Skip to main content
NCBI home page
Journal of Zhejiang University. Science. B logoLink to Journal of Zhejiang University. Science. B
. 2017 Sep;18(9):757–769. doi: 10.1631/jzus.B1600158

Toxicity comparison of different active fractions extracted from radix Sophorae tonkinensis in zebrafish*

Hong-cui Liu 1,2, Xiao-yu Zhu 2, Jiang-hua Chen 1, Sheng-ya Guo 2, Chun-qi Li 2,†,‡, Zhong-ping Deng 1,†,‡
PMCID: PMC5611547

Abstract

Radix Sophorae tonkinensis (RST) is a widely used herb in Traditional Chinese Medicine (TCM) for treating infectious and inflammatory diseases. However, the toxicity data for RST are limited. The aim of this work is to assess and compare the toxicity of the whole RST extract and its five active fractions using the zebrafish model. Five active fractions of RST were prepared using five different types of solvents, which included dealkalized water, ethanol, n-butyl ethanol, dichloromethane, and diethyl ether. The chemical profiles of the active fractions were determined by high-performance liquid chromatography (HPLC), and the toxicity observed in the zebrafish model was confirmed using mouse models. In the zebrafish model, cardiovascular toxicity was observed for the fraction extracted using diethyl ether, and hepatotoxicity was observed for the whole RST extract and the fractions extracted using water and ethanol, whereas both cardiovascular and hepatic toxicities were observed for the fractions extracted using n-butyl ethanol and dichloromethane. The hepatotoxicity of the fractions extracted using n-butyl ethanol and dichloromethane was also observed in mice. Our findings provide the toxicity data for RST and its five active fractions through modeling in a zebrafish, and indicate that the different fractions may each have a different toxicity, which is helpful for the optimal use of RST in clinical practice.

Keywords: Sophorae tonkinensis Gagnep; Active fraction; Hepatotoxicity; Cardiovascular toxicity; Zebrafish

1. Introduction

The dried roots and rhizomes of Sophorae tonkinensis Gagnep, commonly known as radix Sophorae tonkinensis (RST) or “Shandougen” in Chinese, have been used for hundreds of years as a Traditional Chinese Medicine (TCM) for treating acute pharyngolaryngeal infections and sore throats (Tang et al., 2013; CPC, 2015). Previous phytochemical studies of this plant have revealed that quinolizidine alkaloids, flavonoids, and triterpenoids are present as its major constituents (Xiao et al., 1999; Ding and Chen, 2006; Ding et al., 2006; Li et al., 2008a; 2008b; He C.M. et al., 2013; Pan et al., 2015). In the past few years, much more attention has been paid to the analysis of alkaloids in RST, including matrine, oxymatrine, and sophocarpine, due to their various pharmacological activities, which include ameliorating throat inflammation (Li et al., 2012; Yoo et al., 2014; Lee et al., 2015) and suppressing in vitro cancer cell proliferation (Chui et al., 2005; Liu et al., 2006) as well as hepatoprotective effects (Cho et al., 1986; Long et al., 2004; Chai et al., 2012). Unfortunately, some of the alkaloids extracted from RST have been shown to be highly toxic to humans and livestock in ways that damage the nervous system, digestive system, and respiratory system, affecting cardiovascular and liver functions. The frequency of side effects and poisoning from RST including hepatotoxicity and cardiovascular toxicity has increased (Sun et al., 2010; Li et al., 2011; Zhang and Ding, 2013; CPC, 2015) and those toxicities have limited the application of RST and reduced its medical and economic value.

