Exploration of the Molecular Mechanism of Danzhi Xiaoyao Powder in Endometrial Cancer through Network Pharmacology

Endometrial cancer (EC) is a common malignant tumor of the female reproductive system. Current treatments such as surgery and long-term hormone therapy are ineffective and have side effects. Danzhi Xiaoyao powder (DXP) can inhibit the growth of EC cells and induce apoptosis, but the pharmacological and molecular mechanisms of anticancer effects are still unclear. In this study, active components and potential targets of DXP were obtained from public databases. Protein effects and regulatory pathways of common targets were analyzed by protein-protein interaction (PPI), GO and KEGG. The results of network pharmacology showed that there are 87 common targets between EC and DXP. GO enrichment analysis showed that these targets were associated with response to oxidative stress, response to nutrient levels, hormone receptor binding and nuclear hormone receptor binding, etc. The results of KEGG analysis indicated that IL-17, TNF, PI3K/AKT, and RAS/RAF/MEK/ERK (ERK) signaling pathway were enriched in the anti-EC of DXP. Additionally, we cultured HEC-1B and KLE cells for validate experiments. DXP showed an inhibition of proliferation, migration, and cell cycle of both cells. Moreover, the expression of RAS, p-RAF, p-MEK, ERK, and p-ERK related proteins were downregulated. In conclusion, DXP might inhibit the proliferation of EC cells via apoptosis. Furthermore, DXP-induced inhibition of EC development might involve RAS/RAF/MEK/ERK pathway.


Introduction
Endometrial cancer (EC) is one of the malignant tumors of the female reproductive system, and its prevalence and mortality have rapidly risen worldwide [1,2]. Globally, there were an estimated 417,000 incident cases and 97,000 deaths in 2020 [3]. e occurrence of EC is usually associated with overexposure to endogenous or exogenous estrogens [4]. Currently, the main treatment methods for EC are surgery and chemotherapy [5]. Chemotherapy drugs have certain toxicity, and due to the evolutionary nature of tumors, tumors cells may differentiate into subclonal cells with certain drug resistance, resulting in reduced antitumor efficacy of these chemotherapy drugs and possible adverse reactions [6,7]. erefore, it is important to find drugs with low side effects and better efficacy to improve the condition of EC.
With the development of cellular biotechnology, breakthroughs have been made in the research of anticancer herbal medicines. In the past 20 years, natural antitumor drugs have accounted for 66.29% of the anticancer drugs on the market [8]. Traditional Chinese medicine (TCM) has unique advantages in terms of efficacy and safety. Its reasonable application can help control the development of the disease, reduce the toxic effect of radiotherapy and avoid the recurrence of cancer metastasis to a large extent [9]. Danzhi Xiaoyao powder (DXP) is a formula of TCM including Cortex Moutan, Gardeniae Fructus, Radix Bupleuri, Angelicae Sinensis Radix, Paeoniae Radix Alba, Atractylodes macrocephala Koidz., Poria Cocos (Schw.) Wolf., and Licorice [10]. Studies have found that DXP has antidepressant [11] effects, and some studies have found that DXP has some efficacy against breast cancer and precancerous lesions [12,13], especially ER-positive breast cancer [13], suggesting that DXP may also have some anticancer effects. Clinical studies have shown that DXP has good therapeutic effects on patients with endometriosis liver depression and blood heat type [14]. erefore, DXP has the potential to be a drug for the treatment of EC. However, its active components, possible targets, and molecular mechanism have not been elucidated.
TCM plays a therapeutic role by restoring homeostasis in the biological process during disease development through multiple targets [15]. Network pharmacology is a new method to analyze the relationship between drugs, ingredients, targets, and diseases in the network based on biological networks and is gaining importance [16][17][18]. erefore, the combining of TCM and network pharmacology is a new approach for the preferential selection of disease-related genes, revealing the association between diseases and drugs and further elucidating the regulatory role of TCM [19,20].
In this study, network pharmacology combined with in vitro cellular experiment validation was performed to explore the active components, potential targets, and molecular mechanism of the anti-EC effect of DXP.
e duplicate disease targets were merged and deleted.

Collection of Active Components and Potential Targets of DXP.
To collect all the relative components of DXP, we searched the traditional Chinese medicine systems pharmacology database and analysis platform (TCMSP, https:// lsp.nwu.edu.cn/tcmsp.php) [22], Traditional Chinese Medicine Integrated Database (TCMID, https://tcm.cz3.nus.edu. sg/group/tcm-id/tcmid) [23], and Herbal Ingredients' Targets Database (HIT, https://lifecenter.sgst.cn/hit) [24], which includes a great deal of information about Chinese herbal medicine. All predicted targets of active components of DXP were searched from TCM-ID, TCMSP, HIT, and Search Tool for Interacting Chemicals (STITCH, https:// stitch.embl.de) databases and selected according to the component-target correlation degree above 400 from STITCH database.

