Network Pharmacology Analysis and Experimental Verification on Antiangiogenesis Mechanism of Hedyotis diffusa Willd in Liver Cancer

Purpose Hedyotis diffusa Willd (HDW) is one of the most well-known herbs used in the therapy of cancer. However, the potential mechanisms of its antiangiogenic effects have not been fully explored. Here, we applied a network pharmacology approach to explore the potential mechanisms of HDW against liver cancer angiogenesis (LCA) and used a mouse orthotopic liver cancer model for experimental verification accordingly. Methods The effective components, primary active compounds, and possible targets in the therapy of LCA were predicted using network pharmacology and bioinformatics. In vivo testing of the pharmacodynamic foundation of HDW in the treatment of LCA was performed. Hepa1-6 cells were implanted in C57BL/6 mice to establish an orthotopic liver cancer model to evaluate the antitumor and antiangiogenesis effects of the drug. Furthermore, protein levels were evaluated by western blotting, immunofluorescence, and immunohistochemistry. Results We firstly confirmed the therapeutic effect of HDW on LCA and subsequently screened 7 active compounds from HDW according to their pharmacokinetic properties. Network analysis and enrichment analysis indicated that these compounds exhibit antiangiogenic effect by acting on multiple targets and thereby regulating multiple pathways mainly involved in Akt1, IL-6, IL-1β, IL-17, hypoxia inducible factor-1α (HIF-1α), and tumor necrosis factor-α (TNF-α). Importantly, we preliminarily verified the results of the network pharmacology analysis in vivo. Conclusion Collectively, our work initially explored the therapeutic mechanism of HDW on tumor angiogenesis, which lays an experimental reference for further exploring its pharmacological action and its clinical application.


Introduction
Angiogenesis is an essential factor in tumor progression. Tumor cells can get the nutrients and oxygen they need for sustained angiogenesis. Angiogenesis plays a vital role in promoting tumour growth and metastasis [1]. As one of the most common malignant tumors and the fourth leading cause of cancer-related deaths worldwide, liver cancer is characterized by an abundant blood supply and microvessels [2][3][4]. Microvessel density has been reported to be a signifcant predictor of death and closely correlated with tumor stage in liver cancer [4]. Terefore, angiogenesis has become a therapeutic target, and new agents of antiangiogenic therapy are urgently required for liver cancer treatment [5].
Increasing evidence has shown that Chinese medicinal active compounds ofer unique advantages due to the synergy efect of multicompounds, multitargets, and minor adverse reactions in the antitumor aspect [6]. As a traditional medicinal plant, Hedyotis difusa Willd (HDW) has been recorded for thousands of years in clinical applications. It is widely used for heat clearing, detoxifcation, and removal of blood stasis [7]. Moreover, HDW has long been used as an important component in a variety of Chinese medicine formulas to treat various types of cancer [8][9][10][11][12].
Existing research has reported its noticeable antitumor efect on hepatocellular carcinoma (HCC) [10]. However, the efect of HDW on LCA and the potential mechanism of its antiangiogenic efect have not yet been explored.
As an emerging subject, network pharmacology can comprehensively and systematically clarify the mechanism and targets of Chinese medicinal active compounds, which are fully compatible with the characteristics of traditional Chinese medicine (TCM) in terms of multicomponent, multitarget, and multipathway properties [13,14]. In this study, we initially started our analysis by confrming the treatment efect of HDW using the mouse orthotopic liver cancer model. Ten, we applied the network pharmacology approach to explore the pharmacological mechanisms of HDW as an antiangiogenic therapy for HCC. Subsequently, we validated the potential targets and pathways of HDW as a therapy against LCA using molecular biological methods ( Figure 1). Terefore, this study aims to provide a foundation for future clinical applications on the efectiveness of HDW in the treatment of liver cancer.

HDW Bioactive Compound Screening and Target
Identifcation. All chemical compounds of HDW were obtained from the Traditional Chinese Medicine Systems Pharmacology (TCMSP) database [15]. After the data were gathered, oral bioavailability (OB) and drug-likeness (DL) were utilised to screen the bioactive components of HDW. Only compounds with OB ≥ 30% and DL ≥ 0.18 were chosen and identifed as potential bioactive compounds for further investigation. Tese selected compounds were subsequently used for target identifcation and compound-target network establishment. Ten, we utilised the UniProt database (https://www.uniprot.org/) to calibrate the gene names of all targets, using humans as the chosen species [16].