Most medicinal herbs including RST contain different compounds that buffer, modulate, and modify the effects of any “active principles.” The accumulated data suggest that the effects produced by whole plant extracts cannot be mimicked by administering isolated purified constituents of the plant. Herbs generally exert broad actions on a number of whole physiological systems at the same time and these actions are usually oriented in the same general therapeutic direction. The toxicity of Chinese herbs can generally be reduced through traditional processing and the addition of antidote herbs. Recently, the active fractions of Chinese herbs extracted through special solvents have been approved as effective in reducing or removing the toxicity and/or improving the efficacy of the herbs. Matrine and oxymatrine are the major components responsible for the toxicity of RST. There is a certain relationship between the matrine and oxymatrine content and the acute toxicity of the different components of RST, but this relationship is not entirely uniform and consistent (Sun et al., 2010).

In the present study, in combination with conventional TCM extraction technologies, RST extract (RSTE) and the five RST active fractions were prepared using five different types of solvents including ethanol, diethyl ether, dichloromethane, n-butyl ethanol, and dealkalized water. Zebrafish were exposed to each individual sample and their in vivo toxicity was assessed and compared.

2. Materials and methods

2.1. RST and chemicals

The dried roots of RST (lot#: 1207070) were purchased from Sichuan Neautus Traditional Chinese Medicine Co., Ltd., China. Absolute methanol (lot#: 20130204), dichloromethane (lot#: 20120810), diethyl ether (lot#: 20100419), n-butyl ethanol (lot#: T20110518), hydrogen chloride (lot#: T20090312), and anhydrous sodium carbonate (lot#: F20100810) were bought from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China); sophocarpidine (lot#: 01-2002) was from the Research Center of Shanghai Chinese Traditional Medicine; standard oxymatrine (lot#: 091110) was from Shanghai Winherb Medical Technology Co., Ltd. (Shanghai, China); and trifluoroacetic acid (lot#: FC160) and acetonitrile (lot#: I631730212) were from CNW Technologies GmbH (Düsseldorf, Germany) and Merck & Co. (New Jersey, USA), respectively. All chemicals used in this study were of analytical grade. Ultra-pure water (resistance >18 mΩ) was prepared by Millipore Milli-Q purification system (Bedford, USA).

2.2. Zebrafish and mouse handling

Adult AB strain zebrafish were housed in a Hunter Biotechnology, Inc., zebrafish facility and fed with live brine shrimp and dry flake as recommended by Westerfield (1995). The zebrafish were paired for natural mating and 200–300 embryos on average were generated per pair each time. Embryos were maintained at 28 °C in fish water (0.2% (2 g/L) Instant Ocean salt in deionized water, pH 6.9–7.2, conductivity 480–510 μS/cm, and hardness 53.7–71.6 mg/L CaCO3) and washed and staged at 6 and 24 h post fertilization (Kimmel et al., 1995). The Hunter Biotechnology, Inc., zebrafish facility is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) International.

Kunming mice (20–22 g, 4–6 weeks old) were purchased from Shanghai Slac Laboratory Animal Co., Ltd. (certification No.: SCXK (Shanghai) 2012-0002) and housed in large polypropylene cages in a temperature-controlled room ((22±2) °C) and fed with standardized pelleted food (TANUVAS) and clean drinking water ad libitum.

2.3. Preparation of RST extract and active fractions

The RST was weighed and boiled with six times the amount of water for one hour, and the extract was filtered with filter paper to remove suspensions and the extraction was repeated one more time. The filtrates were combined, evaporated under a rotatory evaporator, and lyophilized through quick-freezing to obtain the RSTE.

An amount of RSTE was dissolved in water and separated with 75% ethanol. The precipitate was dried to obtain the ethanol sedimentation fraction from the RST. The supernatant was evaporated under a rotatory evaporator, dissolved in 2% HCl, and extracted four times with a known volume of diethyl ether. The diethyl ether solvent was then removed by evaporation to yield a dried diethyl ether fraction of the RST. The residue was fractionated by a series of solvents, including dichloromethane (dichloromethane fraction), n-butyl ethanol (n-butyl ethanol fraction), and sodium bicarbonate, and a remainder of dealkalized water fraction. All the active fractions were placed in airtight containers for future use.