Drug-Likeness Screening.
In the early stages of drug discovery, drug-likeness is a key consideration in the selection of compounds. A quantitative estimate of druglikeness (QED), reported by Bickerton et al. in 2012, was used to evaluate a component's drug-likeness [25]. In this work, DXP components with QED ≥ 0.3 were included for further study.

Binomial Statistical Mode Analysis.
According to previous studies, a binomial statistical mode was used for secondary screening for active components and potential targets of DXP [26,27]. In this study, we set targets with P < 0.05 that could be used as main targets for the formula.

Common Target Screening and Protein-Protein Interaction (PPI) Construction.
e gene targets of components of DXP and EC were analyzed by Venny (version 2.1) to obtain the common genes, which were the targets of action of DXP for the treatment of EC. e PPI network was constructed by importing the common targets to the String database (https://string-db. org/cgi/input.pl), limiting the species to "Homo sapiens." e results were imported into Cytoscape (version 3.8.0) for further analysis and optimization of the protein interaction network.

Construction of Herb-Components-Target (H-C-T)
Network. An H-C-T network was constructed based on the active components of DXP and common targets. e above information was imported into Cytoscape for further analysis.

Enrichment Analysis.
e clusterProfiler package of R (version 3.15.4) was applied for Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis. e GO analysis enriching for 3 aspects of potential targets is as follows: biological process (BP), cellular component (CC), and molecular function (MF). P < 0.01 was set as a screening threshold and the functional terms and pathways with this value were considered statistically significant and retained.

Cell Culture.
e EC cell line HEC-1B and KLE cells were purchased from the Shanghai Institute for Biological Sciences, Chinese Academy of Sciences. HEC-1B cells were cultured in Dulbecco's Modified Eagle's Medium (DMEM) medium (Gibco, MD, USA) with 10% fetal bovine serum (Gibco); the KLE cells were cultured in DMEM/F12 medium (Gibco) with 10% fetal bovine serum (Gibco), both at 37°C with 95% air and 5% CO 2 . For apoptosis analysis, cells were digested with trypsin without EDTA, washed twice with prechilled PBS, and then resuspended in the binding buffer. Next, cells were incubated with Annexin V-FITC (Becton-Dickinson) and PI for 15 min at room temperature under lightproof conditions. Finally, a flow cytometer (Becton-Dickinson) was used to detect the cell apoptosis of HEC-1B and KLE cells.

Cell Migration.
e ability of migration in HEC-1B and KLE cells was detected by transwell assays. HEC-1B and KLE cells were resuspended in serum-free medium. 200 μL·cell suspensions were added to the upper layer of the transwell chamber (Corning, NY, USA). e lower chamber contained 500 μL of serum-containing medium. After 48 h incubation, the cells in the lower chamber were stained with 0.1% crystal violet for 20 min and counted under a 400 × microscope (Olympus, Japan) in 5 random fields.

Statistical Analysis.
e results were presented as means ± SD and data comparison of multiple groups was adopted by one-way ANOVA followed by Tukey's test. All statistical analyses were obtained by GraghPad 7.0. e results were considered statistically significant if the P value was less than 0.05.

EC-Related Targets.
By retrieving three public databases, 836 EC-related targets were obtained, which included 237 in GeneCards, 197 in OMIM, and 402 in DisGeNET with the key word "endometrial cancer." A total of 679 targets were screened by combining and deleting the repetitive values.

Active Components and Potential Targets of DXP.
A total of 321 DXP-related components were searched from public databases including HIT, TCMSP, and TCMID databases. e potential targets of DXP were obtained from four public databases (HIT, TCMSP, TCMID, and STITCH), and there were 5,558 targets related to active components of DXP.
en, 274 active components were obtained according to drug-likeness screening (QED ≥ 0.3). Moreover, 589 main targets were obtained from binomial statistical mode analysis. Following it, we retrieved 214 active components of DXP. Among 589 screened components-related targets and 679 EC-related targets, 87 common targets were identified and considered as the key targets for further study ( Figure 1). Accordingly, 168 active components were screened and identified as the main active components of DXP that act on EC (Table 1).

Construction of PPI Network.
e 87 common targets of DXP and EC were imported to the String database for analysis, and Cytoscape (version 3.8.0) was used for PPI construction ( Figure 2). It contained 87 nodes and 1,587 edges, and the average degree value was 36.5.

