GK was purchased from MCE co, Ltd. (Shanghai, China). Cell Signaling Technology provided the primary antibodies used in this study: anti-N-cadherin (#13116), anti-Vimentin (#5741), anti-Snail (#3879), anti-p-AKT (#4060), anti-AKT (#4691), anti-GSK-3β (#12456),anti-p-GSK-3β(#5558),anti-actin (#4970), and anti-rabbit IgG HRP (#7074) (Danvers, MA, United States). Anti-c-Myc (ab32072) purchased from Abcam (Cambridge, United Kingdom)). Gibco provided the trypsin-EDTA used in this study (California, United States). CWBio supplied the RIPA Lysis Buffer and the ECL Western blot kit (Taizhou, Jiangsu, China). Hyclone was consulted to acquire the minimum necessary medium (MEM) (Logan, UT, United States). The Beyotime Cell Counting Kit-8 was purchased (Shanghai, China).

We used the PharmMapper (http://www.lilab-ecust.cn/pharmmapper) platform to predict GK targets, backed by a sizable in-house pharmacophore database extracted from all targets in TargetBank, DrugBank, BindingDB, and PDTD, providing storage for and access to >7,000 receptor-based pharmacophore models (covering information on 1,627 drug targets, 459 of which are human protein targets). The top N possible drug targets and the alignment pose for each molecule were generated after the user uploaded molecules (in Tripos Mol2 or MDL SDF format) with the best mapping posture to all targets in PharmTargetDB.

We obtained lung cancer-related targets from MalaCards. The MalaCards (https://www.malacards.org) is a database of diseases specific to humans from GeneCards, containing and organizing >20,000 disease entries from more than 80 data sources, including OMIM and orphanet. The database displays a “disease card” for each disease, with various information regarding the disease, including disease aliases, related diseases, therapeutic agents, anatomical background, and disease-related genes. Using the Uniprot (https://www.uniprot.org) database, we converted the names of drugs’ active components into gene symbols. Finally, we found shared targets between GK and lung cancer using a screening process. All procedures followed were according to the principles of the Declaration of Helsinki.

The Metascape (https://metascape.org) platform was used to conduct pathway enrichment analyses based on the Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) to identify the biological processes and signaling pathways associated with the targets shared by GK and lung cancer. The biomedical researchers’ community may evaluate gene/protein lists and make better data-driven decisions with the support of Metascape’s trustworthy, effective, and user-friendly tool collection.

The STRING database (https://string-db.org) is used to predict network interactions between proteins. The most frequently reported targets for GK and lung cancer were provided on this site to compile PPI network maps and information. Sharing targets and pathways led to establishing a GK network model using Cytoscape 3.9.0.

We download the 3D structure of Ginkgo biflavone in SDF format from PubChem data, import the structure into ChemBio3D Ultra 14.0 for energy minimization, set the Minimum RMS Gradient to 0.001, and save the small molecule in mol2 format. Import the optimized small molecule into AutodockTools-1.5.6 for hydrogenation, charge calculation, charge assignment, and rotatable key setting, and then save it as “pdbqt” format. Download the protein structures of AKT1 (PDB ID: 1UNQ), CTNNB1 (PDB ID: 1JDH) and GSBK3β (PDB ID: 6V6L) from the PDB database; use Pymol 2.3.0 to remove protein crystalline water, original ligands, etc., and import the protein structures into AutoDocktools (v1.5.6) for hydrogenation, charge calculation, charge assignment, and assigning rotatable keys. The protein structure is imported into AutoDocktools (v1.5.6) for hydrogenation, charge calculation, charge assignment, atom type specification and saved in “pdbqt” format. Prediction of protein binding sites using POCASA 1.1 and docking using AutoDock Vina1.1.2.

Lung adenocarcinoma cells (A549, H1299 and LLC) were acquired from the American Type Culture Collection (Manassas, VA, United States) and grown at 37°C in humid air with 5% carbon dioxide (CO₂) in Dulbecco’s Modified Eagle Medium supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin, and 100 g/mL streptomycin.

To test GK’s cytotoxicity against A549 and H1299 cells, we used the CCK-8 assay to determine the viability of the 2 cells. These 2 cells were seeded onto 96-well plates at a density of 3,000 cells/well. Three plates were created under the same conditions as repeat experiments and incubated overnight at 37°C in humid air containing 5% CO₂. The cells were treated for 72 h with GK at the gradient-varying concentrations indicated (0, 2.5, 5, and 10 µM). Subsequently, we added CCK-8 solution to each well and incubated the samples under the same conditions for 1 h. The cell growth vitality in each well was measured at 450 nm using a multi-mode reader (BioTek, VT, United States).

