Mechanism of Erianin anti-triple negative breast cancer based on transcriptomics methods and network pharmacology

Cell culture

The human BC cell line MDA-MB-231 and mouse BC cell line 4T1 were purchased from the Peking Union Medical College Cell Culture Center (Beijing, China). MDA-MB-231 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM; C11995500BT, Thermo Fisher Scientific, USA) supplemented with 10% fetal bovine serum (FBS, 10099141, Thermo Fisher Scientific). The 4T1 cells were maintained in Roswell Park Memorial Institute-1640 medium (C11875500BT, Thermo Fisher Scientific) supplemented with 10% FBS. All cells were cultured in a humidified atmosphere of 5% CO₂ at 37°C.

Reagents

Purified Erianin (HPLC ≥98%) was purchased from Shanghai Yuanye Biotechnology Co., Ltd. (Shanghai, China, B20844). A stock solution at 800 μM was made in DMSO (Sigma, USA) and stored in the dark at −20°C. RIPA (P0013B) and modified sodium citrate antigen repair solution (P0083) were purchased from Beyotime Biotechnology (China), Cell Counting Kit-8 (CCK-8) detection kit was purchased from RiboBio (C-005) (China), goat serum was purchased from Jackson (005-000-121) (USA), immunohistochemical secondary antibody and DAB developer were purchased from Dako (K5007) (Denmark), Antibodies to PI3K (4257p), Phospho-Akt (Ser473, 4060S) and AKT (3788S) were purchased from Cell Signaling Technology (USA). GAPDH (AC001) was purchased from ABclonal (USA). SC79 (HY-18749) was purchased from MedChemExpress (USA).

CCK-8 assay

The CCK-8 experiment was used to evaluate the effect of drug treatment on the proliferation and viability of MDA-MB-231 and 4T1 cells. The cells were inoculated into 48-well plates at a density of 2 × 10⁴ cells per well. When the cells were 70% to 80% confluent, the cells were treated with the drug, and after a certain period of time, 20 μl of CCK-8 solution was added to each well. Then, the optical density (OD) at 450 nm was measured by an enzyme meter. Cell survival rate (%) = (OD experimental group-OD blank group)/(OD negative control group-OD blank group) × 100%.

Wound healing assay

Cells were seeded (4 × 10⁵ cells/well) in a 6-well plate, when the cell density was above 90%, some straight lines were drawn on the bottom of the 6-well plate with a sterile pipette tip. Discarded the old medium, washed it twice with preheated PBS, and added medium containing 1% serum. Then placed the 6-well plate in the incubator (37°C, 5% CO₂) 24 hours. After the experiment, ImageJ software was used for data analysis.

In vivo xenograft assay

Six- to eight-week-old female Balb/c normal mice and Balb/c nude mice were purchased from Kunming Medical University. The mice were housed under a specific pathogen-free (SPF)-grade experimental system. Each Balb/c mouse was injected with 3 × 10⁵ 4T1 cells, and each Balb/c nude mouse was injected with 3 × 10⁵ MDA-MB-231 cells in the second left mammary fat pad. When the tumor size reached approximately 15 mm³, all mice were randomly divided into two groups: the experimental group and the control group (five mice in each group). The mice in the experimental group were injected with 4 mg/kg Erianin (based on previous studies [9, 12, 18–23] and our experimental validation), while the control group received an injection of the same volume of 1% DMSO. Every two days, the tumor volume was measured. The volume was calculated using the following formula: V = 0.5 × length × width². Nude mice were executed after 21 days and Balb/c mice after 17 days and the tumours were photographed and recorded. All protocols were approved by the Animal Ethics Committee of Kunming University of Science and Technology (PZWH K2019-0005).

Transcriptome sequencing and analysis

Cells in the logarithmic growth phase were passaged. After 80% cell confluence, one dish was randomly selected as the experimental group and treated with 40 nM of Erianin. Similarly, another dish was randomly selected as the control group and treated with an equivalent volume of solvent (1‰ DMSO). The cells were then collected after 24 hours, and three samples were replicated three times. Subsequently, the cells were sent to Biomarker Technologies for transcriptome sequencing. Cell samples were sent to Bemac Biotechnology Ltd., for transcriptome sequencing. According to the method described by the company, RNA was initially extracted from the samples. RNA purity and concentration were assessed using a NanoDrop 2000 spectrophotometer. After accurate detection of RNA integrity by Agilent 2100/LabChip GX, library construction occurred upon passing the test. Library quality control was carried out employing the Qubit 3.0 fluorescence quantification instrument, Qsep400, and Q-PCR method. PCR methods were utilized for quality control of the library before sequencing using the Illumina NovaSeq 6000 sequencing platform. The differentially expressed genes (DEGs) were screened for expression differences using the condition | log2Fold Change | >1.

