Sage Journals: Discover world-class research

Abstract

Background

Coptis chinensis Franch. (CCF) is a Traditional Chinese medicine known for its good anti-cancer effects. However, the bioactive compounds and mechanisms underlying its anti-colorectal adenocarcinoma (COAD) remains unclear.

Objectives

This study aims to reveal the bioactive compounds and mechanism of CCF against COAD by utilizing bioinformatics, molecular docking simulation and in vitro experiments.

Materials and Methods

Bioactive compounds and candidate targets of CCF against COAD were collected using the Traditional Chinese Medicine Systems Pharmacology Database, Analysis Platform (TCMSP), Gene Expression Omnibus (GEO), Gene Cards, and SwissTargetPrediction databases, respectively. The Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were performed using the DAVID (v2022q3) database. The “Compounds-Targets-Pathways” (C-T-P) and “Protein-Protein Interaction” (PPI) networks were constructed using Cytoscape (v3.8.0) to identify the hub targets. AutoDock Vina (v1.2.0), UNLCAN, TCGA, MTT, plate cloning, and quantitative real-time PCR (qPCR) assays were used to confirm the results.

Results

The “C-T-P” and “PPI” networks revealed three hub targets: checkpoint kinase 1 (CHEK1), polo like kinase 1 (PLK1), and aurora kinase B (AURKB). Molecular docking simulation results showed that berberine, a candidate bioactive compound of CCF, had high affinity with the hub targets, comparable to that of positive drugs. Both CCF ethanol extracts and berberine significantly down-regulated CHEK1, PLK1, and AURKB, inhibited proliferation and promoted apoptosis of HCT-116 cells.

Conclusion

To summarize, CCF inhibits COAD by down-regulating CHEK1, PLK1, and AURKB, with berberine as its primary bioactive compound.

Keywords

colorectal adenocarcinoma bioinformatics molecular docking simulation

Introduction

Colorectal adenocarcinoma (COAD) is a prevalent and lethal form of digestive tract cancer with high incidence and mortality rates (Dang et al., 2020). Annually, millions of people are diagnosed with COAD, and many die from the disease (Qu et al., 2020). Although treatments like surgery, radiotherapy, chemotherapy, and medication have improved cure rates for early-stage COAD patients, the five-year survival rate remains low for those with locally advanced COAD. Endoscopy and blood screening are crucial in optimizing treatment and reducing COAD mortality (Koster et al., 2020). Unfortunately, most countries currently lack the capacity to conduct large-scale inspections. To tackle this global health issue, it is essential to develop highly effective, low toxicity, and affordable drugs.

In recent years, Traditional Chinese Medicine (TCM) has achieved remarkable curative effects in cancers prevention and treatment, attracting extensive attention worldwide (Wang et al., 2021). However, a gap between modern medicines and TCM is the lack of the bioactive compounds and mechanism of the latter, which has led to lower acceptance of TCM. Nonetheless, this has not affected their wide application in various fields. For instance, Coptis Chinensis Franch. (CCF), first recorded in Shennong’s Classic of Materia Medica, has been used to clear heat, purge fire, and detoxify for more than 3000 years (Zhang et al., 2021). Clinical epidemiology and modern pharmacology reveal that CCF is a highly promising anti-cancer herbal medicine, which shows an impressive inhibition effect on a variety of cancer cells. Gu, Song, et al. (2020) proved that berberine, an essential bioactive compound extracted from CCF, had inhibitory effects on cancer cells. However, the underlying bioactive compounds and mechanisms of CCF against COAD are not yet fully understood.

The emergence of bioinformatics has provided significant support for the in-depth study of TCM (Zhang et al., 2019), with the use of multiple algorithms to comprehensively analyze the internal relationship between drugs, diseases, and targets, which is very consistent with the overall concept of TCM in the treatment of diseases. Identifying the differentially expressed genes (DEGs) between case and normal samples is essential in this process (Guo et al., 2017), and public databases with large clinical samples provide convenience for researchers. Molecular docking simulation (MDS), a tool widely used in drug development, is effective in analyzing the affinity between drugs and targets, providing a very valuable reference for drug development (Pinzi & Rastelli, 2019). MDS has also achieved many substantive achievements (Khelfaoui et al., 2021). A plethora of studies have already proved that it is feasible to study TCM via these bioinformatics tools and in vivo and in vitro experiments synthetically (Gu, Xue, et al., 2020; Song et al., 2020).

This study utilized bioinformatics, public databases and in silico to gain a deeper understanding of the potential bioactive compounds and hub targets of CCF against COAD. After analyzing the function of hub targets and the pathways they participated in, in vitro experiments confirmed the above results. A graphical summary highlights the workflow of this study and is detailed in Figure 1.

Figure 1.

The Workflow of this Study.

Materials and Methods

Bioactive Compounds Collection

To screen potential anti-COAD bioactive compounds in CCF, a crucial step is to collect as many compounds as possible and analyze their oral bioavailability (OB) and drug-likeness (DL). OB is a crucial standard to evaluate the quality of drugs, referring to the relative amount and rate of drugs absorbed by the body into circulation (Guo et al., 2019). DL is derived from the structures and properties of existing drugs and candidate drugs, including chemical stability, solubility, permeability, and metabolism (Wei et al., 2020). Both of them are commonly used to screen potential compounds in the early stage of drug development.

Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP, https://old.tcmsp-e.com/tcmsp.php) is database contains 499 herbs from the Chinese Pharmacopoeia 2010 Edition and over 29,000 compounds, providing a new platform for the systematic study of TCM (Ru et al., 2014). Based on the provisions of the TCMSP database, OB ≥ 30% and DL ≥ 0.18 were used as the thresholds for screening bioactive compounds in this study (Zhang et al., 2020). The PubChem database (https://pubchem.ncbi.nlm.nih.gov/) contains compounds specific information was used to ensure check identification of bioactive compounds (Kim et al., 2021). And three compounds reported to have inhibitory effects on cancer were also added (Lee et al., 2014).

