Cell culture and grouping

Human GC cell lines, including AGS and HGC27 purchased from Pricella Biotchnology Co., Ltd (Wuhan, China), were used in this study. F12K medium, supplemented with fetal bovine serum (FBS, 10%) and 1% streptomycin and penicillin (Gibco, MA, USA), was utilized for the culture of AGS cells. HGC27 cells were cultured in minimum essential medium (MEM), which was supplemented with 1% streptomycin and penicillin, as well as 10% FBS. All the cells were incubated in a humidified incubator 5% CO₂ at 37 °C.

For the grouping, both AGS and HGC27 cells were divided into three groups, including control group, and GPS treatment groups, subjected to GPS with a concentration of 50 µM and 100 µM (Selleckchem, Houston, TX, USA), respectively. The selection of GPS concentrations utilized in this study was decided based on the IC₅₀ test (Fig. 1A).

Open in a separate window

Fig. 1

GPS inhibited the cellular proliferation, migration and invasion of AGS and HGC27 cells in a dose-dependent manner. A Selection of IC₅₀ of GPS in two cell lines. B CCK-8 assay, C Colony formation assay, and D EdU staining kit were performed to measure proliferation ability of GC cells. E Transwell assay and F Wound healing assay were performed to measure migration and invasion ability of GC cells. (***P<0.001)

Cellular viability assay

AGS or HGC27 cells (1×10⁴) were seeded into 96-well plates. Cells were given GPS treatment with a cellular density of reaching 60% or more at 0, 24 h, 48 h and 72 h, respectively. Then, CCK-8 commercial kit was utilized for the evaluation of cellular viability, according to the previous description [14]. After incubating for 2 h, the cellular viability was measured with a microplate reader at an absorbance wavelength of 450 nm.

Colony formation assay

Cells (1×10³) were cultured for 12 days in a 6 cm petri dish. Then, the cells were washed with PBS, followed by fixation with precooled methanol. The colony formation was counted after staining with crystal violet for 20 min.

EdU staining

Commercial EdU staining kit purchased from Beyotime (Shanghai, China) was utilized for the evaluation of cellular proliferation, according to the manufacturer’s instructions. Cells (1×10⁴) in each group were incubated with EdU buffer (10 μM) at 37  for 2 h. Then, the supernatant was discarded and washed with PBS twice, followed by fixing with paraformaldehyde for 20 min. Afterwards, the mixture was incubated with freshly prepared cocktail buffer. For the calculation of proportion of EdU-positive cells, the number of fluorescent stained cells was divided by the number of cells stained with DAPI.

Transwell assay

Transwell assay was conducted to evaluate the migration and invasion of AGS and HGC27 cells. For the migration assay, cells (4×10⁴) suspended in serum-free medium (100 μL) were added to the upper chamber. The lower chamber was added with medium (750 μL) containing 10% FBS, to induce the migration of cells. The invaded cells were washed with PBS thrice, followed by fixation for 30 min using 4% paraformaldehyde and staining with 1% crystal violet. For the invasion assay, the matrigel was added to the upper chamber. The other procedures were similar to those of the migration assay. The images were observed under a magnification of 100×.

Wound healing assay

Cells (6×10⁵) were cultured on 6-well plates. Then, we scratched the plate bottom from top to bottom with a pipettor. The free cells were washed away with PBS buffer. The migration distance was observed at 0 and 24 h, respectively. The area for the wound healing was determined with IMAGE J software.

Flow cytometry

The apoptosis in each cell line was determined with the commercial Annexin V: FITC Apoptosis Detection Kit I (BD Biosciences, USA). Cells (5×10⁵) were seeded in 6-well plates, and were resuspended in 1×binding buffer. Afterwards, the mixture was stained with Annexin V-Alexa Fluor647 (50 µg/mL) and propidium iodide (10 µg/mL) for 30 min. Finally, the cellular apoptosis was determined using flow cytometry.

