Evaluation of hypolipidemic activity of Brassica rapa and its mechanism analysis

Jian-Bing Liu, Feng Lin, Jun Lin, Qiu-Xia Zheng, Han-Xing Su, Li-Yun Wu

Fujian Engineering and Research Center for Microbial Techniques of Hongqu, Fujian Institute of Microbiology, Fuzhou 350007, China

Keywords:Hypolipidemic activity Oleic acid HepG2 cells Network pharmacology Glycosides

ABSTRACT Objective: This study was designed to evaluate the hypolipidemic activity of Brassica rapa and explore its mechanism by network pharmacology approach. Methods: The hypolipidemic activity of Brassica rapa aqueous extract (BRAE) was evaluated by bile salt-binding capacity and oleic acid-induced HepG2 steatosis cell model. The active compounds of Brassica rapa were collected from literature, and targets were predicted from SwissTargetPrediction and SEA Search Server platform. Cytoscape 3.7.2 software was used to construct “compound-target”network. Protein-protein interaction (PPI) network was constructed by String platform. Gene ontology (GO) and Kyoto encyclopedia of genes and genomes (KEGG) enrichment analyses based on DAVID database. Results: In vitro experiment showed that BRAE exhibited excellent bile salt-binding capacity, and the binding rates of sodium glycylcholate, sodium taurocholate and sodium deoxycholate were 36.01%, 28.93% and 78.55% at 7 g.L-1, respectively.BRAE showed a significant hypolipidemic activity on steatosis cells, which can reduce the accumulation of lipid droplets and the levels of triglyceride (TG) and total cholesterol (TC)compared with the model group (P＜0.05). 21 active components of Brassica rapa and 682 potential targets were obtained, among which 55 targets were associated with hyperlipidemia.The “compound-target” network showed that 6-paradol, 6-shogaol, benzyl-beta-Dglucopyranoside, benzyl-alpha-D-fructofuranoside and liquiritigenin were core components.PPI network and KEGG enrichment analysis found that Brassica rapa could treat hyperlipemia by regulated the core targets, such as VEGFA, IL6, EGFR and PPARG, and affected 36 signaling pathways (including starch and sucrose metabolism, galactose metabolism, insulin resistance, HIF-1, PI3K-Akt). Conclusion: This study showed that BRAE had excellent hypolipidemic activity in vitro, and preliminarily revealed the multi-component, multi-target,multi-path mechanism in the treatment of hyperlipidemia by network pharmacology approach,which provides a scientific foundation for further study.

1. Introduction

Brassica rapa L. is also known as rappini, turnip, Nima and Yuangen, which is a one- or two-year old herb in the Brassicaceae family [1-2]. It’s root, stem and leaf can be processed into delicious dishes, especially root, which looks like radish and with a sweet taste. Brassica rapa is a traditional medicinal and edible plant in Tibetan areas of China, and its medicinal use has been recorded in Compendium of Materia Medica and Holy Benevolent Prescriptions[3]. According to literature reports, Brassica rapa contains many bioactive components, such as flavonoids, polysaccharides,triterpenes, glucosinolates and fatty acids, and it has antiinflammatory, antioxidant, antibacterial, anti-fatigue, immune regulation, anti-tumor and hypolipidemic effects [4]. Li, et al. [5]found that Brassica rapa polysaccharides exert hepatoprotective effect against the CCl4-induced liver injury via modulating the apoptotic and inflammatory responses, and could be considered a hepatoprotective medicine. Zhang, et al. [6] studied the effects of n-butyl alcohol extracted from Brassica rapa on the diabetic mice,and found that it could effectively lower blood pressure, lower blood lipid and raise insulin.

Hyperlipidemia is a systemic disease, which is characterized by elevated lipid levels in blood including total cholesterol (TC), total glyceride (TG), and low-density lipoprotein cholesterol (LDL-C)and so on. With the improvement of people's living standards,160 million people have been diagnosed with dyslipidemia in China. Hyperlipidemia is one of the major inducements of cardio cerebrovascular disease, such as diabetes, atherosclerosis, coronary heart disease, stroke. Therefore, the control of hyperlipidemia is important for the prevention of cardiovascular diseases. At present,statins are widely used in the treatment of hyperlipidemia, but these drugs are easy to rebound after withdrawal, and also bring multiple side effects [7].

This study was designed to evaluate the hypolipidemic activity of Brassica rapa aqueous extract (BRAE) by bile salt-binding capacity and cellular-based model assay, and mechanism analysis by network pharmacology approach, which provides a scientific foundation and new idea for its further development and application in functional foods.

