2	0	0
13	M	58	1	0	1	0	0	0
14	W	37	2	0	2	0	0	0
15	M	57	1	1	0	0	0	0
16	W	37	1	0	0	0	0	0
17	M	54	1	0	0	1	0	0
18	M	52	1	0	0	0	0	0
19	W	52	2	0	2	1	0	0
20	M	58	1	0	0	1	0	0
21	W	54	2	0	0	0	0	0
22	M	50	1	0	1	2	0	0
23	M	55	1	0	1	2	0	0
24	W	66	1	0	0	0	0	0
25	M	48	1	0	1	3	0	0
26	W	69	1	0	0	0	0	0
27	W	52	2	2	2	0	0	0
28	M	68	1	0	0	0	0	0
29	M	69	1	0	2	0	0	0
30	M	65	1	1	1	0	0	0
Wilcoxon Signed Rank Test		Z-Value	−4.866		−0.546		−1.000
		p-Value	<0.001		0.585		0.317

Notes: * The severity of pathological changes in gastric mucosa is scored on an ordinal scale as follows: 0 = None, 1 = Mild, 2 = Medium, 3 = Severe.

Fig. 4. HWD Improves the Inflammation of Gastric Mucosa

The CAG patients were under treatment of HWD for 6 months, then the gastric mucosa was analyzed with HE and ICH staining assay. (A) The HE staining assay showed that HWD can decrease mucosal lesion in CAG patients. Scale bar: 100 µm. (B) HWD alter inflammation-related protein expression. The TLR4, p65 NF-κB, pp65 NF-κB, and COX-2 proteins were suppressed in gastric mucosa of CAG patients after HWD treatment. Scale bar: 50 µm.

To evaluate the effect of HWD on predicted targets, we detected the expression levels of TLR4 and NF-κB proteins in the gastric mucosa. Our findings indicated that HWD significantly reduced the expression of TLR4, pNF-κB, and ppNF-κB proteins in the gastric mucosa of CAG (Fig. 4B). Moreover, NF-κB is a key transcription factor regulating COX-2 gene expression.²⁷⁾ The immunohistochemistry result showed that COX-2 protein was suppressed in gastric mucosa of CAG patients after HWD treatment. As previous studies have reported, CAG is associated with the TLR4/NF-κB pro-inflammatory signaling pathway.^28,29) Inhibiting the TLR4/NF-κB signaling pathway is expected to inhibit the inflammatory response in the gastric mucosa, thereby mitigating the progression of CAG. Therefore, upon exposure to inflammatory stimuli, HWD probably inhibits the occurrence and progression of CAG by regulating inflammation associated with the TLR4/NF-κB signaling pathway.

The Active Compounds of HWD Intervened the TLR4 /NF-κB Pathway

To validate the results from network pharmacology analyses, we performed a series of cell biological functional assays using a model cell line. First, GES-1 cells induced by MNNG were used as a model cell line (Model).^30,31) As shown in Fig. 5A, cell viability was inhibited with increasing concentrations of MNNG. At an MNNG concentration of 20 µM, GES-1 cells maintained their viability while exhibiting increased invasion and migration abilities (Fig. 5C). Consistent with previous studies, gastric mucosa epithelial cells can develop precancerous lesions such as CAG following treatment with MNNG. Then, we detected the effects of luteolin, quercetin, and hederagenin on the cell viability of the model cells. The experimental results showed that luteolin, quercetin, and hederagenin decreased cell viability in the MNNG-induced GES-1 cell in vitro model, with respective IC₅₀ values of 646.6 nM, 469.7 nM, and 114.5 nM (Fig. 4B). We also observed that these active compounds inhibited the invasion and migration of the model cells (Figs. 5C, 5D). Additionally, we determined whether luteolin, quercetin, and hederagenin were capable of affecting the expression of the predicted proteins. The immunofluorescent assay showed that the expression levels of TLR4, pNF-κB, ppNF-κB, and COX-2 were higher in the model group compared with the control group, whereas the expression levels in the luteolin, quercetin, and hederagenin group were more similar to those in the control group than to the model group (Fig. 5E). Furthermore, the mRNA expression levels of both TLR4 and NF-κB were significantly reduced in MNNG-induced GES-1 cells following treatment with the active compounds, observed at both short-term (1 and 6 h) and long-term (24 h) (Fig. 5F). These findings indicate that the active compounds exert sustained suppressive effects on TLR4 and NF-κB expression, thereby achieving long-term inhibition of the TLR4/NF-κB pathway. Therefore, luteolin, quercetin, and hederagenin could inhibit cell viability, invasion, migration, and inflammation levels in MNNG-induced model cells.