Transmission electron microscope (TEM)

Cells (4×10⁵) were seeded on 6-well plates, and were treated with GPS until a confluence of 60%. The samples were fixed, dehydrated and permeated successively. Resin blocks were cut into ultrathin sections (80 nm). The sections were stained with uranium acetate, followed by observing under a JEM-1200EX II electron microscope (Tokyo, Japan).

Determination of mitochondrial membrane potential (MMP)

MMP was determined using the commercial JC-1 kit (MK3204, MaokangBio, Shanghai, China), according to the manufacturer’s instructions. In brief, the cells in each group were washed using PBS twice, and mixed with the JC-1 buffer at 37  for 30 min. Finally, the mixture was washed using 1×JC-1 staining buffer and observed under a fluorescent microscope.

Western blot

Total protein from AGS and HGC27 cells was extracted using commercial kit, and the protein concentration was determined with BCA method. The samples were subjected to SDS–PAGE, and were transferred to the PDVF membrane that were placed in 5% BSA solution (ThermoFisher Scientific, MA, USA). The primary antibodies were incubated at four degrees overnight. The primary antibodies utilized in the experiments included anti-BAX (Proteintech, 50599-2-Ig, 1:1000), anti-β-action (Proteintech, 66009-1-Ig, 1:5000), anti-Bcl-2 (Proteintech, 68103-1-Ig, 1:1000), anti-PARP (Cell Signaling Technology, #9532, 1:1000), anti-Cleaved PARP (Cell Signaling Technology, #5625, 1:1000), anti-Caspase-3 (Cell Signaling Technology, #14220, 1:1000), anti-Cleaved Caspase-3 (Cell Signaling Technology, #9664, 1:1000), anti-PI3K (Affinity, AF6241, 1:1000), anti-p-PI3K (Affinity, AF3242, 1:1000), anti-AKT (Cell Signaling Technology, #4691, 1:1000), anti-p-AKT (Cell Signaling Technology, #4060, 1:1000), anti-EGFR (Proteintech, 66455-1-Ig, 1:1000), Anti-Cyt C (Proteintech, 10993-1-AP, 1:1000). The secondary antibodies were added and incubated at room temperature for 1 h. Finally, the bands were visualized using the IMAGE J software, with β-action serving as the internal standard.

Animal, modeling process and grouping

BALB/c nude mice (male, 6 weeks) were purchased from Vital River (Beijing, China) were subjected to induction of xenograft tumor model. Briefly, HGC27 cells (5×10⁶) were given via subcutaneous injection to the position near the right groin. Animals with tumor mass on day 7 were randomly divided into control group (n=8) and GPS group (n=8), treating with PBS and 40 mg/kg GPS via intraperitoneal injection, respectively. The PBS or GPS was given once per 3 days. The volume of the mass and body weight of mice were measured every 3 days. On day 28, the animals were euthanized and the tumor tissues and organs were collected for the following analysis. The study protocols were approved by the Ethical Committee of the First Affiliated Hospital of Shandong First Medical University (Approval No.: 2021-S1039).

Induction of pulmonary metastasis model

To investigate the potential effects of GPS on the metastasis of GC, we first established the pulmonary metastasis model by tail vein injection of HGC27 cells (3×10⁶) with expression of luciferase. Subsequently, the mice were given intraperitoneal injection of GPS (40 mg/kg) or PBS on day 2, the PBS or GPS was given once per 3 days (5 mice in each treatment). On day 28, d-fluorescein (100 mg/kg) was given via intraperitoneal injection after anesthesia, followed by observing the tumor size after in vivo imaging. Finally, the animals were sacrificed to obtain the pulmonary tissues.