Probing LCA Targets and Venn Diagram Establishment.
Te LCA targets were obtained from two diferent sources [17,18]: (1) the GeneCards database (https://www. genecards.org/) and (2) the NCBI gene database (https:// www.ncbi.nlm.nih.gov/). Our investigation searched these human illness target databases for targets associated with LCA using the keywords "liver cancer angiogenesis," "hepatocellular carcinoma angiogenesis," and the species "homo sapiens." After merging and deleting the genes in the two databases, the target genes associated with LCA were obtained; common targets associated with LCA and potential targets of bioactive substances were chosen as HDW's targets against LCA. A Venn diagram of the intersection between the target of the drug and the disease was established using Venny 2.1.

Construction of Protein Interaction Network. Te Search
Tool for the Retrieval of Interacting Genes (STRING, https:// string-db.org/cgi/input.pl) was used to collect possible protein-protein interactions (PPI) by uploading 116 common targets related to LCA and putative targets of active compounds. Species were limited to "homo sapiens" with a confdence score >0.4. Ten, we imported the database from STRING into Cytoscape (version 3.7.1) to construct a PPI network for further analysis [19].

Enrichment Analysis. Enrichment analysis included the Gene Ontology (GO) Enrichment Analysis and the Kyoto
Encyclopedia of Genes and Genomes (KEGG) Pathway Enrichment Analysis. (1) GO enrichment analysis was performed by using the ClusterProfler package of R4.0.3 and mainly included biological process (BP), molecular function (MF), and cellular component (CC). GO terms with the STRING database-corrected p value ≤0.05 were retained for the construction of the clustering network [20]. (2) KEGG pathway enrichment was performed by using the ClusterProfler package of R4.0.3, and KEGG terms with STRING database corrected-p value ≤0.05 were retained for constructing a target-pathway network [21].

Mouse Orthotopic Liver Cancer Model Construction and
Drug Treatment. C57BL/6 mice (male, 4∼6 weeks) were purchased from Beijing SiPeiFu Biotechnology Co., Ltd. and fed in a pathogen-free vivarium under standard conditions. All animal experimental protocols and principles were approved by the Animal Care and Use Committee of Jiangsu Vocational College of Medicine. To establish an orthotopic liver cancer model, 1 × 10 6 Hepa1-6 tumor cells were suspended in 50 μL Dulbecco's Modifed Eagle Medium (DMEM) (containing 20% Matrigel) and orthotopically injected into the liver of C57BL/6 mice under anesthesia with tribromoethanol (240 mg/kg, Sigma, Massachusetts, USA) [22]. After 7 days, tumor-bearing mice were randomized into a vehicle and diferent treatment groups (1 group, 6 mice): (1) Te vehicle group mice were injected with normal saline. (2) Te treatment groups were intraperitoneally injected with HDW injection (50 mg/kg, 100 mg/kg, or 150 mg/kg) every other day 14 times, and the body weight was monitored every 7 days (Figure 2(a)).

Immunofuorescence (IF)
Staining and Immunohistochemistry (IHC). Immunofuorescence (IF) staining has been previously described [23]. Mouse liver cancer tissue samples were immediately frozen in OCT compound or fxed in 4% PFA overnight at 4°C, dehydrated, and embedded in parafn. Samples were blocked with 5% goat serum in PBST (0.3% Triton X-100 in PBS) and then incubated for 1 h at room temperature with the following primary antibodies: anti-CD31 (1 : 200) and anti-α-SMA (1 : 500). Following washing several times, the samples were incubated for 1 h at room temperature with secondary antibodies (diluted 1 : 200) conjugated with FITC and 4,6-diamidino-2-phenylindole (DAPI) for counterstaining nuclei. Te coverslips were mounted on glass slides, and the immunofuorescence staining was visualized and photographed using a Zeiss inverted fuorescence microscope.
An immunohistochemical assay was conducted according to our previous study with slight modifcations [24]. In brief, sections after antigen retrieval were incubated overnight with primary antibodies against HIF-1α (1 : 200) at 4°C, biotin-conjugated secondary antibodies, and horseradish peroxidase-conjugated streptavidin successively. Ten, the sections were stained with diaminobenzidine and counterstained with hematoxylin. Te staining density was assessed as in our previous study [24]. Te details are as follows: negative results of no or weak staining: invisible or light brown staining in <20% of the cell populations; positive results of moderate or strong staining: brown or dark brown staining in >20% of the populations.

