Abstract Background Precancerous lesions of gastric cancer (PLGC) represent critical stages in gastric cancer progression, with a high risk of malignancy. Current treatments, such as Helicobacter pylori eradication, show limited efficacy in reversing precancerous molecular changes. Zuojin Pill (ZJP), a traditional Chinese medicine, has demonstrated potential for treating digestive disorders and may offer a promising approach for PLGC intervention. Objective This study aims to investigate the therapeutic effects and mechanisms of ZJP in treating PLGC, focusing on its active components, target pathways, and molecular interactions. By using advanced analytical techniques, we provide a scientific foundation for ZJP’s potential application in early gastric cancer intervention. Methods Using ultra-high performance liquid chromatography-quadrupole orbitrap high-resolution mass spectrometry (UPLC-Q-Orbitrap HRMS), we identified active components in ZJP. A network pharmacology approach was then applied to construct a “ZJP-compound-target-disease” network. Molecular docking and molecular dynamics simulations were conducted to analyze the stability and interactions of the main active components of ZJP with core protein targets in PLGC. Animal experiments were used to validate significant targets and pathways in vivo. Results Tangeritin, Isorhamnetin, Caffeic Acid, Azelaic Acid, and Adenosine were identified as the main active components of ZJP in the treatment of PLGC, with key targets including PIK3R1, MAPK3, SRC, JAK2, STAT3, and PIK3CA. Molecular docking and molecular dynamics simulations further confirmed the relationship between compounds and target proteins. The potential molecular mechanism of ZJP predicted by network pharmacology analysis was confirmed in PLGC rats. ZJP downregulated IL-6, TNF-α, c-myc, p-MEK1 and p-ERK1/2, effectively reversing the progression of PLGC. Conclusion ZJP can reverse MNNG-induced PLGC, potentially through inhibition of the MEK/ERK/c-myc pathway and regulation of cellular proliferation and apoptosis. Keywords: precancerous lesions gastric cancer, Zuojin Pill, network pharmacology, molecular dynamics simulations, ERK Graphical Abstract [40]graphic file with name DDDT-18-5905-g0001.jpg [41]Open in a new tab Introduction Gastric cancer (GC) is the fifth most common malignancy worldwide and the fourth leading cause of cancer-related deaths.[42]1 Before GC onset, precancerous lesions of gastric cancer (PLGC) often develop in the gastric mucosa, which include chronic atrophic gastritis (CAG), types I, II, and III intestinal metaplasia (IM), and dysplasia (Dys). The progression from CAG to IM to Dys, known as the Correa cascade, is widely recognized as a pathway for intestinal-type non-cardia GC.[43]2 The incidence of GC at each stage of the Correa cascade is higher than in the general population.[44]3 A Swedish study estimated that over 20 years, one in 85 patients with non-atrophic gastritis, one in 50 with CAG, one in 39 with IM, and one in 19 with Dys would develop GC, with no gender differences.[45]4 Although 15–30% of IM may regress after eradication of Helicobacter pylori (Hp),[46]5 it remains unclear if cancer-inducing molecular changes can be reversed. Long-term observational studies show that the overall incidence of GC does not significantly decrease after Hp eradication, especially if pathological changes have advanced.[47]6–10 Therefore, PLGC represents a critical link in the “inflammation-cancer transformation” and is a focal point for GC prevention.[48]11 The clinical manifestations of PLGC are generally non-specific, including symptoms such as epigastric pain, upper abdominal discomfort, acid reflux, bitter taste, bad breath, and reduced appetite. Notably, symptom severity does not correlate with pathology. Hp infection is a primary cause of PLGC.[49]12,[50]13 Environmental factors, immunity, genetics, age, and dietary habits also contribute to PLGC risk.[51]14 Among gastric cancers, Hp infection and other components of the gastric microbiome are significant contributors to GC development.[52]15 Since the mechanisms underlying PLGC remain unclear, proactive intervention is crucial for GC prevention. Traditional Chinese Medicine (TCM) has unique advantages in treating digestive diseases due to its molecular complexity and diversity.[53]16,[54]17 Often used alongside Western treatments, TCM can improve cure and remission rates and inhibit tumor progression.[55]18 Zuojin Pill (ZJP), a formula created by Yuan Dynasty physician Zhu Danxi, contains Coptis chinensis Franch. (Rhizoma Coptidis, HL) and Tetradium ruticarpum (Fructus Evodiae, WZY) at a 6:1 ratio. It has been widely documented in clinical practice for treating gastrointestinal symptoms such as stomach pain, nausea, and vomiting.[56]19–21 Previous studies have confirmed ZJP’s anti-inflammatory, anti-acid, and anti-ulcer effects.[57]22 HL is rich in alkaloids, notably berberine, which is used to treat cancers, digestive disorders, and metabolic, cardiovascular, and nervous system diseases.[58]23,[59]24 Berberine also protects intestinal epithelial barriers by inhibiting toxins and bacteria[60]25 and prevents proliferation and metastasis of various cancer cells.[61]26 WZY contains evodiamine, an alkaloid beneficial for cardiovascular health and for inhibiting inflammation, fibrosis, and tumor cell apoptosis.[62]27–31 ZJP exerts anti-inflammatory effects by reducing gastrin-17 and pro-inflammatory factors IL-8, TNF-α, IL-6, and IL-1β, while inhibiting TGF-β1, PI3K, and downstream signaling proteins like p-Akt and p-mTOR.[63]32 ZJP also targets Hp and protects gastric epithelial cells by modulating pathways such as JMJD2B/COX-2/VEGF and HMGB1/NF-κB.[64]33 It regulates protein stability and transcriptional activity, inhibiting tumor cell cycle progression, migration, and promoting apoptosis, showing potential in pancreatic, liver, and colorectal cancers.[65]34–36 However, the effects and mechanisms of ZJP in reversing PLGC remain to be further studied. This study employs ultra-high-performance liquid chromatography–quadrupole/electrostatic orbitrap high-resolution mass spectrometry (UPLC-Q-Orbitrap HRMS), network pharmacology, molecular docking, and molecular dynamics simulations to explore the active components of ZJP in inhibiting PLGC. Key pathways and targets are validated through in vivo experiments. Initially, UPLC-Q-Orbitrap HRMS identifies ZJP’s main active components, and the SwissTargetPrediction database identifies their targets, creating a dataset. Gene Cards, OMIM, PharmGKB, TTD, and DrugBank databases filter PLGC-related targets, which intersect with ZJP’s targets to build a “ZJP-compound-target-disease” network and a protein-protein interaction (PPI) network. Topological analysis identifies key active components and targets of ZJP for PLGC treatment. