A Network Pharmacology Analysis of the Systems-Perspective Anticancer Mechanisms of the Herbal Drug FDY2004 for Breast Cancer

Abstract

Breast cancer is a malignant tumor with high incidence, prevalence, and mortality rates in women. In recent years, herbal drugs have been assessed as anticancer therapy against breast cancer, owing to their promising therapeutic effects and reduced toxicity. However, their pharmacological mechanisms have not been fully explored at the systemic level. Here, we conducted a network pharmacology analysis of the systems-perspective molecular mechanisms of FDY2004, an anticancer herbal formula that consists of Moutan Radicis Cortex, Persicae Semen, and Rhei Radix et Rhizoma, against breast cancer. We determined that FDY2004 may contain 28 active compounds that exert pharmacological effects by targeting 113 breast cancer-related human genes/proteins. Based on the gene ontology terms, the FDY2004 targets were involved in modulating biological processes such as cell growth, cell proliferation, and apoptosis. Pathway enrichment analysis identified various breast cancer-associated pathways that may mediate the anticancer activity of FDY2004, including the PI3K-Akt, MAPK, TNF, HIF-1, focal adhesion, estrogen, ErbB, NF-kappa B, p53, and VEGF signaling pathways. Thus, our analysis offers novel insights into the anticancer properties of herbal drugs for breast cancer treatment from a systemic perspective.

Introduction

Although cancer research and biomedical technologies have greatly advanced, breast cancer (BC) remains a malignant tumor with high incidence, prevalence, and death rates. ¹ The underlying molecular mechanisms of BC carcinogenesis and progression involve dysregulation of various oncogenes, tumor suppressors, and oncogenic signaling. ² Current primary pharmacological treatment strategies for BC include chemotherapy using cytotoxic agents such as anthracyclines and taxanes, endocrine therapy, and targeted therapy.^3-5 However, such therapies frequently have limited efficacy and give rise to unwanted side effects such as myelosuppression and immunosuppression, cardiovascular toxicities, osteoporosis, gastrointestinal symptoms and dysfunction, and peripheral neuropathy, which can affect the physical health and quality of life of patients.^6-9 Therefore, increased attention has been devoted to the use of herbal drugs, which are therapeutic agents with polypharmacological features, as effective anticancer therapeutics against BC, owing to their promising therapeutic effects and reduced toxicity.^10-12 Results from clinical studies indicate that the combinatorial use of herbal drugs and chemotherapy can increase the tumor response, improve the quality of life, alleviate toxic events, and prolong the survival of cancer patients when compared with the case of using chemotherapy alone.^13,14

FDY2004 is an herbal decoction prepared from three herbal medicines, ¹⁵ Moutan Radicis Cortex (MRC), Persicae Semen (PS), and Rhei Radix et Rhizoma (RRR), which possess potent anticancer properties against different cancer types, including BC.^16-19 Although FDY2004 is known to exert antiproliferative and antioxidant effects in BC cells, ¹⁵ the pharmacological mechanisms of its anticancer activity against BC have not been investigated from a systemic perspective.

Since herbal drugs exert their pharmacological activities in a multicompound-multitarget manner, experimental biological studies often face difficulties in exploring their complex polypharmacological therapeutic mechanisms.^11,12,20-26 To address these challenging issues, network pharmacology, a multidisciplinary field of research that integrates network systems biology, pharmacology, and medicine, has gained much interest as an effective and practical methodology that can facilitate the systematic understanding of the mechanisms of action of herbal medicines.^11,20-26 This convergence science aims to explore the polypharmacological properties of herbal drugs, which involves the synergistic interactions between their chemical components and multiple biological targets, such as disease-related genes, proteins, and metabolites.^11,20-26 Previous studies have shown that the network pharmacology approach is useful for the identification of bioactive compounds in herbal medicines responsible for exerting pharmacological effects, the investigation of their interacting key therapeutic targets, and the analysis of system-perspective polypharmacological features and mechanisms of the herbal drugs.^11,20-26 Here, we investigated the anti-BC mechanisms of FDY2004 from a systemic perspective through a network pharmacology methodology.

