Mechanistic insights into pachymic acid’s action on triple-negative breast Cancer through TOP2A targeting

Abstract

Triple-negative breast cancer (TNBC) is characterized by the absence of estrogen and progesterone receptors, and lack of human epidermal growth factor receptor 2 (HER2) expression. Traditional Chinese medicine (TCM) has demonstrated promising efficacy in treating TNBC. This study explored the mechanisms of pachymic acid (PA) on TNBC by merging network pharmacology with experimental validation. We acquired Microarray data of TNBC from the Gene Expression Omnibus (GEO). The related targets of PA were predicted and screened using the following 6 databases: Swiss Target Prediction, HERB (Herbal Medicine Database), ETCM (Encyclopedia of Traditional Chinese Medicine), BATMAN (Bioinformatics Analysis Tool for Molecular Mechanism of Traditional Chinese Medicine), HIT (Herb Ingredients’ Targets Database), and PharmMapper. The STRING interaction network analysis tool was used to create Protein-Protein Interaction (PPI) networks. Enrichment analysis included Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG). We conducted a pan-cancer analysis, tumor immune microenvironment analysis, and molecular docking. We performed cell experimental, included cytotoxicity assay, apoptosis analysis, proliferation assay, and migration and invasion assays. PA has potential for treating TNBC with the target of TOP2A, and platinum drug resistance possibly serving as the KEGG pathway through which PA exerts its therapeutic effects. PA is involved in processes such as nuclear division, chromosome segregation, mitotic nuclear division, condensed chromosome formation, and protein C-terminus binding. PA probably exert its therapeutic effects through the tumor immune microenvironment, involving elements such as Dendritic cells activated, Eosinophils, Macrophages M0, Macrophages M1, and T cells CD4 memory activated. The therapeutic effects of PA may vary across different subtypes of TNBC such as TNBC-BL1, TNBC-Metaplastic, and TNBC-BL2. This study provides compelling evidence that PA holds significant promise as a therapeutic agent for TNBC, primarily through its action on TOP2A and its influence on the TNBC.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-87286-z.

Triple-negative breast cancer (TNBC) is characterized by the absence of estrogen and progesterone receptors, and lack of human epidermal growth factor receptor 2 (HER2) expression^1,2. TNBC is the most aggressive form of breast cancer (BRCA), marked by higher and earlier recurrence rates, increased mortality in operable stages (I-III), and reduced overall survival (OS) in the inoperable stage (IV)^3–5. Among women with TNBC in the operable stages, 30–40% experience recurrence within 5 years, and this rate increases to around 50% for those who do not attain a pathologic complete response (pCR) following neoadjuvant systemic therapy (NAST)^6,7. In the metastatic stage, even with the latest advancements in systemic treatments, the median OS remains less than two years^4,5.

For many years, chemotherapy has been the primary first-line treatment for patients with TNBC^8,9. In early-stage disease, multi-agent chemotherapy, which often includes fluorouracil, doxorubicin, and cyclophosphamide, is typically administered before surgery as neoadjuvant therapy, and tumors generally show a strong response^10–12. In metastatic TNBC, where the disease remains incurable, chemotherapy continues to be the main treatment option, with a median OS of 2–3 years^2,13. Recent developments, including targeted therapies, have improved outcomes in both early and metastatic TNBC^2,13,14. Additionally, the incorporation of immune checkpoint blockade (ICB) into first- and second-line treatments for various cancers, including solid tumors like TNBC, has shown clinical benefits^15,16. However, despite these advances, TNBC patients, particularly those with metastatic disease, still face poor OS and remain the subtype with the worst prognosis among breast cancers^2,5. Developing new therapeutic approaches, such as novel drug combinations and integrating traditional Chinese medicine (TCM), is crucial to improving patient outcomes. TCM influences immune function by regulating various immune cells in the body, including T lymphocytes, bursa-dependent lymphocytes, natural killer (NK) cells, and macrophages^17,18. These interactions can enhance the body’s anti-tumor immune response¹⁹. In recent years, TCM has demonstrated promising efficacy in treating TNBC²⁰. Pachymic acid (PA), a triterpenoid found naturally in Poria cocos. Recent research has revealed that PA can inhibit cell proliferation and trigger apoptosis across various types of cancer cells²¹. The study by Hong et al. speculate that PA might inhibit breast cancer metastasis through the regulation of PITPNM3²². Ling et al. found that PA inhibits breast cancer cell invasion by targeting NF-κB signaling and reducing the expression of MMP-9²³. However, no studies have yet explored the potential mechanisms by which PA may act in the refractory subtype of breast cancer, TNBC.