2. Materials and methods

2.1 Materials

The Brassica rapa root samples were taken from Linzhi City, Tibet Province, China. HepG2 cells were purchased from iCell Bioscience Inc. (Shanghai, China) and cultured in complete medium (RPMI-1640 medium (Hyclone, USA), supplemented with 10% fetal bovine serum (Hyclone, USA)). trypsin, MTT, oil red O and oleic acid were purchased from Beijing Solarbio Science and Technology Co., Ltd.Triglyceride (TG) and protein concentration determination kits were purchased from Nanjing Jiancheng Biological Engineering Research Institute; Total cholesterol (TC) kit was purchased from Zhejiang Dong'ou Diagnostic Products Co., Ltd.

2.2 Preparation of BRAE

The Brassica rapa root samples were cut into thin sections, which was extracted with water (the ratio of feed-liquid was 3%, w/v) at 90℃ for 60 min to remove impurities. The filtered residues were collected and freeze-dried.

2.3 Determination of bile salt-binding capacity of BRAE

Standard solution of sodium glycine cholate, sodium taurocholate and sodium deoxycholate (Shanghai Macklin Biochemical Co.,Ltd) with different concentrations (0.01, 0.05, 0.1, 0.25, 0.5, 0.75,1 mmol.L-1) were prepared by phosphoric acid buffer solution(0.1 mol.L-1, pH=6.3). 1 mL above standard solution was added into 3 mL H2SO4solution (60%, w/v), after 20 min at 70 ℃, the absorbance at 387 nm was measured. The standard curve was drawn with the bile salt concentration as the horizontal coordinate and the absorbance value as the ordinate.

5ml BRAE aqueous solution with different concentrations (2 g.L-1,5 g.L-1, 7 g.L-1) was added with 3 mL pepsin (10 g.L-1) and 1 mL HCl solution (0.01 mol.L-1), oscillated at 37 ℃ for 1 h. Then, after adjusting pH to 6.3 with NaOH, 4 mL trypsin (10 g.L-1) was added,and oscillated at 37 ℃ for 1 h. Next, 4 mL of sodium glycocholate(0.4 mmol.L-1), sodium taurocholate (0.5 mmol.L-1) and sodium deoxycholate (1 mmol.L-1) were added respectively, oscillated at 37℃ for 1 h to perform bile salt-binding test. After centrifugation, 1 mL supernatant was taken for cholate content measurement, and the concentration of cholate in the sample solution was obtained from the standard curve. Cholate binding rate /%=(C0-C1)/C0*100, C0 is the amount of cholate added and C1 is the remaining amount of cholate.

2.4 MTT assay

HepG2 cells were collected and seeded in 96-well plates. After overnight growth and removal of the old culture medium, fresh medium containing different concentrations of BRAE (0.10、0.35、0.70、1.00、1.40 g.L-1) was respectively added to the experimental groups, while complete medium was added to the control groups,and set up a zero adjustment group with six duplicate wells in each groups. After 24 h, 10% MTT (v/v, 5 mg/mL) solution was added in all groups and incubated continuously for 4 h. Lastly, 100 µL DMSO was added into each and measured at 490 nm. The cell survival rate was calculated as (OD experimental-OD zero)/(OD control -OD zero)×100.

2.5 Oil red O staining

HepG2 cells were seeded in 96-well plates. After overnight growth and removal of the old culture medium, fresh medium containing 0.3 mmol.L-1oleic acid was respectively added to the model group,positive control group (lovastatin) and experimental groups (BRAE), while complete medium was added to the control group. After being continuously incubated for 24 h, model group was replaced with complete medium, positive control group was replaced with fresh medium containing 100 mg.L-1lovastatin, and experimental groups were replaced with fresh medium containing different concentrations of BRAE (0.10、0.35、0.70、1.00、1.40 g.L-1).After another 24 h incubation, the cells were fixed with neutral formaldehyde solution (10%, v/v) for 20 min, followed by oil red O staining for 40 min in the dark. After washing with isopropanol solution (60%,v/v) and PBS solution, the cells were observed under microscope 20x. An additional isopropanol solution was added and stood for 40 min, and then the absorbance value was detected at 528 nm to calculate the intracellular lipid accumulation rate as following formula: intracellular lipid accumulation rate/% =(ODexperimental-ODzero)/(ODcontrol- ODzero)×100.

2.6 Assessment of intracellular TG and TC levels

HepG2 cells were seeded in 6-well plates. The grouping, dosage and administration of this experiment were as described in previous experiment. After treatment, the cells were washed twice with PBS,then collected and separated by ultrasound. TG and TC levels was examined by assay kits.