Materials and Methods

Determination of Active Chemical Components in FDY2004

Comprehensive data regarding the chemical components present in the herbal medicines in FDY2004 were obtained from the Traditional Chinese Medicine Systems Pharmacology (TCMSP), ²⁷ Anticancer Herbs Database of Systems Pharmacology (CancerHSP), ²⁸ Traditional Chinese Medicine Integrated Database (TCMID), ²⁹ and Bioinformatics Analysis Tool for Molecular Mechanism of Traditional Chinese Medicine (BATMAN-TCM) ³⁰ databases. By evaluating the absorption, distribution, metabolism, and excretion (ADME) parameters of the chemical compounds in FDY2004 obtained from those databases, we identified the bioactive compounds in the herbal drug that are primarily responsible for exerting its pharmacological activities. Among the ADME characteristics of the chemical components, we used the Caco-2 permeability, oral bioavailability, and drug-likeness, which are the most widely used pharmacokinetic features in network pharmacological studies,^21,27,31 for the identification of active chemical components. Oral bioavailability is a major determinant of drug design and development; it refers to the fraction of chemical compounds that are successfully absorbed into systemic circulation and delivered to the target sites after oral administration.^27,32 A chemical component is effectively absorbed by the human body if its oral bioavailability is greater than 30%.^27,32 Caco-2 permeability is a parameter that measures the diffusion and absorption rates of a compound passing across Caco-2 human intestinal cells; it is widely used for the evaluation of the intestinal permeability characteristics of a drug. ²⁷ A chemical component with a Caco-2 permeability larger than −0.4 is considered permeable in the intestinal epithelium. ³³ Drug-likeness is a key indicator that qualitatively investigates the suitability of a compound as a drug based on its molecular, pharmacokinetic, physicochemical, and structural properties.^27,34 Interestingly, the average drug-likeness value of all drugs is 0.18, which is generally used as the threshold in network pharmacology studies that evaluate the potential of a compound for pharmacological use.^27,34 The following criteria were used to screen the bioactive compounds of FDY2004, as previously suggested:^21,27,31 drug-likeness ≥ 0.18, oral bioavailability ≥ 30%, and Caco-2 permeability ≥ −0.4.

Investigation of the Targets of the Active Compounds

To explore the targets of the active components of FDY2004 in humans, the canonical simplified molecular input line entry specification (SMILES) information for active chemical components was obtained from the PubChem database. ³⁵ Then, we imported the SMILES information into in silico models and databases to investigate the chemical-protein interactions and obtain the targets for Homo sapiens, which involved the following tools: SwissTargetPrediction, ³⁶ PharmMapper, ³⁷ Search Tool for Interactions of Chemicals (STITCH) 5, ³⁸ and Similarity Ensemble Approach (SEA). ³⁹ The BC-related human targets were investigated using the Medical Subject Headings term “Breast Neoplasms” (ID: D001943) and limiting the search to Homo sapiens in the following databases: Online Mendelian Inheritance in Man, ⁴⁰ Human Genome Epidemiology Navigator, ⁴¹ DisGeNET, ⁴² Therapeutic Target Database, ⁴³ GeneCards, ⁴⁴ Comparative Toxicogenomics Database, ⁴⁵ Pharmacogenomics Knowledgebase, ⁴⁶ and DrugBank. ⁴⁷

Development of Herbal Medicine-Associated Networks

The herbal medicine-active chemical compound-target (H-C-T) network was generated by linking the herbal constituents with their active chemical constituents, and the chemical constituents with their BC-related targets. The H-C-T-pathway (H-C-T-P) network was generated by linking the targets in the H-C-T network and their involved BC-associated signaling pathways. The protein-protein interaction (PPI) network was built based on the interaction pairs of the targets (interaction confidence score ≥ 0.9) obtained from the STRING database. ⁴⁸ The Cytoscape software was used for visualization and analysis of the networks. ⁴⁹ In a network, nodes refer to herbal medicines, chemical compounds, targets, and pathways, and links (or edges) refer to their interaction relationship. ⁵⁰ The degree refers to the number of edges linked to a certain node. ⁵⁰

Survival Analysis

The association between the survival rates of patients with BC and the expression levels of the FDY2004 targets was analyzed using the Kaplan-Meier Plotter. ⁵¹

Functional Enrichment Analysis

Functional enrichment analyses for the targets of interest in terms of gene ontology (GO) and pathway were carried out using g:Profiler ⁵² and the Kyoto Encyclopedia of Genes and Genomes database, ⁵³ respectively. We also used GeneMANIA ⁵⁴ to perform functional association analysis for the targets.