Network pharmacology, an emerging method for investigating drug mechanisms and efficacy, helps elucidate the multi-target effects and comprehensive actions of drugs by building models that map the interactions between drugs, targets, and diseases. This approach allows for a more holistic understanding of how drugs exert their influence across various biological pathways^24–27. In the field of TCM, network pharmacology, when combined with in vitro experiments, facilitates the exploration of the underlying mechanisms of TCM in treating diseases. Therefore, the objectives of this study are to explore the mechanisms of PA on TNBC by merging network pharmacology with experimental validation. This comprehensive new approach enhances the understanding of how PA affects critical processes such as cell proliferation, invasion, and metastasis in TNBC cells, and probably provide novel therapeutic strategies and target identification for TNBC management. Figure 1 presented the workflow.

Methods

Network pharmacology analysis

Microarray data collection

We gained Microarray data of TNBC from the Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/). We used the search strategy “Triple negative breast cancer” OR “TNBC” AND “Homo sapiens”, with “Entry type” limited to “Series” and “Study type” limited to “Expression profiling by array.” We excluded Microarray data with fewer than 3 samples per group. Finally, nine GEO datasets were included and downloaded for analysis (Table 1).

Table 1.

The GEO datasets for triple-negative breast cancer.

GEO datasets	GPL	Upload year	Update year	TNBC sample size	Normal sample size
GSE65216	GPL570	2015	2019	55	11
GSE76250	GPL17586	2015	2021	165	33
GSE61723	GPL16686	2014	2021	33	17
GSE61724	GPL6244	2014	2018	16	4
GSE81838	GPL6244	2016	2018	10	10
GSE113865	GPL10558	2018	2023	3	3
GSE53752	GPL7264	2014	2014	51	25
GSE124648	GPL96	2019	2019	108	10
GSE10893	GPL1390	2008	2017	31	8

GEO, Gene Expression Omnibus; GSE, Gene Expression Series; GPL, Gene Expression Omnibus Platform; TNBC, Triple-Negative Breast Cancer.

Identification of DEGs

We first performed batch correction on all datasets because they used different GPL platforms. Preprocessing steps for the datasets also included log² conversion and normalization. We analyzed the data using the “limma” software program (RStudio, version 1.3.1093) to identify differentially expressed genes (DEGs) between TNBC and normal group, considering P < 0.05 and |log²FoldChange| > 1 as significant. We then used the RStudio software packages “ggplot2” and “pheatmap” to create the volcano plots and heat maps.

Screening of PA targets

The 3D structure of PA was obtained from PubChem (a comprehensive chemical database maintained by the NCBI; https://pubchem.ncbi.nlm.nih.gov/). The related targets of PA were predicted and screened using the following six databases: Swiss Target Prediction (a database for predicting the potential targets of small molecules based on their chemical structure; http://www.swisstargetprediction.ch/)²⁸, HERB (a database for herbal medicine-related bioactive compounds and their targets; http://herb.ac.cn/)²⁹, ETCM (a database of TCM with detailed information about the chemical components and therapeutic targets; http://www.tcmip.cn/ETCM/)³⁰, BATMAN (a bioinformatics tool designed to predict the targets of TCM compounds and explore the relationship between TCM and diseases; http://bionet.ncpsb.org.cn/batman-tcm/#/home)³¹, HIT (a database focused on herb-target interactions, providing information on the molecular interactions between herbs and their target proteins; http://www.badd-cao.net:2345/)³², and PharmMapper (a web-based tool for predicting potential protein targets of small molecules, based on their molecular structure; http://www.lilab-ecust.cn/pharmmapper/)³³.