Condidate Targets of CCF Against COAD

Swiss Target Prediction (https://labworm.com/tool/swisstargetprediction) is a database for predicting targets and the off-targets of bioactive compounds, which helps to understand the molecular mechanism of biological activity (Daina et al., 2019). Consequently, after setting “humans” as species, the smiles formats of the above compounds were imported into SwissTargetPrediction to catch their related targets in reverse.

A wealth of clinical data can provide effective data support for researchers to mine the core targets of diseases. Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/) is a free database that provide huge transcriptome, genome, and other multi-omics data from clinical samples (Clough & Barrett, 2016). DEseq2 was used to classify the samples and screen the DEGs from GSE41657 with adjusted p < 0.05 and |log (Fold Change) | > 2 as the thresholds. Gene Cards database (https://www.genecards.org/) is a recognized free database for comprehensive screening disease-related targets (He et al., 2020). With COAD as the keyword, the genes related to it were obtained from Gene Cards. The intersection targets of compounds and COAD were compared and visualized with Venn (v2.1.0) database (https://bioinfogp.cnb.csic.es/tools/venny/) (Jia et al., 2021).

Enrichment Analysis

The Database for Annotation, Visualization and Integrated Discovery (DAVID, https://david.ncifcrf.gov/) is a widely used bioinformatics database for systematic biological function annotation and pathways enrichment analysis of genes (Huang da et al., 2009). After importing the candidate targets into DAVID, the identifier was set to “GENE OFFICIAL SYMBOL” and the species to “Homo sapiens” for gene ontology biology process (GO BP), gene ontology cellular compound (GO CC), gene ontology molecular function (GO MF) and KEGG pathways enrichment. The top 10 GO entries (BP, MF, CC) and KEGG pathways (p < 0.05) were visualized as bubble plot and circus via the Omicshare database (http://www.omicshare.com), respectively (Wang, Su, et al., 2020).

Acquisition of Hub Targets

Comprehensive analysis of the “Compounds-Targets-Pathways” (C-T-P) and “Protein-Protein Interaction” (PPI) networks is crucial for screening the potential bioactive compounds and core targets of CCF against COAD. A “C-T-P” network was constructed using Cytoscape (v3.8.0, https://cytoscape.org/download.html), an analytical software used to visualize and analyze the association among drugs, candidate targets, and pathways (Shannon et al., 2003). The relationship between nodes is directly proportional to their corresponding degrees. In general, the compounds and targets with higher degrees are considered to be more critical.

STRING (https://string-db.org/) is a powerful open-access database containing over 52 million protein interactions (Szklarczyk et al., 2021). In this study, “Homo sapiens” was chosen as the species to create the “PPI” network of core targets. The TSV format of the “PPI” network was downloaded and visualized by the CytoHubba plug-in. According to the median principle, based on the degrees of nodes from the “PPI” and “C-T-P” networks, the potential bioactive compounds and hub targets of CCF against COAD were identified.

Molecular Docking Simulation

MDS is a widely used computational method for screening drugs based on their affinity to proteins. AutoDock Vina, a molecular docking software based on semi-flexibility docking, is one of the best software for this purpose (Eberhardt et al., 2021). To ensure the accuracy of the results, we compared the affinity between positive drugs and proteins, bioactive compounds, and proteins.

The PDB formats of the five hub targets’ structureswere downloaded by RSCBPDB (https://www1.rcsb.org/) database, and repaired via Chimera software (v1.16) (Burley et al., 2017). Then, Avogadro software was used to construct the 3D structures of berberine, berberrubine, palmatine and positive drugs, and the MMF94 force field was used to minimize their energies. Using the getbox plugin in PyMOL, the original combination pockets of hub targets and created config files (num modes = 20, energy range = 5, exhaustiveness = 20) were obtained. The compounds and targets were then hydrogenated and charged via Mgltools (v1.5.6). Batch docking of the above compounds and proteins was carried out and visualized as heat maps via python (v3.8.8). The 3D and 2D docking results of compounds and the hub targets were visualized using PyMOL and Ligplus software (Siraj et al., 2015). Comprehensive analysis from the molecular docking results helped verified the hub targets of CCF against COAD.

UNLCAN and TCGA Analysis

To further validate our experiments and establish a connection between research and the clinic, the public clinical data from UALCAN and TCGA were used. UALCAN (ualcan.path.uab.edu/home) is a free database that provides publicly available cancer omics data and allows users to identify biomarkers or to perform in silico validation of potential genes of interest (Chandrashekar et al., 2017). The Cancer Genome Atlas (TCGA, https://www.cancer.gov/about-nci/organization/ccg/research/structural-genomics/tcga), a huge free public database, molecularly characterized over 20,000 primary cancer and generated over 2.5 petabytes of genomic, epigenome, and proteome data (Wang et al., 2016). The keyword as “COAD” was selected to search for the expression of hub genes in COAD and normal tissues from these two databases, with a significant difference found when p < 0.05.

Acquisition of CCF Ethanol Extracts and Berberine

An amount of 10 g CCF was purchased from Bozhou City (Anhui, China), crushed and passed through a sieve, and extracted twice using 75% ethanol for 30 min each time via ultrasonication. The ethanol solution of CCF was recovered using a rotary evaporator, and its dry paste yield was calculated. Berberine was purchased from Chengdu Pusi Biotechnology Co., Ltd., (Sichuan, China).

MTT Assay

HCT-116 cells in logarithmic growth phase were incubated in two 96-well plates at a density of 7.0×10³ cells/well for 24 h until all cells were adherent. After that, they were treated with varying concentrations of CCF ethanol extracts (0, 10, 20, 40, 80, 160, and 320 µg/mL) and berberine (0, 10, 20, 40, 80, and 160 µg/mL) for 24 h, respectively. Subsequently, 10 µL MTT (Sigma, USA) was added to each well of both plates. After 4 h, the culture medium in each well was sucked away, and the absorbance of wells was read at 490 nm using a Spark 10 M microplate reader (Tecan, Mannedorf, Switzerland).