Analysis of Molecular Docking Activity

The chemical structures of the compounds were obtained from PubChem, ³⁵ and the 3D protein structures of targets were retrieved from the RCSB Protein Data Bank. ⁵⁵ Based on the structural information, the molecular docking scores of compound-target pairs were evaluated using the Autodock Vina. ⁵⁶ The resulting compound-target pairs were considered to have potent binding affinity if their docking scores were ≤ −5.0 as previously described. ⁵⁷

Results

Overall Research Flow of the Network Pharmacological Analysis of the Anti-Breast Cancer Mechanisms of FDY2004

To conduct a network pharmacology analysis of the anticancer mechanisms of FDY2004 against BC, we first searched for its phytochemical components using herbal medicine-associated databases (Figure 1). Then, we screened for active compounds using the ADME features of individual phytochemical compounds (Figure 1). Next, we identified the targets of the bioactive components by investigating their chemical-protein interactions (Figure 1). In addition, BC-associated targets were obtained from diverse biomolecular databases (Figure 1). Finally, based on the comprehensive biomedical information regarding FDY2004, we performed a network pharmacology analysis of its anticancer molecular mechanisms against BC from system-level perspective (Figure 1).

Figure 1.

Schematic illustration of the research flow of the network pharmacology analysis exploring anticancer molecular mechanisms of FDY2004 against breast cancer.

Bioactive Chemical Compounds in FDY2004

Comprehensive information on the chemical compounds in FDY2004 was obtained from TCMSP, ²⁷ CancerHSP, ²⁸ TCMID, ²⁹ and BATMAN-TCM ³⁰ (Supplementary Table S1). Compounds with drug-likeness ≥ 0.18, oral bioavailability ≥ 30%, and Caco-2 permeability ≥ −0.4 were classified as active based on their ADME parameters, as previously suggested.^21,27,31 In addition, some chemical constituents that did not satisfy the criteria were included as active constituents due to their potent anti-BC activities and their considerable quantities among the herbal constituents of FDY2004. In total, 35 bioactive components were identified in FDY2004 (Supplementary Table S2).

Exploration of FDY2004 Targets

The following in silico databases and algorithms used for the protein-chemical interaction analysis were employed to identify the targets of the bioactive compounds in FDY2004: PharmMapper, ³⁷ SwissTargetPrediction, ³⁶ STITCH 5, ³⁸ and SEA. ³⁹ In total, 113 human BC-related and 96 non-BC-related targets were retrieved for FDY2004 (Supplementary Table S3).

Network Pharmacological Investigation of FDY2004 for Breast Cancer

By connecting the herbal constituents to their active chemical constituents and the active chemical constituents to their BC-associated targets, we built an H-C-T network for the herbal decoction to explore the anti-BC mechanisms of FDY2004 from a systemic perspective (Figure 2). The network consisted of 144 nodes (3 herbal constituents, 28 active chemical constituents, and 113 BC-related targets) and 244 edges (Figure 2 and Supplementary Table S3). The bioactive chemical components with the largest number of BC-associated targets were quercetin, kaempferol, gallic acid, (+ )-catechin, ( + )-epicatechin, (-)-catechin, and emodin (Figure 2 and Supplementary Table S3), suggesting their potential roles as key anti-BC pharmacological compounds in FDY2004. Furthermore, 41.59% (47 out of 113 nodes) of BC-associated human genes/proteins were targeted by two or more phytochemical components of FDY2004 (Figure 2), which demonstrated the multiple compound-multiple target nature of herbal drugs.

Figure 2.

The herbal medicine-active chemical compound-target network of FDY2004. Green hexagonal nodes, herbal medicines; red rectangular nodes, active chemical components; blue elliptical nodes, breast cancer-related targets.

