Systemic Pharmacological Approach to Identification and Experimental Verification of the Effect of Anisi Stellati Fructus Extract on Chronic Myeloid Leukemia Cells

Anisi stellati fructus (ASF) is the dried fruit of the Illicium verum Hook.f. tree. The aim of this research was to evaluate the antileukemic effect of ASF on chronic myeloid leukemia (CML) cells, which was hypothesized from the systemic pharmacological analysis of ASF, focusing on the combined effect of ASF extract (ASFE) and imatinib (IM). The compounds of ASF were identified using the Traditional Chinese Medicine Systems Pharmacology database and analysis platform. The target gene information was acquired from the UniProt database. The compound and target interaction network was generated from Cytoscape 3.7.1. Using this analysis, 10 compounds effective against CML cells were obtained. ASFE was prepared and analyzed by high-pressure liquid chromatography to provide experimental proof for the relationship between ASF and CML. The anti-p210Bcr-Abl effects of ASFE and ASFE + IM combination were evaluated by western blotting. Either ASFE alone or in combined treatment with IM on K-562 CML cells resulted in a significant reduction of the Bcr-Abl levels. As expected from the systemic analysis results, ASF had antileukemic activity, showing that it is a potential therapy for CML.


Introduction
Anisi stellati fructus (ASF) is fruit that comes from the evergreen aromatic tree, Illicium verum Hook. f. According to e Great Pharmacopoeia written by Li Shi Zhen, published in 1596, the fruit of Illicium verum Hook. f. (Chinese star anise) has been used as a remedy for infectious diseases. ASF has long been used as a food as well as for medical purposes because of its effect on eliminating odor and relieving symptoms such as high fever, diarrhea, and vomiting [1]. Additionally, Tamiflu, a treatment for swine-origin influenza A, was developed based on the components of ASF [2].
Chronic myeloid leukemia (CML) is considered to be a myeloproliferative neoplasm characterized by the expansion of a clone of hematopoietic cells that carries the abnormal Philadelphia chromosome and fused Bcr-Abl gene [3]. Bcr-Abl is a constitutively active cytoplasmic tyrosine kinase [4]. Bcr-Abl fusion protein is present in 95% of CML cases and 20%-30% of acute lymphoblastic leukemia cases [5,6]. For these reasons, p210 Bcr-Abl is a major target of tyrosine kinase inhibitors (TKIs) [7]. CML is diagnosed in the chronic phase and is transformed into the acute phase in an average of 2-3 years, resulting in mortality within a few months despite antileukemic combination therapy [8,9]. Imatinib (brand name: Gleevec, Novartis; formerly called STI571) is a relatively specific inhibitor of the Bcr-Abl tyrosine kinase, which has efficacy in CML [10]. Although survival from CML was increased by imatinib (IM), drug resistance by mutated forms of Bcr-Abl led to the development of new drugs such as dasatinib, nilotinib, bosutinib, and ponatinib [11]. e K-562 CML cell line is highly resistant to various stimuli. However, the development of new drugs often causes resistance to these drugs [12].
To date, some studies have shown a correlation between infection and cancer, and approximately 2 million emerging cancer cases are due to infections from microorganisms [13][14][15]. Numerous studies have shown that antimicrobial peptides have cytotoxic effects on cancer cells and have the possibility of being used in anticancer therapy either alone or in combination with other conventional drugs. An increasing number of studies have shown that some of the cationic antimicrobial peptides, which are toxic to bacteria but not to normal mammalian cells, exhibit a broad spectrum of cytotoxic activities against cancer cells [16,17]. Our previous study has shown that ASF contains 49 identified compounds, of which 35 compounds were analyzed to have target genes related to antimicrobial activities [18]. However, there has been minimal research on the anticancer effects of ASF. Only ASF has been reported to inhibit the metastasis and angiogenesis of malignant cancer cells [19]. In the present study, the 49 identified compounds of ASF were further investigated for their efficacy against cancer using systemic pharmacological analysis. e results showed that 20% of the total ASF compounds were associated with CML. Based on this result, we characterized the antileukemic potential of ASF against K-562 CML cells.
e results showed that the ASF extract (ASFE) either alone or in combination with IM significantly reduced the amount of p210 Bcr-Abl protein, suggesting further studies to confirm ASF as an effective antileukemic adjuvant.