Plate Cloning Assay

HCT-116 cells were inoculated into two 6-well plates at 2.0×10³ cells/well for 24 h until all cells were adherent. After culturing cells with different concentrations of CCF ethanol extracts (0 and 160 µg/mL) and berberine (0 and 80 µg/mL) for 24 h, photos were taken under fluorescence microscope (Olympus Corporation, Tokyo, Japan). Then, the culture medium in each well was replaced with normal culture medium and cultured for 14 days. The cells in 6-well plates were fixed with 4% paraformaldehyde, and reacted for 20 min at room temperature in the dark. The colony formation in each well was observed under a microscope and photographed.

Cell Cycle Detection by Flow Cytometry

After treating the cells with CCF ethanol extracts (0, 80, and 160 µg/mL) and berberine (0, 40, and 80 µg/mL) for 24 h, the cells were digested, and the cell suspension was collected. The cell suspension was centrifuged at 1000 rpm for 5 min, and the supernatant was discarded. After resuspension twice in PBS, cells were fixed overnight in 70% ethanol. PI dye was added to the cell samples, and they were incubated for 30 min at room temperature in the dark. The cell cycle of the samples was detected using a flow cytometer (BD, FACSCalibur, US).

Hoechst 33342 Staining Assay

Logarithmic growth phase HCT-116 cells were inoculated into two 6-well plates at 5.0×10⁵ cells/well. After cells adhered to the walls, they were incubated for 24 h in different concentrations of CCF ethanol extracts (0, 80, and 160 µg/mL) and berberine (0, 40, and 80 µg/mL). The cells were fixed with 4% polyoxymethylene, washed twice with PBS, and incubated with 8 µg/mL Hoechst 33342 (Beyotime Biotechnology, Shanghai, China) for 10 min at 4℃. After washing the cells thrice with PBS, they were observed using a fluorescence microscope (Olympus, Nanjing, China).

Quantitative Real-time PCR Assay

A qPCR was conducted on HCT-116 cells in the logarithmic growth phase, which were seeded in 6-well plates with 5.0×10⁵ cells/well and treated with CCF ethanol extracts (0, 80, and 160 µg/mL) and berberine (0, 40, and 80 µg/mL) for 24 h. The primers for the three hub targets were designed and synthesized by Siwega Biotechnology Co., Ltd., (Wuhan, China), as shown in Table 1. The reverse-transcribed samples were detected on Bio-Rad CFX Manager software (Bio-Rad, CFX Connect, US) using a qPCR machine (Bio-Rad, CFX Connect, US).

Table 1.

Primers for the Three Hub Targets.

Primer name	Sequence (5’ to 3’)
Hu-ACTIN-F	CATGTACGTTGCTATCCAGGC
Hu- ACTIN –R	CTCCTTAATGTCACGCACGAT
Hu-PLK1-F	CTTCGTGTTCGTGGTGTTG
Hu-PLK1-R	CGGTTTCGGTGCAGGTA
Hu-CHEK1-F	AACCAAGTTTCAGGGGACA
Hu-CHEK1-R	TGGCTTCGCTTCACAGA
Hu-AURKB-F	CTTGGCTCGGGAGAAGAA
Hu-AURKB-R	GACGCAGGATGTTGGGA

Statistical Analysis

Statistical significance was assessed using a two-tail, equal variance independent-samples t-test. GraphPad Prism 7.0 (GraphPad Software Inc., San Diego, CA) was used to perform data analysis and p < 0.05 was set as a significant threshold.

Results

Acquisition of Bioactive Compounds

Based on OB ≥ 30%, DL ≥ 0.18 and literatures, a total of 16 potential bioactive compounds of CCF against COAD were collected. These compounds contained 10 alkaloids, 1 dianthrone, 1 flavonoid, 1 triterpenoid, 1 amide, 1 lactone, and 1 polyphenol. Numerous studies have reported that many alkaloids anti-inflammatory and anti-tumor bioactive properties, which supports the objectives of this study. Further information on the 16 compounds is shown in Table 2.

Table 2.

Information on Bioactive Compounds Contained in CCF.

Type	MOL ID	Compounds	OB (%)	DL	CAS	Degree
Alkaloids	MOL 001454	Berberine	36.86	0.78	2086-83-1	17
	MOL 000785	Palmatine	64.6	0.65	3486-67-7	16
	MOL 002894	Berberrubine	35.74	0.73	15401-69-1	15
	MOL 002898	Groenlandicine	28.42	0.72	38691-95-1	14
	MOL 002897	Epiberberine	43.09	0.78	6873-09-2	14
	MOL 002904	Berlambine	36.68	0.82	549-21-3	13
	MOL 001458	Coptisine	30.67	0.86	3486-66-6	13
	MOL 002639	Obamegine	2.55	0.11	479-37-8	11
	MOL 002668	Worenine	45.83	0.87	38763-29-0	11
	MOL 002903	Canadine	55.37	0.77	522-97-4	9
Lactone	MOL 000622	Magnograndiolide	63.71	0.19	92618-98-9	14
Amide	MOL 008647	Moupinamide	86.71	0.26	66648-43-9	13
Dianthrone	MOL 000762	Palmidin a	35.36	0.65	17062-55-4	13
Flavonoid	MOL 000098	Quercetin	46.43	0.28	117-39-5	13
Triterpenoid	MOL 013352	Obacunone	43.29	0.77	751-03-1	10
Polyphenol	MOL 002902	Ethyl caffeate	103.85	0.07	66648-50-8	9

Acquisition of Candidate Targets

The GES41657 dataset (12 normal and 76 COAD samples) in the GEO database was analyzed for DEGs, and the results were displayed as heat and volcano maps. A total of 2639 DEGs were obtained, including 1565 up-regulated genes and 1,074 down-regulated genes (Figure 2A and B). Additionally, 1,293 genes from Gene cards were collected. By comprehensively analyzing the targets of 16 compounds obtained from the SwissTargetPrediction database, and removing the redundancy between them, 556 targets were obtained. Finally, 55 candidate targets of CCF against COAD were obtained using the Venn 2.1.0 database (Figure 2C). The symbols of the 55 targets are listed in Table 3.

Figure 2.

Note: (A) Volcano plot of 2693 DEGs from GSE41657. (B) Microarray map of the top 100 DEGs from GSE41657, where red represents up-regulated genes and blue represents down-regulated genes. (C) Venn diagram of CCF and COAD related targets.