Understanding the biological features underlying the complex interactions between genes and proteins is important for the investigation of the treatment mechanisms of drug therapies. ⁵⁸ To this end, we built a PPI network (86 nodes and 212 edges) for the BC-associated FDY2004 targets and analyzed its topological characteristics to investigate hub nodes in the network (Figure 3). Of note, hubs refer to nodes that have a higher number of links compared to others in a network, and they are reported to serve as attractive therapeutic targets owing to their crucial biological roles. ⁵⁹ In the analysis, nodes were considered as hubs if their degree was equal to or higher than two times the average value of the degree of all nodes in the network as previously described. ⁶⁰ The results demonstrated that TP53 (degree = 25), PIK3R1 (degree = 20), HSP90AA1 (degree = 17), VEGFA (degree = 16), TNF (degree = 14), AKT1 (degree = 14), JUN (degree = 13), CYP1A1 (degree = 11), ESR1 (degree = 11), PTK2 (degree = 11), and EGFR (degree = 11) were determined to be hubs (Figure 3). We further found that these BC-associated hub targets of FDY2004 may serve as prognostic factors for the survival of patients with BC. High expression levels of AKT1, CYP1A1, EGFR, ESR1, JUN, PIK3R1, PTK2, TNF, and TP53, along with low expression levels of HSP90AA1 and VEGFA were correlated with an increased survival rate of patients with BC (Figure 4), thus implying their clinical significance.

Figure 3.

The protein-protein interaction network for breast cancer-related targets of FDY2004.

Figure 4.

Survival analysis of breast cancer-associated targets of FDY2004. Kaplan-Meier curves for the survival of patients with breast cancer with respect to the expression levels of the indicated FDY2004 targets.

Overall, the results revealed the network-level polypharmacological mechanisms that mediate the anticancer properties of FDY2004 against BC.

Functional Enrichment Analysis of FDY2004 Mechanisms

To assess the molecular mechanisms of the anti-BC activity of FDY2004, we investigated the GO enrichment of its targets. These targets were associated with GO terms for modulating various biological functions, which included cell proliferation, cell growth, and apoptotic cell death (Supplementary Figure S1), implying the molecular-level pharmacological property of FDY2004. GeneMANIA ⁵⁴ analysis for FDY2004 targets further showed that they might functionally interact via various biological mechanisms, such as co-expression, physical interaction, and co-localization (Supplementary Figure S2).

Because diverse signaling processes are implicated in the tumorigenesis, development, and progression of BC, ⁶¹ we explored pathway enrichment of the BC-related FDY2004 targets (Figure 5 and Supplementary Figure S1). The following pathways were shown to be enriched with the FDY2004 targets: “Pathways in cancer”, “PI3K-Akt signaling pathway”, “Endocrine resistance”, “MAPK signaling pathway”, “TNF signaling pathway”, “Apoptosis”, “HIF-1 signaling pathway”, “Breast cancer”, “EGFR tyrosine kinase inhibitor resistance”, “Focal adhesion”, “Steroid hormone biosynthesis”, “Cellular senescence”, “Drug metabolism - cytochrome P450”, “Estrogen signaling pathway”, “PD-L1 expression and PD-1 checkpoint pathway in cancer”, “Platinum drug resistance”, “ErbB signaling pathway”, “NF-kappa B signaling pathway”, “p53 signaling pathway”, and “VEGF signaling pathway” (Figure 5 and Supplementary Figure S1).

Figure 5.

The herbal medicine-active chemical compound-target-pathway network of FDY2004. Green hexagonal nodes, herbal medicines; red rectangular nodes, active chemical components; blue elliptical nodes, breast cancer-related targets; orange diamond nodes, breast cancer-associated signaling pathways.

Together, these functional analyses revealed the molecular- and pathway-perspective properties of the anti-BC activity of FDY2004.

Molecular Docking Analysis of Active FDY2004 Constituents and Their Targets

To examine the potential binding activity of the phytochemical components of FDY2004 and their targets, we performed a molecular docking analysis (see Materials and Methods). The analysis results showed that the docking scores of 95.11% of the active compound-hub target pairs were ≤ −5.0 (Figure 6 and Supplementary Table S4), suggesting their potential binding interactions.

Figure 6.