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses elucidate the association between ZJP components and signaling pathways. Molecular docking and dynamics simulations confirm the binding stability and interactions between key components and core targets.[66]37 Finally, an N-methyl-N’-nitro-N-nitrosoguanidine (MNNG)-induced PLGC rat model is used, with Western blotting verifying relevant targets and pathways. This study lays a foundation for further molecular research on ZJP’s therapeutic potential in PLGC treatment. Materials and Methods Materials and Reagents Rhizoma Coptidis (Huanglian, HL) (batch number: 1230311) and Fructus Evodiae (Wuzhuyu, WZY) (batch number: 20230101) were both purchased from China Shaohuatang National Medicine Co., Ltd. and were certified by experts to meet the 2020 edition of the Chinese Pharmacopoeia. The Ultimate 3000 ultra-high-performance liquid chromatography system and Q-Exactive high-resolution mass spectrometer from Thermo Fisher Scientific, USA, equipped with an electrospray ion source. Micro-ultracentrifuge (D3024R, Beijing Dalong Xingchuang Experimental Instrument Co., Ltd)., pure/ultrapure water system (Direct-Q^®5, Merck Millipore, Germany), methanol, and formic acid (Shanghai Aladdin Biochemical Technology Co., Ltd.) were all of chromatographic grade. Ranitidine hydrochloride capsules (batch number H11022491) were purchased from Beijing Cain Technology Co., Ltd., MNNG solution was obtained from Sigma, USA, Vitacoenzyme Tablets (batch number H41024769) were acquired from Luoyang Yilong Pharmaceutical Co., Ltd.; BCA protein assay kit was from Beyotime Institute of Biotechnology, PVDF transfer membrane from Millipore, and ECL DualVue WB Marker from GE Healthcare. Preparation of High and Low Dose ZJP Solutions HL (288 g) and WZY (48 g) were decocted twice with water (each time for 1 h, using 10 and 8 times the weight of the herbs respectively as the solvent volume).[67]38 The decoctions were combined and concentrated to about 336 mL of stock solution by atmospheric evaporation and stored at 4 °C. When used, this concentration served as the high dose (1 g/mL), and was diluted to 0.5 g/mL to serve as the low dose. UPLC-Q-Orbitrap HRMS Analysis Preparation of ZJP Sample for Testing 100 µL of high-dose ZJP solution was taken and 400 µL of methanol was added, vortexed for 10 minutes. Centrifuged at 4°C for 10 minutes at 20,000 × g, and the supernatant was collected. Mass Spectrometry Conditions A heated electrospray ionization source was used, with alternating positive and negative ion scanning. Ion source spray voltage was 3.2 kV, capillary temperature 300°C, sheath gas flow rate 40 Arb; nebulizer temperature 350°C. Scanning was done using full scan/data-dependent secondary scan (Full MS/dd-MS2) mode, primary resolution 70,000 FWHM, secondary resolution 17,500 FWHM. Chromatographic Conditions An AQ-C18 chromatography column (150mm × 2.1mm, 1.8 µm) was used, with a flow rate of 0.3 mL/min and column temperature set at 35°C. The injection volume was 5 µL. The mobile phase consisted of 0.1% formic acid in water (A) and methanol (B). The chromatographic gradient is shown in [68]Table 1. Table 1. The Elution Gradient Time (min) A (%) B (%) 0 98 2 1 98 2 5 80 20 10 50 50 15 20 80 20 5 95 27 5 95 28 98 2 30 98 2 [69]Open in a new tab Network Pharmacology Analysis Selection of ZJP Active Components Using UPLC-Q-Orbitrap HRMS, ZJP was separated and identified by injecting the test sample solution and collecting the total ion flow charts in both positive and negative ion modes. Compound Discoverer 3.3 was used for retention time correction, peak identification, and peak extraction. Substances were identified using the Thermo mzCloud online database based on secondary mass spectrometry information. Components with a score of 99 or higher were considered as ZJP active components. Prediction of ZJP-Related Targets Active components of ZJP were searched in the PubChem Database ([70]https://pubchem.ncbi.nlm.nih.gov/) to obtain their Compound CID and Canonical SMILES. Using the Swiss Target Prediction Database ([71]http://www.swisstargetprediction.ch/) with “Homo sapiens” and “Probability>0” as filters, effective action targets corresponding to ZJP active components were retrieved. The UniProt ([72]https://www.uniprot.org) protein database was used to standardize gene names of the retrieved targets to establish a dataset. Prediction and Integration of PLGC Disease Targets Using keywords “Precancerous lesions of gastric cancer”, “Gastric precancerous lesions”, and “Chronic atrophic gastritis”, disease-related targets were filtered from databases such as Gene Cards, OMIM, PharmGKB, TTD, and DrugBank. After deduplication, potential disease targets were obtained. Using R 4.3.0 (venn package) software, the intersection of the component action targets and PLGC-related targets was plotted in a Venn diagram, representing the action targets of ZJP for treating PLGC. Construction of the “ZJP-Compound-Target-Disease” and PPI Network The ZJP action targets and PLGC disease-related targets were imported into Cytoscape 3.7.0 software to construct the “ZJP-compound-target-disease” regulatory network. Co-expressed target information was uploaded to STRING v11.5 ([73]https://cn.string-db.org/), setting the species to “Homo sapiens” and interaction threshold to >0.9 while hiding disconnected nodes in the network to generate a PPI network graph. The file was downloaded in TSV format and imported into Cytoscape to construct the PPI network. Parameters such as degree centrality (DC), betweenness centrality (BC), closeness centrality (CC), eigenvector centrality (EC), local average connectivity-based method (LAC), and network centrality (NC) were used as topology importance parameters to filter out ZJP core components and key targets for visual analysis using Cytoscape. GO and KEGG Enrichment Analysis Using the Cluster Profiler package in R4.3.0, GO functional enrichment and KEGG pathway enrichment analyses were conducted. GO functional enrichment included molecular function (MF), biological process (BP), and cellular component (CC). KEGG pathway enrichment analysis elucidated the pathways of ZJP action in inhibiting PLGC. A q-value <0.05 was used as a selection criterion to identify significantly meaningful results for visualization, creating bubble charts and circle diagrams. Molecular Docking Key pharmacological substances (small molecules) among the top five by degree value were selected for molecular docking validation with core targets. Core components’ 2D structures were downloaded from the PubChem database in SDF format. The 3D structures of core target proteins were downloaded from the