Drug and disease target acquisition

Exploring the genes associated with targeting a certain disease is an important foundation in modern small molecule drug discovery and development. To explore the genes associated with the anti-TNBC progression of Erianin, we analyzed the target genes of Erianin against TNBC by using web-based pharmacology. First, we obtained the 2D structure and 3D structure of Erianin from PubChem through BATMAN-TCM [24] (http://bionet.ncpsb.org.cn/batman-tcm/index.php), PharmMapper [25] (http://www.lilab-ecust.cn/pharmmapper/), Swiss Target Prediction (http://swisstargetprediction.ch/), TargetNet [26] (http://targetnet.scbdd.com/), PubChem (https://pubchem.ncbi.nlm.nih.gov/) and CTD [27] (https://ctdbase.org/) databases to obtain the targets of action of Erianin. Where the BATMAN-TCM database was obtained for Erianin-related targets and screened for “Score cutoff >2.608” targets. Through the UniProt database, target names were standardized, and duplicates were removed and merged to create the Erianin target dataset. “Triple negative breast cancer” or “TNBC” were used as the search terms, and the data were obtained from the GeneCards (https://www.genecards.org/), OMIM (https://www.omim.org/) and DisGenet [28] (https://www.disgenet.org/) databases. Among them, TNBC-related gene information was retrieved from the GeneCards database, and the genes with GeneCards GIFtS greater than 40 were screened. Subsequently, target names were standardized, and duplicates were removed and merged to establish the TNBC-related gene dataset through the UniProt database.

Construction of the protein-protein interaction and hub gene network

The online network analysis platform, STRING 11.5 [29] (http://string-db.org/) was used to analyze the network topology. The common targets were inputted into the STRING 11.5 database for creation, with the selection of “Homo sapiens” as the species. Network was then visualized and analyzed using Cytoscape 3.8.0. The Analyze Network and CytaHubba were then used to screen the hub genes based on degree in the PPI network.

Molecular docking analysis

Molecular docking can effectively predict the binding abilities between small molecule compounds and target genes. The 2D structure of Erianin was downloaded from the PubChem database. The 2D structure was then converted to a 3D structure in Chem3D software. The 3D crystal structures of 8 hub genes were retrieved from the PDB database. AutoDockTools 1.5.6 software was utilized to remove all ligands from the protein receptors and add hydrogens and charges before molecular docking simulation. Ligands and protein receptors were recorded in PDBQT format. Then, the binding sites of Erianin to the target gene receptor protein were examined using PyMOL 2.4.1. The binding energy was determined from the affinity. The higher the absolute affinity value, the stronger the binding affinity of Erianin with proteins.

Gene expression and survival analysis

The BC gene expression dataset, comprising tumor tissue and paraneoplastic normal tumor samples, was procured from the TCGA database, along with patient-related clinical data. Patients were screened based on clinical information to obtain cases of TNBC with PR-negative, ER-negative, and Her-2-negative conditions. R software was utilized to analyze the expression matrices of both the normal and tumor groups. R software was utilized to extract expression data of potential target genes of Erianin acting on TNBC in the TCGA database. The edge R and DEGseq packages were employed for analyzing and screening DEGs in TNBC tissue samples and normal tissue samples. Histograms were plotted with loglFCl >1 and p < 0.05 thresholds. Additionally, we performed a survival analysis on possible target genes of Erianin in TNBC using the Kaplan-Meier Plotter web platform. We selected the best nodes while setting negative ER, PR, and Her-2 statuses. We regarded the genes with p < 0.05 in the results as influencing TNBC patient survival.

Gene function and pathway enrichment analysis

Gene Ontology (GO) [30] is internationally used to analyze the functional enrichment of genes and proteins, while Kyoto Encyclopedia of Genes and Genomes (KEGG) [31] pathway enrichment analysis provides a deeper explanation of different gene and protein functions. We used the R 4.3.1 programming language packages “clusterProfiler”, “org.Hs.eg.db”, “enrichplot” and “ggplot2” to analyze the GO and KEGG enrichment of target genes. The top 10 biological processes (BPs), cellular localizations (CCs) and molecular functions (MFs) in GO analysis and the top 15 pathways in KEGG analysis were screened at p < 0.05.