Molecular docking analysis for the active chemical compounds of FDY2004 and their targets. (A) Aloe-emodin and HSP90AA1 (score = −7.2). (B) Aloe-emodin and TP53 (score = −7.1). (C) Emodin and ESR1 (score = −7.8). (D) Emodin and TNF (score = −6.8). (E) Emodin and TP53 (score = −7.3). (F) Emodin and VEGFA (score = −7.0). (G) Gallic acid and JUN (score = −5.1). (H) Pentagalloylglucose and EGFR (score = −8.9). (I) Pentagalloylglucose and ESR1 (score = −7.0). (J) Quercetin and AKT1 (score = −6.4). (K) Quercetin and CYP1A1 (score = −7.5). (L) Quercetin and EGFR (score = −7.9). (M) Quercetin and PIK3R1 (score = −6.3). (N) Quercetin and PTK2 (score = −6.1). (O) Rhein and VEGFA (score = −7.1).

Discussion

In recent years, increasing attention has been given to herbal drugs as anticancer treatments for BC, a malignant tumor with high incidence, prevalence, and mortality rates, ¹ based on their promising therapeutic activity and reduced toxicity. ¹¹ However, their pharmacological properties and mechanisms have not been fully understood at the systemic level. Here, we analyzed the molecular mechanisms of the herbal anticancer drug FDY2004 ¹⁵ against BC from a systemic perspective. We found that FDY2004 may contain 28 active compounds that target 113 BC-related human genes/proteins to exert its anticancer activity. GO enrichment analysis showed that the FDY2004 targets were involved in the functional modulation of biological functions, involving cell proliferation, cell growth, and apoptosis. Pathway enrichment analysis revealed key BC-associated pathways that may confer the anticancer effects of FDY2004; these include PI3K-Akt, MAPK, TNF, HIF-1, focal adhesion, estrogen, ErbB, NF-kappa B, p53, and VEGF signaling pathways. These findings provide information about the molecular- and pathway-level anticancer properties of herbal medicines for BC treatment.

The hub targets of FDY2004 have been found to be associated with the pathological process of BC. The loss-of-function mutation of TP53 contributes to the malignant progression of BC, and its expression may act as a useful prognostic factor in BC patients.^62-66 High expression levels of the PIK3R1 and HSP90AA1 genes are related to reduced survival rates in patients with BC.^67,68 Vascular endothelial growth factor (VEGF)-A (encoded by VEGFA) modulates the angiogenesis, metastasis, and proliferation of BC cells, and its activation may result in chemotherapeutic resistance.^69-71 The inflammatory cytokine tumor necrosis factor (TNF)-α (encoded by TNF) modulates proliferation, motility, migration and invasion, and metastasis of BC cells, and its activity is associated with the aggressiveness and progression of BC.^72-76 Dysregulated activation of the oncogenic kinase AKT (encoded by AKT1), c-Jun (encoded by JUN), epidermal growth factor receptor (EGFR; encoded by EGFR), and focal adhesion kinase (FAK; encoded by PTK2) is implicated in cellular processes of BC (eg, proliferation, survival, migration, invasion, metastasis, cancer stemness, and epithelial-to-mesenchymal transition [EMT]), and their activities correlate with the reduced survival of patients with BC.^77-101 Previous studies have further reported that these kinases are involved in the development of resistance against anticancer therapies and that their targeting can enhance the efficacy of chemotherapy and radiotherapy for BC.^{79-81,84-86,88,90,95,96,102-109} Cytochrome P450 1A1 (encoded by CYP1A1), a major regulator of estrogen metabolism, is linked to the proliferation and survival of BC cells.^110-113 Genetic polymorphisms and aberrant activity of estrogen receptor α (ERα; encoded by ESR1) is one of the major causes of BC pathogenesis, which makes it the most common therapeutic target for BC treatment.^111-116