Table 3.

Candidate Targets of CCF Against COAD.

Gene Official Symbol
AURKB	CHEK2	CSF1R	BCL2	HSP90AB1	AXL
PLK1	ABCG2	PRKCB	EPHB4	TGM2	FASN
CHEK1	CCNE1	CDK1	PTGES	TTK	PLA2G2A
PIK3CG	MMP3	DPP4	MPO	TEK	EPHX1
AURKA	PHLPP2	BIRC2	TLR7	BMP4	CASP7
MMP1	ABCC1	PTGS1	PRKDC	ATR	TOP2A
KIT	MELK	MIF	PDCD4	EPHA2	PTP4A3
PTGS2	ABCB1	CA9	NR3C1	MMP7	PRMT1
HPGD	NOS2	AKR1C3	HMOX1	EPHB2	JUN
ALDH1A1

Functional Enrichment Analysis

Using the DAVID database, 18 KEGG pathways (52, 94.5%), 139 GO BP (55, 100%) entries, 28 GO CC (55, 100%) entries, and 28 GO MF (55, 100%) entries were obtained. Among them, 18 KEGG pathways, 98 GO BP entries, 21 GO CC entries, and 24 GO MF entries were significant (p < 0.05). The visualization results of the GO entries showed that the above candidate targets were widely distributed from the cell surface to the nuclear membrane (Figure 3A). And they regulated the activities of protein kinase, tyrosine kinase, and transmembrane receptor protein. Therefore, these targets affected COAD cell growth through multiple BP entries involved in cell apoptosis, proliferation, cell cycle, inflammation, and drug responses. These results were also confirmed in the KEGG pathways enrichment (Figure 3B). The top ten pathways, such as pathways in cancer, PI3K-Akt signaling pathway, Cell cycle, Ras signaling pathway, and other pathways had been widely reported to be significantly associated with the refractory tumor.

Figure 3.

Note: (A) The circos of the top 10 GO BP, GO MF and GO CC entries. (B) The bubble plot of the top 10 KEGG pathways. (C) The “C-T-P” network. (D) The “PPI” network, where different colors represent different modules in it.

Construction and Analysis of “C-T-P” and “PPI” Networks

A “C-T-P” network consisting of 81 nodes and 280 edges was constructed using Cytoscape. The red nodes in the network represented bioactive compounds, greed nodes represented KEGG pathways, while blue nodes represented candidate targets (Figure 3C). Using the STRING and Cytoscape, a “PPI” network with 55 nodes and 174 edges was obtained. The module analysis of the 55 targets was carried out using the MCODE enhancement package of Cytoscape, and a total of 11 modules were obtained (Figure 3D). With comprehensive comparison and analysis of the degrees of nodes in both networks, we identified the top 5 targets based on the median principle (median = 11) as prostaglandin-endoperoxide synthase 2 (PTGS2, degree = 28), jun proto-oncogene (JUN, degree = 27), checkpoint kinase 1 (CHEK1, degree = 26), polo like kinase 1, (PLK1, degree = 26), and aurora kinase B (AURKB, degree = 25).

Molecular Docking Simulation

The docking affinity heatmap (Figure 4A) showed that berberine had the best potential bioactivity. The particularity of the structure of JUN (5t01) results in that it did not have a good binding pocket on its surface, so several bioactive compounds could not form a stable binding model with it. To consolidate the accuracy of the docking results, AURKB (2vgo), CHK1 (3jvs) and PLK1 (4j52) were considered as potential targets since berberine and positive drugs had similar affinities to these targets (affinity numerical difference was less than 2). The docking pockets and 2D modes of compounds with targets are shown in Figure 4. Undeniably, further validation of the expression of these three targets in COAD and normal colon tissues using clinical data was necessary to confirm their importance.

Figure 4.

Note: (A) Affinity heatmap of berberine and positive drugs docking with hub targets. (B) Docking results of positive drug and AURKB. The left side of the picture is the docking 3D diagram and the docking pocket, and the right side is the binding force diagram of the docking. (C) Docking results of berberine and AURKB. (D) Docking results of positive drug and CHK1. (E) Docking results of berberine and CHK1. (F) Docking results of positive drug and PLK1. (G) Docking results of berberine and PLK1.

Clinical Dataset Validation

Analysis of 41 normal and 286 patient samples recorded in the UNLCAN database found that the expressions of CHK1 (p < 1×E-12), PLK1 (p = 1.62×E-12), and AURKB (p = 1.62×E-12) were significantly higher in tumor samples than in normal samples (Figure 5A–C). Simultaneous analysis of 455 normal and 820 patient samples found that the expression of CHK1 (p =5.5×E-262), PLK1 (p = 0), and AURKB (p = 0) in tumor was also significantly higher than that in normal samples (Figure 5D–F). These results were consistent with our previous findings. Multiple studies had shown that the high expression of CHK1, PLK1, and AURKB caused changes in the cycle of HCT-116 cells and promoted their proliferation.

Figure 5.

Note: (A) Expression of CHEK1 in COAD based on UNLCAN database. (B) Expression of PLK1 in COAD based on UNLCAN database. (C) Expression of AURKB in COAD based on UNLCAN database. (D) Expression of CHEK1 in COAD based on TCGA database. (E) Expression of PLK1 in COAD based on TCGA database. (F) Expression of AURKB in COAD based on TCGA database.

CCF Ethanol Extracts and Berberine Decreased the Proliferation of HCT-116 Cells

In vitro experiments revealed a significant decrease in cell viability of the treatment groups compared to the control group (^*p < 0.05 or ^** p < 0.01) as the concentrations of CCF ethanol extracts and berberine increased (Figure 6A and B). Notably, the inhibitory effect of berberine on HCT-116 cells was better than that of CCF ethanol extracts, with their half-maximal inhibitory concentrations (IC₅₀) for HCT-116 cells at 33.61 µg/mL and 130.5 µg/mL. The results of plate cloning showed that HCT-116 cells treated with CCF ethanol extracts and berberine did inhibit the proliferation of HCT-116 cells in a concentration-dependent manner (Figure 6C and D).