Identification of effective targets

To identify effective targets, we performed a Venn analysis using the R software packages “VennDiagram” and “ggvenn” to intersect the targets of PA with those of TNBC. The common targets identified through this intersection are considered as the effective targets.

Verification of Effective Targets by TCGA

TCGA is a publicly available database that contains multi-omics data for various types of cancer, including TNBC³⁴. The gene expression data and clinical data for TNBC were obtained from TCGA database through UCSC XENA website (http://xena.ucsc.edu/). Then, 115 TNBC samples and 113 normal tissue samples were selected based on immunohistochemical standards. The “ggplot2” package was used to display the expression of effective targets in TNBC and normal tissues, and the “survival” package was used to evaluate the association between effective targets and TNBC prognosis.

PPI analysis

The STRING interaction network analysis tool (https://string-db.org/) was used to create Protein-Protein Interaction (PPI) networks. Effective targets validated by TCGA were uploaded to STRING’s official website to explore the protein interrelationships. Subsequently, the interaction data were downloaded, and the PPI network was constructed and visualized using Cytoscape v3.9.1.

Enrichment analysis

The Gene Ontology (GO) enrichment analysis is a common method for examining biological processes (BP), cellular components (CC), and molecular functions (MF)³⁵. Kyoto Encyclopedia of Genes and Genomes (KEGG) is a database that helps understand biological processes and disease mechanisms through genome, chemical, and pathway annotations. To explore the biological functions and specific mechanisms of effective targets validated by TCGA, “cluster Profiler” and “org.Hs.eg.db” R package was used for GO and KEGG analyses, and an adjusted P < 0.05 was used as the cutoff criterion.

Pan-cancer analysis and immune microenvironment investigation

We conducted a pan-cancer analysis using the TISCH2 (http://tisch.comp-genomics.org) database, in order to observe the effective genes expression across different types of cancer. Subsequently, we used the CCLE (https://sites.broadinstitute.org/ccle/) database to investigate the differential expression of the effective gene across various breast cancer subtypes. To explore the expression of this effective gene in the tumor immune microenvironment and different immune cell types, we utilized the “estimate” and “CIBERSORT” packages in RStudio. In addition, we analyzed the correlation between this effective gene and immune checkpoint genes, as well as the tumor mutation burden in TNBC.

Molecular docking simulations and cell line selection

To prepare for subsequent in vitro and in vivo experiments, we performed molecular docking simulations. The targets of the effective gene were searched in the PDB database (http://www.wwpdb.org/) and saved as PDB format, and the ligands were stored in mol2 format for PA. We used the online platform CB-Dock2 (https://cadd.labshare.cn/cb-dock2/php/index.php) for this predictive analysis. Similarly, to provide a reference for selecting cell lines for subsequent cellular experiments, we first downloaded relevant data from the CCLE (https://sites.broadinstitute.org/ccle/) database and then analyzed the expression of effective gene in breast cancer cell lines using the “ggplot2” package in R.

Vitro experiments.

Previous studies, along with the analysis presented above, have established the therapeutic potential of PA in TNBC, and identified MDA-MB-231 cells (BL1 subtype) as one of the most appropriate cell for vitro experiments. To further substantiate this, we conducted preliminary cell viability assays, which validated the efficacy of PA prior to initiating formal experiments.

Cell Line and culture

.

For the in vitro experiments, the MDA-MB-231 cell line, a widely used model for TNBC, was selected due to its relevance to our research objectives and alignment with previous findings. The cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS) to support cell growth and metabolism, as well as antibiotics to prevent bacterial contamination-100 U/mL penicillin and 100 µg/mL streptomycin were added to the medium. The cultures were maintained in a humidified incubator at 37 °C with 5% CO₂ to simulate the physiological conditions. The cells were regularly monitored for confluency, and medium was changed every 2–3 days to ensure optimal growth conditions. Upon reaching 80–90% confluency, the cells were passaged using trypsin-EDTA solution to detach them from the culture surface, followed by resuspension in fresh culture medium for further experimentation.