Identification of ASF Compounds.
To search for identified compounds in ASF and to investigate the relationship between the compounds and cancers, the Traditional Chinese Medicine Systems Pharmacology (TCMSP, http://lsp.nwu.edu.cn/) database and analysis platform were used. e 49 compounds were found in ASF, and all diseases related to each compound were recorded. After screening a total of 49 compounds, 33 compounds were found to have relationships with cancer categories. Among them, 10 CMLrelated compounds were identified. To screen bioactive compounds of these 10 CML-related compounds, five parameters related to absorption, distribution, metabolism, and excretion were used: molecular weight (MW), oral bioavailability (OB), Caco-2 permeability (Caco-2 cells), drug likeness (DL), and drug half-life (HL) [20][21][22].

Target
Collection. Molecular targets of the 10 CMLrelated compounds and the official gene information were acquired and listed using the UniProt database (http://www. uniprot.org).

Network Construction.
To visualize the 10 CML-related compounds and their possible targets, a compound (C)-target (T) network was constructed using Cytoscape 3.7.1, an open-source bioinformatics platform for visualizing molecular interaction networks (https://cytoscape.org/). In the network, the node denoted either compounds or target proteins, and the edges represented compound-target connections.

Preparation of Standard Solutions.
e standard stock solutions of 4-anisaldehyde and 4-prophenylanisole were prepared in methanol and stored at 4°C. A working standard solution was prepared by serial dilution of stock solutions with methanol. All calibration curves were obtained from assessments of peak areas of standard solutions in the 2 Evidence-Based Complementary and Alternative Medicine following concentration ranges: 4-anisaldehyde, 0.10-2.00 μg/mL; 4-prophenylanisole, 0.10-2.00 μg/mL.

Preparation of Sample Solution.
e resulting solution of ASFE was filtered through a 0.2 μm syringe filter and injected into the HPLC instrument.
2.6. Cell Viability Assay. Cell viability was determined using the CCK-8 assay according to the manufacturer's instructions. K-562 cells were seeded at a density of 1 × 10 4 / well in a 96-well plate. After incubation for 24 h, the cells were treated with dimethyl sulfoxide or various concentrations of ASFE for 24 h at 37°C in an incubator with a 5% CO 2 atmosphere. e cells were then incubated with 10 μL of CCK-8 reagent for 4 h at 37°C. Absorbance was measured at 450 nm using an ELISA microplate reader (Tecan, Mannedorf, Switzerland). Data are presented as the percentage viabilities of untreated cells (100%). e assay was independently repeated three times.

Western Blot Analysis.
After treatment with ASFE, IM, and ASFE + IM by either a dose-dependent or time-dependent manner, K-562 cells were harvested by centrifugation at 4°C, washed once with ice-cold D-PBS, resuspended in lysis buffer (50 mM Tris (pH7.5), 2 mM EDTA, 100 mM NaCl, and 1% NP-40) containing protease inhibitor cocktail (Sigma-Aldrich), and incubated for 30 min at 4°C. Following centrifugation for 15 min at 13,000 rpm at 4°C, the protein concentration of the lysate was determined using a bicinchoninic acid (BCA) Protein Assay Kit ( ermo Fisher Scientific, Waltham, MA, USA), according to the manufacturer's instructions. 30 μg of total protein was separated on a 6% or 7% sodium dodecyl sulfate-polyacrylamide gel and electroblotted onto polyvinylidene difluoride (PVDF) membranes (Bio-Rad, Hercules, CA, USA). e blots were incubated with a blocking solution (5% skim milk) for 1 h at room temperature, followed by incubation overnight with the primary antibodies (anti-Bcr, 1 : 3,000 dilution; anti-beta-actin, 1 : 1,000 dilution) at 4°C. e blots were then washed three times with Tween-20/Trisbuffered saline (TTBS), followed by incubation with the appropriate horseradish peroxide-conjugated secondary antibodies (1 : 3,000 dilution) for 1 h at room temperature, and then washed with TTBS. Immunoreactive bands were visualized using the Enhanced Chemiluminescence Detection Solution (ECL Plus; ermo Fisher Scientific). ImageJ software was used to quantitate the band densities.