IHC and HE staining

Tumor tissues were fixed using paraformaldehyde and were embedded in paraffin. The sections were incubated at 70 and dewaxed in xylene, followed by rehydration in ethanol. Upon heating in a high pressure for 10 min, the sections were treated with sodium citrate buffer (10 mmol/L, pH=6.0). To inhibit the activity of endogenous peroxidase, the mixture was exposed in 3% hydrogen peroxide for 30 min at 37 . Then, the mixture was blocked in goat serum for 30 min, and incubated with the primary antibodies including overnight including anti-EGFR (Proteintech, 66455-1-Ig, 1:300), anti-PI3K (Affinity, AF6241, 1:100), anti-p-PI3K (Affinity, AF3242, 1:100), anti-AKT (Cell Signaling Technology, #4691, 1:200), anti-p-AKT (Cell Signaling Technology, #4060, 1:100), anti-Ki67 (Proteintech, 27309-1-AP, 1:1000). Then, the HRP-conjugated secondary antibodies were given at 37 °C for 30 min, together with observation under a microscope. For the HE staining, sections (5 μm) were dewaxed and dehydrated, followed by incubating with hematoxylin at room temperature for 5 min. The sections were incubated with hydrochloric acid alcohol (1%) for 30 s, and were dipped in eosin for 2 min. After dehydration, the sections were sealed with neutral resin and observed under a microscope according to the previous description [15].

GPS induced GC cells apoptosis

Flow cytometry indicated that GPS induced apoptosis in a dose-dependent manner (Fig. 2A). Mitochondrial dysfunction can trigger intracellular signaling cascades that initiate programmed cell death, leading to apoptosis in a variety of tumour cells [16]. To further validate whether GPS could induce mitochondrial apoptosis, TEM was conducted to measure the mitochondrial injury after GPS treatment. As shown in Fig. 2B, GPS triggered mitochondrial injuries featured by excessive mitochondrial swelling and rupture of mitochondrial cristae. Besides, GPS triggered the decrease of MMP in AGS and HGC27 cells, featured by significant decrease of JC-1 aggregates and significant increase of JC-1 monomers (Fig. 2C). In addition, Western blot analysis was performed to determine the expression of apoptosis-related proteins. As shown in Fig. 2D, GPS could induce significant downregulation of the anti-apoptotic Bcl-2. In addition, it significantly induced the up-regulation of the pro-apoptotic proteins Cleaved Caspase-3, Cleaved PARP, Bax and Cyt C. Taken together, these results suggest that GPS-induced apoptosis is mediated by mitochondrial dysfunction, thereby activating the mitochondrial apoptotic pathway in GC cells.

Open in a separate window

Fig. 2

GPS promoted the cellular apoptosis, which was related to the rupture of mitochondrial integrity. A GPS induced cellular apoptosis in AGS and HGC27 cells. B TEM showed that GPS induced rupture of mitochondrial structure. C GPS induced decrease of JC-1 aggregate and increase of JC-1 monomer in cells. D GPS regulated the expression of apoptosis-related proteins (*P<0.05, **P<0.01, ***P<0.001)

Network pharmacological analysis predicts GPS targets and pathways affecting GC

Through taking the intersection of the action targets obtained from the prediction of GPS by SwissTargetPrediction (Fig. 3A) and the action targets obtained from the retrieval of OMIM and Genecards database for GC, a total of 64 cross-targets potentially related to the treatment of GC by GPS were identified (Fig. 3B). In order to identify the core targets of GPS for GC treatment, we used STRING to construct the cross-targets of the PPI network and analysed the results using Cytoscape. The PPI network is shown in Fig. 3C. The top 5 interacting hub proteins were obtained based on the basis of the high and low p value (Fig. 3D). The related KEGG pathways were listed according to the ascending order of p value (Fig. 3E). The results showed that the PI3K-AKT signaling pathway had the highest significance and the highest number of enriched genes. Further analysis of the top five scoring hub proteins in the PPI network revealed that EGFR was the highest scoring protein excluding the internal reference protein GAPDH. Intriguingly, molecular docking analyses of the three-dimensional structure of GPS and EGFR showed that there is an interaction between GPS and EGFR (Fig. 3F). Based on a large number of previous studies suggesting that EGFR/PI3K/AKT are important signals that regulate cell proliferation and apoptosis [17], we deduced that the EGFR/PI3K/AKT axis may be the key to the function of GPS in GC.