Western blotting (WB)

BC cells were cultured in 6-well plates at a density of 1 × 10⁶/ml per well. After treatment with Erianin for 24 hours, cells were lysed in precooled RIPA buffer. The protein concentrations were determined by a BCA protein assay kit (Beyotime, P0009). Equal amounts of protein were separated by SDS-PAGE and transferred to 0.45 μm PVDF membranes (Millipore, IPVH00010). After blocking with 2% BSA for 2 hours at room temperature, the membranes were incubated with primary antibodies at 4°C overnight. Then, the membranes were washed with TBST buffer 3 times and incubated with HRP-conjugated secondary antibody. The results were developed and recorded by a chemiluminescence analysis system (Tanon-5200, Tanon Science and Technology, China).

Immunohistochemistry (IHC)

Tumors were fixed for 24 hours in 4% paraformaldehyde and rinsed in running water. Then, paraffin-embedded tissue was cut into 5 μm thick sections. Sodium citrate buffer was used to boil the slides for 10 minutes to retrieve the antigens. After blocking with goat serum solution for 30 min, the sections were incubated overnight at 4°C with primary antibodies. A secondary antibody was added the next day, DAB chromogenic reagent was added, and the sections were counterstained with hematoxylin. Scores were assigned based on the intensity of the immunohistochemical signal and the area of the tissue sections stained positively. Finally, Image-Pro Plus 6.0 was used for quantitative analysis of the picture.

Statistical processing

All statistical data were analyzed using GraphPad Prism 8.0 software and R 4.3.1. The data were first tested for normal distribution, followed differences between the two groups were measured by T-test, and comparisons differences between the two groups were evaluated by one-way analysis of variance (ANOVA). The experimental data are expressed as the mean ± SD. Statistical significance was defined as p < 0.05. ^*p < 0.05, ^***p < 0.001, ^****p < 0.0001.

Data availability statement

The data that support the findings of this study are available on request from the corresponding author, upon reasonable request.

Erianin inhibits the progression of TNBC cells

To investigate the antiproliferation of Erianin in TNBC cells, MDA-MB-231 cells and 4T1 cells were treated with different concentrations of Erianin (0, 20, 40, 80, 120, 160 nM) for 24 hours, and the viability of the cells was measured by CCK-8 assay. The results indicated that Erianin had a significant inhibitory effect on the proliferation of MDA-MB-231 and 4T1 cells in a dose-dependent manner (Figure 2A, 2B). Furthermore, we evaluated the effect of Erianin on cell migration. Considering the killing effect of Erianin on TNBC cells, we used 10 nM and 16 nM Erianin to treat MDA-MB-231 and 4T1 cells, respectively. In wound healing assay, Erianin significantly inhibited the migratory ability of MDA-MB-231 and 4T1 cells, compared with the control group (Figure 2C, 2D). In addition, two xenograft transplantation models were used to investigate the antitumor effect of Erianin in vivo. MDA-MB-231 cells and 4T1 cells were injected into Balb/c nude mice and Balb/c mice in situ. When the tumor size reached approximately 15 mm³, the mice were separated randomly into two experimental groups: the control group (DMSO) and the Erianin group. We found that 4 mg/kg Erianin treatment significantly inhibited the tumours growth of MDA-MB-231 and 4T1 cells (Figure 2E–2H).

Figure 2. Erianin inhibits the progression of TNBC cells in vitro and in vivo. MDA-MB-231 (A) and 4T1 cells (B) were treated with different concentrations of Erianin for 24 hours, and then the viability was determined using the CCK-8 assay. (C) The migration capacity of MDA-MB-231 and 4T1 cell with Erianin was detected by wound-healing assay. (D) Statistical analysis of wound-healing assay. (E) MDA-MB-231 cells were injected into Balb/c nude mice in situ, and xenograft tumors were imaged after Erianin treatment. (F) Tumor volume in each group (N = 5). (G) 4T1 cells were injected into Balb/c normal mice in situ, and xenograft tumors were imaged after Erianin treatment. (H) Tumor volume in each group (N = 5).

Transcriptomic sequencing analysis of TNBC cells after Erianin treatment

By comparing the gene expression profile data of the Erianin group and Control group, we identified 1350 DEGs out of 13629 transcriptome annotated genes, following the threshold standard of |log2-fold change| > 1 and p < 0.05. The analysis revealed 1124 upregulated genes and 226 downregulated genes. The volcano plot and heatmap of all DEGs were shown in Figure 3A, 3B. To facilitate analysis, we excluded genes that were not expressed in any group from the transcriptome sequencing expression matrix. The top 10 upregulated and downregulated genes were shown in Figure 3C, 3D.