Enzyme-Linked Immunosorbent Assay (ELISA).
Te infammatory factor levels in serum were detected with enzyme-linked immunosorbent assay kits (PeproTech, Rocky Hill, NJ) according to the manufacturer's instructions.

Western Blotting (WB).
Te western blotting assay was done as previously described in our study [25]. In brief, total lysates were prepared from tumor tissue. Te protein concentration was calculated by the BCA (bicinchoninic acid) Protein Assay Kit (Beyotime, Shanghai, China). Ten, the protein samples experienced sodium dodecyl sulphatepolyacrylamide gel electrophoresis (SDS-PAGE) and membrane transfer. Te membrane was blocked overnight, then the primary antibodies (Akt1, p-Akt1, mTOR, p-mTOR, STAT3, p-STAT3, and GAPDH (1 : 2000); HIF-1α (1 : 1000)) and secondary antibodies (1 : 5000) were added for  electrochemiluminescence (ECL) coloration, and the image was semiquantitatively analyzed with alpha SP image analysis software. GAPDH was used as an invariant control for equal loading of total proteins. Te proposed blots are representative of three independent experiments.

Statistical
Analysis. Data were presented as means ± standard deviation (SD). Te statistical signifcance of the diference was evaluated by the unpaired Student's t-test or one-way ANOVA, as appropriate. Values of p < 0.05 were considered statistically signifcant.

HDW Suppressed Tumor Growth and Angiogenesis in the Orthotopic Mouse Model.
To investigate the treatment efect of HDW on LCA, we established a mouse orthotopic liver cancer model and subsequently evaluated the attenuation efect of HDW on tumor growth and angiogenesis. Te in vivo experimental schema is depicted in Figure 2(a). As shown in Figure 2(b), HDW induced the most marked regression in the orthotopic tumors. In addition, we observed a signifcant decrease in angiogenesis with IF staining on the morphology of tumor tissue after HDW treatment, indicating its potential efect on attenuating tumor angiogenesis in HCC (Figure 2(e)). Meanwhile, the mouse's body weight was remarkedly declined in the vehicle group rather than in the HDW-treated group on day 28 (Figure 2(c)). In addition, the liver index (liver-to-body weight) was also signifcantly increased in the HDW group and this efect tended to be dose-dependent (Figure 2(d)). Te above results manifest that HDW suppressed liver cancer growth and angiogenesis in mouse orthotopic models.

Active
Compounds of HD Screening. TCMSP yielded a total of 7 active compounds of HDW after ADME screening with OB ≥ 30% and DL ≥ 0.18. All 7 compounds were validated as promising bioactive molecules for future research. Detailed information is presented in Table 1.

Common Target
Analysis of HDW and LCA. 174 putative targets linked to 7 compounds of HDW were collected. A total of 2116 genes related to LCA were obtained. Trough Venny analysis, 116 common targets of HDW in the treatment of LCA were obtained (Figure 3(a)).

Common Target Network Analysis and PPI Network of
Common Targets. 116 common targets were uploaded to the STRING database, and the PPI network was generated with the following conditions: combined score (≥0.4) and species limited to "homo sapiens." After that, the PPI network was established, as shown in Figure 3(b). Te network graph has 116 nodes and 1,992 edges, and the average node degree is 34.3. In addition, we then imported the PPI results into Cytoscape for topology analysis of PPI networks. A closer node to the center and the darker color represent the protein that has a greater degree in this network; then, the top seven core targets were extracted, which were successively Akt1, IL6, IL-1β, HIF-1α, EGFR, JUN, and CASP3 (Figure 3(c)).

Enrichment Analysis of GO and KEGG Pathway Analysis.
To further explore the underlying mechanisms of HDW against LCA, we performed GO enrichment and KEGG pathway analysis, executed using the DAVID database (https://david.ncifcrf.gov/home.jsp). GO analysis includes biological processes (BP), cellular components (CC), and molecular functions (MF), which together describe the functions of gene products. Te top 10 signifcantly enriched GO targets are presented (adjusted p value <0.05) in Figure 4. As shown in Figure 5, the KEGG pathway enrichment analysis of HDW against LCA mainly involves the TNF-α signalling pathway, IL-17 signalling pathway, and so on.