Fig. 5. The Active Components of HWD Inhibit Proliferation, Migration, Invasion, and Inflammation Level in MNNG-Induced GES-1 Cells

(A) GES-1 cells show a dose-dependent proliferation when treated with MNNG. The effect of MNNG on the proliferation of GSE-1 cells was measured using the MTT assay. (B) The cell viability in the MNNG-induced model was assessed after treatment with active components for 48 h. Luteolin, quercetin, and hederagenin decreased cell viability, with the respective IC₅₀ values of 646.6, 469.7, and 114.5 nM. (C, D) The migration and invasion capacities were assessed by the Transwell assay after treatment with luteolin (600 nM), quercetin (400 nM), and hederagenin (100 nM). Scale bar: 50 µm. (E) Effect of active components on the expressions of TLR4, NF-κB, ppNF-κB, and COX2 protein in the MNNG-induced model were analyzed. Scale bar: 50 µm. (F) Relative mRNA levels of TLR4 and NF-κB were determined by RT-PCR after treatment with luteolin (600 nM), quercetin (400 nM), and hederagenin (100 nM).

DISCUSSION

CAG is associated with chronic inflammation, which promotes the transformation of normal gastric mucosa to atrophic gastritis. The inflammatory microenvironment of the gastric mucosa plays a crucial role in the progression of CAG, making anti-inflammatory treatment one of the important intervention strategies. Currently, TCM has been proven to inhibit inflammatory responses and effectively improve the inflammatory microenvironment of gastric mucosa.

HWD can significantly improve the clinical symptoms of CAG. It is formulated primarily as Si-Ni-San, complemented by HQ and HTG to strengthen the spleen and supplement qi, HP to promote qi circulation, SQ, SL, and EZ to activate blood circulation and resolving stasis, and BHSSC, BZL to detoxification and anticancer effects. In terms of TCM syndrome differentiation, HWD emphasizes

strengthening the spleen and supplementing qi (Jianpi Yiqi),” and “resolving stasis and detoxifying (Huayu Jiedu).” Network pharmacology analysis has revealed that the 165 chemical compounds in HWD primarily act on the TLR signaling pathway. Through molecular docking, we identified that luteolin, quercetin, and hederagenin demonstrate strong binding affinity with TLR4 and NF-κB proteins.

TLRs are type I transmembrane glycoproteins that activate signaling pathways such as NF-κB and MAPK, which regulate the expression of cytokines and chemokines, including TNF-α, IL-2, IL-6, IL-12, and COX2, either in a MyD88-dependent or a MyD88-independent manner.^23,24) Previous studies have shown that miR-365 targets and inhibits TLR4 expression, thereby reducing IRF3 phosphorylation and YAP-mediated CDX2 transcription, which in turn prevents the progression of CAG.^5,25) The TLR signaling pathway may server as a promising target for CAG treatment.

In terms of experimental validation, we observed a significant improvement in gastric mucosal lesions before and after treatment with HWD in CAG patients. The TLR4/NF-κB mediated inflammatory response in the gastric mucosa was also significantly suppressed following HWD treatment. Furthermore, we utilized MNNG to induce human gastric mucosal GES-1 cells, establishing a cell model for CAG.^30,31) In this model, luteolin, quercetin, and hederagenin could inhibit the proliferation, migration, and invasion of MNNG-induced GES-1 cells, and effectively suppressed the expression of TLR4, NF-κB, and COX-2 proteins. Notably, we identified the anti-inflammatory active compounds in HWD, providing a valuable foundation for further drug development.

In a conclusion, HWD, a TCM formula, demonstrated efficacy in inhibiting the inflammatory responses in the gastric mucosa of CAG patients. By integrating TCM network pharmacology with experimental validation, our study identified that luteolin, quercetin, and hederagenin in HWD can inhibit the expression of the inflammatory factor COX-2 through the suppression of the TLR4/NF-κB signaling pathway. This effectively elucidates the underlying biological molecular mechanisms of HWD in the treatment of CAG. This study provides a scientific approach to further guide the experimental verification and clinical application of HWD. In future research, we will conduct additional experiments to explore the anti-inflammatory mechanisms of luteolin, quercetin, and hederagenin, offering theoretical guidance for the treatment of CAG.

Acknowledgments

This work was supported by the National Nature Science Foundation of China (Grant No. 82404576); the Project of Administration of Traditional Chinese Medicine of Guangdong Province of China (Grant No. 20231023); and the Guangdong Provincial Key R&D Program (Grant No. 2020B1111100011).

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

This article contains supplementary materials.

