Selection of solvent for extraction of antioxidant components from Cynanchum auriculatum, Cynanchum bungei, and Cynanchum wilfordii roots

Abstract In east Asia, “Baishouwu” has been used as an herbal drug and functional dietary supplement for hundreds of years. Actually, “Baishouwu” is the common name of the roots of Cynanchum auriculatum, Cynanchum bungei, and Cynanchum wilfordii. In the present study, roots of these three specie were extracted and then fractionated using petroleum ether (PE), dichloromethane (DCM), ethyl acetate (EA), and water. DPPH scavenging experiments revealed high antioxidant activity of DCM and EA fractions of C. bungei and the EA fraction of C. wilfordii. Treatments with these three fractions significantly reduced malondialdehyde content in heat‐stressed Daphnia magna, validating in vivo antioxidant activity. Gas chromatography–mass spectrometer (GC‐MS) analyses demonstrated that the chemical components of fractions extracted from C. bungei, C. bungei, and C. wilfordii were different. Further determination of total phenol and total flavonoids contents showed that DCM and EA fractions of C. bungei and EA fraction of C. wilfordii had much higher contents of total phenol and total flavonoids, which might be the reason to explain their strong antioxidant activity. Overall, the present study suggested that these three plants have different chemical components and biological activities. They could not be used as the same drug.

herbal drugs or sources of modern drugs, but also applied as healthcare food or functional food supplements.
Similarly, the same and novel C21-steroidal glycosides were also detected in C. auriculatum and C. wilfordii (Li et al., 2008;Lu, Xiong, Hong, Guang, & Zhi, 2011;Xiang, Ma, & Hu, 2010;Ye, Wang, Yang, & Zhang, 2013;Zhang, Ye, Shen, & Liang, 2000). Besides, novel compounds, such as Wilfoside K1N inhibiting angiogenesis (Kim et al., 2005), cynandione A protecting from cytotoxicity (Lee et al., 2015), and amino-substituted p-benzoquinone (Yeo & Kim, 1997), were isolated from roots of C. wilfordii. Obviously, due to the different chemical components among C. auriculatum, C. bungei, and C. wilfordii, the healthcare effects of these three species should be different. Misuse of these three plants might reveal no effects or even trigger adverse effects to human body. To develop healthcare food and functional dietary supplements from "Baishouwu," it is necessary to investigate the bioactivity of three species.
The relative antioxidant activity of C. wilfordii, C. auriculatum, and C. bungei has not been compared. Moreover, solvent was proved to significantly affect the effectiveness of antioxidant compound extraction from plants (Spigno, Tramelli, & De Faveri, 2007;Sultana, Anwar, & Ashraf, 2009;Turkmen, Sari, & Velioglu, 2006). Due to the variation of chemical components in C. auriculatum, C. bungei, and C. wilfordii, the best solvent for the extraction of antioxidant compounds should be investigated. These pieces of information would promote the development of antioxidant dietary supplements from "Baishouwu" extracts.
In the present study, to compare the antioxidant activity of three "Baishouwu" species and to search the most efficient solvent for antioxidant compound extraction, roots of C. auriculatum, C. bungei, and C. wilfordii were extracted using ethanol solution and then fractionated using petroleum ether (PE), dichloromethane (DCM), ethyl acetate (EA), and water. The antioxidant activity of these fractions was tested by calculating the scavenging rate of 1,1-diphenyl-2-picrylhydrazyl (DPPH) radicals and monitoring changes of antioxidant indices is heatstressed Daphnia magna. Gas chromatography-mass spectrometer (GC-MS) was applied to analyze the chemical components of these fractions, and contents of total phenol and total flavonoids were determined. These results might be useful for the development of functional dietary supplements from "Baishouwu."

| Materials
Dry roots of C. auriculatum, C. bungei (produced in Binhai, Jiangsu Province, China), and C. wilfordii (produced in Tai'an, Shandong Province, China) were purchased from the local planting bases. To identify species, genomic DNA was extracted from these samples using BioFastSpin plant genomic DNA extraction kit (Bioer, Hangzhou, China). Ribosomal DNA internal transcribed spacer (ITS) seuqences were amplified using BioReady Hot Start Taq (Bioer, Hangzhou, China) and primers of 5'-GTCGAATTCGTAGGTGAACCTGCGGAAGGATC A-3' and 5'-CCTGCAGTCGACAKATGCTTAARTTCAGCRGG-3', sequenced and then blasted against Genbank database. The final results confirmed the species of these three samples. The roots were dried at 60°C for 24 hr to completely remove moisture and then sealed in plastic bags before analysis. All reagents used in the present study were HPLC grade.

