The main active ingredients of Paridis Rhizoma were obtained through searching the literature. The PubChem database (https://pubchem.ncbi.nlm.nih.gov/) was used to obtain the canonical simplified molecular-input line-entry system strings of the active ingredients, the targets of active ingredients were predicted with the Swiss Target Prediction database (http://www.swisstargetprediction.ch/; probability>0) and PharmMapper database (http://lilab-ecust.cn/pharmmapper/index.html) (21,22), and the target names were converted into gene symbols using UniProt (https://www.uniprot.org). The GeneCards database (https://www.genecards.org/), Comparative Toxicogenomics Database (https://ctdbase.org/) and Online Mendelian Inheritance in Man database (https://omim.org/) were used to identify breast cancer-related target genes using the keyword ‘breast cancer’. Intersecting targets between active ingredients of Paridis Rhizoma and breast cancer were identified using the Venny online platform (https://bioinfogp.cnb.csic.es/tools/venny/).

The ClusterProfiler package (version 3.16.1) in R software (version 4.0.2) was used for Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis with a q value <0.05 to identify the cancer-related pathways affected by PRS. The DAVID tool (https://david.ncifcrf.gov/) was used to obtain the KEGG enrichment information of the targets. The top 15 KEGG pathways were selected for further analysis.

The protein structures were obtained from the Protein Data Bank (https://www.rcsb.org/) and the compound structures were downloaded from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/). PyMol (https://pymol.org/2/) was used to remove water and solvent molecules from proteins, and process receptors and ligands further using AutoDock Tool 1.5.7 before docking was performed. Using PyMol, the docking data with the lowest binding energies was chosen. The binding sites were visualized using a protein-ligand interaction profiler (https://projects.biotec.tu-dresden.de/plip-web/plip/index).

MDA-MB-231 human breast cancer cells were obtained from a cell bank/stem cell bank (Center for Excellence in Molecular Cell Science, Chinese Academy of Sciences) and cultured in Leibovitz's L-15 medium (Adamas-Beta, Ltd.) supplemented with 10% FBS (Adamas-Beta, Ltd.) and 1% penicillin/streptomycin (Adamas-Beta, Ltd.), with 100% air in an incubator with 90±5% humidity at 37°C.

All mouse experiments were carried out according to protocols authorized by The Scientific Research Ethics Committee of Shanghai University of Medicine and Health Sciences (Shanghai, China; approval no. 2023-XKY-07-32092419880308342X) and in accordance with UK guidelines (23). A total of ten 8-week-old female BALB/cA nude mice (19±3 g; Shanghai Jiesijie Laboratory Animal Co., Ltd.) were maintained in an SPF setting (22±2°C and 55±5% humidity) on a 10 h light, 14 h dark cycle with food and water supplied ad libitum throughout the experimental period.

MDA-MB-231 cells (5×10⁶ cells) suspended in 100 µl L-15 media were injected into the mice through the subcutaneous axilla. When the tumor volume reached 50–100 mm³, the mice were randomly divided into two groups (n=5/group): i) The PPII group [0.5 mg/kg/day; CAS: 50773–42-7; Beijing Solarbio Science & Technology Co., Ltd.; high-performance liquid chromatography ≥98%, dissolved in DMSO (Beijing Solarbio Science & Technology Co., Ltd.) and diluted in complete media to the final concentration]; and ii) the control group (0.9% saline). The mice were administered PPII or 0.9% saline every other day for 14 days by intraperitoneal injection. Tumor growth was calculated according to the following equation: Volume=(width² × length)/2 and the maximum tumor volume permitted in the tumor-bearing mice was ≤1,500 mm³. On the 14th day, the mice were euthanized by cervical dislocation, and the tumors were removed for further analysis. Tumor growth inhibition of tumor weight (TGItw) was assessed using the following formula: TGItw=(tumor weight of control group-tumor weight of PPII group)/tumor weight of control group ×100%.