REFERENCES

• 1)  Shah SC, Piazuelo MB, Kuipers EJ, Li D. AGA clinical practice update on the diagnosis and management of atrophic gastritis: expert review. Gastroenterology, 161, 1325–32.e7 (2021).
• 2)  Li J, Pan J, Xiao D, Shen N, Wang R, Miao H, Pu P, Zhang H, Yv X, Xing L. Chronic atrophic gastritis and risk of incident upper gastrointestinal cancers: a systematic review and meta-analysis. J. Transl. Med., 22, 429 (2024).
• 3)  Neumann WL, Coss E, Rugge M, Genta R. Autoimmune atrophic gastritis--pathogenesis, pathology and management. Nat. Rev. Gastroenterol. Hepatol., 10, 529–541 (2013).
• 4)  Zheng SY, Zhu L, Wu LY, Liu HR, Ma XP, Li Q, Wu MD, Wang WJ, Li J, Wu HG. Helicobacter pylori-positive chronic atrophic gastritis and cellular senescence. Helicobacter, 28, e12944 (2023).
• 5)  Jia J, Zhao H, Li F, Zheng Q, Wang G, Li D, Liu Y. Research on drug treatment and the novel signaling pathway of chronic atrophic gastritis. Biomed. Pharmacother., 176, 116912 (2024).
• 6)  Yang T, Wang R, Liu H, Wang L, Li J, Wu S, Chen X, Yang X, Zhao Y. Berberine regulates macrophage polarization through IL-4-STAT6 signaling pathway in Helicobacter pylori-induced chronic atrophic gastritis. Life Sci., 266, 118903 (2021).
• 7)  Liu K, Huang H, Xiong M, Wang Q, Chen X, Feng Y, Ma H, Chen W, Li X, Ye X. IL-33 accelerates chronic atrophic gastritis through AMPK-ULK1 axis mediated autolysosomal degradation of GKN1. Int. J. Biol. Sci., 20, 2323–2338 (2024).
• 8)  Zhou W, Zhang H, Wang X, Kang J, Guo W, Zhou L, Liu H, Wang M, Jia R, Du X, Wang W, Zhang B, Li S. Network pharmacology to unveil the mechanism of Moluodan in the treatment of chronic atrophic gastritis. Phytomedicine, 95, 153837 (2022).
• 9)  Huang M, Li S, He Y, Lin C, Sun Y, Li M, Zheng R, Xu R, Lin P, Ke X. Modulation of gastrointestinal bacterial in chronic atrophic gastritis model rats by Chinese and west medicine intervention. Microb. Cell Fact., 20, 31 (2021).
• 10)  Xie D, Wu C, Wang D, Nisma Lena BA, Liu N, Ye G, Sun M. Wei-fu-chun tablet halted gastric intestinal metaplasia and dysplasia associated with inflammation by regulating the NF-kB pathway. J. Ethnopharmacol., 318 (Pt. B) 117020 (2024).
• 11)  Guo DA, Lu A, Liu L. Modernization of traditional Chinese medicine. J. Ethnopharmacol., 141, 547–548 (2012).
• 12)  Yang L, Liu X, Zhu J, Zhang X, Li YL, Chen J, Liu H. Progress in traditional Chinese medicine against chronic gastritis: from chronic non-atrophic gastritis to gastric precancerous lesions. Heliyon, 9, e16764 (2023).
• 13)  Liu Y, Huang T, Wang L, Wang Y, Liu Y, Bai J, Wen X, Li Y, Long K, Zhang H. Traditional Chinese Medicine in the treatment of chronic atrophic gastritis, precancerous lesions and gastric cancer. J. Ethnopharmacol., 337, 118812 (2025).
• 14)  Cai Y, Xu B, Zhou F, Wu J, Li S, Zheng Q, Li Y, Li X, Gao F, Dong S, Liu R. Si-Ni-San ameliorates chronic colitis by modulating type I interferons-mediated inflammation. Phytomedicine, 84, 153495 (2021).
• 15)  Zhang P, Zhang D, Zhou W, Wang L, Wang B, Zhang T, Li S. Network pharmacology: towards the artificial intelligence-based precision traditional Chinese medicine. Brief. Bioinform., 25, bbad518 (2023).
• 16)  Zhao L, Zhang H, Li N, Chen J, Xu H, Wang Y, Liang Q. Network pharmacology, a promising approach to reveal the pharmacology mechanism of Chinese medicine formula. J. Ethnopharmacol., 309, 116306 (2023).
• 17)  Ma M, Liang L, Lin M, Luo C, Deng X, Yu C. Identification of bioactive compounds and potential mechanisms of fuzi in the treatment of ulcerative colitis by integrating network pharmacology and experimental validation. Biol. Pharm. Bull., 48, 246–261 (2025).