2.7 Statistical analysis

The data was expressed as mean ± S.D. and was analyzed by oneway ANOVA of SPSS20.0. The values with p＜0.05 were considered significant.

2.8 Construction and screening of chemical components databases in Brassica rapa

The components of Brassica rapa were obtained by searching PubMed and Chinese knowledge net databases literature. All chemical components were input into admetSAR platform (http://lmmd.ecust.edu.cn/admetsar2/) to predict the molecular properties.The lipinski rules were as the screening criteria for the active components of Brassica rapa (the molecular weight is less than 500, the number of hydrogen bond donors is less than 5, the number of hydrogen bond acceptors is less than 10, the distribution coefficient of fat water is less than 5 and the number of rotatable bonds is no more than 10). Swiss Target Prediction (http://www.swisstargetprediction.ch/) and SEA Search Server(https://sea.bkslab.org/)databases were used to retrieve the gene targets for active components, selecting the Homo sapiens for the species, and collecting the targets for further analysis.

2.9 Potential target genes for hyperlipidemia

GeneCards databases (http://www.genecards.org/) were used to retrieve hyperlipidemia -related genes. The keywords used in the search were limited to "hyperlipidemia". The targets with a “relevance score” value greater than 2 were selected as hyperlipidemia related targets.

2.10 Construction of "compound-target" network

The targets obtained were compared to hyperlipidemia related targets were selected. The “compound-target” network was established and visualized by Cytoscape 3.6.1 software.

2.11 GO gene enrichment analysis and KEGG pathway annotation

The DAVID(https://david.ncifcrf.gov/)database was used to analyze the GO function and KEGG pathway enrichment. The items of cellular components (CC), molecular functions (MF) and biological processes (BP)were selected the top 10 according to FDR (false discovery rate) to draw the histogram. The results of KEGG pathway analysis were set as P ＜ 0.01 and count ＞ 3 as the significant signal pathway.

2.12 Construction and analysis of target protein-protein interaction (PPI) network

The genes were uploaded to STRING database, the species was selected as “Homo sapiens”, and the “minimum required interaction score” was set as 0.7 (high confidence). The obtained protein interaction data were submitted to Cytoscape3.6.1 to build a PPI network.

3. Results

3.1 Analysis of bile salt-binding capacity of BRAE

Drugs can indirectly reduce serum cholesterol by binding bile salts in the digestive tract [8]. As shown in Figure 1, BRAE exhibited excellent bile salt-binding capacity of sodium glycylcholate, sodium taurocholate and sodium deoxycholate, and increased with increase of its concentration. The binding rates of these three bile salts reached the maximum of 36.01%, 28.93% and 78.55% at 7 g.L-1,respectively.

Figure 1 The effect of Brassica rapa aqueous extract (BRAE) on bile salt-binding capacity (sodium glycylcholate, sodium taurocholate and sodium deoxycholate)

3.2 Hypolipidemic effect of BRAE on oleic acid-induced HepG2 steatosis cell

3.2.1 Effect on the growth of HepG2 cellsAs shown in Fig. 2A, within the concentration range of 0.1～1.4 g.L-1, BRAE had no significant effect on the growth of HepG2 cells(p＞0.05), and the cell viability was above 90%.

3.2.2 Effect of BRAE on intracellular lipid accumulation

The effect of BRAE to intracellular lipid accumulation were analyzed by means of Oil Red O staining (2C) and semi-quantitative detection (Fig. 2B). The results showed that there were dense red lipid droplets in cells of the model group, and the steatosis rate was significantly higher than that of control, which indicated that the oleic acid-induced HepG2 steatosis cell model was made successful.

3.2.3 Effect of BRAE on TG and TC levels

The plasma concentrations of triglyceride (TG) and total cholesterol(TC) are the main biomarker for diagnosing hyperlipidemia. This experiment analyzed the effects of 1.4 g.L-1BRAE on the levels of TG (Fig. 3A) and TC (Fig. 3B) in oleic acid-induced HepG2 steatosis cells. The results showed that the levels of TG and TC in the model group were significantly higher than control (P ＜ 0.01), while those in the treatment group intervened by BRAE were significantly lower than model group (P ＜ 0.01).

Table 1 Information of active components of Brassica rapa

Figure 2 Effect of Brassica rapa aqueous extract (BRAE) to intracellular lipid accumulation.A: Effects of BRAE on on HepG2 cell viability; B: semiquantitative detection of Lipid suppression rate of BRAE, n=3,±s D,*p ＜0.05, **p ＜ 0.01 vs. model group; C: Lipid droplets in the cells observed by Oil Red O staining, (Originalmagnification: × 200, a was control group, b was model group, h was lovastatin, c～g in turn were 0.10, 0.35, 0.70, 1, and 1.40 g.L-1 BRAE treatment group.)