PDB database, and small molecule structures (in PDB format) were drawn using ChemDraw 8.0.3 for docking. Water molecules and small molecule ligands were removed using PyMOL 2.2.0 software, and hydrogen was added with AutoDockTools. Molecular docking was conducted between the receptor and ligand, and binding activity was evaluated by docking scores, with the results visualized using PyMOL 2.2.0 and Discovery Studio Client v19.1.0. Molecular Dynamics Simulation Molecular dynamics simulation was used to verify the stability and dynamic interactions between proteins and ligands after molecular docking. Proteins and small molecule ligands were separated, and small molecule force field files were generated using the antechamber tool in Ambertools software, then converted into Gromacs force field files using the acpype software tool. Small molecules used the GAFF force field, proteins used the AMBER14SB force field and TIP3P water model. The protein and small molecule ligand files were merged to construct the simulation system for the complex. Molecular dynamics simulations were carried out using Gromacs2022 software under constant temperature, constant pressure, and periodic boundary conditions. During the MD simulation, all interactions involving hydrogen bonds were constrained using the LINCS algorithm, with an integration step of 2 fs. Electrostatic interactions were calculated using the Particle-mesh Ewald method, with a cutoff value set at 1.2 nm. Non-bonded interactions had a cutoff value set at 10 Å, updated every 10 steps. The simulation temperature was controlled at 298 K using the V-rescale temperature coupling method, and pressure was controlled at 1 bar using the Berendsen method. At 298 K, a 100 ps NVT and NPT equilibration simulation was conducted, and the complex system underwent a 100 ns MD simulation, saving a conformation every 10 ps. After the simulation, the simulation trajectory was analyzed using VMD and PyMOL, and the MMPBSA binding free energy analysis between proteins and small molecule ligands was conducted using the g_mmpbsa program. Animal Experiment Validation Establishment and Identification of Animal Models Specific pathogen-free (SPF) male Sprague–Dawley rats (200 ± 20 g) were provided by Chengdu Dashuo Experimental Animals Co., Ltd. (license number: SCXK (Chuan) 2020–030) and housed in a controlled environment with a constant temperature (22±2°C), humidity (55%±10%), and a 12-hour light/dark cycle. All studies were conducted in accordance with the Principles of Laboratory Animal Care (Guide for the Care and Use of Laboratory Animals: Eighth Edition. Washington, DC: The National Academies Press). This study was also approved by the the Ethics Committee of Shaanxi University of Chinese Medicine (Ethics Number: SUCMDL20220602001-1, [74]Supplementary 1). After one week of acclimatization, 50 SD rats were randomly divided using a random number table into a normal group (10 rats) and a model group (40 rats). The normal group was given a standard diet and water, while the model group underwent MNNG compound modeling to establish a PLGC rat model. This involved free access to a light-protected aqueous solution of MNNG (100 µg/mL), alternating gavage every other day with 20% ethanol solution and hot salt water ranitidine (2.25 g/L, 50°C) (2mL/d), combined with irregular feeding patterns (one day fasting, two days feeding) continuously for 28 weeks.[75]39,[76]40 Rats were randomly selected from the model group at the end of weeks 20, 22, 24, 26, and 28 for euthanasia and immediate abdominal dissection to collect gastric tissues, which were fixed in 4% polyformaldehyde, processed into routine paraffin sections, and stained with HE to observe the degree of gastric mucosa and gland atrophy. Successful modeling was confirmed if intestinal metaplasia occurred in both rats. Animal Grouping and Intervention The PLGC model rats were randomly divided into model, Vitacoenzyme, high-dose ZJP (ZJP-H), and low-dose ZJP (ZJP-L) groups, with 8 rats in each group. The ZJP-H and ZJP-L groups were administered (through gavage) oral doses of 1 g kg^−1 and 0.5 g·kg^−1 (at a 1:1 ratio), respectively, based on the conversion of body surface area between rats and humans. From week 29, ZJP-H and ZJP-L groups were given the traditional Chinese medicine decoction prepared in Preparation of High and Low Dose ZJP Solutions (2mL/d), and the Vitacoenzyme group received a dosage of 0.30g/kg, all for a continuous 6 weeks. During this period, the model group was given an equal volume of saline by gavage, while the normal group maintained their standard diet and water intake. Gastric Tissue Pathological Changes After the last dose, rats were fasted for 24 hours before sample collection. Euthanasia was performed via intraperitoneal injection of 2% pentobarbital sodium (0.225mL/kg), and gastric tissues from the gastric body-antral junction were collected. Part of the tissues were fixed in 4% polyformaldehyde solution, embedded in paraffin, sectioned, and stained with H&E for microscopic observation of morphological changes in the gastric wall. The remaining tissues were snap-frozen in liquid nitrogen and then transferred to a −80°C freezer for storage. Western Blot 2mL of RIPA lysis buffer containing 100uL of PMSF and 100uL of phosphatase inhibitor was added to EP tubes containing 100mg of gastric tissue along with two steel balls to homogenize the tissue. The homogenate was then centrifuged at 12,000 rpm at 4°C for 15 minutes. The supernatant was transferred to a new 1.5mL EP tube for BCA protein content determination. 50μg protein samples were loaded onto SDS-PAGE gels for electrophoresis and transferred to PVDF membranes. The membranes were blocked with 5% skim milk at room temperature for 2 hours, then incubated overnight at 4°C with primary antibodies against MEK (1:1000), p-MEK (1:1000), ERK (1:1000), p-ERK (1:200), c-myc (1:1000), IL-6 (1:1000), and TNF-α (1:1000). After washing with TBST, membranes were incubated with secondary antibodies (1:5000) at room temperature for 2 hours. Development was carried out according to the ECL chemiluminescence detection kit instructions, and protein expression bands were quantitatively analyzed using Image J software. Statistical Methods Data analysis was performed using GraphPad Prism 8.0. Count data that met the normal distribution were expressed as mean±SD. For comparisons between groups, one-way analysis of variance (ANOVA) was used. When variances were homogeneous, pairwise comparisons between groups were conducted using the Least Significant Difference (LSD) method. When variances were not equal, Dunnett’s T3 test was used. A p-value of less than 0.05 was considered to indicate statistical significance. Results Active Components of ZJP Identified