Cytotoxicity assay

To investigate the cytotoxic effects of PA on MDA-MB-231 cells, the cells were treated with escalating concentrations of PA (0 µM, 5 µM, 10 µM, 20 µM, 30 µM, 60 µM, 80 µM, and 100 µM) over a 48-hour period. Post-treatment, cell viability was measured using the MTT (Thiazolyl Blue Tetrazolium Bromide) assay. MDA-MB-231 cells were exposed to MTT solution (0.5 mg/mL) for 4 h at 37 °C, allowing for the formation of formazan crystals. These crystals were subsequently dissolved in dimethyl sulfoxide (DMSO), and absorbance was recorded at 570 nm using a microplate reader. The half-maximal inhibitory concentration (IC₅₀) for PA was determined utilizing GraphPad Prism software.

Apoptosis analysis

The induction of apoptosis in MDA-MB-231 cells was assessed via flow cytometry. Cells were treated with PA at concentrations of 10 µM, 30 µM, and 60 µM at various time intervals (0 h, 12 h, 24 h, and 48 h). After the treatment period, cells were collected and subjected to staining with Annexin V-FITC and propidium iodide (PI), following the manufacturer’s protocol (BD Biosciences). Flow cytometric analysis was conducted to quantify late apoptosis rates (Q2).

Proliferation assay

To evaluate the impact of PA on the proliferation of MDA-MB-231 cells, the EdU (5-Ethynyl-2’-deoxyuridine) incorporation assay was employed. Cells were exposed to 30 µM PA for 48 h, followed by EdU staining as per the manufacturer’s instructions (Ribobio). The cells were subsequently fixed, permeabilized, and incubated with the EdU reaction cocktail. Visualization was performed using a fluorescence microscope, and the proliferation rates were quantified.

Migration and invasion assays

The effects of PA on the migration and invasion capabilities of MDA-MB-231 cells were assessed using both scratch and Transwell assays. Scratch Assay: MDA-MB-231 cells were seeded into 6-well plates and allowed to achieve confluence. A sterile pipette tip was utilized to create a uniform scratch in the cell monolayer. Cells were treated with 30 µM PA or a control (placebo group) and incubated for 48 h. Images were captured at the start and after 48 h to quantify migration by measuring the scratch area. Transwell Assay: For invasion analysis, MDA-MB-231 cells were placed in the upper compartment of a Transwell insert coated with Matrigel (BD Biosciences) in serum-free medium. The lower compartment contained DMEM supplemented with 10% FBS as a chemoattractant. After 48 h of incubation, non-invasive cells were removed from the upper chamber. The invasive cells that migrated to the lower chamber were fixed, stained with crystal violet, and counted under a microscope.

Statistical analysis

All experiments were conducted in triplicate, and results were presented as mean ± standard deviation (SD). Statistical significance was assessed using one-way ANOVA followed by Tukey’s post hoc test, with a p-value of < 0.05 considered significant. The description of all software, databases, and analytical platforms shown in Supplement Table S1.

Results

Exploring potential therapeutic targets

Identification of DEGs in TNBC

A total of 472 TNBC tissue samples and 121 normal tissue samples were analyzed. From this, 195 DEGs were identified, including 98 genes with downregulated expression and 97 genes with upregulated expression. Figure 2 shows the volcano plot of the DEGs (with the heatmap presented in Supplement Figure S1), and the detailed expression of each DEG is provided in Supplement Table S2. In Fig. 2, each node represents an individual gene. Blue nodes indicate genes with no significant difference in expression. The red nodes on the left correspond to genes that exhibit low expression in tumor tissue (e.g., AR, EDNRB and KIT), whereas the nodes on the right represent genes with elevated expression in tumor tissue (e.g., CCNA2, KIF11 and TOP2A).

Fig. 2.