Figure 6.

Note: (A) Viability of HCT-116 cells after treatment with CCF ethanol extracts at different concentrations (0, 10, 20, 40, 80, 160, and 320 µg/mL) for 24 h. (B) Viability of HCT-116 cells after treatment with berberine at different concentrations (0, 10, 20, 40, 80, and 160 µg/mL) for 24 h. (C) 160 µg/mL of CCF ethanolic extract inhibited the cloning of HCT-116 Cells. (D) 80 µg/mL of berberine inhibited the cloning of HCT-116 Cells. (E) CCF ethanol extracts at different concentrations (80 and 160 µg/mL) down-regulated the expression of the three hub targets in HCT-116 cells. (F) Berberine at different concentrations (40 and 80 µg/mL) down-regulated the expression of the 3 hub targets in HCT-116 Cells. ^*p < 0.05 and ^**p < 0.01 versus the control group.

CCF Ethanol Extracts and Berberine Blocked the Cell Cycle of HCT-116 Cells

As shown in Figure 7D, the cell cycle of HCT-116 cells treated with different concentrations of berberine (40 and 80 µg/mL) was significantly inhibited in G2 phase in a dose-dependent manner compared with the control group. However, the effects of CCF ethanol extracts on the cell cycle of HCT-116 cells did not exhibit regularity (Figure 7C and D), possibly due to its extremely complex compounds, and the multi-targets and multi-pathways it acts on. Further analysis will be conducted to investigate the reasons for this phenomenon in future works.

Figure 7.

Note: (A) Compared with the control group, CCF ethanol extracts at different concentrations (80 and 160 µg/mL) promoted the apoptosis of HCT-116 cells. (B) Compared with the control group, berberine at different concentrations (40 and 80 µg/mL) promoted the apoptosis of HCT-116 cells. (C) CCF ethanol extracts at different concentrations (80 and 160 µg/mL) blocked the cell cycle of HCT-116 cells compared with the control group. (D) Berberine at different concentrations (40 and 80 µg/mL) blocked the cell cycle of HCT-116 cells compared with the control group.

CCF Ethanol Extracts and Berberine Promoted the Apoptosis in HCT-116 Cells

To determine whether CCF ethanol extracts and berberine promoted apoptosis in HCT-116 cells, the cells were preincubated with them for 24 h and stained with Hoechst 33342 for 10 min. The cells were then observed under the microscope, and with the increase in drug concentration, the number of cells decreased and the number of cells with bright blue nuclei increased (Figure 7A and B), confirming that different concentrations of CCF ethanol extracts and berberine inhibited the growth of HCT-116 cells and promoted their apoptosis.

CCF Ethanol Extracts and Berberine Down-regulated the Expression of the 3 Hub Targets

The expression of PLK1, CHK1, and AURKB was down-regulated in the CCF ethanol extracts and berberine-treated groups compared to the control group (^**p < 0.01), as revealed by qPCR analysis, thus confirming the accuracy of the above results (Figure 6E and F).

Discussion

COAD is the leading cause of lethality globally, while the development of therapeutic drugs in this area remains a daunting challenge. Over the past few hundred years, TCMs have been explored as a potential source of anti-cancer drugs (Wang, Qi, et al., 2020). As a classic TCM, CCF is often used to clear heat, dry dampness, purge fire and detoxify, and has been proven to have anti-cancer effects by modern pharmacological studies (Yu et al., 2020; Zhang, Ma, et al., 2021). However, the bioactive compounds and mechanism of CCF against COAD are not well understood. With the advancement of science and technology, integrated bioinformatics has become an important way to identify core genes and pathways related to disease pathogenesis. As representatives of bioinformatics, network pharmacology and molecular docking have been widely used in the research and development of TCM (Taha et al., 2020; Xia et al., 2020). Therefore, this study aimed to explore the bioactive compounds and mechanism of CCF against COAD through comprehensive bioinformatics analysis and in vitro experiments.

Bioinformatics analysis of CCF revealed that it treated COAD by modulating the expression of 55 targets. The distribution of these targets in cells had no obvious regularity, but they were widely involved in the proliferation and apoptosis of cancer cells, which coincided with the results of the KEGG pathway analysis. Interestingly, in the “PPI” and “C-T-P” networks, the degree values of the above targets were different. In order to ensure the rationality of the experiments, the degree values of both networks were integrated and found that PTGS2 (degree=28), JUN (degree=27), CHEK1 (degree=26), PLK1 (degree=26), and AURKB (degree = 25) were the core targets of CCF anti-COAD. The semi-flexible docking results of Autodock Vina showed that the binding pocket of JUN could not fit well with small molecules. Berberine, a bioactive compound of CCF, was found to have the highest degree in the “C-T-P” network and a good affinity with the core targets, making it a promising candidate drug for further study. For the sake of accuracy, three targets (CHEK1, PLK1, and AURKB) whose binding affinity was closer to that of the positive drugs were selected as hub targets. More than 1,000 clinical samples in TCGA and UNLCAN databases also confirmed their significantly higher expression in COAD than in normal samples.

Numerous studies have hypothesized that CHEK1, PLK1, and AURKB are extensively involved in the regulation of tumor cell proliferation and apoptosis, which is consistent with GO and KEGG enrichment results (Liu et al., 2017; Õsz et al., 2019; Wang, Yu, et al., 2020). As hallmarks of tumor cells, rapid proliferation and resistance to apoptosis are major obstacles in the treatment of cancer (Wang et al., 2018; Wei et al., 2019). The plate cloning and morphological analysis of HCT-116 cells before and after administration indeed confirmed the inhibitory effect of CCF ethanol extracts and berberine on cell proliferation. From the results of Hoechst 33342 staining experiments, we found that the different concentrations of CCF ethanol extracts and berberine promote apoptosis and reduce the number of HCT-116 cells. Under the light microscope, the morphology of the cells after drug intervention did change and they were basically eliminated after washing with PBS. It cannot be ignored that the effects of CCF ethanol extracts and berberine on the cell cycle of HCT-116 were not completely consistent, and other chemical compounds in the extract might be the cause of this phenomenon. In HCT-116 cells, the three hub targets with high docking affinity to berberine, namely CHEK1, PLK1, and AURKB, were significantly downregulated after drug intervention, which verified the accuracy of our results.