by UPLC-Q-Orbitrap HRMS Under both positive and negative ion detection modes, total ion chromatograms were obtained as shown in [77]Figure 1 (black for negative ion mode and red for positive ion mode). After screening, 45 active components of ZJP were identified, with their CID numbers, active components, peak times, and peak areas listed in [78]Table 2, and further details provided in [79]Supplementary 2. Figure 1. [80]Figure 1 [81]Open in a new tab Total ion current spectrum of ZJP in positive and negative ion modes. Table 2. The Active Ingredients of ZJP Compound CID Name Formula Calc. MW m/z RT [min] mzCloud Best Match Reference Ion Area: CD069_ZJW-high.raw (F434) 60961 Adenosine C10H13N5O4 267.09672 268.10397 4.619 100 [M+H]+1 4379624074 8988 D-Proline C5H9NO2 115.06359 116.07087 1.41 99.9 [M+H]+1 4556832075 6925665 L-Phenylalanine C9H11NO2 165.07898 166.08626 5.318 99.9 [M+H]+1 6059953330 1148 DL-Tryptophan C11H12N2O2 204.08979 188.0705 7.353 99.9 [M+H-NH3]+1 6859290834 6306 L-Isoleucine C6H13NO2 131.0947 132.10197 2.561 99.9 [M+H]+1 7208200682 643798 Citraconic acid C5H6O4 130.02553 129.01825 5.774 99.9 [M-H]-1 616526938.9 3026 Dibutyl phthalate C16H22O4 278.15186 279.15915 18.534 99.9 [M+H]+1 1527050316 10257 Bufotenin C12H16N2O 204.12641 205.13368 4.994 99.9 [M+H]+1 1269260347 6089 N,N-Dimethyltryptamine C12H16N2 188.13133 189.13861 7.111 99.9 [M+H]+1 416382975.4 243 Benzoic acid C7H6O2 122.03554 121.02827 9.651 99.9 [M-H]-1 117544828.3 45027869 Lariciresinol 4-O-glucoside C26H34O11 539.23668 540.24396 11.639 99.8 [M+H]+1 602967512.7 6088 N-Methyltryptamine C11H14N2 174.11574 175.12302 7.005 99.8 [M+H]+1 486975727.6 158980 L-Methionine sulfoxide C5H11NO3S 165.04606 166.05321 1.326 99.8 [M+H]+1 301101986.7 5281654 Isorhamnetin C16H12O7 316.05812 317.0654 13.75 99.8 [M+H]+1 367496041.7 938 Nicotinic acid C6H5NO2 123.03234 124.03962 2.068 99.8 [M+H]+1 587922745.9 338 Salicylic acid C7H6O3 138.03058 137.0233 13.214 99.7 [M-H]-1 509981454.1 13730 2’-Deoxyadenosine C10H13N5O3 251.10195 252.10922 4.854 99.7 [M+H]+1 183199251.7 89272 Dipropyleneglycol dibenzoate C20H22O5 342.14654 365.1358 18.181 99.7 [M+Na]+1 577651132.9 10334120 Luotonin A C18H11N3O 303.10085 286.09756 16.079 99.7 [M+H-H2O]+1 513426613.6 689043 Caffeic acid C9H8O4 180.04194 163.03891 10.155 99.7 [M+H-H2O]+1 1787564623 9799386 (1S,3R,4R,5R)-1,3,4-trihydroxy-5-{[(2E)-3-(4-hydroxy-3-methoxyphenyl)pr op-2-enoyl]oxy}cyclohexane-1-carboxylic acid C17H20O9 368.11046 369.11774 8.249 99.6 [M+H]+1 374519401.2 68077 Tangeritin C20H20O7 372.12065 373.12793 17.495 99.6 [M+H]+1 346011944.3 107541 4-Oxoproline C5H7NO3 129.04141 128.03412 2.709 99.6 [M-H]-1 2952389165 6274 L-Histidine C6H9N3O2 155.06949 156.07677 1.221 99.6 [M+H]+1 170236557.2 64961 Norharman C11H8N2 168.0687 169.07597 8.827 99.6 [M+H]+1 358155079.7 1183 Vanillin C8H8O3 152.04626 151.03899 10.521 99.6 [M-H]-1 114918122 736316 D-Pipecolinic acid C6H11NO2 129.07918 130.08646 1.728 99.6 [M+H]+1 2053165475 10349 Methylsuccinic acid C5H8O4 132.04115 131.03387 6.133 99.6 [M-H]-1 269639700.5 3469 Gentisic acid C7H6O4 108.0201 153.0183 9.378 99.5 [M+FA-H]-1 146226113.7 2776555 2,3-Dihydro-1-benzofuran-2-carboxylic acid C9H8O3 164.04629 163.03902 11.479 99.5 [M-H]-1 85167182.39 5281643 Hyperoside C21H20O12 464.09584 463.08878 12.943 99.5 [M-H]-1 974858298.8 6253 Cytarabine C9H13N3O5 243.08545 244.09268 1.459 99.5 [M+H]+1 185839815.3 333703 4-Indolecarbaldehyde C9H7NO 145.05162 144.04434 11.827 99.5 [M-H]-1 62019678.55 237332 5-Hydroxymethyl-2-furaldehyde C6H6O3 126.03197 127.03922 5.957 99.4 [M+H]+1 1679043946 33032 L-Glutamic acid C5H9NO4 147.05306 148.06033 1.305 99.4 [M+H]+1 1104763165 439227 L(-)-Pipecolinic acid C6H11NO2 129.07911 130.08638 1.464 99.4 [M+H]+1 3284595518 2266 Azelaic acid C9H16O4 188.10427 187.09691 13.383 99.3 [M-H]-1 282025853.3 135398634 Guanine C5H5N5O 134.02292 152.05675 5.376 99.3 [M+NH4]+1 1667295486 7059 6-Methylquinoline C10H9N 143.07352 144.0808 8.614 99.3 [M+H]+1 2478819120 5570 Trigonelline C7H7NO2 154.07421 138.05493 1.46 99.3 [M+H-NH3]+1 5427913459 439165 Glucose 1-phosphate C6H13O9P 278.04022 261.03693 1.637 99.3 [M+H-H2O]+1 480567714.8 10177048 (3R,5R)-1,3,5-Trihydroxy-4-{[(2E)-3-(4-hydroxy-3-methoxyphenyl)-2-prope noyl]oxy}cyclohexanecarboxylic acid C17H20O9 350.10054 349.09326 12.109 99.2 [M-H]-1 100427490.6 1001 Phenethylamine C8H11N 104.06301 122.09679 4.893 99.2 [M+NH4]+1 189669916 6760 4,7,8-trimethoxyfuro[2,3-b]quinoline C14H13NO4 259.08452 260.0918 15.068 99.2 [M+H]+1 230583133.3 6508 D-(-)-Quinic acid C7H12O6 192.06341 193.07074 1.639 99.2 [M+H]+1 1316411596 [82]Open in a new tab Network Pharmacology Results Identification of ZJP Targets for PLGC Treatment Using the Swiss Target Prediction Database for searching and filtering, and standardizing gene names, 560 targets related to ZJP components were identified. From the DrugBank database, 18 disease targets were identified, from Gene Cards 2582, from OMIM 185, from PharmGKB 312, and from TTD 1 target was identified. After deduplication, a total of 3015 PLGC disease targets were obtained ([83]Figure 2a), and intersecting these with the ZJP-related targets yielded 200 potential action targets ([84]Figure 2b). Figure 2. [85]Figure 2 [86]Open in a new tab Network pharmacology target analysis. (a) Venn diagram of PLGC targets; (b) Venn diagram of active ingredient targets of ZJP and PLGC targets; (c) Active ingredient-target network of ZJP treatment in PLGC; (d) PPI network of targets of ZJP treatment in PLGC; (e) PPI network of targets of ZJP treatment in PLGC. “ZJP-Compound-Target-Disease” Network The “ZJP-compound-target-disease” network was constructed using ZJP active components and PLGC-related targets ([87]Figure 2c). The network includes 243 nodes and 528 edges, where circular green nodes (200) represent action targets, yellow diamond nodes (42) represent active drug components, and a red “V” shaped node (1) represents ZJP. The links between nodes indicate relationships among targets. The network demonstrates that a single active component in ZJP can act on multiple PLGC targets and that multiple active components can share the same target, showcasing ZJP multi-component, multi-target characteristics. PPI Analysis of Key Targets for ZJP Treatment of PLGC Two hundred intersecting genes were uploaded to the STRING database, excluding targets with a confidence level below 0.9 and disconnected nodes, to construct a PPI network graph ([88]Figure 2d). Through topological analysis and three rounds of filtering, six core genes were identified (filtering conditions shown in [89]Table 3): PIK3R1, MAPK3, SRC, JAK2, STAT3, and PIK3CA. These genes exhibit strong interactions within the network and are considered potential core targets for ZJP treatment of PLGC ([90]Figure 