Volcano plots for DEGs between TNBC and normal tissues.

Fig. 1.

The flowchart of this study.

Effective targets of PA for TNBC

A total of 282 potential effective targets of PA were identified (Supplement Table S3). Then, we identified 10 potential targets of PA for TNBC (Supplement Figure S2). Figure 2 displays the names of these genes along with their expression levels, which include CCNA2, KIF11, MMP12, MMP13, TOP2A, AR, EDNRB, KIT, PGR, and PPARG.

​Identification the key therapeutic target-TOP2A

The expression of 10 genes was compared between 115 TNBC samples and 113 normal samples, as shown in Supplement Figure S3. This results found that, CCNA2, KIF11, MMP12, MMP13, and TOP2A were highly expressed in TNBC, while AR, EDNRB, KIT, PGR, and PPARG were lowly expressed in TNBC. Among these ten genes, only TOP2A was found to be associated with TNBC patient OS (Fig. 3). In the 114 TNBC patients, the difference in OS between those with high and low expression of TOP2A was statistically significant, with a p-value of 0.038 (Fig. 3).

Fig. 3.

The relationship between the expression of ten genes in the TCGA database and overall survival (OS) of TNBC.

Mechanistic insights of TOP2A

Analysis of PPI network

Validation analysis suggests that TOP2A probably play a significant role in the PA treatment of TNBC. To further explore this, we performed a PPI analysis of TOP2A, with the results shown in Supplement Figure S4. The PPI analysis revealed that TOP2A may interact with several proteins, including ASPM, BUB1, BUB1B, CCNA2, CCNB1, CDC20, CDK1, MKI67, NUSAP1, and TOP1.

Enrichment analysis of GO and KEGG

The results showed that PA affected nuclear division, organelle fission, chromosome segregation, mitotic nuclear division, regulation of mitotic nuclear division, and other BPs (Supplement Figure S5). The main CCs included spindle, condensed chromosome outer kinetochore, condensed chromosome, condensed nuclear chromosome kinetochore, and condensed nuclear chromosome, centromeric region et al. (Supplement Figure S5). The top five MFs included histone kinase activity, protein serine/threonine kinase activity, protein C-terminus binding, cyclin-dependent protein serine/threonine kinase regulator, and activity (Supplement Figure S5). We also analysis the potential BPs, CCs and MFs process of PA therapeutic effects on TNBC through TOP2A (Supplement Figure S6). The results found that PA therapeutic effects on TNBC through its involvement in various cellular processes, including nuclear division, chromosome segregation, mitotic nuclear division, condensed chromosome, protein C-terminus binding, and others. Supplement Figure S7 showed the eight important pathways were analyzed using KEGG enrichment, and PA exerted anti-TNBC effects through the pathway of platinum drug resistance and TOP2A (Supplement Figure S8). The findings from the PPI network, as well as the GO and KEGG analyses, suggest that processes such as nuclear division, chromosome segregation, mitotic nuclear division, condensed chromosome formation, and protein C-terminus binding, along with the platinum drug resistance pathway, could be potential processes of PA in the treatment of TNBC through TOP2A.

Pan-cancer analysis

The results indicate that TOP2A is highly expressed in multiple types of cancer, included breast cancer (BRCA), compared to normal tissue (Supplement Figure S9). The results also shown that TOP2A expression is inconsistent across various subtypes of breast cancer, such as basal-BRCA, Herb2-BRCA, LumA-BRCA, and LumB-BRCA (Supplement Figure S9). The expression of TOP2A in different subtypes of TNBC are shown in Supplement Figure S10, and the expression levels of TOP2A, from highest to lowest, are as follows: TNBC- Basal-Like 1 (BL1), TNBC- Metaplastic, TNBC- Unclassified, TNBC-Basal-Like 1 (BL2), TNBC- Immunomodulatory, TNBC- Mesenchymal Stem-Like (MSL), and TNBC-Luminal Androgen Receptor (LAR).