Reckoning the study findings, CCF and its bioactive compounds berberine are potential adjuvant drugs for COAD. Nevertheless, further in vitro and in vivo experiments are necessary to confirm the efficacy and safety of these compounds.

Conclusions

In conclusion, this study demonstrated that CCF exhibits anti-COAD activity by down-regulating the expression of CHEK1, PLK1, and AURKB, with berberine identified as its main bioactive compound. The findings suggest that CCF and berberine have potential as adjuvant drugs for the treatment of COAD. However, further in vitro and in vivo experiments are needed to confirm these results and explore the clinical applications of CCF and berberine in COAD therapy.

Abbreviations

CCF: coptis chinensis franch.; COAD: colorectal adeno‑carcinoma; TCMSP: traditional chinese medicine systems pharmacology database, analysis platform; MDS: molecular docking simulation; GEO: gene expression omnibus; KEGG: kyoto encyclopedia of genes and genomes; GO: gene ontology; C-T-P: compounds-targets-pathways; PPI: protein-protein interaction; qPCR: quantitative real-time PCR; CHEK1: checkpoint kinase 1; PLK1: polo like kinase 1; AURKB: aurora kinase B; TCM: traditional chinese medicines; DEGs: differentially expressed genes; OB: oral bioavailability; DL: drug-likeness; DAVID: database for annotation, visualization and integrated discovery; GO BP: gene ontology biology; GO CC: gene ontology cellular compound; GO MF: gene ontology molecular function.

Footnotes

Acknowledgment

The authors declare that this work was done by the authors named in this article and all liabilities pertaining to claims relating to the content of this article will be borne by the authors.

The Statement of Informed Consent and Ethical Approval

Necessary ethical clearances and informed consent was received and obtained respectively before initiating the study from all participants.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

This study was supported by the research projects of clinical medicine in Wuhan in 2019 [Wuhan Municipal Health Commission, WZ19Q15]. Thanks to everyone who had contributed to this research.

Summary

CCF significantly down-regulated CHEK1, PLK1 and AURKB, inhibited proliferation, and promoted apoptosis of HCT-116 cells. Berberine is the main bioactive compound of CCF against COAD.

ORCID iD

Han Tu

References

Burley

S.K.

, Berman

H.M.

, Kleywegt

G.J.

, Markley

J.L.

, Nakamura

, & Velankar

(2017). Protein Data Bank (PDB): The single global macromolecular structure archive. Methods in Molecular Biology , 1607, 627–641. https://doi.org/10.1007/978-1-4939-7000-1_26. PMID: 28573592

Chandrashekar

D.S.

, Bashel

, Balasubramanya

S.A.H.

, Creighton

C.J.

, Ponce-Rodriguez

, Chakravarthi

, & Varambally

(2017). UALCAN: A portal for facilitating tumor subgroup gene expression and survival analyses. Neoplasia (New York N. Y.) , 19(8), 649–658. https://doi.org/10.1016/j.neo.2017.05.002. PMID: 28732212

Clough

, & Barrett

(2016). The gene expression omnibus database. Methods in Molecular Biology , 1418, 93–110. https://doi.org/10.1007/978-1-4939-3578-9_5

Daina

, Michielin

, & Zoete

(2019). SwissTargetPrediction: Updated data and new features for efficient prediction of protein targets of small molecules. Nucleic Acids Research , 47(W1), W357–W364. https://doi.org/10.1093/nar/gkz382. PMID: 31106366

Dang

, Gao

, Huang

, & Qi

(2020). Clinical complete regression after local radiotherapy combined with chemotherapy for stage IV rectal cancer: A case report. Molecular and Clinical Oncology , 13(2), 186–190. https://doi.org/10.3892/mco.2020.2049. PMID: 32714544

Eberhardt

, Santos-Martins

, Tillack

A.F.

, & Forli

(2021). AutoDock Vina 1.2.0: New docking methods, expanded force field, and python bindings. Journal of Chemical Information and Modeling , 61(8), 3891–3898. https://doi.org/10.1021/acs.jcim.1c00203. PMID: 34278794

, Song

, Xie

, Ouyang

, Xie

, Li

, Su

, Xu

, Huang

, & Liang

(2020). Berberine inhibits cancer cells growth by suppressing fatty acid synthesis and biogenesis of extracellular vesicles. Life Sciences , 257, 118122. https://doi.org/10.1016/j.lfs.2020.118122. PMID: 32702446

, Xue

, Gao

, Shen

, Zhang

, Chen

, Xue

, Pan

, Tang

, Zhu

, Wu

, & Dou

(2020). Mechanisms of indigo naturalis on treating ulcerative colitis explored by GEO gene chips combined with network pharmacology and molecular docking. Scientific Reports , 10(1), 15204. https://doi.org/10.1038/s41598-020-71030-w. PMID: 32938944

Guo

, Huang

, Wang

, Tan

H.Y.

, Cheung

, Chen

, & Feng

(2019). Integrating network pharmacology and pharmacological evaluation for deciphering the action mechanism of herbal formula Zuojin Pill in suppressing hepatocellular carcinoma. Frontiers in Pharmacology , 10, 1185. https://doi.org/10.3389/fphar.2019.01185. PMID: 31649545

10.

Guo

, Bao

, Ma

, & Yang

(2017). Identification of key candidate genes and pathways in colorectal cancer by integrated bioinformatical analysis. International Journal of Molecular Sciences , 18(4), 722. https://doi.org/10.3390/ijms18040722. PMID: 28350360

11.

, Ou

, Chen

, & Ding

(2020). Network pharmacology-based study on the molecular biological mechanism of action for compound Kushen injection in anti-cancer effect. Medical Science Monitor , 26, e918520. https://doi.org/10.12659/msm.918520. PMID: 31892693

12.

Huang da

, Sherman

B.T.

, & Lempicki

R.A.