The BC-associated pathways targeted by FDY2004 are the key signaling mechanisms for the malignancy. The impaired regulation of focal adhesion, mitogen-activated protein kinase (MAPK), erythroblastic leukemia viral oncogene homolog (ErbB), and PI3K-Akt pathways are important signaling mechanisms for controlling various cancerous phenotypes of BC cells, involving survival, proliferation, metastasis, mortality, migration, invasion, cancer stemness, tumorigenic potential, and angiogenesis, which substantially contribute to BC initiation and progression.^96,117-121 The TNF and nuclear factor kappa-light-chain-enhancer of activated B cells (NF-kB) pathways function as key regulators of inflammatory processes and their activities correlate with the promotion, progression, metastasis, therapeutic resistance, and prognosis of BC.^122-128 Uncontrolled activation of the estrogen signaling pathway is the most crucial risk factor for BC and the pathway serves as the major target for anti-BC therapeutics.^129-131 The expression status of programmed death-ligand 1 (PD-L1) may act as an indicator of prognosis for BC patients, and targeting the programmed cell death protein 1 (PD-1)/PD-L1 pathway can augment anticancer immunity.^132-135 The hypoxia-inducible factor (HIF)-1 signaling pathway controls a variety of cellular behaviors involving proliferation, stemness, and metastasis of BC cells.^136,137 Malfunction of the tumor suppressor p53 pathway is commonly seen in the BC carcinogenic process, and its status is associated with the clinical efficacy and prognosis of patients with BC.^64,138 The VEGF pathway promotes the disease progression of BC via the stimulation of angiogenesis, survival, and metastasis of BC cells.^71,139 The disrupted coordination of key cellular responses, such as apoptosis and cellular senescence, is associated with diverse BC cell behaviors and also with the development of chemo- and radio-therapeutic resistance.^140-145 Resistance to endocrine, EGFR-targeted, and platinum-based therapies is responsible for the ineffectiveness and failure of BC treatment.^146-150

The therapeutic activities of the herbal and chemical components of FDY2004 against BC have been reported previously. PS suppresses estrogen signaling activity, which has anti-aromatase, antiproliferative, and cytotoxic effects on BC cells. ¹⁹ RRR blocks the migration, invasion, motility, and survival of BC cells by targeting PI3K-Akt, MAPK, and NF-kB signaling.^17,151 Catechins inhibit the growth and cell cycle progression of BC cells through MAPK regulation. ¹⁵² Epicatechin induces reactive oxygen species (ROS)-mediated apoptosis of BC cells.^153,154 Aloe-emodin exerts anticancer stemness, anti-invasive, anti-migratory, antimetastatic, antiproliferative, and cytotoxic effects on BC cells through the modulation of ERα, Ras/extracellular-signal-regulated kinase (ERK), human epidermal growth factor receptor 2 (HER2), and PI3K/mammalian target of rapamycin (mTOR) oncogenic kinases and also the estrogen, PI3K-Akt, ErbB, and MAPK pathways.^155-159 Caffeic acid exerts growth-inhibitory, pro-apoptotic, anti-migratory, and chemosensitizing activities against BC cells by regulating p53, transforming growth factor-β (TGF-β), and insulin-like growth factor-1 receptor (IGF1R)-PI3K-Akt pathways.^160-163 Campesterol reduces the growth capability of BC cells by stimulating apoptosis. ¹⁶⁴ Chrysophanol regulates NF-kB, PI3K-Akt, and MAPK signaling and promotes ROS production, along with endoplasmic reticulum stress, which leads to growth suppression and apoptosis in BC cells.^165,166 Daucosterol arrests the cell cycle progression and proliferation while promoting apoptotic death of BC cells, and these effects involve the regulation of the PI3K-Akt pathway. ¹⁶⁷ Emodin suppresses proliferation, growth, survival, migration, invasion, metastasis, stemness, angiogenesis, EMT, and drug-resistant behaviors of BC cells through the pharmacological intervention of multiple BC-associated regulators and pathways, including the p53, estrogen, ErbB, PI3K-Akt, focal adhesion, senescence, MAPK, NF-kB, and VEGFR signaling pathways.^156,168-177 Gallic acid has anticancer activities for the suppression of proliferation, survival, cell cycle progression, invasion and migration of BC cells, which are mediated through the oncogenic EGFR, focal adhesion, MAPK, ErbB, PI3K-Akt, NF-kB, p53, Fas/FasL, cell cycle regulatory, and mitochondrial pathways.^178-183 Kaempferol and quercetin possess diverse anticancer mechanisms, including the induction of apoptosis and the suppression of proliferation, migration, invasion, cell cycle progression, metastasis, angiogenesis, and cancer stemness in BC cells; these therapeutic effects are mediated via pharmacological modulation of diverse crucial BC-associated signaling processes.^184-193 In addition, these chemical compounds may sensitize BC cells to various anticancer drugs.^194-199 Furthermore, the combination of kaempferol and quercetin produced more enhanced anticancer activity than either compound alone. ²⁰⁰ Mairin (betulinic acid) can inhibit the inflammatory response, proliferative and growth capability, survival capacity, angiogenesis, metastasis, cell cycle progression, and aerobic glycolysis of BC cells, and further chemosensitizes cells to anti-BC therapeutic agents.^201-206 Paeoniflorin functions as an anticancer compound that blocks proliferation, invasion, and EMT of BC cells via the modulation of Notch-1, PI3K-Akt, and HIF-1 signaling.^207-209 Pentagalloylglucose exerts antiproliferative effects by inactivating the cell cycle regulators, ERα, EGFR, Akt1, along with the ErbB, estrogen, and PI3K-Akt pathways.^210,211 Previous studies have reported that paeonol can induce growth suppression and apoptosis of BC cells by regulating the chemokine, NF-kB, and MAPK signaling, and enhance the anticancer efficacy of chemotherapeutic agents while reducing the adverse toxic effects.^212-215 Physcion has antiproliferative and anti-survival properties that promote apoptosis and cell cycle arrest of BC cells, mediated by the modulation of cell cycle and apoptotic regulatory proteins. ²¹⁶ Rhein can control the functional activities of p53, NF-kB, VEGFR, TNF, PI3K-Akt, MAPK, HIF-1, HER2, and apoptotic and cytokine signaling pathways to play its pro-apoptotic, antiproliferative, chemosensitizing, and anti-angiogenic roles in BC cells.^217-219 β-Sitosterol promotes the activation of major apoptosis-associated pathways, including the Fas and caspase signaling pathways, which results in the reduction of BC cell viability.^164,220,221 Moreover, β-sitosterol may enhance the anticancer efficacy of tamoxifen, an endocrine therapeutic agent widely used in clinics for BC. ²²² Thus, these data further support the experimental evidence for the anticancer mechanisms of FDY2004 in BC treatment.