Figure 3 Effect of BRAE on TG and TC Levels in HepG2 cells. n=3,±s D,*p ＜ 0.05, **p ＜ 0.01 vs. model group.

3.3 Mechanism of Brassica rapa in hypolipidemic based on network pharmacology

3.3.1 Filtering of active components of Brassica rapaAs shown in Table 1, 21 active components of Brassica rapa were identified in this study.

3.3.2 Screening of hypolipidemic targetsAfter removing repeated targets, a total of 682 potential targets were obtained from Swiss Target Prediction and SEA Search Server databases. The target genes related to hyperlipidemia were searched in the GeneCard databases, and 55 overlapping genes were identified by matching the aforementioned 682 genes with disease-associated genes (Fig.4). 15 active components of Brassica rapa matched to hyperlipidemia related targets, including 6 glycosides (red compounds), 3 flavonoids (yellow compounds), 2 phenolic (blue compounds), 2 isothiocyanates (turquoise compounds) and 2 other compounds (green compounds). The “compound-target” network has 71 nodes and 137 edges (Fig.4). Among the 5 compounds(6-paradol, 6-shogaol, benzyl-beta-D-glucopyranoside, benzyl-alpha-D-fructofuranoside and liquiritigenin) were connected to more than 10 genes, which can be inferred that these compounds might be core components.

Figure 4 compound-target network. Ellipse represents the target,quadrilateral represents the compound and size is defined by its degree value

3.3.3 Construction and analysis of target PPI Network

55 target genes associated with hypolipidemic were imported into the STRING database for PPI network construction (Fig.5).As shown in Fig.5, the PPI protein interaction network includes a total of 55 nodes, 118 edges, of which nodes represent the target protein and each edge represents the Protein-Protein interaction. In this network, the average shortest path length is 1.979, the average clustering coefficient is 0.250 and the average node degree is 5.2.The degree of each node represents the number of targets that are connected to the target, the greater the degree, the stronger the role of the target protein corresponding to this node in the network, which indicates that the target plays a core role in the entire interaction network. Therefore, according to the ranking of degree values, the degree values of VEGFA (degree = 18), IL6 (degree = 17), EGFR(degree = 13), PPARG (degree = 12) were more than 10, which were presumed to be the key targets for hypolipidemic in Brassica rapa.

Figure 5 PPI protein interaction network

3.3.4 Go enrichment analysis of targets

GO enrichment analysis was performed on the analyzed target proteins, and the top 10 GO entries (FDR ＜ 0.05) were selected based on false discovery rate (FDR). As shown in Fig. 6, the results indicate that these target proteins are related to biological process (BP), cellular component (CC) and molecular function(MF). 172 BP entries were enriched, accounting for 69.35%, mainly including carbohydrate metabolic process, glycogen catabolic process, cholesterol homeostasis, cholesterol metabolic process.

Figure 6 GO enrichment analysis

3.3.5 KEGG pathway analysis of targetsTo further verify that the biological process related to the target protein is associated with the occurrence of hyperlipidemia, a total of 36 signal pathways were screened using KEGG pathway analysis(Fig. 7), which mainly including starch and sucrose metabolism,galactose metabolism, insulin resistance, insulin signaling pathway,HIF-1 signaling pathway and PI3K-Akt signaling pathway.

Figure 7 KEGG enrichment analysis.

4. Discussion

Brassica rapa has a long planting and application history in China,it is not only a delicious vegetables, but also a health Food with multifaceted biological activities, such as anti-inflammatory, antioxidation, antibacterial, antifatigue, immunomodulatory, anti-tumor and hypolipidemic. Chen, et al. [16] investigated the hypolipidemic effects of total saponins from Brassica rapa. Compared with the control group, the activities of SOD, GSH-Px, LPL and HL in experimental groups were given of total saponins from Brassica rapa were significantly increased, which suggested that total saponins from Brassica rapa had excellent hypolipidemic effects in highfat diet induced obese rats. Ethanol extract from Brassica rapa also had hypolipidemic effects. Pi, et al. [17] showed that the ethanol extract from Brassica rapa significantly reduced TC, TG levels,and increased HDL-C levels in hyperlipidemic rats. This study was evaluated the hypolipidemic activity of Brassica rapa aqueous extract(BRAE) by bile salt-binding capacity and oleic acid-induced HepG2 steatosis cell model. The results showed that BRAE exhibited excellent bile salt-binding capacity in sodium glycylcholate, sodium taurocholate and sodium deoxycholate, and had a significant hypolipidemic activity on steatosis cells, which coincide with reports in the literature .