2e), with their topological parameters listed in [91]Table 4. Table 3. Filter Conditions of CytoNCA Filter 1 Filter 2 Filter 3 BC 35.33 13.497 4.069 CC 0.067 0.521 0.732 DC 4 8.5 9.5 EC 0.019 0.109 0.247 LAC 1.367 4.25 6 NC 2 5.144 7.336 [92]Open in a new tab Table 4. Information of Topology Parameters of Core Targets PIK3R1 MAPK3 SRC JAK2 STAT3 PIK3CA BC 4.382 5.16 13.72 5.176 21.887 4.382 CC 0.75 0.75 0.938 0.75 1 0.75 DC 10 10 14 10 15 10 EC 0.263 0.259 0.342 0.257 0.355 0.263 LAC 6.4 6.2 8 6.4 8.133 6.4 NC 7.631 7.595 13.307 7.928 15 7.631 [93]Open in a new tab GO and KEGG Enrichment Analysis Using the R 4.3.0 software with the clusterProfiler package, GO functional enrichment analysis was performed on the 200 common targets, resulting in 3012 entries (q<0.05), including 2666 for biological processes (BP), 113 for cellular components (CC), and 233 for molecular functions (MF). The top 10 enriched categories were visualized in bubble charts and circle diagrams. KEGG pathway enrichment analysis yielded 175 entries (q<0.05), from which the top 30 enriched pathways were selected for bubble chart visualization ([94]Figure 3a and [95]b). Pathway diagrams were created for the PI3K-Akt signaling pathway, the MAPK signaling pathway, and gastric cancer-related pathways, with red nodes indicating genes at the intersection of ZJP and PLGC ([96]Figure 3c-e). Figure 3. [97]Figure 3 [98]Open in a new tab Enrichment analysis of network pharmacological targets and docking binding affinities. (a) GO enrichment analysis of core targets; (b) KEGG pathway enrichment analysis of core targets; (c) PI3K-Akt signaling pathway; (d) MAPK signaling pathway; (e) GASTRIC CANCER signaling pathway; (f) The results of the docking binding affinities. Molecular Docking Results Generally, a docking score less than −4.25 k/mol indicates some binding activity, less than −5.0 k/mol indicates good binding activity, and less than −7.0 k/mol indicates very strong binding activity. The docking binding energies are displayed in [99]Figure 3f. For instance, the binding energy between Tangeritin and MAPK3 protein is −8.7 k/mol, indicating strong binding activity. Isorhamnetin shows binding energies of −9.4 k/mol with MAPK3 and −8.7 k/mol with SRC protein, indicating strong binding activity with both proteins. Adenosine has a binding energy of −8.6 k/mol with PIK3CA protein, also showing strong binding activity. These results suggest that these active drug components can stably bind to core target molecules. The docking results visualized using PyMOL 2.2.0 software are shown in [100]Figure 4. From the figure, it can be seen that Tangeritin forms hydrogen bonds with MAPK3 at SER-225, LYS-224, GLY-374, ARG-370, and PRO-373. Isorhamnetin forms hydrogen bonds with MAPK3 at ARG-116, ASP-117, ARG-108, ASP-37, ASP-105, and GLY-39; and with SRC at PHE-307, GLU-310, and GLY-406. Adenosine forms hydrogen bonds with PIK3CA at GLU-259, ASP-258, ARG-662, VAL-166, and GLN-661. Figure 4. [101]Figure 4 [102]Open in a new tab Molecular docking results. (a) Tangeritin binds with MAPK3; (b) Isorhamnetin binds with MAPK; (c) Isorhamnetin binds with SRC; (d) Adenosine binds with PIK3CA. Molecular Dynamics Simulation Results Based on the results of molecular docking, the MAPK3-Isorhamnetin complex was selected for molecular dynamics simulation. Stability Analysis of the Complex Root Mean Square Deviation (RMSD) Analysis RMSD is a crucial metric for evaluating the similarity of molecular mimicry or protein conformation to the target structure, providing a quantitative assessment of system stability by calculating the root-mean-square deviation of atomic positions. As shown in [103]Figure 5a, the horizontal axis represents time, while the vertical axis represents the overall deviation values. The RMSD of small molecules, proteins, and complexes fluctuated during the initial phase but stabilized after 60 ns, indicating that the complexes gradually reached equilibrium. Figure 5. [104]Figure 5 Continued. Figure 5. [105]Figure 5 Molecular dynamics simulation results. (a) RMSD of the complex, protein, and small molecule ligand; (b) Rg of the complex; (c) RMSF of the protein in the complex; (d) Distance between the docking site and the small molecule on the protein surface; (e) Superimposed conformations in simulation. The black circle indicates the position of the small molecule’s overlaid conformation. (f) Buried SASA; (g) Hbond number; (h) Occupancy; (i) VDW and ELE between the small molecule and the protein; (j) Amino acid contribution to binding energy; (k) Interactions between the protein and the small molecule. [106]Open in a new tab Radius of Gyration (Rg) Analysis Rg is used to describe changes in the overall structure and is an important measure of the compactness of a protein structure. Significant changes in Rg suggest an increase in the system’s expansion. According to [107]Figure 5b, the Rg value of the complex gradually stabilizes, indicating increasing stability in its overall structure. Root Mean Square Fluctuation (RMSF) Analysis RMSF reflects the degree of flexibility by calculating the root-mean-square fluctuation of each amino acid residue in a protein, serving as a key indicator of the local dynamics within the protein structure. As shown in [108]Figure 5c, the horizontal axis represents the residues, while the vertical axis represents the fluctuation values. The peaks and valleys in the figure correspond to the maximum and minimum RMSF values, respectively. Residues at the peaks may exhibit higher flexibility and are more likely to undergo conformational changes or participate in interactions crucial for protein function. Conversely, residues at the troughs tend to be more stable Certain regions with higher RMSF values may be critical for protein function and could be involved in enzyme active sites, protein interaction interfaces, or in regulating conformational changes in the protein. Analysis of Small Molecule Binding Stability to the Protein Centroid Evolution Analysis To explore the behavior of the small molecule on the protein surface, the initial docking position of the small molecule was obtained, and the distance between the residue centroid at the initial docking site and the centroid of the small molecule was measured. Additionally, the distance between the small molecule and the entire protein centroid was analyzed. These distances help judge the binding of the small molecule to the protein. Data from [109]Figure 5d shows that the distance between the small molecule and the protein, as well as with the initial binding site, remains stable throughout the simulation, indicating good stability of the binding. Notably, there was a slight fluctuation in the distance to the binding site towards the end of the simulation (95–100 ns), possibly reflecting minor adjustments in the binding site of the small molecule. Analysis of Small Molecule Binding Sites By superimposing the simulated conformations of the small molecule and the protein, it can be observed, as shown in [110]Figure 5e, that the small molecule in the complex remains near the initial binding site with a high degree of overlap. This phenomenon indicates that the small molecule maintains a stable binding state with the protein throughout the simulation. Analysis of Buried Solvent-Accessible Surface Area (Buried SASA) Buried SASA reflects the size of the binding interface between the small molecule and the protein, revealing their binding state. Analysis of Buried SASA showed initial fluctuations, mainly due to conformational adjustments during the binding process of the small molecule to the protein. As shown in [111]Figure 5f, the Buried SASA value gradually stabilizes as the simulation progresses, indicating that the small molecule reaches a stable binding state with the protein after conformational adjustments. Analysis of Hydrogen Bond Interactions Between Small Molecule and Protein Evolution of Hydrogen Bond Numbers Hydrogen bonds are one of the key forces in the binding of proteins to ligands and are closely related to electrostatic interactions, thus reflecting the strength of electrostatic interactions. According to data in [112]Figure 5g, the number of hydrogen bonds between the small molecule and the protein is relatively low and stable, generally maintained between 1 and 5. Additionally, a decrease in hydrogen bond numbers was observed towards the end of the simulation, indicating minor conformational adjustments in the small molecule during the binding process. Analysis of Hydrogen Bond Occupancy (Frequency) To gain a deeper understanding of which protein residues form hydrogen bonds with the small molecule and the stability of these hydrogen bonds, the occupancy (ie, the frequency of hydrogen bond formation) of hydrogen bonds between the small molecule and the protein was analyzed. As shown in [113]Figure 5h, the left side lists the acceptors, donors, and their occupancy rates, while the right side shows the formation frequency of the corresponding hydrogen bonds, with the density of the lines reflecting the frequency of hydrogen bond formation. The graph indicates that the hydrogen bonds formed between the small molecule and the protein generally show high stability, particularly bonds such as those between Ligand and 37ASP, 116ARG and Ligand, 117ASP and Ligand. Interaction Analysis of Small Molecule Binding to Protein Analysis of Electrostatic and Van der Waals Interactions Ignoring solvation effects, the Van der Waals forces (VDW) and electrostatic interaction forces (ELE) between the small molecule and the protein in the complex were calculated, and their changes during the simulation were analyzed. VDW represents Van der Waals interaction forces, ELE represents electrostatic interactions, and the Binding value, which is the sum of VDW and ELE, is used to represent the binding energy of the small molecule to the protein without considering solvation effects. According to data in [114]Figure 5i, VDW and ELE in the complex gradually stabilize during the simulation, reflecting the stable binding between the small molecule and the protein. Binding Energy Analysis Considering solvation energy, a comprehensive analysis of parameters such as RMSD, Rg, Distance, Buried SASA, and interaction energy was conducted, and a trajectory of the complex in a stable state was selected. Molecular Mechanics-Poisson Boltzmann Surface Area (MM-PBSA) was applied to calculate related energy terms of binding energy. Results are shown in the table: ΔE[ele] represents the electrostatic interaction energy between the small molecule and the protein, ΔE[vdw] represents the Van der Waals interaction energy, ΔE[pol] represents polar solvation energy, ie, electrostatic potential energy, and ΔE[nonpol] represents nonpolar solvation energy, indicating hydrophobic interactions. ΔE[MMPBSA] is the sum of these energy terms, and ΔG[bind] is the sum of ΔE[MMPBSA] and -TΔS (since the calculation of -TΔS has a high degree of uncertainty, it is usually not considered in the comparison of binding energies, and ΔE[MMPBSA] is directly used to represent binding energy). According to the analysis in [115]Table 5, the Van der Waals interaction ΔE[vdw] in the complex is significantly higher than the electrostatic interaction ΔE[ele] and the hydrophobic interaction ΔE[nonpol], with ΔE[vdw] being five times that of ΔE[ele] and 8.4 times that of ΔE[nonpol]. This indicates that Van der Waals forces are the dominant forces, electrostatic interactions play a secondary role, and hydrophobic interactions are complementary. Therefore, the binding energy of the small molecule to the protein is high, showing strong affinity. Table 5. Binding Energy and Its Components at Stable State (Unit: kJ/Mol) Complex ΔE[vdw] ΔE[ele] ΔE[pol] ΔE[nonpol] ΔE[MMPBSA] -TΔS ΔG[bind]* Protein-ligand −156.673±0.68 −31.57±0.916 153.003±3.575 −18.506±0.238 −53.745±3.871 25.293±1.34 −28.452±2.608 [116]Open in a new tab Notes: *ΔG[bind]=ΔE[vdw]+ΔE[ele]+ΔE[pol]+ΔE[nonpol]-TΔS. Residue Contribution Analysis By decomposing the binding energy ΔE[MMPBSA], the contribution of each amino acid residue to the total binding energy was analyzed, thus assessing key amino acids in the protein. As shown in [117]Figure 5j, residues that contribute significantly to the overall binding energy include 375VAL, 107LEU, 61ALA, 119TYR, among others. These residues are key amino acids in the binding of the small molecule and play a significant role in maintaining the stable binding between the small molecule and the protein. Structural Analysis The last frame of the simulation was selected for structural and interaction analysis. As shown in [118]Figure 5k, ARG41 and GLY39 in the protein form hydrogen bonds with the small molecule. Additionally, PRO40 and VAL38 form carbon-hydrogen bond interactions with the small molecule, while ILE107 forms two types of hydrophobic interactions with the small molecule: Pi-Alkyl and Amide-Pi Stacked. Meanwhile, PRO373, TYR42, VAL375, TYR119, and other amino acids engage in Van der Waals interactions with the small molecule, collectively maintaining the stable binding of the small molecule to the protein. Animal Experiment Validation Results for ZJP Treatment of PLGC Improvement of General Conditions in PLGC Rats by ZJP Before establishing the PLGC rat model, there was no significant difference in body mass among the groups (P>0.05). As the modeling time increased, compared to the normal group, the model rats showed a gradual reduction in body weight (P<0.01) and exhibited symptoms