2.4. Treatment of RSTE and active fractions

Stock solutions of RSTE and RST active fractions, including dealkalized water extract, ethanol sedimentation extract, n-butyl ethanol extract, dichloromethane extract, and diethyl ether extract were prepared in fish water and serial dilutions were made before each experiment. The zebrafish were placed into six-well plates (Nest Biotech, Shanghai, China), 30 zebrafish per well in 3 ml of RSTE solution at designated concentrations for the treatment period (Westerfield, 1995; McGrath and Li, 2008). Fish-water-treated zebrafish served as the control. To protect the RST solutions from light-induced decomposition, all experiments were performed at a constant temperature (28 °C) in the dark. After treatment, the zebrafish were visually observed and photographed with a dissecting stereomicroscope (Olympus Co., Tokyo, Japan) equipped with a high-speed video camera (JVC, Japan). Images were quantified using image-based morphometric analysis (NIS-Elements D 3.1, Japan). At the end of the experiments, all the zebrafish were sacrificed with 0.25 g/L tricaine methanesulfonate, which conforms to the American Veterinary Medical Association (AVMA) requirements for euthanasia by anesthetic (Shen et al., 2015).

2.5. Determination of maximum non-lethal concentration and 10% lethal concentration

To determine the maximum non-lethal concentration (MNLC) and 10% lethal concentration (LC10) of RSTE and RST active fractions for toxic target organ identification and cardiovascular toxicity and hepatotoxicity assessment, 30 zebrafish per condition were exposed to RSTE or the RST active fractions for 3 d from 2 to 5 d post fertilization (dpf) (for toxic target organ identification), for 1 d from 2 to 3 dpf (for cardiovascular toxicity assessment), and for 2 d from 3 to 5 dpf (for hepatotoxicity assessment), respectively, following the standard procedures we reported earlier (He J.H. et al., 2013; Zhu et al., 2014; Shen et al., 2015). In the initial experiments, five testing concentrations ranging from 0.1, 1, 10, 100, and 1000 µg/ml were used for each sample and mortality was recorded daily. Zebrafish that lacked an observable heartbeat under a dissecting stereomicroscope were counted as dead zebrafish. If MNLC and LC10 were not estimated from the initial tests, additional testing concentrations up to 10 000 µg/ml and down to 0.01 µg/ml were tested. Zebrafish mortality curves were generated using Origin 8.0 (OriginLab, USA) and MNLC and LC10 were calculated from this curve (Shen et al., 2015).

2.6. Target organ identification

To identify the target organs of toxicity induced by RSTE and RST active fractions, 30 zebrafish per condition were treated with each testing sample from 2 to 5 dpf at the nominal concentrations of 1/10 MNLC, 1/3 MNLC, MNLC, and LC10 as summarized in Table 1. After treatment, 15 zebrafish were randomly selected from each group for visual observation and photographs. The major organs and tissues of the zebrafish were visually assessed and toxic target organs were identified based on morphological changes and morphometric analyses.

Table 1.

Tested concentrations used for RST toxicity studies (n=10)

Toxicity study RST component Concentration (µg/ml)
1/10 MNLC 1/3 MNLC MNLC LC10
Target toxicity RSTE 300.0 1000.0 3000.0 4323.4
Dealkalized water extract 71.0 236.7 710.2 820.9
Ethanol sedimentation extract 72.3 240.8 722.5 907.0
n-Butyl ethanol extract 35.8 119.2 357.5 380.5
Dichloromethane extract 14.0 46.7 140.0 150.0
Diethyl ether extract 6.8 22.5 67.5 68.3
Cardiovascular toxicity n-Butyl ethanol extract 47.9 159.6 478.9 538.3
Diethyl ether extract 8.3 27.6 82.9 93.6
Dichloromethane extract 45.0 150.0 450.0
Hepatotoxicity RSTE 543.2 1810.7 5432.0 7510.8
Dealkalized water extract 96.0 319.9 959.7 1009.1
Ethanol sedimentation extract 340.9 1136.2 3408.5 4367.6
n-Butyl ethanol extract 70.0 262.2 700.0
Dichloromethane extract 340.9 1136.2 3408.5 4367.6
Open in a new tab