Immune microenvironment investigation

In TNBC tissues with high and low expression of TOP2A (Supplement Figure S11), there were statistically significant differences in StromalScores, ImmuneScores, and the Estimation of Stromal and Tumor Immune Malignant Tissues Expression Score (ESTIMATScore). Investigating gene expression within tumors and its association with immune cell infiltration is essential for elucidating the regulatory within the tumor microenvironment. In TNBC, we found that the expression of TOP2A is positively correlated with the infiltration of Dendritic cells activated, Eosinophils, Macrophages M0, Macrophages M1, and T cells CD4 memory activated (Supplement Figure S12-13). Conversely, it is negatively correlated with the infiltration of B cells memory, Macrophages M2, Mast cells resting, and T cells CD8 (Supplement Figure S12-13). As shown in Supplement Figure S14, the expression of TOP2A is negatively correlated with the expression of the immune checkpoint genes C10orf54 and TNFRSF14. And we also found that the expression of TOP2A is positively correlated with tumor mutation burden in TNBC (Supplement Figure S15).

Molecular docking simulations and cell line selection

PA was utilized to conduct molecular docking of TOP2A. We found 2 maybe CurPocket and the binding energy of PA with TOP2A were − 8.1 kcal/mol and − 8.0 kcal/mol (Table 2), respectively (Supplement Figure S16). Supplement Figure S17 illustrates the expression of TOP2A in different BRCA cell lines. We found that the top 5 cell lines with the highest expression levels were UACC812, HCC38, EFM192A, MDA-MB-231, and SKBR3. The first 4 BRCA cell lines are all classified as TNBC, while SKBR3 is a HER2-positive BRCA cell line. MDA-MB-231 is the most commonly used TNBC cell line, widely utilized in studies of cancer biology, drug response, and metastatic mechanisms.

Table 2.

Docking scores of PA with TOP2A.

CurPocket ID	Cavity volume	Center (x, y, z)	Docking size (x, y, z)	Binding energy/(kcal/mol)
C1	5762	-58, 61, -40	31, 32, 35	-8.1
C2	4433	-73, 44, -10	25, 25, 35	-8.0

Experimental results in vitro

To investigate the cytotoxicity of PA on breast cancer cell, MDA-MB-231 cell was exposed to increase concentrations (0 µM, 5 µM, 10 µM, 20 µM, 30 µM, 60 µM, 80 µM, 100 µM) of PA for 48 h. Cell viability following PA treatment was assessed using the MTT (Thiazolyl Blue Tetrazolium Bromide) assay. We found that PA reduced the cell growth of MDA-MB-231 cells dose-dependently with the IC₅₀ value 28.08 µM (Fig. 4).

Fig. 4.

Flow cytometry results of MDA-MB-231 cell apoptosis.

We used flow cytometry to analyze the apoptosis of MDA-MB-231 cells treated with different concentrations of PA (10 µM, 30 µM, 60 µM) at various time points (0 h, 12 h, 24 h, 24 h). See results in Fig. 5. At 0 h, the late apoptosis rates (Q2) of MDA-MB-231 cells in the control group for the 10 µM, 30 µM, and 60 µM concentrations were 2.56%, 3.53%, and 4.26%, respectively; in the PA group, the late apoptosis rates (Q2) of MDA-MB-231 cells at 10 µM, 30 µM, and 60 µM were 6.46%, 5.65%, and 5.32%, respectively. At 12 h, the late apoptosis rate (Q2) of control group were 4.97%, 4.40%, and 4.36%, respectively; in the PA group, the late apoptosis (Q2) were 16.62%, 16.52%, and 13.27%, respectively. At 24 h, the control group (Q2) were 5.45%, 5.22%, and 4.29%, respectively; the PA group (Q2) were 13.78%, 22.58%, and 17.98%, respectively. At 48 h, the control group (Q2) were 5.76%, 5.92%, and 5.68%, respectively; the PA group (Q2) were 16.51%, 27.62%, and 22.29%, respectively. The results indicated that compared to the control group, PA can accelerate the apoptosis of MDA-MB-231 cells, with the 30 µM group appearing to have the strongest effect.