| Chemical extraction
All the three samples were ground into powers using an electronic grinder. Then, approximately 1 kg of each species was weighed. For chemical extraction, samples were soaked in 4 L of 80% ethanol for 1 hr with sonication. After centrifugation at 5,980 g for 5 min, the supernatant was collected and the pallets were further extracted in 4 L of 60% ethanol for 1 hr with sonication. This step using 60% ethanol was repeated once. All the supernatants were pooled, filtered through a filter paper, and concentrated by rotary evaporation (Heidolph Hei-VAP, Germany).
The extracts were dispersed in 300 ml of Milli-Q water and sequentially extracted using 400 ml of petroleum ether (PE), 400 ml of dichloromethane (DCM), and 400 ml of ethyl acetate (EA). Extraction with each solvent was repeated for three times, and the two layers were separated using separating funnels. Extracts obtained by the same solvent were pooled and dried by rotary evaporation (Heidolph Hei-VAP). Finally, four factions were collected, including PE fraction, DCM fraction, EA fraction, and water fraction. All fractions were dried, precisely weighed, and then dissolved in DMSO to prepare 50 mg/ml stock solution.

| Determination of DPPH radical scavenging activity
The DPPH radical scavenging activity was determined following Wong, Chai, and Hoo (2012) with modifications. To prepare DPPH solution, 8 mg of DPPH was dissolved in 50 ml of methanol and then further diluted by adding 50 ml of Milli-Q water. The butyl hydroxy anisd (BHA) solution was prepared by dissolving 1 mg of BHA in 40 μl of methanol. These two solutions were freshly prepared just before use.
The tests of DPPH radical scavenging activity were performed in 24-well plates. In each well, 1 ml of DPPH solution and 1 μl of sample solution were mixed and then placed at room temperature in dark for 30 min. The final concentrations of extracts included 50, 25, 20, 12.5, 10, and 6.25 mg/L by diluting the stock solution using DMSO. Next, the absorbance at 517 nm was determined using a microplate reader (FlexStation3, Molecular Devices, USA). The sample solution was replaced by DMSO as the negative control. BHA solution was used instead of sample solution as the positive control. The rate of scavenging DPPH radicals was calculated using the following formula.
where A s represents the absorbance of reaction containing sample solution, and A 0 is the absorbance in the negative control.
Each test was repeated three times and the EC 50 was calculated using the Probit method in SPSS 23.

| Effects of extracts on antioxidant indices in Daphnia magna
To validate antioxidant activity detected by DPPH radical scavenging experiments, D. magna individuals were treated with heat shock and extract fractions. Changes of antioxidant indices were compared. Briefly, D. magna was maintained at 20°C using EPA medium and fed 5.0 × 10 5 cells/ml Scenedesmus obliquus. Experiments were carried out in 50-ml glass beakers. In each beaker, 150 individuals were placed and then treated with 10 mg/L EA and DCM fraction of C. bungei extract and 10 mg/L EA fraction of C. wilfordii extract.
The culture temperature increased from 20 to 30°C to trigger heat stress. After incubation at 30°C for 24 hr, all animals were harvested to detect activities of superoxide dismutase (SOD, Dong, Zhao, Huang, & Wen, 2011), peroxidase (POD, Upadhyaya, Sankhla, Davis, Sankhla, & Smith, 1985), catalase (CAT, Johansson & Borg, 1988), and content of malondialdehyde (MDA, Draper & Hadley, 1990) as described previously. Concentration of total protein was assayed using a Bradford protein assay kit (Tiangen, Beijing, China). One unit of SOD activity was defined as the amount of enzyme required for 1 mg of tissue proteins in 1 ml of reaction mixture to achieve SOD inhibition rate of 50%. One unit of CAT activity was defined as 1 mg of tissue proteins consumed 1 μmol H 2 O 2 in 1 s. One unit of POD activity was defined as increase in absorbance at 420 nm per 1 mg of tissue proteins in 1 min at 37°C. Enzyme activities were expressed as U/mg proteins, and content of MDA was expressed as nmol/mg proteins. Each assay was repeated three times.
Helium was used as the carrier gas with the flow rate of 1.5 ml/min, and the temperature of the injection port and the detector was set at 220°C. All data were obtained by collecting the full-scan mass spectra. The spectra information was aligned with the database to identify chemicals.

| Determination of total phenol (TP)
Contents of TP were determined using a plant total phenol test kit (Solarbio, Beijing, China) following the manufacturer's protocol.
Briefly, 50 μl of sample solution was mixed with 250 μl of Folin-Ciocalteu reagent for 2 min at room temperature and then was neutralized using 250 μl of sodium carbonate solution (7.5%, w/v). The reaction mixture was incubated at room temperature for 10 min with DPPH radical scavenging rate = [(A 0 − A s )∕A 0 ] × 100% F I G U R E 1 DPPH radical scavenging activities of dichloromethane (DCM) and ethyl acetate (EA) fractions of Cynanchum bungei and C. wilfordii (mean ± SD) intermittent shaking for color development. Afterward, the reaction solution was diluted with 450 μl of distilled water and absorbance at 760 nm was determined using a UV2800 UV-Vis spectrophotometer (Unico, Shanghai, China). Gallic acid was used to prepare the standard curve. Each assay was repeated three times.