(2009). Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nature Protocols , 4(1), 44–57. https://doi.org/10.1038/nprot.2008.211. PMID: 19131956

13.

Jia

, Xu

, & Wang

(2021). Venn diagrams in bioinformatics. Brief in Bioinformatics , 22(5), 1–17. https://doi.org/10.1093/bib/bbab108. PMID: 33839742

14.

Khelfaoui

, Harkati

, & Saleh

B.A.

(2021). Molecular docking, molecular dynamics simulations and reactivity, studies on approved drugs library targeting ACE2 and SARS-CoV-2 binding with ACE2. Journal of Biomolecular Structure and Dynamics , 39(18), 7246–7262. https://doi.org/10.1080/07391102.2020.1803967. PMID: 32752951

15.

Kim

, Chen

, Cheng

, Gindulyte

, He

, Li

, Shoemaker

B.A.

, Thiessen

P.A.

, Yu

, Zaslavsky

, Zhang

, & Bolton

E.E.

(2021). PubChem in 2021: New data content and improved web interfaces. Nucleic Acids Research , 49(D1), D1388–D1395. https://doi.org/10.1093/nar/gkaa971. PMID: 33151290

16.

Koster

T.D.

, Klooster

, Ten Hacken

N.H.T.

, van Dijk

, & Slebos

D.J.

(2020). Endobronchial valve therapy for severe emphysema: An overview of valve-related complications and its management. Expert Review of Respiratory Medicine , 14(12), 1235–1247. https://doi.org/10.1080/17476348.2020.1813571. PMID: 32842819

17.

Lee

H.N.

, Kim

J.K.

, Kim

J.H.

, Lee

S.J.

, Ahn

E.K.

, Oh

J.S.

, & Seo

D.W.

(2014). A mechanistic study on the anti-cancer activity of ethyl caffeate in human ovarian cancer SKOV-3 cells. Chem-Biological Interactions , 219, 151–158. https://doi.org/10.1016/j.cbi.2014.05.017. PMID: 24892518.

18.

Liu

, Sun

, & Wang

(2017). PLK1, a potential target for cancer therapy. Translational Oncology , 10(1), 22–32. https://doi.org/10.1016/j.tranon.2016.10.003. PMID: 27888710

19.

Õsz

., Aszódi

, Vajda

, Keserû

M. G.

, Moll

H.P.

, Casanova

, & Gyõrffy

(2019). [CHEK1 expression and inhibitors in TP53 mutant cancer]. Magyar Onkologia , 63(4), 345–352. (A checkpoint kinase 1 expressziója és gátlása TP53-mutációt hordozó rosszindulatú daganatokban.), PMID: 31821389

20.

Pinzi

, & Rastelli

(2019). Molecular docking: Shifting paradigms in drug discovery. International Journal of Molecular Sciences , 20(18), 4331. https://doi.org/10.3390/ijms20184331. PMID: 31487867

21.

, Chen

, Zhang

, & Tian

(2020). Construction of prognostic predictor by comprehensive analyzing alternative splicing events for colon adenocarcinoma. World Journal of Surgical Oncology , 18(1), 236. https://doi.org/10.1186/s12957-020-02010-7. PMID: 32883335

22.

, Li

, Wang

, Zhou

, Li

, Huang

, Li

, Guo

, Tao

, Yang

, Xu

, Li

, Wang

, & Yang

(2014). TCMSP: A database of systems pharmacology for drug discovery from herbal medicines. Journal of Cheminformatics , 6, 13. https://doi.org/10.1186/1758-2946-6-13. PMID: 24735618

23.

Shannon

, Markiel

, Ozier

, Baliga

N.S.

, Wang

J.T.

, Ramage

, Amin

, Schwikowski

, & Ideker

(2003). Cytoscape: A software environment for integrated models of biomolecular interaction networks. Genome Research , 13(11), 2498–2504. https://doi.org/10.1101/gr.1239303. PMID: 14597658

24.

Siraj

F.M.

, Natarajan

, Huq

M.A.

, Kim

Y.J.

, & Yang

D.C.

(2015). Structural investigation of ginsenoside Rf with PPARγ major transcriptional factor of adipogenesis and its impact on adipocyte. Journal of Ginseng Research , 39(2), 141–147. https://doi.org/10.1016/j.jgr.2014.10.002. PMID: 26045687

25.

Song

, Zhang

, Dai

, Wang

, & Du

(2020). Prediction of triptolide targets in rheumatoid arthritis using network pharmacology and molecular docking. International Immunopharmacology , 80, 106179. https://doi.org/10.1016/j.intimp.2019.106179. PMID: 31972422

26.

Szklarczyk

, Gable

A.L.

, Nastou

K.C.

, Lyon

, Kirsch

, Pyysalo

, Doncheva

N.T.

, Legeay

, Fang

, Peer Bork

, Jensen

L.J.

, & von Mering

(2021). The STRING database in 2021: Customizable protein-protein networks, and functional characterization of user-uploaded gene/measurement sets. Nucleic Acids Research , 49(D1), D605–D612. https://doi.org/10.1093/nar/gkaa1074. PMID: 33237311

27.

Taha

K.F.

, Khalil

, Abubakr

M.S.

, & Shawky

(2020). Identifying cancer-related molecular targets of Nandina domestica Thunb. by network pharmacology-based analysis in combination with chemical profiling and molecular docking studies. Journal of Ethnopharmacology , 249, 112413. https://doi.org/10.1016/j.jep.2019.112413. PMID: 31760157

28.

Wang

, Liu

, Jia

, Chu

, Zhou

, He

, Guo

, Chen

, & Xu

(2018). Nanostructured lipid carriers for MicroRNA delivery in tumor gene therapy. Cancer Cell International , 18, 101. https://doi.org/10.1186/s12935-018-0596-x. PMID: 30008618

29.

Wang

, Qi

, Wang

, Zhang

, Pan

, Huai

, Qu

, & Zhao

(2020). A review of traditional Chinese medicine for treatment of glioblastoma. Bioscience Trends , 13(6), 476–487. https://doi.org/10.5582/bst.2019.01323. PMID: 31866614

30.