2.7. Cardiovascular toxicity assessment

To confirm cardiovascular toxicity found in the target organ toxicity identification study, 30 zebrafish per condition were treated with

n-butyl ethanol extract, diethyl ether extract, or dichloromethane extract from 2 to 3 dpf at the nominal concentrations of 1/10 MNLC, 1/3 MNLC, MNLC, and LC10 as summarized in Table 1. At the end of the experiment, 15 zebrafish were randomly selected from each group for visual observation and image acquisition. Six specific phenotypic endpoints were validated and used in the cardiovascular toxicity analyses: heart rate, heart rhythm, pericardial edema, circulation, hemorrhage, and thrombosis as described in our earlier report (Zhu et al., 2014).

To confirm the cardiovascular toxicity induced by dichloromethane extract, a zebrafish histopathology was performed. After treatment, the zebrafish were fixed and dehydrated before paraffin embedding. The embedded zebrafish were longitudinally sectioned at 5 µm and stained with hematoxylin and eosin (H&E). Histological images were obtained using a histological microscope (Leica, Germany) and a pathological diagnosis was performed by a certified pathologist (Zhu et al., 2016).

2.8. Hepatotoxicity assessment

The hepatotoxicity observed in the target organ toxicity identification study was further assessed in zebrafish treated with RSTE, dealkalized water extract, ethanol sedimentation extract, dichloromethane extract, or n-butyl ethanol extract. Thirty zebrafish per condition were treated with a testing concentration from 3 to 5 dpf at the nominal concentrations of 1/10 MNLC, 1/3 MNLC, MNLC, and LC10 as summarized in Table 1. After treatment, 15 zebrafish from each group were randomly selected for visual observation and quantitative image acquisition performed using image-based morphometric analysis (NIS-Elements D 3.1; Japan). Three specific phenotypic endpoints (liver size, liver degeneration, and yolk sac retention) were used for assessing hepatotoxicity as reported by our group (He J.H. et al., 2013; Shen et al., 2015). Liver size, liver degeneration, and yolk sac retention were calculated based on the formulas below: liver relative size (% of control)=liver area (sample)/liver area (control)×100%; liver degeneration (%)=[1−liver brightness (sample)/liver brightness (control)]×100%; yolk sac retention (% of control)=yolk sac area (sample)/yolk sac area (control)×100%.

2.9. Mouse study

The mice were divided into three groups of 12 mice each (half male and female). Dichloromethane extract and diethyl ether extract of RST (8 g/kg body weight) were administered to the mice via oral administration. Group I served as the control group, and received deionized water only; Group II and Group Ш received dichloromethane extract and diethyl ether extract once a day, respectively, for seven consecutive days. All the mice were sacrificed at the end of the eighth day after drug administration, and blood was drawn from the eyeball of the mouse and the serum was separated for liver function assays. Total bile acid (TBA), total bilirubin (TBil), alanine aminotransferase (ALT), aspartate aminotransferase (AST), and alkaline phosphatase (ALP) in the blood serum were quantified by an autoanalyzer (Shimadzu CL-7080, Shimadzu, Japan). The mouse livers were immediately removed, sliced, and washed in saline. Liver pieces were fixed in 10% formalin, embedded in paraffin wax, sectioned, and stained with H&E for histopathological analysis.