Figure 3. RNA-seq data analysis. (A) A volcano plot of 1350 DEGs, including 226 downregulated genes and 1124 upregulated genes in response to Erianin treatment, was drawn. (B) Heatmap of 1350 DEGs was created. (C) Heatmap of the top 10 upregulated or downregulated genes based on the ranking of expression changes. (D) Table of the top 10 upregulated or downregulated genes based on the ranking of expression changes.

Network pharmacological screening of Erianin anti-TNBC target genes

We analyzed the target genes of Erianin against TNBC by using network pharmacology. First, we obtained the 2D structure and 3D structure of Erianin from PubChem (Figure 4A, 4B). Then, we obtained a total of 1002 Erianin-related targets from six drug databases and 5156 TNBC-related targets from the GeneCards, OMIM and DisGenet databases (Figure 4C), and the intersection of these two databases led to the identification of 622 Erianin targets in the TNBC-related target genes (Figure 4D). Furthermore, we combined the potential target genes obtained from the network pharmacological analysis and the differential genes analyzed by transcriptome sequencing, and a total of 51 intersections were obtained (Figure 4E). We carried out subsequent analyses of these 51 common targets.

Figure 4. Screening of target genes for anti-TNBC activity of Erianin. (A) 2D structure of the Erianin molecule. (B) 3D structure of the Erianin molecule; (C) Flowchart of the screening process of target genes of Erianin anti-TNBC obtained by network pharmacological analysis. (D) Venn diagram of target genes of Erianin in anti-TNBC obtained by network pharmacological analysis; (E) Venn diagram of target genes identified in network pharmacology analysis and differentially expressed genes (DEGs).

PPI network analysis of common targets

The clarity of drug targets is crucial for the in-depth development of drugs, and although our preliminary studies have shown that Erianin plays an important regulatory role in anti-TNBC progression, its target of action in TNBC has not yet been reported in the literature. Therefore, using the STRING database, we explored the interactions between the 51 common targets obtained from the above transcriptomics and network pharmacology to mine the interactions, selecting a confidence score >0.15. A complete interaction network was constructed using Cytoscape, and a total of 51 nodes and 314 edges were obtained. The darker the color, the higher the corresponding degree value. In the PPI network, we employed the degree method to identify the top 8 hub genes. These include CCND1, APOE, PDGFRB, PLAU, CDK1, PPARA, ROCK2 and MUC1 (Figure 5A). Most of these proteins had significant expression differences in TNBC patients (Figure 5B). For example, APOE, PLAU, CDK1 and MUC1 were significantly higher in TNBC tissues than in cancer-parasite tissues, whereas PDGFRB, RET, ROCK2, CCND1 and PPARA were significantly lower in TNBC tissues than in cancer-parasite tissues (Figure 5C–5J). In addition, high expression of ROCK2 and PLAU in TNBC patients predicted poorer survival, whereas low expression of PDGFRB and CCND1 in TNBC patients predicted poorer survival (Figure 5K–5N).

Figure 5. Bioinformatic analysis of PPI-enriched and hub genes. (A) PPI of potential hub target genes. (B) The expression of hub genes in TNBC and the correlation analysis with the overall survival time of patients. (C–J) Expression maps of MUC1, APOE, PLAU, CDK1, PDGFRB, CCND1, ROCK2, and PPARA in TNBC patients and normal tissues. (K–N) Impact of ROCK2, PLAU, PDGFRB, and CCND1 expression on the survival of TNBC patients.

To further clarify the specific targets of Erianin anti-TNBC, we performed molecular docking of these hub genes as target targets with Erianin, and the specific docking binding results were shown in Figure 6A. As shown in the table, Erianin has the lowest binding energy to PPARA, followed by ROCK2, PDGFRB, CCND1, MUC1 and CDK1. The lower docking binding energy indicates that Erianin has a higher binding affinity to these targets, suggesting that it has a greater possibility to act on these targets, affecting the structural changes of the targets and then regulating the corresponding signaling pathways. Docking structures of Erianin with the target proteins PPARA, ROCK2, PDGFRB, CCND1, MUC1 and CDK1 were shown in Figure 6B–6G.

Figure 6. Molecular docking results of Hub protein and Erianin. (A) Table of molecular docking results of Erianin and hub genes. (B–G) The molecular docking visualization results of Erianin binding to PPARA, ROCK2, PDGFRB, CCND1, MUC1 and CDK1.