Wang

, Fu

J.L.

, Hao

H.F.

, Jiao

Y.N.

, Li

P.P.

, & Han

S.Y.

(2021). Metabolic reprogramming by traditional Chinese medicine and its role in effective cancer therapy. Pharmacological Research , 170, 105728. https://doi.org/10.1016/j.phrs.2021.105728. PMID: 34119622

31.

Wang

, Su

, Zhong

, Yang

, Chen

, Liu

, Wu

, Zhong

, Li

, Mao

, & Lu

(2020). An Eight-CircRNA Assessment model for predicting biochemical recurrence in prostate cancer. Frontiers in Cell and Developmental Biology , 8, 599494. https://doi.org/10.3389/fcell.2020.599494. PMID: 33363156

32.

Wang

, Jensen

M.A.

, & Zenklusen

J.C.

(2016). A practical guide to the cancer genome atlas (TCGA). Methods in Molecular Biology , 1418, 111–141. https://doi.org/10.1007/978-1-4939-3578-9_6. PMID: 27008012

33.

Wang

, Yu

, Wang

G.H.

, Zhou

Y.M.

, Deng

J.P.

, Feng

, Chen

J.Q.

, & Tian

(2020). AURKB promotes the metastasis of gastric cancer, possibly by inducing EMT. Cancer Management and Research , 12, 6947–6958. https://doi.org/10.2147/cmar.s254250. PMID: 32801915

34.

Wei

, Wen

J.Y.

, Chen

, Ma

X.K.

, Wu

D.H.

, Chen

Z.H.

, & Huang

J.L.

(2019). Oncogenic ADAM28 induces gemcitabine resistance and predicts a poor prognosis in pancreatic cancer. World Journal of Gastroenterology , 25(37), 5590–5603. https://doi.org/10.3748/wjg.v25.i37.5590. PMID: 31602160

35.

Wei

, Li

, Qiao

, Ma

, Xie

, Wang

, Liu

, Wei

, Li

, Zheng

, Sun

, & Yu

(2020). Based on network pharmacology to explore the molecular targets and mechanisms of gegen qinlian decoction for the treatment of ulcerative colitis. BioMed Research International , 2020, 5217405. https://doi.org/10.1155/2020/5217405. PMID: 33299870

36.

Xia

Q.D.

, Xun

, Lu

J.L.

, Lu

Y.C.

, Yang

Y.Y.

, Zhou

, Hu

, Li

, & Wang

S.G.

(2020). Network pharmacology and molecular docking analyses on Lianhua Qingwen capsule indicate Akt1 is a potential target to treat and prevent COVID-19. Cell Proliferation , 53(12), e12949. https://doi.org/10.1111/cpr.12949. PMID: 33140889

37.

, Ren

, Liang

, Zhang

, Jiang

, Ma

, Li

, & Ye

(2020). Effect of epiberberine from Coptis chinensis Franch on inhibition of tumor growth in MKN-45 xenograft mice. Phytomedicine , 76, 153216. https://doi.org/10.1016/j.phymed.2020.153216. PMID: 32534357

38.

Zhang

, Yuan

, Zhou

, Qin

, Xu

, Men

, & Lin

(2020). Network pharmacology analysis of Chaihu Lizhong Tang treating non-alcoholic fatty liver disease. Computational Biology and Chemistry , 86, 107248. https://doi.org/10.1016/j.compbiolchem.2020.107248. PMID: 32208163

39.

Zhang

, Zhu

, Bai

, & Ning

(2019). Network pharmacology databases for traditional Chinese medicine: Review and assessment. Frontiers in Pharmacology , 10, 123. https://doi.org/10.3389/fphar.2019.00123. PMID: 30846939

40.

Zhang

, Ma

, & Zhang

(2021). Berberine for bone regeneration: Therapeutic potential and molecular mechanisms. Journal of Ethnopharmacology , 277, 114249. https://doi.org/10.1016/j.jep.2021.114249. PMID: 34058315

41.

Zhang

, Yao

, Fu

, Yuan

, Wu

, Wang

, Hong

, Yang

, & Wu

(2021). Inhibition effect of oxyepiberberine isolated from Coptis chinensis franch. On non-small cell lung cancer based on a network pharmacology approach and experimental validation. Journal of Ethnopharmacology , 278, 114267. https://doi.org/10.1016/j.jep.2021.114267. PMID: 34087401

Bioinformatics,Molecular Docking Simulation and in vitro Experiments Reveal the Bioactive Compounds and Mechanism of Coptis chinensis Franch. Against Colorectal Adenocarcinoma

Abstract

Background

Objectives

Materials and Methods

Results

Conclusion

Keywords

Introduction

The Workflow of this Study.

Materials and Methods

Bioactive Compounds Collection

Condidate Targets of CCF Against COAD

Enrichment Analysis

Acquisition of Hub Targets

Molecular Docking Simulation

UNLCAN and TCGA Analysis

Acquisition of CCF Ethanol Extracts and Berberine

MTT Assay

Plate Cloning Assay

Cell Cycle Detection by Flow Cytometry

Hoechst 33342 Staining Assay

Quantitative Real-time PCR Assay

Primers for the Three Hub Targets.

Statistical Analysis

Results

Acquisition of Bioactive Compounds

Information on Bioactive Compounds Contained in CCF.

Acquisition of Candidate Targets

Candidate Targets of CCF Against COAD.

Functional Enrichment Analysis

Construction and Analysis of “C-T-P” and “PPI” Networks

Molecular Docking Simulation

Clinical Dataset Validation

CCF Ethanol Extracts and Berberine Decreased the Proliferation of HCT-116 Cells

CCF Ethanol Extracts and Berberine Blocked the Cell Cycle of HCT-116 Cells

CCF Ethanol Extracts and Berberine Promoted the Apoptosis in HCT-116 Cells

CCF Ethanol Extracts and Berberine Down-regulated the Expression of the 3 Hub Targets

Discussion

Conclusions

Abbreviations

Footnotes

Acknowledgment

The Statement of Informed Consent and Ethical Approval

Declaration of Conflicting Interests

Funding

Summary

ORCID iD

References