Experimental Validation of the Efect of HDW on Multipathway In Vivo.
To further better validate the putative multipathway mechanisms of the HDW efect according to the results of network pharmacological prediction. We have further studied the efects of HDW on the production of infammatory factors and the protein expression of multiple key signalling pathways in vivo experiments. As a result, ELISA analyses showed that the infammatory factors in the serum were decreased in the HDW treated group, including IL-6, IL-1β, IL-17, and TNF-α, as predicted above (Figure 6(a)). Moreover, western blot analyses showed that HDW treatment decreased the rates of p-Akt1/Akt1, p-mTOR/mTOR, and p-STAT3/STAT3 (Figure 6(b)). All the above results showed signifcant diferences at doses greater than 100 mg/kg. As a well-known core molecule of tumor angiogenesis potential, HIF-1α was also detected by WB analyses and IHC staining. As shown in Figures 6(b) and 6(c), both WB and IHC analyses revealed that HDW could downregulate the protein expression level of HIF-1α compared with the vehicle group. Collectively, these results confrmed our previous putative consequence that HDW exerted a multipathway therapeutic efect on LCA through regulation of anti-infammatory processes and key signalling pathways such as those of Akt1, STAT3, and the core molecule HIF-1a.

Discussion
HDW is one of the most efective medications for clearing heat and detoxicating, with the efects of dispersing the mass, relieving pain, promoting urination, and removing dampness [7,26]. Modern research has shown that HDW has anticancer properties via multiple pathways and is a commonly used TCM in clinical cancer treatment [7,10,11]. In light of previous reports, HDW can suppress tumor angiogenesis [27,28], but the exact mechanism is unknown.
According to the results of GO and KEGG enrichment analysis, the targets of HDW's antiangiogenic efect in liver cancer are primarily involved in signalling pathways such as TNF-α and IL-17. TNF-α is a cytokine produced by macrophages or activated monocytes and plays a signifcant role in infammatory response and tumor angiogenesis [32]. IL-17 is a highly versatile proinfammatory cytokine that has proven to be essential for tumor angiogenesis [33,34]. It can upregulate IL-6 expression and activate the STAT3 pathway to exert protumour efects [35].
Tis research adopts the in vivo experiment to further validate the pharmacodynamic efects of HDW on anti-liver cancer and angiogenesis. Te result shows that the injection of HDW could signifcantly reduce the liver lesion area and the serum levels of IL-6, IL-17, TNF-α, and IL-1β in the orthotopic liver cancer mouse model, and these results were dosedependent to some extent. It suggests that HDW may attenuate tumor angiogenesis by lowering the levels of infammatory factors. However, the exact mechanism needs further study.
In addition, it was known that the phosphorylation of Akt1 and STAT3 regulates the transcription of HIF-1α, a core molecule that induces tumor angiogenesis [36][37][38]. Phosphorylated Akt1 activates mTOR, which enhances HIF-1α transcription. Similarly, phosphorylated STAT3 in the nucleus enhances HIF-1α transcription, which in turn promotes the transcription of several proangiogenic genes such as VEGF, ultimately contributing to tumor angiogenesis [37,38]. By detecting the expression of key proteins related to signalling pathways in tumor tissue, we found that the expression levels of p-Akt1 and p-STAT3 in the HDW groups were signifcantly lower than those of the model group. Tese fndings suggest that HDW may regulate HIF-1α via modulating the level of phosphorylated Akt1 and STAT3 in several key signalling pathways, which ultimately afect tumor angiogenesis.
However, the shortcomings of this study inevitably exist. Specifcally, the limitations of this study are as follows: the selection of the database is not comprehensive enough, the positive control group is lacking, and the validation of in vivo targets is not deep enough. In the next study, these shortcomings need to be further improved.

Conclusion
In conclusion, HDW may afect key targets such as Akt1, HIF-1, IL6, and IL-1β, as well as signalling pathways such as IL-17 and TNF-α via key active components such as DMQ, MMA, HHTCA, poriferasterol, stigmasterol, β-sitosterol and quercetin. It refects the antitumor and antiangiogenesis effects of HDW that are multicomponent, multitarget, and multipathway. Te result of the in vivo experiment further confrms the antiangiogenic efect of HDW, and it is supposed that its mechanism may be related to the inhibition of infammatory factor production, which afects Akt1 and STAT3 phosphorylation, providing a reference for future studies on HDW's antiangiogenic mechanism.

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