Quantification of Apoptosis by Annexin V Labeling.
Apoptosis was quantified using an annexin V and 7-AAD Kit (Millipore, Hayward, CA, USA) according to the manufacturer's instructions. Briefly, after 1 × 10 6 K-562 cells were seeded in a 60 mm cell culture dish, ASFE (2 μg/mL), IM (2.0 μmol), or their combination was added. Following incubation for 24 h at 37°C in a 5% CO 2 incubator, the cells were collected and incubated with annexin-V and 7-AAD for 20 min at room temperature in the dark. e events for live, dead, and early (annexin V+/7-AAD+) cells were counted using the Muse cell analyzer (Merck Millipore, Billerica, MA, USA).

Statistical
Analysis. Statistical analysis was performed using SPSS statistical software for Windows, version 23 (IBM, Armonk, NY, USA). Data are expressed as the mean ± standard deviation (SD) of triplicate determinations. Differences in means between groups were subjected to oneway analysis of variance followed by the least-significant multiple comparison test and independent t-test. P values <0.05 were considered significant differences.

Screening 33 Cancer-Related Compounds.
According to the TCMSP database, ASF contained 49 identified compounds. Every compound and its related diseases were investigated, and 33 cancer-related compounds were screened (Table 1).

Collecting 10 CML-Related Active Compounds and Systemic Pharmacological
Analyses. Ten CML-related compounds and the analyzed results are shown in Table 2. To search bioactive compounds, the criteria of each parameter were suggested. Molecular weights from 180 to 500 Daltons were recognized as comparatively more druggable [23]. Oral bioavailability (OB) ≥30% and drug likeness ≥0.18 were used as ideal drug characteristics [24,25]. Caco-2 cell permeability was tested using an in vitro experiment with the Caco-2 human intestinal cell line. Drug half-life (t1/2) represented how long it took for the drug to be internalized and reduced by half; a drug half-life ≤4 h was classified as the fast elimination group, between 4 and 8 h as the midelimination group, and ≥8 h as the slow-elimination group [20,21].

Network Construction of 10 CML-Related Compounds and eir Related Targets.
To visualize the target genes of 10 CML-related compounds, a compound (C)-target (T) network was established. A total of 313 target proteins were collected from the TCMSP, and the official gene names of the targets were obtained from the UniProt database. Figure 1 shows that the C-Tnetwork revealed interactions between 10 CML-related compounds and 313 target genes. e nodes showed the 10 CML-related compounds and the target genes, and the edges represented the interaction of compounds and targets. e size of nodes was related to the degrees. Quercetin (degree � 77) had the greatest interaction with targets, followed by luteolin (degree � 56), kaempferol (degree � 36), salicylic acid (degree � 35), 2-methyl-N-phenylmaleimide (degree � 29), R-linalool (degree � 20), honokiol (degree � 18), astragalin (degree � 16), CHEBI:7 (degree � 15), and (+)-catechin (degree � 10).

HPLC Analysis of ASFE.
e main components of ASFE were determined using an HPLC system to evaluate the quality of ASFE. Because 4-anisaldehyde and 4-prophenylanisole [(E)-anethole] have been reported to be the main compounds of ASFE, we used them as standards and conducted the analysis. As shown in the representative chromatographic fingerprint of ASFE, the main components of ASFE were 4-anisaldehyde and 4-prophenylanisole (Figure 2).