Open in a separate window

Fig. 3

Network pharmacological analysis predicted GPS targets and pathways affecting GC. A Predicting the intersection of targets of action of GPS. B Venn diagram of the cross-targets associated with the target of GC and the target of GPS. C PPI network of Gentianoside on GC targets. D List of hub genes for GPS on GC by degree method. E Analysis of KEGG enrichment of GC target genes by GPS. F Molecular docking analyses of the three-dimensional structure of GPS and EGFR

GPS inhibited the activation of EGFR/PI3K/AKT

EGFR/PI3K/AKT signaling pathway plays an important role in cell proliferation, migration and apoptosis. EGFR stimulates tumour growth by modulating dimerisation of Ras proteins, which leads to a phosphorylation cascade that activates the PI3K/AKT signaling pathway [18]. Therefore, to confirm whether EGFR was regulated in GPS-treated GC cells, we found by Western blot assay that EGFR levels in GC cells decreased in a concentration-dependent manner after GPS treatment (Fig. 4A). At the same time, we also detected the expression levels of proteins related to PI3K/AKT signaling pathway, one of the downstream pathways of EGFR. As shown in the figure, GPS had a significant dose-dependent inhibitory effect on the phosphorylated expression of PI3K and AKT in GC cells, whereas the total PI3K and AKT protein levels remained essentially unchanged.

Open in a separate window

Fig. 4

GPS inhibited the activation of EGFR/PI3K/AKT. A The protein expression levels of PI3K/AKT signaling in GPS-treated GC cells were determined by Western blot. B The protein expression levels of PI3K/AKT signaling in GPS-treated GC cells after addition of overexpressed EGFR plasmid were determined by Western blot. (***P<0.001; ns, non-significant)

To further validate the relationship between GPS and EGFR, we added an overexpression plasmid of EGFR (Sequence: NM-005228) in GC cells treated with GPS, and then detected the protein level of this pathway (Fig. 4B). The analysis showed that the addition of overexpressed EGFR significantly attenuated the inhibitory effect of GPS on the phosphorylation of PI3K and AKT downstream of the EGFR. These results suggest that GPS inhibits the PI3K/AKT pathway by reducing EGFR expression.

GPS suppressed malignant behaviour of GC cells by inhibiting the EGFR/PI3K/AKT axis

To verify whether GPS affects biological behaviours such as proliferation and apoptosis of GC cells by inhibiting EGFR/PI3K/AKT, we performed rescue experiments by adding overexpression plasmid of EGFR to GPS (100 μM)-treated cells. It was found by EdU as well as colony formation assay that overexpression plasmid of EGFR-treated GC cells partially reversed the proliferation inhibition induced by GPS (Fig. 5A, B). It was also seen by wound healing assay that the GPS-induced diminished migration ability of GC cells was also partially restored after the application of overexpression plasmid of EGFR (Fig. 5C). In the JC-1 mitochondrial membrane potential assay, the GPS-induced decrease in MMP in overexpression plasmid of EGFR-treated GC cells was also suppressed (Fig. 6A). The results of apoptosis detection by flow cytometry also showed that overexpression plasmid of EGFR treatment partially reversed the apoptosis induced by GPS in GC cells (Fig. 6B). From the above results, we conclude that GPS inhibits the malignant behaviour of GC cells by inhibiting EGFR/PI3K/AKT.

Open in a separate window

Fig. 5

GPS inhibited GC cells proliferation and migration by inhibiting the EGFR/PI3K/AKT axis. A EdU staining kit, B Colony formation assay and C wound healing assay were performed to measure the proliferation and migration ability of GC cells treated with GPS (100 μM) and GC cells treated with GPS (100 μM) co-overexpressing EGFR plasmid. (***P<0.001; ns, non-significant)

Open in a separate window

Fig. 6

GPS promoted GC cells apoptosis by inhibiting the EGFR/PI3K/AKT axis. A The tendency of GPS (100 μM)-induced decrease in JC-1 aggregates and increase in JC-1 monomers in GC cells could be reversed after co-application of overexpressing EGFR plasmid. B The tendency of GPS (100 μM) to promote apoptosis in GC cells could be reversed after co-application of overexpressing EGFR plasmids. (***P<0.001; ns, non-significant)

Not applicable.