For immunohistochemistry, tumor tissues were fixed with 4% paraformaldehyde at room temperature for 24 h (cat. no. P0099; Beyotime Institute of Biotechnology). After fixation, the tissues were dehydrated in ethanol with different concentrations (70, 80, 90 and 100%; Sinopharm Chemical Reagent Co., Ltd.) successively for 10 min each. After which, the tissues were soaked in paraffin (Sinopharm Chemical Reagent Co., Ltd.) with a temperature not exceeding 60°C for 1–2 h, and cut into 3 µm-thick sections. The sections were baked at 60°C for 20 min, infiltrated with xylene twice (10 min each time; Sinopharm Chemical Reagent Co., Ltd.), and placed into ethanol of different concentrations (100, 95 and 85%; Sinopharm Chemical Reagent Co., Ltd.) for infiltration (5 min each time). The sections were rinsed with distilled water 5 min for rehydration, and in a pressure cooker antigen retrieval was performed on the sections with citric acid repair solution (cat. no. P0081; Beyotime Institute of Biotechnology) for 2.5 min. Subsequently, the tissues were blocked with 5% bovine serum albumin (Beijing Solarbio Science & Technology Co., Ltd.) at 37°C for 10 min. Then the sections were incubated with Ki67 antibody (1:200; cat. no. AF0198; Affinity Biosciences) and Caspase 3 antibody (1:200; cat. no. AF6311; Affinity Biosciences) overnight at 4°C. After which the samples were incubated with Goat Anti-Rabbit IgG (H+L) HRP (1:200; cat. no. S0001; Affinity Biosciences) at 37°C for 1 h and stained with DAB Horseradish Peroxidase Color Development Kit (cat. no. P0203; Beyotime Institute of Biotechnology) and redyed with hematoxylin (cat. no. C0107; Beyotime Institute of Biotechnology). Images were obtained under a fluorescence inverted microscope (IX73P2F; Olympus Corporation) and quantified using ImageJ 1.48V (National Institutes of Health).

A Cell Counting Kit-8 (CCK-8; cat. no. C8022-500T; Adamas-Beta, Ltd.) was used to evaluate the cytotoxic effect of PPII on MDA-MB-231 cells, and all steps were performed according to the manufacturer's instructions. MDA-MB-231 cells (6×10³ cells/well) were seeded in 96-well plates. After overnight incubation, the culture media were changed to new media containing different concentrations of PPII (0, 0.09765625, 0.390625, 1.5625, 6.25, 25 or 100 µmol/l) and 0.1% DMSO, and the cells were then incubated for 0, 12, 24, 36, 48 or 72 h at 37°C in an incubator. Afterwards, the culture media were replaced with 100 µl/well 10% CCK-8 solution, and the cells were incubated for 2 h at 37°C. Subsequently, the absorbance values were measured at 450 nm on a microplate reader (MULTISKAN FC; Thermo Fisher Scientific, Inc.). The inhibition ratio (%) was calculated using the following formula: Inhibition ratio (%)=[1-(A PPII-A Blank)/(A Control-A Blank)] ×100%.

MDA-MB-231 cells were seeded in 6-well plates (6×10⁵ cells/well) and incubated until they adhered to the plate. Subsequently, the cells were treated as follows: The control group was treated with 0.1% DMSO and the PPII group was treated with PPII (2 µmol/l) for 48 h at 37°C. The supernatant was discarded and the cells were washed three times with PBS and digested with 0.25% trypsin (Adamas-Beta, Ltd.). The cells were centrifuged at 300 × g for 10 min at 4°C, the supernatant was discarded and PBS was added for washing three times. A total of 100 µl binding buffer was added to each tube to prepare the cell suspension. The cells were stained singly with annexin V-FITC and doubly with PI, and incubated at room temperature away from the light for 15 min. After the incubation, 400 µl binding buffer was added to each tube. Cell apoptosis was detected by flow cytometry (CytoFLEX S; Beckman Coulter, Inc.) using an Annexin V-FITC Apoptosis Detection Kit (cat. no. C1062M; Beyotime Institute of Biotechnology). CytExpert 2.4 (Beckman Coulter, Inc.) was used to evaluate the ratios of apoptotic and non-apoptotic cells in each of the four populations: i) Live cells, annexin V-negative and PI-negative; ii) early apoptotic cells, annexin V-positive and PI-negative; iii) late apoptotic or dead cells, annexin V-positive and PI-positive; iv) and dead non-apoptotic cells, annexin V-negative and PI-positive. Apoptosis rate=ratios of early apoptosis + ratios of late apoptosis.