Construction of Herb-Components-Target (H-C-T) Network.
e H-C-T network was obtained through the import of 168 active components and 87 potential common targets of DXP into Cytoscape (version 3.8.0) and is shown in Figure 3. ere were 262 nodes and 1159 edges. e node size was proportional to its degree value. e red rhombus stood for different herbs which were contained in DXP. e cyan hexagon represented key targets of DXP acting on EC. Other different colored rhombuses represented the compounds contained in different herbs. Among them, the orange rhombus stood for the components that were contained in many kinds of herbs. e result showed that the top 5 active components nodes with the largest size of degree were quercetin, ethanol, kaempferol, IFP, and palmitic acid,

Enrichment Analysis.
In order to illuminate the biological functions and signaling pathways during the process of anti-EC of DXP, we employed the clusterProfiler package of the R language to perform GO and KEGG enrichment analysis on the 87 targets. A total of significant GO 1123 terms were obtained, including 1066 terms of BP, 18 terms of CC, and 39 terms of MF (P < 0.01). e top 15 enriched terms of BP, CC, and MF were selected and visualized. e results showed that the targets were related to response to oxidative stress, response to nutrient levels, response to steroid hormone, and other biological processes; enriched in the transcription factor complex, membrane raft, membrane microdomain, and other cellular components; and related to RNA polymerase II transcription factor binding, hormone receptor binding, ubiquitin-like protein ligase binding, and other molecular functions (Figures 4(a)-4(c)).
In KEGG enrichment analysis, a total of 143 pathways of anti-EC targets of DXP were identified according to the P value less than 0.01. e top 15 significantly enriched pathways with the highest gene counts were shown in Figure 4(d), and the specific information was shown in Table 2. e results showed that the targets may be associated with the AGE-RAGE signaling pathway in diabetic complications, the IL-17 signaling pathway, the TNF signaling pathway, and other signaling pathways. Among them, the endometrial cancer, apoptosis, EGFR tyrosine kinase inhibitor resistance, and proteoglycans in the cancer signaling pathway map showed that the PI3K-AKT and RAS-RAF-MEK-ERK (ERK) signaling pathways may play a crucial role in the treatment of EC with DXP ( Figure 5). e network of target-BP-pathways was obtained by Cytoscape (version 3.8.0) and shown in Figure 6, which contained 105 nodes and 713 edges and included 75 common targets, the top 15 BP terms, and the top 15 signaling pathways. e square stood for disease targets, the triangle stood for signaling pathway, and the arrow represented biological processes.

Correlation Analysis of Targets with BP and Signaling
Pathways. To explore the correlation of the hub targets with the signaling pathway and BP more clearly, a hub target network was constituted with the top 10 targets, including TP53 (n � 71), AKT1 (n � 71), INS (n � 70), IL6 (n � 67), EGFR (n � 66), CASP3 (n � 65), ESR1 (n � 65), VEGFA (n � 65), MAPK3 (n � 64), and JUN (n � 63) (Figure 7(a)). e correlations were analyzed by combining them with the top 15 pathways and BP terms. e results showed that AKT1 had a higher correlation with the BP, and its response to oxidative stress was more relevant to the targets (Figure 7(b)). In the analysis of correlation with signaling pathways, AKT1 and MAPK3 were more important to the pathways (Figure 7(c)).

DXP Suppresses the Viability, Cell Cycle, and Metastasis
While Triggers the Apoptosis of EC Cells. Different concentrations of DXP were used to treat HEC-1B and KLE cells and assay their survival rates. As shown in Figure 8 Figure 4: Enrichment analysis of GO and KEGG pathways of key target genes. e size of the circle represents the number of genes, and the color of the circle represents the corrected P value.   response to oxidative stress cellular response to oxidative stress response to nutrient levels aging response to steroid hormone response to reactive oxygen species neuron death regulation of neuron death cellular response to external stimulus response to peptide hormone regulation of apoptotic signaling pathway cellular response to reactive oxygen species regulation of smooth muscle cell proliferation response to metal ion muscle cell proliferation   Cell cycles results showed DXP significantly increased the proportion of G0/G1 phase cells (P < 0.05), and DXP with a high-dose have a better inhibitory effect than lowdose (P < 0.01) (Figure 8(b)). Meanwhile, DXP also inhibited the migration of EC cells. e apoptosis rate in high-dose was significantly higher than low-dose (P < 0.01) (Figure 8(c)). Additionally, the apoptosis rate of EC cells was promoted by DXP addition (P < 0.05) (Figure 8(d)).
ese results indicated that DXP suppressed the malignant progress of EC cells in vitro.

DXP Inhibited the Protein Expression of ERK Signaling
Pathway. Network pharmacology analysis indicated that the RAS pathway is the most important pathway in the treatment of DXP on EC. us, the expression levels of RAS, p-RAF, MEK, p-MEK, ERK, and p-ERK were examined by Western blot. As shown in Figure 9, DXP decreased the expression of RAS, p-RAF, p-MEK, and p-ERK proteins, but had no effect on MEK and ERK protein expression. It suggested that DXP against EC through ERK signaling pathway.

