G-Rh2 was purchased from Chengdu Phytoelite Bio-Technology Co., Ltd. (Chengdu, Sichuan, China) and the purity (>98%) was determined by high-performance liquid chromatography (Figure 1). 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) and Annexin V-FITC apoptosis detection kit were purchased from Sigma-Alorich (St. Louis, MO, United States). The fetal bovine serum (FBS) was purchased from Corning (Medford, MA, United States). Antibiotics (100× penicillin/streptomycin) and 0.25% Trypsin-EDTA were purchased from Gibco (California, United States). Minimum essential medium (MEM) was purchased from Hyclone (Logan, UT, United States). RIPA Lysis Buffer and eECL Western Blot kit were purchased from CWBio (Taizhou, Jiangsu, China). The primary antibodies against p-AKT, AKT, p-PDK1, PDK1, p-p85, p85, p-IκBα, IκBα, and anti-rabbit IgG HRP (#7074) were from Cell Signaling Technology (Danvers, MA, United States).

The chemical structure of G-Rh2 was imported into PharmMapper server (http://www.lilab-ecust.cn/pharmmapper/, version 2017) (Xia et al., 2017), STITCH database http://stitch.embl.de/, version 5.0) (Szklarczyk et al., 2016), SwissTargetPrediction (http://www.swisstargetprediction.ch/) (David et al., 2014), and Similarity ensemble approach (http://sea.bkslab.org/) (Keiser et al., 2007) to obtained the related targets of G-Rh2.

The potential targets of Lung cancer were identified by the Genetic Association Database (https://geneticassociationdb.nih.gov/), which is a database of genetic association data from complex diseases and disorders. Lung cancer was imported as a keyword and the disease targets associated with it were provided in the database.

In order to explore the relationship between G-Rh2 related targets and lung cancer disease related targets, protein-protein interaction (PPI) were analysed by the Database of Interacting Proteins (DIP^TM), Biological General Repository for Interaction Datasets (BioGRID), Human Protein Reference Database (HPRD), IntAct Molecular Interaction Database (IntAct), Molecular INTeraction database (MINT), and biomolecular interaction network database (BIND) using the plug-in Bisogenet (Martin et al., 2010) of Cytoscape 3.7.1 software. The PPI networks of G-Rh2 putative targets and lung cancer-related targets were established and visualized by the plug-in Bisogenet of Cytoscape 3.7.1 software.

GO analysis with the biological process, cellular component, and molecular function was carried out using the Database for Annotation, Visualization and Integrated Discovery (DAVID, https://david.ncifcrf.gov/, v6.8) (Huang et al., 2009). Functional categories were enriched within genes (FDR <0.005) and the top 10 GO functional categories were selected. KOBAS 3.0 (http://kobas.cbi.pku.edu.cn/) that assigned Kyoto Encyclopedia of Genes and Genomes (KEGG) database was used for pathway analysis (Xie et al., 2011). The significantly changed pathways which corrected p value <0.005 were selected and genes regulated these pathways were enriched by gene-pathway network analysis. The gene-pathway network was constructed to screen the key target genes for G-Rh2 against lung cancer.

The A549 lung cancer cell line was purchased from the China Center for Type Culture Collection (Wuhan, Hubei, China) and cultured in MEM supplemented with 10% FBS, 100 U/mL penicillin and 100 μg/ml streptomycin at 37°C in a humidified atmosphere with 5% CO₂.

The viability of A549 cells was measured by MTT assay. Briefly, cells were seeded into 96-well plates at 2 × 10⁴ cells/well. After adhered to the plates for overnight, cells were treated with different concentrations of G-Rh2 for 24 and 48 h. The medium was then removed and supplemented with 100 µL MTT solution for 4 h, following with 100 µL stopping buffer. The absorbance was determined at 550 nm using a microplate reader (Tecan Infinite M200 Pro, Männedorf, Switzerland).