MDA-MB-231 cells (2×10⁵ cells/well) were seeded in a 6-well plate and grown to 90% confluence. A scratch was made in each well using a 200-µl pipette tip, and the cells were subsequently washed with PBS. The cells were treated with 0.1% DMSO and 2 µmol/l PPII in serum-free medium. Wound healing was observed at 0 and 48 h under a microscope (XD-202; Nanjing Jiangnan Novel Optics Co., Ltd.). The areas of the scratches were measured using ImageJ 1.48v software (National Institutes of Health) and the cell migration was calculated according to the following formula: Migration rate (%)=(final scratch area-initial scratch area)/initial scratch area ×100%.

The Transwell plate (Costar; Corning, Inc.) insert membrane was coated with Matrigel (BD Biosciences) to assess cell invasion. On ice, the Matrigel was diluted to 1 mg/ml with serum-free medium. After which, 60 µl of diluted Matrigel was evenly spread on the bottom of the Transwell chamber and incubated at 37°C for 3 h. After which, the Matrigel was removed and 100 µl serum-free culture medium was added and incubated at 37°C for 30 min for hydration. Finally, the liquid in the chamber was removed and cell inoculation was carried out. MDA-MB-231 cells (5×10⁴ cells/well) were seeded in Transwell chambers, and the cells were starved in serum-free media for 12 h. Subsequently, 700 µl L-15 medium supplemented with 20% FBS (Adamas-Beta, Ltd.) was added to the lower wells, and 500 µl serum-free medium supplemented with 2 µmol/l PPII and 0.1% DMSO were added to the upper Transwell chambers. The cells were incubated for 48 h at 37°C. Subsequently, the cells and Matrigel on the upper chamber were scraped with a cotton swab. The filter was fixed with 4% paraformaldehyde fixative solution (cat. no. P0099; Beyotime Institute of Biotechnology) for 20 min and stained with 0.1% crystal violet (Shanghai Aladdin Biochemical Technology Co., Ltd.) at room temperature for 30 min. Finally, the number of cells that adhered to the lower surface of the insert membranes was counted under a microscope (XD-202; Nanjing Jiangnan Novel Optics Co., Ltd.).

The cell treatment conditions were the same as those for flow cytometry. The intracellular pyruvate and lactic acid levels were detected using pyruvate assay kits (cat. no. BC2205; Beijing Solarbio Science & Technology Co., Ltd.) and lactic acid assay kits (cat. no. A019-2-1; Nanjing Jiancheng Bioengineering Institute), respectively. All steps were performed according to the manufacturer's instructions. Subsequently, the absorbance values were measured at 520 and 530 nm on a microplate reader.

Western blotting was performed according to standard procedures (24). The cell treatment conditions were the same as those for flow cytometry. Cells were collected and lysed in lysis buffer supplemented with protease inhibitors (cat. no. BL504A; Biosharp Life Sciences) to extract proteins. The protein concentration was measured using a BCA Protein Assay kit (cat. no. G2026-1000T; Wuhan Servicebio Technology Co., Ltd.). Proteins were separated by 10% SDS-PAGE (cat. no. Ba1012; Wuhan Baiqiandu Biotechnology Co., Ltd.) with a protein loading amount of 40 µg/lane and transferred to a PVDF (cat. no. IPVH00010; MilliporeSigma) membrane. After blocking the membranes with 5% nonfat milk for 1 h at room temperature, the membranes were incubated overnight at 4°C with the primary antibodies, followed by incubation with the HRP-Goat anti Rabbit and HRP-Goat anti Mouse at room temperature for 1 h. β-actin was used as the internal control. Finally, proteins were detected with ECL reagent (cat. no. MA0186; Dalian Meilun Biology Technology Co., Ltd.) and the membranes were developed by the image analyzer (cat. no. 4800; Tanon Science and Technology Co., Ltd.).