ASFE Treatment on K-562 Cells Decreases the Amount of p210
Bcr-Abl in a Time-dependent Manner. ASFE or IM was    used to treat K-562 cells, and the p210 Bcr-Abl protein steadystate levels in the total lysate were analyzed by western blotting (Figure 4). 2 μmol IM was subjected based on previous studies [41]. e p210 Bcr-Abl protein levels were decreased after treatment with ASFE or IM on K-562 cells in a time-dependent manner ( * P < 0.05; * * P < 0.01). Figure 5 shows that when both ASFE and IM were added together to K-562 cells, the level of p210 Bcr-Abl was diminished compared to that of the lanes containing K-562 cells after treatment with IM or ASFE, indicating that the combined treatment of K-562 cells by ASFE and IM was more potent than each one separately.

Assessment of ASFE-and IM-Inducing Cell Death
Mechanisms on K-562 Cells. Apoptosis, necrosis, and autophagy are well-known cell death mechanisms. To observe apoptosis and necrosis, we conducted flow cytometry using annexin V and 7-AAD staining. After 16 h treatment of IM, ASFE, or IM + ASFE, the cells were stained with both annexin V and 7-AAD. e results showed that, in the control, >85% of the cells survived; the cell population decreased to 82.79% for the IM, 9.54% for the ASFE, and 0% for the IM + ASFE. e late apoptotic/necrotic cell population increased to 13.03% (IM), 36.33% (ASFE), and 42.89% (ASFE + IM). e early apoptotic cell population increased to 0.93% (IM), 0.00% (ASFE), and 0.00% (ASFE + IM) ( Figure 6). e cell populations of late apoptosis/necrosis significantly increased, but increased cell population during early apoptosis was rare. Based on these results, we postulated that K-562 cells were involved in an alternate pathway and were not involved in apoptosis. ese results also showed that K-562 cell killing was significant after treatment with ASFE.