The apoptosis ratio of cells was analyzed by Annexin V-FITC Apoptosis Detection Kit according to the instruction of the manufacturer. In brief, the cells treated with different concentration of G-Rh2 for 24 h were harvested and centrifuged at 1,500 rpm for 3 min to remove the medium. The precipitation was then resuspended in 100 µL 1 × binding buffer. Subsequently, the cells were stained with Annexin-FITC and PI for 15 min in the dark. After added 400 µL 1 × binding buffer, the samples were evaluated by a Accuri C6 Plus flow cytometer (Becton, Dickinson and Company, CA, United States).

A total of 1 × 10⁶ A549 cells were seeded into 40 mm Petri dish and grown for overnight. Then the cells were treated with G-Rh2 for different time points. After collection, the cells were lysed by RIPA lysis buffer and the protein content was determined by Bradford reagent, using bovine serum albumin (BSA) as a standard. Total protein were separated by 20% SDS-polyacrylamide gels for 1 h at 100 V, transferred to PVDF membranes, and blocked with 5% BSA in Tris-buffered saline containing Tween 20 (1 × TTBS) for 2 h. After washed with 1 × TTBS, the PVDF membranes were then incubated with primary antibodies (1:500–1:2000) in 5% BSA at 4°C for overnight, followed by washing and incubated for 1 h with HRP-conjugated secondary anti-IgG (1:500). The bands were then visualized by the eECL Kit and photographed using Tanon 5200 Multi imaging system (Tanon, China).

The results have been represented as the mean ± SD. Variances among two groups were analyzed by Student’s t-test. Data analysis was completed using SPSS 20.0 (SPSS Inc., Chicago, IL, United States). p < 0.05 indicated significant differences.

We evaluated the ability of G-Rh2 to inhibit the viability and induce apoptosis of A549 cells using a network pharmacology approach. The compound-target network was created by one approach. The network of G-Rh2 and its targets from PharmMapper server, SwissTargetPrediction, a similarity ensemble approach, and the STITCH database was constructed as shown in Figure 2. Ninety-one targets were obtained, among of which, 70 targets were obtained from PharmMapper server, 14 targets from SwissTargetPrediction, 5 from the similarity ensemble approach, and 2 from the STITCH database. After removing the duplicates, A total of 91 compound-targets was obtained.

Biological networked systems include functional units or protein complexes in PPI networks and disease- or drug-target genes can be analyzed according to complex network theory (Peng et al., 2018). PPIs are important in the regulation of biological systems and are the targets of an increasing number of drugs (Scott et al., 2016). PPI networks of G-Rh2 putative targets and lung cancer-related targets were structured with PPI data. The PPI network of G-Rh2 putative targets contained 3444 nodes and 86990 edges, which represented 3444 interacting proteins and 86990 interactions (Figure 3A). The PPI network of lung cancer-related targets contained 7493 interacting proteins and 179815 interactions (Figure 3B).

The interaction network of G-Rh2 comprised 91 putative targets of G-Rh2 and 431 lung cancer-related targets. The interaction network encompassed 512 interacting proteins and 1,589 interactions (Figure 3C). The structured interaction network was merged with the PPI networks of G-Rh2 putative targets and lung cancer-related targets to identify targets for G-Rh2 in lung cancer. The new network had 217 nodes and 1,028 edges (Figure 3D); thus, 217 targets for G-Rh2 in lung cancer were identified.