• 18)  Li J, Wu K, Zhong Y, Kuang J, Huang N, Guo X, Du H, Guo C, Li R, Zhu X, Zhang T, Gong L, Sheng L, Sun R. Si-Ni-SAN ameliorates obesity through AKT/AMPK/HSL pathway-mediated lipolysis: network pharmacology and experimental validation. J. Ethnopharmacol., 302 (Pt. A), 115892 (2023).
• 19)  Wang H, Wu R, Xie D, Ding L, Lv X, Bian Y, Chen X, Nisma Lena BA, Wang S, Li K, Chen W, Ye G, Sun M. A combined phytochemistry and network pharmacology approach to reveal the effective substances and mechanisms of Wei-Fu-Chun Tablet in the treatment of precancerous lesions of gastric cancer. Front. Pharmacol., 11, 558471 (2020).
• 20)  Zhan XZ, Wei TH, Yin YQ, Xu JQ, Yu H, Chen XL, Kong XT, Sun SL, Li NG, Ni HW. Determination and mechanism of Xiao-Ai Jie-Du decoction against diffuse large B-cell lymphoma: in silico and in vitro studies. J. Ethnopharmacol., 319, 117271 (2024).
• 21)  Pinzi L, Rastelli G. Molecular docking: shifting paradigms in drug discovery. Int. J. Mol. Sci., 20, 4331 (2019).
• 22)  Wang H, He H, Zhang T, Jiang J. Application of reverse docking in the research of small molecule drugs and traditional Chinese medicine. Biol. Pharm. Bull., 45, 19–26 (2022).
• 23)  Kawai T, Akira S. The role of pattern-recognition receptors in innate immunity: update on Toll-like receptors. Nat. Immunol., 11, 373–384 (2010).
• 24)  Kaur A, Baldwin J, Brar D, Salunke D, Petrovsky N. Toll-like receptor (TLR) agonists as a driving force behind next-generation vaccine adjuvants and cancer therapeutics. Curr. Opin. Chem. Biol., 70, 102172 (2022).
• 25)  Zhang T, Zhang K, Ji K, Zhang C, Jiang Y, Zhang Q, Tian Z, Wang X, Zhang M, Li X. microRNA-365 inhibits YAP through TLR4-mediated IRF3 phosphorylation and thereby alleviates gastric precancerous lesions. Cancer Cell Int., 20, 549 (2020).
• 26)  Wang X, Liu Q. Dexmedetomidine relieved neuropathic pain and inflammation response induced by CCI through HMGB1/TLR4/NF-κB signal pathway. Biol. Pharm. Bull., b21-00329 (2021).
• 27)  Luo P, Chen T, Huang S, Peng F, Huang Y, Pan W. Potential COX-2 inhibitors modulating NF-κB/MAPK signaling pathways: Design, synthesis and evaluation of anti-inflammatory activity of Pterostilbene-carboxylic acid derivatives with an oxime ether moiety. Bioorg. Med. Chem., 118, 118022 (2025).
• 28)  Hold GL, Rabkin CS, Chow WH, Smith MG, Gammon MD, Risch HA, Vaughan TL, McColl KE, Lissowska J, Zatonski W, Schoenberg JB, Blot WJ, Mowat ASG, Fraumeni JF Jr, El-Omar EM. A functional polymorphism of toll-like receptor 4 gene increases risk of gastric carcinoma and its precursors. Gastroenterology, 132, 905–912 (2007).
• 29)  Zhang H, Liu H. Mechanism of Weiwei granules in the treatment of chronic active Helicobacter pylori gastritis with atrophy based on the TLR4/NF-κB/COX-2 inflammatory signaling pathway. Histol. Histopathol., 39, 761–769 (2024).
• 30)  Xu J, Shen W, Pei B, Wang X, Sun D, Li Y, Xiu L, Liu X, Lu Y, Zhang X, Yue X. Xiao Tan He Wei Decoction reverses MNNG-induced precancerous lesions of gastric carcinoma in vivo and vitro: regulation of apoptosis through NF-κB pathway. Biomed. Pharmacother., 108, 95–102 (2018).
• 31)  Xie H, Hu M, Yu J, Yang X, Li J, Yu N, Han L, Peng D. Mass spectrometry-based metabolomics reveal Dendrobium huoshanense polysaccharide effects and potential mechanism of N-methyl-N’-nitro-N-nitrosoguanidine-induced damage in GES-1 cells. J. Ethnopharmacol., 310, 116342 (2023).