In order to explore the effect of GK on human lung adenocarcinoma cells (A549 and H1299) migration, we did the wound healing experiment, using a marker pen and a straightedge to draw horizontal lines evenly across the back of the 6-well plate, approximately every 0.6 cm, across the wells, with five lines across each well, adding approximately 5×10⁵ cells to the wells, and after the cells were spread across the 6-well plate, we pretreated the plate with 1 μg/ml mitomycin C for 1 h, and then performed the scratching experiment with a 200 μL gun tip with five lines perpendicular to it. The cells were then washed three times with phosphate-buffered saline to remove the scratched-down cells; the adherent cells were cultured in FBS-free, a free medium containing the indicated GK concentrations (0,2.5, 5, and 10 µM). The cells’ migration distance on the wound was observed at 0 and 24 h, measured using a microscope and photographed for documentation.

To demonstrate further modifications of GK-induced migratory capability on human lung adenocarcinoma cells (A549 and H1299), we performed Transwell migration research using Transwell chambers (Corning, NY, United States). Cells were pretreated with 10 μg/mL of mitomycin C for 1 h in serum-free medium, the upper chamber was injected with 2 × 10⁴ cells and kept at 37°C.Dimethyl sulfoxide or GK was applied to the cells at the appropriate concentrations (0, 2.5, 5, and 10 µM). To induce cell migration to the lower chamber, media containing 20% FBS was added to the bottom of the cell chamber. Cells were fixed with paraformaldehyde and stained with 0.5% crystal violet after 24 h. Lastly, a light microscope (×100 magnification, Olympus, Tokyo, Japan) was used to view, photograph, and count the cells.

Total cellular proteins treated with various GK doses were extracted using RIPA lysis buffer and subjected to protein electrophoresis using SDS-PAGE, followed by protein transfer to PVDF membranes (Merck Millipore Ltd., Tulla Green, Ireland). TBST solution with 5% skimmed milk was used to block the PVDF membranes. The PVDF membranes were treated with a primary antibody (1:1,000) overnight at 4°C, followed by three washes with TBST and incubation with an HRP-conjugated secondary antibody (1:5,000) for 1 h. Finally, the PVDF membranes were photographed using a Tanon 5200 imager (Shanghai, China). To discover Immune complexes, utilize an ECL kit (Share Bio, Shanghai, China).

Male C57BL/6 wild-type mice (7 weeks old, weighing 18–20 g, Shanghai. China) were purchased from Shanghai Jihui Laboratory Animal Care Co.,Ltd. Mice were housed and maintained in an animal facility that adhered to specific pathogen-free (SPF) standards. All animal experiments were performed in accordance with the Guide for Ethical Review of Laboratory Animal Welfare (GB/T 35892). All animal experiments were conducted in accordance with the “Guide for Ethical Review of Laboratory Animal Welfare (GB/T 35892-2018)”. All experimental procedures were approved by the Animal Care and Use Committee of Shanghai University of Traditional Chinese Medicine. All procedures were approved by the Animal Care and Use Committee of Shanghai University of Traditional Chinese Medicine (PZSHUTCM220919016). 1.5 × 10⁵ LLC cells were injected through the tail vein, and C57 mice were given the next day intraperitoneally with GK (15 mg/kg or 30 mg/kg) once daily or cisplatin (2 mg/kg). Intraperitoneal injection was given once every 2 days. Mice were euthanized after 2 weeks, and lungs were taken for follow-up investigation.

Lung tissue from mice was immersed and fixed in 10% formalin for 24 h and embedded in paraffin embedded and cut into 4 µm thick sections. h&E staining was performed in 4 µm paraffin sections were stained using standard H&E staining protocols. Final. Slides were sealed with neutral adhesive; tissue morphology was then observed and photographed under light microscopy.

The statistical findings are reported as the mean ± standard deviation. The Student’s t-test was performed to determine whether or not there was a significant difference between the two groups. We determined the statistical significance of group differences using GraphPad Prism8 software (GraphPad Software, Inc., San Diego, CA, United States). p-values <0.05 were considered statistically significant.

We utilized network pharmacology to identify the GK targets that decreased lung adenocarcinima cells (A549, H1299, and LLC) activity and migration. Figure 1 showed the molecular structure of ginkgetin. We constructed a network of GK and their targets using the PharmMapper server. We obtained 297 targets from the PharmMapper server (Figure 2).

On the PharmMapper platform, we acquired 297 GK-related targets. Similarly, we obtained 875 target genes related to lung cancer using the MalaCards database. We obtained 79 similar targets by crossing the active ingredient targets with disease-related ones (Figure 3).

The p-value was set to 0.01, and the lung cancer and GK’s shared targets were submitted to the Metascape platform. For the enrichment study of GO biological processes, cellular components, and molecular functions, 1,051 GO biological processes, 64 GO cellular components, 91 GO molecular functions, and 154 KEGG pathways were employed.

The KEGG pathway analysis uncovered 154 highly-enriched pathways, including cancer pathways, proteoglycans in cancer, transcriptional dysregulation in cancer, viral oncogenesis, apoptosis-polymorphic, and p53 signaling pathways. All KEGG routes were chosen for mapping and analysis. Figure 4 shows the top 20 most enriched routes. Potential GK treatment targets in lung cancer were primarily concentrated along the cancer process (hsa05200) (Figure 4).