Functional and pathway enrichment analyses of common targets

Furthermore, we analyzed GO with KEGG enrichment for 51 common targets and found that there were 2427 results related to BPs, of which 1007 results were significantly different (p < 0.05), 220 results related to CCs, of which 72 results were significantly different (p < 0.05), and 298 results related to MFs, of which 95 results were significantly different (p < 0.05). The GO results are shown in Figure 7A (TOP10). In terms of BP, these genes were enriched in positive regulation of ion transport, regulation of body fluid levels, response to oxygen levels, positive regulation of kinase activity and positive regulation of MAPK cascade. CCs alterations were mainly involved in neuronal cell bodies, cyclin-dependent protein kinase holoenzyme complexes, synaptic clefts, desmosomes and serine/threonine protein kinase complexes. The MFs included peptide binding, amide binding, alpha-catenin binding, amyloid-beta binding and sulfur compound binding. The KEGG results showed that these potential target genes were enriched in a total of 206 pathways, of which 60 pathways were significantly different (p < 0.05), including focal adhesion, the PI3K-Akt signaling pathway, the Rap1 signaling pathway, microRNAs in cancer and human papillomavirus infection (Figure 7B). Among them, focal adhesion and the PI3K-AKT signaling pathway were significantly enriched. In addition, we conducted KEGG enrichment analysis on the differentially expressed genes (DEGs) obtained from transcriptome sequencing and found that the PI3K-AKT signaling pathway was significantly enriched among the DEGs (Figure 7C).

Figure 7. The results of GO and KEGG pathway enrichment. (A) Top 10 results of GO for target genes of Erianin against TNBC. (B) Top 15 enriched KEGG pathways of target genes of Erianin against TNBC. (C) KEGG results for DEGs obtained by transcriptome sequencing.

The PI3K-AKT signaling pathway holds significance in responding to extracellular signals and regulating biological processes, including cell proliferation, apoptosis, and migration. Therefore, we explored the regulatory effects of Erianin on the PI3K-AKT signaling pathway by WB and immunohistochemical experiments. We treated two types of TNBC cells with concentrations of Erianin at 40 nM and 80 nM, respectively, which correspond to the half maximal inhibitory concentration for each cell type. Western blot results showed that Erianin significantly inhibited the expression of PI3K proteins and the phosphorylation of AKT after treating MDA-MB-231 and 4T1 cells for 24 hours (Figure 8A–8C). Furthermore, the expression of key proteins in the PI3K-AKT signaling pathway was also detected by immunohistochemistry in both transplantation tumor models, and the results were consistent with those of Western blot analysis (Figure 8D–8F). To further investigate whether Erianin can regulate the progression of TNBC through the PI3K-AKT signaling pathway, rescue experiments were conducted. We incubated TNBC cells with the Akt agonist SC79, CCK-8 assay revealed that 2 ug/ml SC79 treatment significantly reduced the inhibitory effect of Erianin on the proliferation of MDA-MB-231 and 4T1 cells, suggesting that the activation of the PI3K/Akt pathway was associated with the antitumor effect of Erianin (Figure 8G, 8H). Western blot experiments also showed that SC79 significantly enhanced AKT phosphorylation and attenuated the inhibitory effect of Erianin on AKT phosphorylation (Figure 8I–8K). The above results indicated that Erianin may inhibit TNBC cell proliferation by affecting the PI3K-AKT pathway, which confirmed the findings of network pharmacology and transcriptome sequencing.

Figure 8. Erianin inhibits the PI3K-AKT signaling pathway in TNBC. (A) Western blot detection of PI3K and AKT expression as well as AKT phosphorylation after Erianin treatment of MDA-MB-231 and 4T1 cells. (B) Quantitative graphs of PI3K expression after Erianin treatment of MDA-MB-231 and 4T1 cells. (C) Quantitative graphs of p-AKT expression after Erianin treatment of MDA-MB-231 and 4T1 cells. (D) Immunohistochemistry detection of PI3K and AKT expression as well as AKT phosphorylation in two transplantation tumor models. (E) Quantitative graphs of PI3K expression after Erianin treatment of two transplantation tumor models. (F) Immunohistochemical detection of PI3K and AKT expression as well as AKT phosphorylation in two transplantation tumor models. (G) MDA-MB-21 cells were treated with 40 nM of Erianin and 2 ug/ml of SC79 for 24 hours, and then the viability was determined using the CCK-8 assay. (H) 4T1 cells were treated with 80 nM of Erianin and 2 ug/ml of SC79 for 24 hours, and then the viability was determined using the CCK-8 assay. (I) Western blot detection of AKT expression and AKT phosphorylation after Erianin and SC79 treatment of MDA-MB-231 and 4T1 cells. (J) Quantitative graphs of PI3K expression in MDA-MB-231 and 4T1 cells. (K) Quantitative graphs of AKT phosphorylation expression in MDA-MB-231 and 4T1 cells.