DAVID was used to carry out GO analysis to elucidate the function of 217 candidate targets in biological process, cellular component, and molecular function. One hundred and three GO terms with FDR <0.05 were significantly enriched: 73 in biological process, 11 in cellular component, and 19 in molecular function. The top 10 GO terms enriched in each sub-ontology are shown in Figure 4. Regulation of apoptotic process, regulation of transcription, nucleoplasm, nucleus, protein binding, and damaged DNA binding were the highly enriched GO terms. Anticancer agents activate several pathways simultaneously, positively or negatively regulating the death process (Solary et al., 2000). Regulation of apoptosis is the mechanism by which most chemotherapeutic drugs induce tumor cell death. Genetic alterations induce cancer and always result in dysregulated transcriptional programs. Almost every DNA, RNA, and protein component controlled by normal transcription is influenced by recurrent somatic mutations in tumor cells (Bradner et al., 2017). DNA binding of a new compound is an important aspect of its therapeutic potential for anticancer (Thangavel et al., 2018). Because of reversible binding or formation of covalent bonds with deoxyribonucleic acid, small DNA-interacting anticancer drugs abrogate the interaction between DNA and transcription factors in gene promoters (José, 2018). Therefore, G-Rh2 may reduce A549 cell viability by intervening in biological processes and affecting cellular components and molecular functions.

The KEGG pathway analysis was performed by KOBAS and 143 significantly enriched pathways (corrected p < 0.005) including pathways in cancer, the PI3K-Akt signaling pathway, proteoglycans in cancer, focal adhesion, the Jak-STAT signaling pathway, the FoxO signaling pathway, and apoptosis were obtained. Figure 5 shows the top 20 enriched pathways. The most significantly enriched pathways were related to pathways in cancer, followed by the PI3K-Akt signaling pathway. The PI3K-Akt signaling pathway is crucial in the development of many types of tumors (Lorusso, 2016; Zhang et al., 2017). Cell proliferation, growth, cell cycle, apoptosis, and protein synthesis are regulated by the PI3K-Akt signaling pathway (Zhu et al., 2018). Also, activation of the PI3K-Akt signaling pathway promotes cancer cell proliferation, survival, and angiogenesis. The PI3K-Akt signaling pathway is activated in cancer and is a potential therapeutic target (Chen et al., 2017). Xie et al. indicated that the PI3K-Akt signaling pathway is important in lung cancer and found that ginsenoside Rg3 promoted apoptosis by inhibiting the ratio of p-PI3K/PI3K and p-Akt/Akt in A549 cells (Xie et al., 2017). Our results suggest that G-Rh2 inhibits the viability of A549 cells by inducing apoptosis via the PI3K-Akt signaling pathway. In addition, the Wnt/β-catenin, mTOR, VEGF, EGFR, and metabolic signaling pathways are important in lung cancer (Cho, 2013).

The significantly enriched pathways and genes were used to construct a gene-pathway network. As shown in Figure 6, 143 pathways and 191 genes were identified. Betweenness centrality (BC) was used to carry out the topological analysis. In the network, metabolic pathways had the highest BC, followed by pathways in cancer, the neurotrophin signaling pathway, HTLV-1 infection, and the PI3K-Akt signaling pathway. PIK3CA had the highest BC and several other genes including IRS1, AKT1, NFKB1, MAPK9, and POLD1 had larger BC. PIK3CA is one of the most commonly mutated oncogenes in human cancer and is used in the development of PI3 kinase inhibitors, which can be used as targeted therapies for cancers with these mutations (Jelovac et al., 2014). The PI3K signaling pathway is activated in cancers and is concerned with oncogenesis and cancer progression (Hennessy et al., 2005). PIK3CA mutations have been focused on as potential biomarkers of PI3K pathway activation (Ito et al., 2017). Activation of mutations and genomic amplification of the PIK3CA gene are closely related to increased PI3K activity in lung cancer (Yamamoto et al., 2008). Somatic mutations in the IA PI3K catalytic subunit p110α, encoded by PIK3CA, activate the PI3K signaling pathway (Sawa et al., 2017). IRS1 regulates many cancer-cell processes and PI3K within malignant cells (Houghton et al., 2010; Porter et al., 2013). A549 cell survival, proliferation, malignancy, and metastasis are suppressed by Akt1 knockdown (Jere et al., 2009). Also, constitutional activation of the PI3K-Akt signaling pathway can be caused by oncogenic mutations in the AKT1 gene, increasing the malignant potential of the affected cells. Our results indicate that G-Rh2 induces A549-cell apoptosis mainly by regulating the expression of PIK3CA, thus inhibiting activation of the PI3K-Akt signaling pathway. Also, regulation of IRS1, AKT1, NFKB1, MAPK9, and POLD1 may explain the G-Rh2-mediated inhibition of the viability of A549 cells.