| Determination of total flavonoids (TF)
Contents of TF were determined using a plant total flavonoids test kit (Solarbio, Beijing, China) following the manufacturer's protocol.
Briefly, 0.2 ml of sample solution was reacted with 0.3 ml of ethanol, 0.05 ml of 10% aluminum chloride, 0.05 ml of 1 M potassium acetate, and 0.4 ml of distilled water at 37°C for 45 min. Absorbance of the reaction mixture was measured at 450 nm using a UV2800 UV-Vis spectrophotometer (Unico, Shanghai, China). Gallic acid was used to prepare the standard curve. Each assay was repeated three times.

| RE SULTS AND D ISCUSS I ON
The antioxidant activity of PE, DCM, EA, and water fractions extracted from C. auriculatum, C. bungei, and C. wilfordii was tested using the DPPH radical scavenging tests. At the concentration of 50 mg/L, all fractions of C. auriculatum, PE and water fractions of C. bungei as well as PE, DCM, and water fractions of C. wilfordii did not reveal obvious DPPH radical scavenging activity. DCM and EA fractions of C. bungei as well as EA fraction of C. wilfordii showed significantly lower absorbance at 517 nm in comparison to the blank control (DMSO), thus these three fractions were considered as antioxidant fractions.
The fractions with preliminary antioxidant activity were further tested with more concentrations. In the range of tested concentrations (6.25-50 mg/L), all treatments significantly scavenged DPPH radicals. Along with increasing concentration of extracts, the scavenging rate of EA fraction of C. bungei increased significantly.
The scavenging rate of DCM fraction of C. bungei and EA fraction of C. wifordii also increased with elevating concentration. However, when the concentration arrived 20 mg/L, the scavenging rate of DPPH radicals became stable (Figure 1) (Table 1).
Heat stress effectively induced reactive oxygen species (ROS) in D. magna (Becker, Brinkmann, Zeis, & Paul, 2012;Klumpen et al., 2017) and peroxidation of membrane lipids, resulting in elevation of MDA content. Under stress environments, D. magna generally increases activities of SOD, POD, and CAT to resist harmful effects of ROS (Kim, Kim, Kim, Won, & Lee, 2018). Treatment with C. bungei extract decreased MDA content and increased SOD activity in mouse brain, displaying antioxidant and neuroprotective effects (Chen et al., 2017). In the present study, partially similar results were displayed. Compared with the control, treatments with EA, DCM fractions of C. bungei and EA fraction of C. wilfordii significantly decreased MDA content but did not affect activities of CAT, SOD, and POD (Table 2), suggesting that heat stress-induced membrane lipid peroxidation might be directly suppressed by these fractions, which was consistent with the results of DPPH assays. Overall, these results suggested that C. bungei and C. wilfordii had antioxidant activity and might be applied as healthcare food and functional food supplements. For manufacturing the functional food supplements, C. bungei could be extracted using DCM and/or EA, but only EA could be used to extract the antioxidant components from C. wilfordii. Zhang, Tao, Cui, and Ao (2006) optimized the extraction procedure of antioxidant compounds from C. bungei and found that 75% ethanol solution was the best solvent. The present study further TA B L E 1 EC 50 values of DPPH radical scavenging activity of C. auriculatum, C. bungei, and C. wilfordii extracts  Only triphenyl phosphate and 9-octadecenoic acid (Z)-, methyl ester were detected in EA fractions of both C. bungei and C. wilfordii but not in C. auriculatum. However, it has been reported that triphenyl phosphate was toxic to D. magna and zebrafish (Isales et al., 2015;Lin, 2008), which should not be the reason for the antioxidant activity in EA fractions of C. bungei and C. wilfordii. 9-octadecenoic acid (Z)-, methyl ester is easy to be autoxidized (Deatherage & Mattill, 1939). It was possible that this compound absorbed radicals produced by DPPH and appeared as a possible reason for the anti-  (Table 3).
Overall, these results demonstrated that DCM and EA fractions of C. bungei and EA fraction of C. wilfordii showed significant in vitro and in vivo antioxidant activity probably through phenol and/or flavonoids compounds.

ACK N OWLED G M ENTS
We thank the Shenzhen Nobel Science and Technology Service

CO N FLI C T O F I NTE R E S T
The authors declare that they do not have any conflict of interest.

E TH I C A L S TATEM ENTS
This study does not involve any human or mammal testing.
Experiments on Daphnia magna did not require any approval in P.