such as dull fur, reduced eating, and decreased activity. After drug intervention, compared to the model group, rats in the ZJP and Vitacoenzyme groups showed improvements in body weight, eating habits, and overall vitality to varying degrees (P<0.01), with the high-dose ZJP group showing the most significant improvement (P<0.01) ([119]Figure 6). Figure 6. [120]Figure 6 [121]Open in a new tab Effects of ZJP on Body Weight of PLGC Rats (mean ± SD, n=8). **compared to normal group (P<0.01); #compared to model group (P<0.05); ##compared to model group (P<0.01). (a) Changes in rat body weight during model establishment; (b) Changes in rat body weight during drug treatment period. Effects of ZJP on Gastric Tissue Pathology in PLGC Rats In the normal group, the gastric mucosa was clear, the mucosal glands were abundant and tightly arranged, and there was no infiltration of inflammatory cells. In the model group, there were observable losses of epithelial cells, thinning of the gastric mucosal layer, reduction in the number of intrinsic glands, and in some areas, intestinal metaplasia and infiltration of inflammatory cells were visible. After treatment with ZJP and Vitacoenzyme, all groups showed significant improvement in gastric mucosal damage. The ZJP-L group still had some damage and inflammatory signs, whereas the ZJP-H and Vitacoenzyme groups showed relatively organized gland structures, less infiltration of inflammatory cells, more normal cell morphology, and improved mucosal lesions without any goblet cells (see [122]Figure 7). Figure 7. [123]Figure 7 [124]Open in a new tab HE staining of gastric tissue in Normal, Model, ZJP-L, ZJP-H, and Vitacoenzyme groups (×200, mean ± SD, n=8). Impact of ZJP on Gastric Tissue Cytokine Levels in PLGC Rats Compared to the normal group, the model group showed significantly increased levels of inflammatory cytokines TNF-α and IL-6 in gastric tissues (P<0.01). Compared to the model group, all treatment groups showed a significant reduction in TNF-α and IL-6 levels (P<0.01), with the ZJP-H group showing the most pronounced effects (see [125]Figure 8a–c). Figure 8. [126]Figure 8 [127]Open in a new tab Changes in TNF-α, IL-6, MEK1, p-MEK1, ERK1/2, p-ERK1/2 and c-myc levels in gastric tissues of different groups (mean ± SD, n=8). (a) Representative Western blot images of the TNF-αand IL-6 levels; (b and c) Quantification of the TNF-αand IL-6 levels; (d) Representative Western blot images of the MEK1, p-MEK1, ERK1/2, p-ERK1/2 and c-myc levels; (e-i) Quantification of the p-MEK1, p-ERK1/2, p-MEK1/MEK1, p-ERK1/2 / ERK1/2 and c-myc levels. **** P < 0.0001; *** P < 0.001; ** P < 0.01; * P < 0.05; ns P > 0.05. Changes in p-MEK1, p-ERK1/2, Ratios of p-MEK1/MEK1, p-ERK1/2/ERK1/2 and c-Myc Protein Levels in Gastric Tissues of Different Groups [128]Figure 8d–i shows Western-blot images of p-MEK1, p-ERK1/2, Ratios of p-MEK1/MEK1, p-ERK1/2/ERK1/2 and c-myc proteins in the normal, model, Vitacoenzyme, low-dose ZJP, and high-dose ZJP groups. Compared to the normal group, the expression of p-MEK1 and p-ERK1/2 was significantly upregulated in the model group’s gastric tissues (P<0.01). After drug intervention, the levels of p-MEK1/MEK1, p-ERK1/2/ERK1/2 decreased (P<0.05), with the most significant reduction in the high-dose ZJP group (P<0.01). Additionally, compared to the normal group, the level of c-myc protein in the gastric tissues of the model group was significantly elevated (P<0.01). Compared to the model group, all treatment groups showed a significant decrease in c-myc protein levels in gastric tissues (P<0.01), with the ZJP-H group showing the most pronounced effects. Discussion Network pharmacology elucidates the mechanism of drugs through the intersection and interaction of multi-dimensional targets, aligning well with the holistic view of traditional Chinese medicine. Combined with modern pharmacological techniques such as liquid chromatography-mass spectrometry (LC-MS), it enables more accurate assessment of the rationality and interaction of traditional Chinese medicine. This study leverages LC-MS and network pharmacology to thoroughly explore the anti-inflammatory, anti-cancer factors, and signaling pathways influenced by ZJP in PLGC. Molecular docking and molecular dynamics simulation techniques further verify the spatial and energy match between targets and components, providing a theoretical basis for the core components of ZJP in reversing PLGC, and corroborating the effective components’ role in improving PLGC through in vivo experiments. This research, utilizing LC-MS and network pharmacology, identifies active components in ZJP such as Tangeritin, Isorhamnetin, Caffeic acid, Azelaic acid, and Adenosine as the main pharmacologically active components for treating PLGC. Tangeritin, for instance, can inhibit inflammation and extracellular matrix degradation in cartilage cells and animal models through the Nrf2/NF-κB and MAPK/NF-κB pathways, used to treat osteoarthritis.[129]41 Isorhamnetin, widely found in fruits and herbs, has high pharmacological activity,[130]42 significantly inhibiting platelet aggregation induced by collagen and TRAP-6, thereby protecting the cardiovascular system and preventing related diseases.[131]43,[132]44 It also affects cytokines and kinase expressions by regulating pathways like PI3K/Akt/PKB, NF-κB, and MAPK, conferring anti-inflammatory, antioxidant, organ-protective, and anti-obesity effects.[133]42 It can inhibit tumor cell proliferation and induce apoptosis by modulating tumor suppressor genes, oncogenes, and signaling pathways,[134]45 showing anti-tumor activity against gastric cancer, liver cancer, and colorectal cancer.[135]46–48 Caffeic acid, a naturally occurring phytochemical, has been proven to possess robust antimicrobial properties.[136]49 Several studies demonstrate that these main active components exhibit varying degrees of anti-inflammatory, antioxidant, and anti-cancer effects.