The antibodies in the present experiments included: Akt (1:2,000; cat. no. 60203-I–Ig; Wuhan Sanying Biotechnology, Inc.), phosphorylated (p-)Akt (1:10,000; cat. no. T40067F; Abmart Pharmaceutical Technology Co., Ltd.), PI3K (1:2,000; cat. no. 60225-1-Ig; Wuhan Sanying Biotechnology, Inc.), p-PI3K (1:1,000; cat. no. 341468; Chengdu Zen-Bioscience Co., Ltd.), hexokinase 2 (HK2; 1:2,000; cat. no. DF6176; Affinity Biosciences), pyruvate kinase M2 (PKM2; 1:2,000; cat. no. AF5234; Affinity Biosciences), β-actin (1:5,000; cat. no. bs-0061R; BIOSS), HRP-Goat anti Rabbit (1:50,000; cat. no. 5220-0336; SeraCare Life Sciences) and HRP-Goat anti Mouse (1:50,000; cat. no. 5220-0341; SeraCare Life Sciences).

The results were analyzed using GraphPad Prism 9.0.0 (Dotmatics). Statistical comparisons between two groups were performed using unpaired t-test and the analysis of the control group and the experimental group at different time points was performed using two-way ANOVA and Sidak multiple comparisons test. P<0.05 was considered to indicate a statistically significant difference.

Previous studies have demonstrated that polyphyllin I, PPII, polyphyllin III, polyphyllin V, polyphyllin VI, polyphyllin VII, polyphyllin H and gracillin are the important components in Rhizoma Paridis, which have antitumor properties (25–32). A total of 280 potential targets of PRS were identified from databases, and there were 9,343 disease-related targets obtained from the three databases. The comparison of breast cancer-related targets and potential targets of PRS yielded a total of 128 shared targets according to the Venny online platform analysis (Fig. 1A). The protein-protein interaction network of PPII had more edges than that of other saponins, indicating that PPII is a core component in the treatment of breast cancer (Fig. 1B). Fig. 1C shows the chemical structural formula of PPII. Through further analysis, several essential genes were identified. The top 26 important targets (Fig. 1D) included AKT1, ESR1, EGFR, CASP3, MMP9, SRC, PPARG, MMP2, STAT1, GSK3B, MDM2, CDC42, JAK2, PPARA, IL2, CCL5, KIT, HMOX1, RXRA, MET, CASP1, PPARD, ESR2, ACE, NQO1 and GSTP1. Furthermore, the KEGG pathway enrichment analysis revealed that the effect of PRS on breast cancer was closely related to the PI3K/Akt signaling pathway (Fig. 1E).

To verify the effect of PPII on breast cancer growth in vivo, a xenograft tumor model was established by subcutaneously injecting MDA-MB-231 cells into nude mice. Next, PPII was injected intraperitoneally (Fig. 2A), and the tumor volume was measured every 2 days. The results demonstrated that, compared with that in the control group, the tumor volume growth in the PPII group was slowed down, indicating that PPII inhibited the growth of tumors in vivo. Compared with the control group, tumor growth was significantly inhibited starting from the 8th day (P<0.01; Fig. 2B). After 14 days of PPII intervention, the mice were euthanized by cervical dislocation, and the tumors were removed and weighed. The tumor weight of the PPII group was lower than that of the control group (P<0.05; Fig. 2C), and the TGItw was 35.74%.

To determine the effects of PPII on the proliferation of cancer cells, immunohistochemical analysis was performed. The Ki67 staining results revealed that the percentage of Ki67-positive cells was decreased in the PPII group compared with the control group (P<0.05). However, the Caspase 3 staining results revealed that the percentage of Caspase 3-positive cells was significantly increased in the PPII group compared with the control group (P<0.01; Fig. 2D).