Discussion
Traditionally, ASF has been used in Chinese medicine for the treatment of skin inflammation, stomach aches, and rheumatic pain [1]. For decades, herbal medicine has been used as an adjunct in chemotherapy for cancer treatment, demonstrating their synergistic roles in enhancing efficacy, ameliorating side effects, and reducing drug resistance [42]. However, there have been few scientific studies describing the anticancer effects of ASF.
CML is a clonal myeloproliferative disorder characterized by the presence of a constitutive tyrosine kinase activity of the fusion oncogene, Bcr-Abl [11]. Bcr-Abl protein induces cellular transformation by activating the signaling molecules, STAT5 and Akt [43]. As a first-line treatment, TKI-targeted Bcr-Abl was expected to treat CML effectively, but it has shown drug resistance, so the next generation of drugs from imatinib has been required [44]. Bcr-Abl T315I is the most frequent resistant mutation to tyrosine kinase inhibitors [45]. Following imatinib, the second-generation inhibitors such as dasatinib, nilotinib, and bosutinib and third-generation inhibitor such as ponatinib have been used to treat resistant cases of CML. Ponatinib has been developed to overcome the T315I mutation. e next candidates of tyrosine kinase inhibitors, including bafetinib, danusertib, rebastinib, tozasertib, HG-7-85-01, GNF-2, and 1,3,4-thiadiazole derivatives are presently undergoing approval as the next generation of therapeutic agents [46]. However, the issue of drug resistance is a rapidly increasing problem, and alternatives that are less toxic are urgently needed. Accordingly, this study aimed at providing the potential for exploring new therapies that are less harmful and more effective.
In the present study, we used a systemic pharmacological approach, filtered active compounds of ASF, constructed the compound-target network, and screened cancer-related compounds of ASF. ASF was found to contain 49 identified compounds, and 33 compounds were active in biological pathways related to cancers. Moreover, 10 of the 33 were classified as being associated with CML. Additionally, the compound-target network showed multiple compounds and multiple targeting characteristics of ASF. Multiple compounds and multiple targeting actions of other herbal medicines such as licorice [47], Xiao-Chaihu Decoction, Da-Chaihu Decoction [48], Pulsatillae Radix, Baekduong-tang [49,50], and Bulsusan [51] have also been studied. Based on our results, we propose that ASF is a potential antileukemic agent.
Experiments were conducted to confirm this possibility. First, the active components of ASF were extracted. According to previous studies, the extraction of ASF could be conducted using several extraction techniques such as hydrodistillation, steam distillation, solvent extraction, supercritical fluid CO 2 extraction, hydrodistillation-headspace solvent microextraction, and microwave-assisted extraction [52,53]. In our study, a combination of solvent extraction and sonication, followed by pressured gas blowing concentration, was used. To obtain the maximum extraction ASF, methanol was used as a solvent [54], followed by sonication to increase the extraction efficiency. e 4prophenylanisole was one of the main compounds of ASF and accounted for 94% of the essential oil [55]. Another major component of ASF was 4-anisaldehyde, which is used as a food fragrance and reported to have antifungal and saliva-enhancing effects [56]. In this study, the main compounds of 4-prophenylanisole and 4-anisaldehyde were identified in the ASFE using HPLC. e ASFE and ASFE + IM combination showed potent antileukemic effects by decreasing the p210 Bcr-Abl levels.
Because leukemic stem cells do not respond to TKI, when the drug supply is stopped, the cells can continuously multiply to allow the disease to persist. Moreover, the combination of treatment targets results, more efficiently, in p210 Bcr-Abl β-Actin * * * * * Figure 5: ASFE + IM combined therapy led to a decrease of p210 Bcr-Abl . K-562 cells were treated with IM (2 μmol), ASFE (2 μg/mL), and IM + ASFE combination for 12 h at 37°C. e p210 Bcr-Abl protein levels in the total lysate were analyzed by western blotting ( * P < 0.05; * * P < 0.01).   Evidence-Based Complementary and Alternative Medicine alternative surviving mechanisms of cancer cells such as autophagy [16].
Our results showed that ASFE combined with IM resulted in the significant death of K-562 cells. e results suggested the feasibility of further research to identify cell death mechanisms following a decrease in Bcr-Abl. In addition, our next phase of research will be to verify the effects of 10 CML-related compounds as well as two representative compounds of ASF, 4-anisaldehyde and 4-prophenylanisole, on CML. e workflow scheme of our study is summarized in Figure 7.

Conclusion
We have identified 10 CML-related compounds of ASF using systematic pharmacological analyses. e ASF extract significantly induced cell death of K-562 cells. In addition, combined treatment with ASFE and IM led to a significant decrease of Bcr-Abl protein levels. e results showed that ASF is a promising source of effective chemotherapeutic adjuvants and is a candidate, in combination with other chemotherapeutic drugs, for the treatment of CML patients, to ameliorate the side effects and overcome resistance. Although we do not know the exact mechanisms of the 10 compounds of ASF and their antileukemic effects, this study is the first to report the antileukemic effects of ASF on K-562 CML cells, which should prompt further mechanistic studies of these antileukemic compounds.

Data Availability
All data used to support the findings of this study are included within the article.

Conflicts of Interest
e authors declare that they have no conflicts of interest.

Authors' Contributions
Won G. An designed the experiments. Youn Sook Kim performed the experiments, and Su Yeon Suh analyzed the data. Youn Sook Kim and Won G. An wrote the paper. Chul Won Lee, Sang Yull Lee, and Yong Tae Ahn reviewed and edited the manuscript. All the authors read and approved the final manuscript. Philadelphia chromosome Figure 7: A workflow scheme illustrating the hypothesis that compounds of ASF having antileukemic potential can be identified using systematic pharmacological approaches; the experimental proof was that ASF was effective in killing of K-562 chronic myeloid leukemia cells.