Previous preclinical and clinical studies have suggested that the use of herbal drugs can increase the efficacy of anticancer therapies, alleviate the development of adverse reactions, and prolong the survival of cancer patients.^13,14 Further studies that can verify and advance the findings obtained from the network pharmacological investigation are warranted to design and develop effective anticancer herbal drugs and therapies. First, the therapeutic effects of FDY2004 should be widely assessed in other cancer types, such as gastric, colorectal, lung, liver, and cervical cancer. In addition, different potential mechanisms of action of the herbal drug against cancer should be investigated, including proliferation, survival and apoptosis, migration and invasion, metastasis, metabolism, cancer stemness, differentiation of cancer cells, angiogenesis, immunomodulatory activity, and tumor microenvironment.^223,224 Integrative research that combines network pharmacology and advanced research methodologies, such as high-throughput technologies, is also needed for an in-depth investigation of the distinct pharmacological properties of herbal medicines, which can vary depending on their treatment time and dose. Moreover, the therapeutic efficacy, safety, and toxicity of the combinatorial treatment of FDY2004 with standard anticancer strategies (eg, chemotherapeutics, molecularly targeted therapies, immune checkpoint inhibitor therapeutics, and radiotherapy) should be examined in further preclinical and clinical studies. These in-depth studies may facilitate the improvement of herbal medicinal strategies for BC.

To summarize, we explored the molecular mechanisms of the anticancer herbal formula FDY2004 against BC from a systems perspective based on a network pharmacological investigation. We found that the herbal drug contains 28 active components that interact with 113 BC-related targets to achieve its pharmacological effects. The classification based on GO terms indicated that the FDY2004 targets were involved in modulating biological processes, including cell proliferation, cell growth, and apoptosis. Pathway enrichment analysis further identified various important BC-associated pathways that may mediate the anticancer activity of FDY2004. Thus, the overall data offer novel insights into the anticancer properties of herbal drugs as therapies for BC from a systemic perspective. Future preclinical and clinical studies are warranted for the design and development of advanced herbal therapy-based cancer treatment strategies.