Figure 5 showed the top 20 enriched GO keywords in each sub-ontology. Highly enriched GO keywords included cell migration, signaling pathways, epithelial cell proliferation, receptors, kinases, mRNA translation inhibitors, and cell adhesion molecules. Further analysis of the top 20 GO biological processes, cellular components and molecular function enrichment revealed that the potential GK targets for lung cancer treatment primarily involve investigating its mechanism of action on lung cancer cell migration and invasion. Using the STRING database, we developed a protein interaction network of crossover genes, submitted the highest confidence protein interaction scores, and evaluated the key target, AKT.

The PPI interaction network information was gathered by uploading targets to the STRING database and combining the data in Cytoscape 3.90. There are 99 vertices (drug: 1 node; target: 79 nodes; pathway: 19 nodes) (Figure 6).

The interaction pattern analysis of the docking results using PyMOL2.3.0 showed that the binding energy of ginkgo biflavone with AKT1 was −7.6 kcal/mol, which proved to have a good binding effect. Ginkgo biflavone interacted with AKT1 mainly through the formation of hydrogen bonds as well as hydrophobic forces, forming hydrogen bonds with GLU-91 with a length of 3.2 Å; and hydrophobic interactions with HIS-13, PRO-24, TRP-11 and LEU-12 (Figure 7A). The binding energy of ginkgo biflavone with CTNNB1 was −8.8 kcal/mol, which proved to have a good binding effect. Ginkgo biflavones interacted with CTNNB1 mainly through the formation of hydrogen bonds as well as hydrophobic forces with SER-425, ASN-430, ARG-474, SER-473, ASN-516, ARG-469 with hydrogen bond lengths of 3.0Å, 3.1Å, 3.2Å, 3.0Å, 3.3Å, respectively; with ARG-469, GLY-512 have hydrophobic interaction (Figure 7B). The binding energy of ginkgo biflavone with GSK3β was −8.9 kcal/mol, which proved to have a good binding effect. Ginkgo biflavone interacted with GSK3β mainly through the formation of hydrogen bonds as well as hydrophobic forces, forming hydrogen bonds with PRO-136 with a length of 3.0 Å; and hydrophobic interactions with TYR-134, ALA-83, ILE-62, PHE-67, GLN-185, TYR-140, VAL-139 (Figure 7C).

We carried out the Cell Counting Kit-8 assay to verify the anti-proliferative effect of GK on A549 and H1299 cells. We treated these two types of human lung adenocarcinoma cells with cultures containing different GK concentrations for 72 h. The results demonstrated that GK significantly inhibited the proliferative activity of A549 and H1299 cells in a concentration-dependent manner. According to CCK-8 test data, GK’s IC₅₀ on A549 and H1299 cells for 72 h was determined to be 10.05 μM and 2.789 μM (Figure 8).

The wound healing experiment revealed that the cell wound width was considerably lesser after 24 h or 48 h than at 0 h in the control group. In contrast, there was no significant difference between that at 0 and 24 or 48 h in the GK-treated group’s wound breadth (Figure 9A). In addition, we performed the transwell assay, and the results indicated that GK effectively attenuates migratory capacity of A549 and H1299 cells. GK markedly and dose-dependently inhibited the migration of A549 and H1299 cells (Figure 9B).

After Western blotting analysis, we found that GK markedly inhibited N-cadherin and vimentin expression at the protein level in A549 and H1299 cells (Figure 10).

EMT is the process by which epithelial cells undergo a phenotypic switch to that of a mesenchymal cell type. As a result of EMT, epithelial cells lose their polarity and attachment to the basement membrane but gain the ability to migrate, invade, resist apoptosis, and degrade the extracellular matrix. Western blotting analysis showed that the expression levels of p-Akt (ser473) and snail protein were significantly downregulated in A549 and H1299 cells treated with indicated concentrations of GK (p < 0.05; Western blotting). These results indicate that GK blocks the activation of AKT/GSK-3β/Snail in A549 and H1299 cells. Furthermore, In A549 and H1299 cells treated with the indicated concentrations of GK, the expression level of c-Myc protein was also significantly decreased (p < 0.05; Western blotting). These results indicate that GK can block Wnt/β-catenin activation in A549 and H1299 cells (Figure 10).

To confirm the anti-metastatic efficacy of GK. LLC cells were injected into 7-week-old C57 mice via the tail vein and GK was injected into the mice. We found that the number of lung tumor nodules was lower in both the GK group than in the control group (Figure 11A). Meanwhile, HE staining showed that the lung tumor burden in the GK group was significantly lower than that in the control group (Figure 11B). These findings suggest that GK can effectively prevent tumor cells from spreading to the lungs (Figures 11C–E).