Discussion
As an epithelial malignant tumor originating from the endometrium, EC ranks first in the incidence of malignant tumors of the female reproductive system, with a trend of younger incidence [28][29][30]. In recent years, the extraction of effective antitumor components with low toxicity from TCM has become a new research orientation [31,32]. DXP is attracting attention for its anticancer and immune-enhancing activities [12]. In this work, we conducted network pharmacology and cellular experiments to identify the active components, core targets, and potential mechanisms of DXP in EC.
In the present study, a total of 168 active components of DXP and 87 DXP-EC targets were screened from public databases. In the H-C-T network, we found that quercetin, associated with the largest number of targets, is the most important component in DXP for EC. Quercetin is a polyhydroxyflavonoid with various biological functions and has antioxidant, anti-inflammatory, and antiviral properties [33]. A New Jersey survey reported that lean women who consume quercetin have a lower risk of developing EC [34]. It has also been found that quercetin inhibits the proliferation of Ishikawa cells (an EC cell line) by downregulating EGF and cyclin D1 [35]. In addition, kaempferol also played a crucial role in the suppression of EC. Kaempferol is a dietary bioflavonoid with anticancer, anti-inflammatory, and antioxidant properties that inhibits human cancer cell proliferation through various mechanisms, including induction of the tumor suppressor p53 and inhibition of Erα [36]. Additionally, Lei's study has substantiated that kaempferol against EC by inducing cell apoptosis, arresting the cell cycle in G2/M phase, suppressing cell invasion and upregulating the m-TOR/PI3K signaling pathway [37]. ese reports combined with the present research provide evidence for the therapeutic potential of quercetin and kaempferol in DXP to cure EC. PPI analysis showed that the hub targets of DXP for EC  were TP53, AKT1, INS, IL6, EGFR, CASP3, ESR1, VEGFA,  MAPK3, and JUN. To annotate the functions and related pathway of 87 targets, we conducted the GO and KEGG signal pathway analysis. GO analysis results suggested that targets were mainly related to responses to oxidative stress, nutrient levels, steroid hormones, and other biological processes.
e KEGG analysis results suggested that the pathways involved in DXP treating EC were mainly related to the IL-17 signaling pathway, TNF signaling pathway, PI3K-AKT signaling pathway, and ERK signaling pathway. Among them, IL-17 is a proinflammatory cytokine and is significantly expressed in a variety of cancers [38][39][40]. Lai's study demonstrates that IL-17 may regulate chemokines and cytokines in gynecologic cancers through activation of IκBα and phosphorylation of extracellular signal-regulated kinase 1/2 [41]. PI3K-AKT signaling pathway is a key intracellular signaling pathway that regulates cell proliferation and cell cycle, and is closely related to tumorigenesis [42,43]. Additionally, numerous studies have shown that the PI3K-AKT signaling pathway is involved in the malignant process of EC, such as picropodophyllin inhibiting EC cell proliferation by disrupting the PI3K-AKT signaling pathway [44]; NLRC5 promoting EC progression through activation of the PI3K/ AKT pathway [45]. e ERK signaling pathway is part of the mitogen-activated protein kinase (MAPK) signaling pathway, which is involved in various cellular activities, such as cell proliferation, differentiation, apoptosis, inflammation, and other cellular activities, and is closely related to tumorrelated immune escape mechanisms [46,47]. Studies have reported that the ERK signaling pathway is one of the important pathways involved in endometrial carcinogenesis and is involved in tumor cell metabolism, growth, proliferation, survival, and angiogenesis [48].
In vitro experiments were performed to further verify the molecular mechanism of DXP against EC.
e cell experiments demonstrated that DXP for the treatment of EC suppressed cell proliferation, arrested the cell cycle in G0/G1 phase, induced cell apoptosis, weakened the ability of cell migration, downregulated RAS expression and RAF, MEK, and ERK phosphorylation levels.
ese findings verified the predicted result of the network pharmacology approach.

Conclusion
In summary, we combined network pharmacology and in vitro experiments to verify the potential mechanism of the bioactive components and targets of DXP in treating EC. We demonstrated that DXP could inhibit the proliferation and migration of EC cells and enhance EC cell apoptosis. Moreover, DXP protects against EC malignant progression via downregulating the ERK signaling pathway. is study provides an effective drug for the treatment of EC. Meanwhile, network pharmacology analysis reveals the molecular mechanisms involved in the treatment of EC with DXP, providing a research basis for subsequent studies.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.