Fig. 5.

(a) EdU assay; (b) The migration; (c) Transwell assay.

We utilized the EdU (5-Ethynyl-2’-deoxyuridine) incorporation assay to investigate the effects of PA on MDA-MB-231 cells proliferation. The results indicated that, compared to the control group, the proliferation rate of cancer cells in the 30 µM PA group was significantly lower than that of the placebo group (14.3% vs. 27.16%; Fig. 6a). The 48-hour scratch assay indicated that, compared to the placebo group, 30 µM PA can inhibit the migration of MDA-MB-231 cells (Fig. 6b). The Transwell assay results demonstrated that, compared to the placebo group, 30 µM PA significantly inhibits the invasion of MDA-MB-231 cells (Fig. 6c).

Fig. 6.

.

Discussion

In this study, integrating various research methods, we found that PA has potential for treating TNBC, with TOP2A likely being its target, high TOP2A expression in TNBC is linked to poor OS, and platinum drug resistance possibly serving as the KEGG pathway through which PA exerts its therapeutic effects. By targeting TOP2A, PA is involved in processes such as nuclear division, chromosome segregation, mitotic nuclear division, condensed chromosome formation, and protein C-terminus binding. We found that PA probably exert its therapeutic effects through the tumor immune microenvironment, involving elements such as Dendritic cells activated, Eosinophils, Macrophages M0, Macrophages M1, and T cells CD4 memory activated. The therapeutic effects of PA probably vary across different subtypes of TNBC. We found that three subtypes, TNBC-BL1, TNBC-Metaplastic, and TNBC-BL2, are likely to respond more favorably. In recent years, improvements in medical technology and scientific research have propelled the fundamental study of TCM for disease treatment from cellular investigations to more advanced molecular and genetic levels^36,37. This advancement has been particularly rapid in the field of oncology³⁸.

PA could potentially be developed as a novel therapeutic agent targeting TOP2A in TNBC, and TOP2A was prioritized. TOP2A is a well-established target in cancer therapy, especially in breast cancer, where it plays a pivotal role in DNA replication and repair. As an essential enzyme, it facilitates the unwinding of DNA during replication and is a target for several chemotherapeutic agents, including anthracyclines and epipodophyllotoxins^39,40. Through molecular docking simulations, we found that the binding affinity between PA and TOP2A is similar to that of some chemotherapeutic drugs with TOP2A, such as Etoposide^41,42. Our study also found that PA’s interaction with TOP2A probably inhibit critical cellular processes such as nuclear division, chromosome segregation, and mitotic nuclear division, ultimately leading to tumor cell apoptosis⁴³. This aligns with previous studies demonstrating that inhibition of TOP2A can induce significant cytotoxic effects in cancer cells, particularly those exhibiting high proliferation rates⁴⁴.

Resistance to platinum-based therapies remains a significant challenge in treating TNBC⁴⁵. The KEGG pathway analysis in our study suggests that PA may influence platinum drug resistance mechanisms. Resistance to platinum drugs often involves several cellular processes, including DNA damage repair, drug efflux, and alterations in apoptosis pathways^46,47. By potentially enhancing the sensitivity of TNBC cells to platinum agents, PA could address one of the critical barriers in the effective treatment of this aggressive cancer subtype. This findings highlights the importance of integrating molecular targeting with established chemotherapeutic regimens, a strategy that could lead to improved clinical outcomes for patients with TNBC.

The TME plays a crucial role in tumor progression and response to therapy⁴⁸. Our findings indicate that PA may enhance the activity of various immune cells, including activated dendritic cells, eosinophils, M0 macrophages, M1 macrophages, and activated CD4 memory T cells. Dendritic cells are essential for orchestrating anti-tumor immunity, as they present antigens to T cells and activate adaptive immune responses⁴⁹. By promoting dendritic cell activation, PA could facilitate a more robust immune response against TNBC cells.