Network pharmacology is a comprehensive approach based on traditional pharmacology, bioinformatics, chemoinformatics, and network biology, which aims to analyze the mechanism of drugs on diseases[18]. Compared to the traditional drug development model“one drug, one gene, one disease ”, network pharmacology is more suitable for the therapeutic characteristics of “multi-component,multi-target and multi-channel” in Chinese medicine.

We further explore the mechanism of the hypolipidemic activity of Brassica rapa by network pharmacology approach. A total of 21 active components of Brassica rapa were screened out through lipinski rules. By analyzing the “compound-target” network,we concluded that glycosides and phenolic components were considered to play an important role in hypolipidemic effect, and 6-paradol, 6-shogaol, benzyl-beta-D-glucopyranoside, benzylalpha-D-fructofuranoside and liquiritigenin were core components.6-paradol and 6-shogaol are common active components in ginger,and have anti-inflammatory, antitumor, antioxidant, hypolipidemic,and neuroprotective activities. Wu, et al. [14] isolated 6-paradol and 6-shogaol from the fresh root of Brassica rapa by silica gel column and Sephadex LH-20 column separation. 6-paradol is converted from 6-shogaol during samples drying and processing, and they are also the main pungency substances in Brassica rapa[19]. Through cell and animal experiments, Wei, et al. [20] confirmed that 6-paradol and 6-shogaol had hypolipidemic effect, which could increase glucose utilization in 3T3-L1 adipocytes and C2C12 myotube cells by enhancing the phosphorylation of AMPK enzyme.

Based on the PPI protein interaction network, we concluded that VEGFA, IL6, EGFR and PPARG 4 are the key targets of Brassica rapa in hypolipidemic effect. VEGFA is fully known as vascular endothelial growth factor, is an important angiogenic growth factor. It is reported that VEGFA is involved in the development and progression of coronary heart disease, cancer, diabetes,hyperlipidemia and other many diseases, especially closely related to hyperlipidemia. VEGFA can exert hypolipidemic effects by downregulating the expression of endothelial lipase (EL), lipoprotein lipase (LPL), and element binding protein-2 (SREBP-2). In this study, 6 glycosides and 2 flavonoids (liquiritigenin and 4,4`-dihydroxy-3`-methoxychalcone ) in Bracssica rapa interacted with VEGFA targets, which indicated that Bracssica rapa could regulate VEGFA in a variety of ways. IL6 (Interleukin-6) is an important pro-inflammatory factor secreted by macrophages,adipocytes and T cells, which can regulate the progress of cell proliferation, vascular inflammatory and inflammatory reactions [21].Hyperlipidemia patients generally suffer from low-grade chronic systemic inflammation. It is reported that IL6 is closely related to dyslipidemia and plays an important role in regulating blood lipids and body weight. IL6 can regulate blood lipid level by downregulate the expression of TNF-α (tumor necrosis factor-alpha) and reduce insulin resistance. 6-paradol and liquiritigenin in Bracssica rapa can interacted with PPARG targets, which is a key transcription factor in regulating fatty acid metabolism and is clinically used as a target gene for the treatment of fatty acid disorders [23].

This study confirmed that BRAE had excellent hypolipidemic effect in vitro, and can reduce the accumulation of lipid droplets and the levels of triglyceride (TG) and total cholesterol (TC).Based on network pharmacology approach, we preliminarily revealed the multi-component, multi-target, multi-path mechanism in Bracssica rapa treatment of hyperlipidemia. The active components in Bracssica rapa (6-paradol, 6-shogaol, benzyl-beta-Dglucopyranoside, benzyl-alpha-D-fructofuranoside and liquiritigenin) could exert hypolipidemic effects by regulated the core targets(VEGFA, IL6, EGFR and PPARG) and affected multiple signaling pathways (starch and sucrose metabolism, galactose metabolism,insulin resistance, HIF-1, PI3K-Akt). This study provides a scientific foundation for pharmacological mechanisms of Bracssica rapa, but the results are based on the prediction of network pharmacology,which cannot ensure its correctness. Further systematic rigorous biological validation based on this results is also needed.

Authors’ contribution

Jian-bing LIU: Project design, Project administration, Writingoriginal draft, Writing-review & editing

Feng LIN: Project design, Resources, Theoretical guidance

Jun LIN: Implementation section experimental content

Qiu-xia ZHENG: Implementation section experimental content

Han-xing SU: Implementation section experimental content

Li-yun WU: Project design, Theoretical guidance