The viability of MDA-MB-231 cells treated with PPII at different concentrations for different durations was determined using a CCK-8 assay. PPII effectively decreased the viability of MDA-MB-231 cells, suggesting that PPII could inhibit breast cancer cell proliferation. As shown in Fig. 3A and B, MDA-MB-231 cell viability was inhibited by ~50% at 48 h, and the IC₅₀ was 2.131 µmol/l. Therefore, the PPII concentration used in subsequent experiments was 2 µmol/l for 48 h. The apoptotic effect of PPII on breast cancer cells was determined by flow cytometry. The results showed that the percentage of apoptotic cells was 8.98% in the control group and 12.16% in the PPII group. This result indicated that PPII treatment could increase apoptosis in MDA-MB-231 cells (P<0.05; Fig. 3C).

Transwell and scratch assays were performed to evaluate the effects of PPII treatment on cell invasion and migration. When counting the number of cells that migrated across the Transwell insert, the number of invasive cells in the control group was greater than that in the PPII group (P<0.05; Fig. 3D). The results of the scratch assay revealed that the migration rate of the control group was higher than that of the PPII group (P<0.01; Fig. 3E). These results demonstrated that PPII could inhibit the migration and invasion of MDA-MB-231 cells.

KEGG enrichment analysis of the targets identified by network pharmacology prediction methods demonstrated that the effect of PRS on breast cancer was closely related to the PI3K/Akt signaling pathway, which is an important signaling pathway for tumor development, and can regulate cell survival, metastasis and metabolism (33). To investigate the effect of PPII on the PI3K/Akt signaling pathway in breast cancer cells, the binding of PPII to expected critical targets in the PI3K/Akt signaling pathway was evaluated by molecular docking analysis. The binding of PPII to Akt was evaluated. PyMol was used, and the binding energy between PPII and Akt was −7.9405 kcal/mol. The molecular docking stability is associated with the binding energy, and it is generally considered that a binding energy <-5 kcal/mol represents stable binding (34). Therefore, the results of the analysis suggested that the binding of PPII to Akt is stable. The binding sites were visualized and the results showed that the binding of PPII and Akt involved hydrogen bond interactions with TYR18, ARG86, THR82, GLN79, LYS14 and LEU52 (Fig. 4A). These docking results indicated that PPII binds to the target via hydrogen bonding. In summary, the KEGG results were further supported by molecular docking analysis results, and PPII likely exerts its anticancer effect via the PI3K/Akt signaling pathway.

To verify the effect of PPII on PI3K/Akt signaling in breast cancer cells, the expression and phosphorylation levels of PI3K and Akt were detected by western blotting. The results demonstrated that the levels of PI3K, p-PI3K and p-Akt were decreased (Fig. 4B). These results confirmed that PPII could inhibit breast cancer cell proliferation via the PI3K/Akt signaling pathway.

Lactic acid and pyruvate are metabolites of aerobic glycolysis. To explore the effect of PPII on aerobic glycolysis in breast cancer cells, the levels of lactic acid and pyruvate were measured with the appropriate detection kits. The results revealed that the level of lactic acid in the control group was 0.086 mmol/gprot and that of pyruvate was 0.136 µg/mg prot. In the PPII group, the lactic acid level was 0.054 mmol/gprot and the pyruvate level was 0.069 µg/mgprot (Fig. 4C). This indicated that PPII inhibited the production of lactic acid and pyruvate in breast cancer cells.

To further verify the effect of PPII on aerobic glycolysis in breast cancer, the expression levels of the key enzymes involved in aerobic glycolysis, namely HK2 and PKM2, were evaluated. The results indicated that the expression levels of HK2 and PKM2 were decreased (Fig. 4D). Overall, PPII was capable of inhibiting the expression of key enzymes and downregulating aerobic glycolysis.