Furthermore, the polarization of macrophages from the M0 to M1 phenotype is associated with enhanced anti-tumor activity⁵⁰. M1 macrophages secrete pro-inflammatory cytokines and can directly kill tumor cells, while M0 macrophages are typically immunosuppressive⁵¹. Our study suggests that PA may shift the balance towards a more immune-active state within the TME, potentially improving therapeutic efficacy. The involvement of CD4 memory T cells indicates that PA may also help sustain long-term immune responses against TNBC, an area that warrants further exploration in clinical settings⁵².

One of the significant findings of our study is the differential therapeutic effects of PA across various TNBC subtypes, particularly TNBC-BL1, TNBC-Metaplastic, and TNBC-BL2⁵³. The heterogeneity of TNBC is a major challenge in developing effective treatments⁵⁴. The BL1 subtype is characterized by high proliferation and increased DNA repair mechanisms, making it particularly susceptible to TOP2A inhibitors⁵⁵. In contrast, the Metaplastic subtype often exhibits aggressive behavior and may respond favorably to therapies that modulate the immune environment⁵⁶. The BL2 subtype, known for its immune-rich microenvironment, may benefit from PA’s immunomodulatory effects⁸. Understanding these subtype-specific responses is critical for optimizing therapeutic strategies and developing personalized medicine approaches for TNBC patients.

This study presents several significant strengths. Firstly, we systematically collected and standardized a comprehensive array of GEO datasets, ensuring a robust foundation for our analyses. We also validated gene expression data using the TCGA database, facilitating a thorough examination of the relationship between gene expression profiles and patient prognosis. Moreover, we performed in vitro cellular experiments to substantiate some of the findings derived from our network pharmacology analysis, thereby enhancing the translational relevance of our results. However, this study is not without limitations. Specifically, we did not conduct extensive validation of all results obtained from the network pharmacology analysis. This endeavor represents a considerable undertaking, requiring additional time and resources for comprehensive exploration. Furthermore, while in vitro experiments provided valuable insights, the lack of in vivo validation limits our ability to fully assess the clinical applicability of PA in therapeutic settings. The complex tumor microenvironment and its interaction with potential therapeutic agents could result in outcomes that differ from those observed in vitro. Addressing these gaps will be a primary focus of our ongoing research initiatives to bridge the transition from preclinical findings to clinical applications.

Conclusion

In summary, our study provides compelling evidence that PA shows promise as a foundation for targeted TNBC therapies, either as a standalone drug leveraging its cytotoxic and immunomodulatory effects or in combination with existing chemotherapeutics to overcome drug resistance. By addressing platinum drug resistance and enhancing anti-tumor immune responses, PA could contribute to improved treatment outcomes in TNBC patients. Further research is warranted to validate these findings and explore the potential of PA in combination with existing therapies, particularly focusing on its effects across various TNBC subtypes, such as particular in vivo research or clinical trial.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

Triple-negative breast cancer

TNBC

Traditional Chinese medicine

TCM

Pachymic acid

PA

Gene Expression Omnibus

GEO

Protein-Protein Interaction

PPI

Gene Ontology

GO

Kyoto Encyclopedia of Genes and Genomes

KEGG

Human epidermal growth factor receptor 2

HER2

Breast cancer

BRCA

Neoadjuvant systemic therapy

NAST

Overall survival

OS

Immune checkpoint blockade

ICB

Natural killer

NK

Differentially expressed genes

DEGs

Biological processes

BP

Cellular components

CC

Molecular functions

MF

Dulbecco’s Modified Eagle Medium

DMEM

Fetal bovine serum

FBS

Standard deviation

SD

Author contributions

ML and JHT designed the study; ML and LZ conducted the experiments; YZ and JHT supervised the experiments; ML wrote the initial draft; ML, YZ, and JHT revised the manuscript.

Data availability

The datasets generated and/or analysed during the current study are available in the GEO database and TCGA database repository, [https://www.ncbi.nlm.nih.gov/geo/; https://www.cancer.gov/ccg/].

Declarations

Competing interests

The authors declare no competing interests.