G-Rh2 suppresses cell proliferation, causes G1-phase arrest, enhances the activity of capase-3, and induces apoptosis in A549 cells (An et al., 2013). To validate the anti-proliferation effect of G-Rh2 on A549 cells, a MTT assay was carried out. A549 cells were treated with G-Rh2 at different concentrations for 24 h or 48 h. As shown in Figure 7, G-Rh2 significantly inhibited A549 cells proliferation in a dose-dependent manner. Based on the results of MTT assay, IC₅₀ of G-Rh2 on A549 cells at 24 and 48 h were calculated respectively as 42.75 and 36.25 µM which was consist with previous report (Zhang et al., 2011).

Apoptosis, an orderly process of programmed cell death, is central to development of cancers. It involves of the activation, expression and regulation of a series of genes, and inhibits the growth of tumor cells. Apoptosis may not the main way for the death of many cancers response to common treatments (Brown and Attardi, 2005). G-Rh2 were reported to induces apoptosis of many cancer cells, such as human epidermoid carcinoma A431cells (Park et al., 2010), human malignant melanoma A375-S2 cells (Fei et al., 2002), and hepatocellular carcinoma HepG2 cells (Zhang et al., 2019). In order to determine whether G-Rh2 can regulates A549 cell death by inducing apoptosis in this study, A549 cells treated with 10, 20, and 40 μM G-Rh2 were stained by Annexin-PI based on a flow cytometry. As shown in Figure 8, the apoptosis rate increased significantly from 4.4% in the control group to 15.92, 23.72, and 78.7% at 10, 20, and 40 μM G-Rh2, respectively. It is indicate that G-Rh2 can significantly induce A549 cells apoptosis to regulate this lung cancer cell death, and is consistent with literatures (Cheng et al., 2005; Wang et al., 2018).

The PI3K/Akt signaling pathway operates by phosphorylation and dephosphorylation of the substrate-level (Sang and Li, 2007; Yu et al., 2021). PI3K, phosphoinositide 3-kinase, a lipid kinase family protein, which was constructed by two subunit including a regulatory subunit (P85) and a catalytic subunit (P110). When the growth factors binds to their receptor tyrosine kinase (RTK) or G protein-coupled receptors (GPCR), PI3K isoforms were stimulated and catalyzes the production of phosphatidylinositol-3,4,5-triphosphate (PIP3) at the cell membrane. PIP3, serves as a secondary messenger, is in turn to help the phosphorylation and activation of PDK1 and AKT. AKT is a serine/threonine kinase, once activated, it controls key cellular processes including the inhibition of apoptosis (Liu et al., 2009), and promotes cell survival by activating CREB and NF-κB activity (Park et al., 2010). Therefore, PI3K/Akt as a most commonly activated signaling pathway in human cancer, presents both an opportunity and a challenge for cancer therapy (Liu et al., 2009). In the present study, G-Rh2 was confirmed to down regulate the phosphorylation of P85, PDK1, Akt and IκBα in A549 cells by western blot analyses (Figure 9). The inactivation of P85, PDK1 and Akt by G-Rh2 treatment indicating that G-Rh2 can induces cancer cells to apoptosis via inhibiting the PI3K/Akt signaling pathway. Similarly, the inactivation of IκBα means G-Rh2 suppress the cell survival via inhibiting the NF-κB signaling pathway and to induces cancer cells death. These results are consistent with that deduced from network pharmacology analyses.