[137]50–54 Molecular dynamics simulations have shown that Isorhamnetin binds stably at the initial binding site, forming numerous hydrogen bonds with MAPK3, indicating high overall stability. The Gibbs free energy of binding between the two is −28.452±2.608 kJ/mol, suggesting a high affinity of Isorhamnetin for MAPK3. In the complex, Van der Waals interactions play a dominant role over electrostatic and hydrophobic interactions, supporting its potential as a key pharmacologically active component of ZJP for treating PLGC. Topological analysis of the PPI diagram identified core targets such as PIK3R1, MAPK3, SRC, JAK2, STAT3, and PIK3CA in the prevention and treatment of PLGC by ZJP. Coupled with molecular docking and dynamics, it is suggested that MAPK3 has the best affinity with the main pharmacological components and is part of the MAPK/ERK signaling pathway, aligning with KEGG enrichment analysis results, making it a potential significant target and pathway in the treatment of PLGC. The MAPK pathway, or mitogen-activated protein kinase, is a crucial signaling pathway widely present in various human cells. The MAPK protein family plays an important role in regulating intracellular gene transcription and protein expression, affecting physiological processes like cell proliferation, differentiation, and apoptosis. It is closely associated with cancer, involved in the onset and development of various tumors such as colorectal cancer, lung cancer, cervical cancer, gastric cancer, hypertension, non-ossifying fibromas, ovarian tumors, rectal cancer, neuroblastoma, nasopharyngeal carcinoma, and renal cysts. Activation of MAPK proteins in these diseases often impacts cell functions, potentially a crucial pathogenic mechanism.[138]55 As a classical pathway for GC and PLGC, the MAPK/ERK signaling pathway primarily follows this sequence: Ras→activates Raf→activates MEK→activates ERK1/2→activates RSK or transcription factors to produce biological effects (as shown in [139]Figure 9), which can be activated upstream by growth factors and downstream can regulate nuclear transcription factors like c-fos, Elk-1, c-myc, etc., thereby participating in inflammatory responses and cell proliferation, playing a significant role in the development of gastric cancer.[140]56 Previous studies have shown that activation of the MAPK/ERK pathway is involved in TGF-β1-induced epithelial-mesenchymal transition (EMT),[141]57 now included in molecular targeted therapy studies for gallbladder cancer.[142]58 An animal study confirmed that inhibiting the pERK1/2 signal in a gastric inflammation model created in C57BL/6 mice with ethanol can reduce MMP-10 expression induced by ethanol in gastric inflammation, improve gastric mucosal damage, and alleviate inflammatory responses.[143]59 Studies have also observed significantly increased activation of STAT3 and ERK in Helicobacter pylori-dependent gastritis, with the expression of pro-inflammatory cytokines activated by the STAT3 and ERK pathways related to the progression of GC.[144]60 Multiple studies confirm that inhibiting the ERK pathway displays strong anti-inflammatory activity,[145]61,[146]62 and downregulating ERK1/2 and PI3K/Akt signal pathways to suppress tumor activity by expressing RhoA/MMP-9,[147]63 with tMMP9 as a secreted protein playing an important role in local proteolysis of the extracellular matrix, participating in promoting neutrophil-mediated gastric tissue inflammatory responses.[148]64 c-Myc, a downstream protein of the ERK pathway, is one of the first oncogene products discovered related to tumor cell proliferation and apoptosis, capable of malignant transformation. It binds with DNA in the nucleus to promote protein expression, leading to cell growth; it can also act as a regulatory protein, controlling the transition of cells from G0/G1 to S phase, and regulating cell proliferation and apoptosis, playing a significant role in the transformation from normal cell growth to tumor growth.[149]65,[150]66 Experimental results show that the expression rate of C-myc increases progressively from groups with atrophic gastritis, superficial gastritis, gastric ulcers, to gastric cancer groups, suggesting that high expression of the C-myc gene might lead to cell apoptosis.[151]67 Additional studies have shown that over 61% of patients with gastric cancer have significantly higher levels of ERK than normal individuals,[152]68 and the abnormal signaling pathway of MEK/ERK/c-myc is closely linked to the proliferation and apoptosis of gastric cancer cells. Further observation revealed that the overactivation of the MEK/ERK/c-myc pathway plays a key role in tumor onset and development, not only regulating the cancer process but also controlling the invasive capabilities and metastatic behavior of tumor cells.[153]69 By inhibiting this pathway, it is possible to stop the proliferation of gastric cancer cells and promote apoptosis, interrupting the malignant cycle of tumor cell growth and survival, providing new targets and directions for treatment. Figure 9. [154]Figure 9 [155]Open in a new tab MAPK/MEK/ERK signaling pathway. Thus, this study further used Western Blot to examine key indicators on the MAPK/ERK signaling pathway and downstream c-myc protein and inflammatory cytokines TNF-α, IL-6 expressions. Results indicate that ZJP can significantly inhibit the activation of the MEK/ERK/c-myc pathway, downregulate the expression of inflammatory cytokines TNF-α, IL-6 in gastric tissues, improve local symptoms of the gastric mucosa, and reverse PLGC. The findings of ZJP’s reversal effects on PLGC suggest that traditional Chinese medicine holds certain advantages in the early prevention and treatment of malignant diseases, providing new insights and methods for treating pre-cancerous gastric lesions with traditional Chinese medicine, and offering new experimental evidence for the clinical application of ZJP, expanding the research field of modernization of traditional Chinese medicine. However, this study still has certain limitations. First, although the main pharmacologically active components in ZJP were explored using LC-MS, there is still a lack of experimental validation of the exact therapeutic effects of individual components, and further cell experiments are needed to improve the level of evidence. Second, as a clinically applicable drug, ZJP should undergo related clinical trials in the future to increase its evidence base. Due to certain limitations, we were unable to validate the MAPK and PI3K-AKT pathways in this study. These pathways will also be an area of focus for us in subsequent studies. Lastly, ZJP could serve as a promising therapeutic agent for early intervention in human gastric lesions. Future clinical studies are needed to further validate its efficacy, safety, and optimal dosage, supporting the application of ZJP in gastric cancer prevention and advancing the modernization of traditional Chinese medicine in oncology. Conclusion ZJP can significantly improve the pathological damage in gastric tissues of rats with MNNG-induced PLGC, and reduce the levels of cytokines IL-6 and TNF-α, achieving a reversal effect on PLGC. The mechanism is likely related to the inhibition of the MEK/ERK/c-Myc pathway and the regulation of cell proliferation and apoptosis. These interactions have been validated through network pharmacology and animal experiments. These findings not only provide a new drug candidate for the clinical management of PLGC, but also highlight an avenue of the potential for ZJP as a therapeutic agent in human gastric lesions. Acknowledgments