2.10. Fingerprint analysis of the dichloromethane fraction and the diethyl ether fractions

To establish the fingerprints, 10 mg of dichloromethane fraction or diethyl ether extract was weighed into a 10-ml polypropylene centrifugal tube, and 2 ml of water was added for sonication, filtered with a 0.22-μm Millipore filter, and then a total volume of 10 μl was loaded into the high-performance liquid chromatography (HPLC) column. The analyses were performed on a Waters Alliance e2695 HPLC (Waters Corporation, Milford, USA) coupled with vacuum degasser, quaternary pump, column compartment, and diode-array detector. The instrument was fitted with a Waters XBridge C18 Column (250 mm×4.6 mm, 5 μm), operated at 30 °C. HPLC analysis was performed using a gradient method. Gradient analysis was performed using a mobile phase A (acetonitrile) and B (aqueous solution of 0.1% (1 g/L) trifluoroacetic acid). The gradient profile was as follows: 5% A and 95% B were held for 20 min, followed by a decrease of 70% B over 15 min and 65% B for 15 min, and then 5% A and 95% B were held for a further 10 min. The flow rate used was 1.0 ml/min and chromatograms were collected at 210 nm. The similarity evaluation of the different fractions was performed using the Similarity Evaluation System for Chromatographic Fingerprint of TCM (Version 2004A) software, published by the National Pharmacopoeia Committee of China.

2.11. Statistical analysis

Sigmoidal regression for concentration-response curves was used for estimating MNLC and LC10 (Origin 8.0) and one-way analysis of variance (ANOVA) followed by Dunnett’s test was conducted for comparing differences between the groups. All statistical analyses were performed on SPSS Version 16.0 (SPSS Inc., Chicago, USA) and P<0.05 was considered statistically significant. Quantitative data were expressed as mean±standard error (SE) and all experiments were repeated at least three times to confirm the reproducibility of all data (Shen et al., 2015).

2.12. Quality control standard

According to our laboratory quality control standard, successful experiments must meet these milestones: (1) zebrafish natural death in untreated (control) groups was ≤10%; and (2) intraplate and interplate coefficients of variation (CV) were ≤25% (Zhou et al., 2014).

3. Results

3.1. Toxic target organs

We assessed toxicity and identified toxic target organs for RSTE and RST active fractions in zebrafish, concentration-dependent mortality was demonstrated, and the MNLC and LC10 are summarized in Table 1. Pericardial edema and/or reduced heart rates were observable in the zebrafish treated with dichloromethane extract, n-butyl ethanol extract, and diethyl ether extract in a dose-dependent manner, but not in zebrafish treated with RSTE, dealkalized water extract, or ethanol sedimentation extract. Liver degeneration was found in zebrafish treated with RSTE, dealkalized water extract, ethanol sedimentation extract, dichloromethane extract, and n-butyl ethanol extract, but not in zebrafish treated with diethyl ether extract. These findings suggest that RST diethyl ether extract induced cardiovascular toxicity, RSTE, RST dealkalized water extract and ethanol sedimentation extract induced hepatotoxicity, whereas RST n-butyl ethanol extract and dichloromethane extract induced both cardiovascular and liver injuries.

3.2. Cardiovascular toxicity

To confirm and characterize cardiovascular toxicity induced by RST n-butyl ethanol extract, diethyl ether extract, and dichloromethane extract, these RST active fractions were exposed to zebrafish at 2 dpf and the atrial and ventricular rates were counted at 4 h after treatment and the systematic cardiovascular toxicity was assessed after 24 h exposure. As summarized in Table 2, n-butyl ethanol extract, dichloromethane extract, and diethyl ether extract induced pericardial edema (Fig. 1a) and slowed the blood circulation. As indicated in Fig. 1b, n-butyl ethanol extract did not affect heart rate, whereas diethyl ether extract and dichloromethane extract treatment resulted in slower heartbeats in a dose-dependent manner. Statistically significant bradycardia was found in zebrafish treated with diethyl ether extract or dichloromethane extract at 1/3 MNLC and the above concentrations (P<0.05 or P<0.001). In addition, dichloromethane extract caused zebrafish atrioventricular conduction block at MNLC. No hemorrhage or thrombosis was found from any treatment. Cardiovascular toxicity was confirmed by histopathological examination in zebrafish treated with dichloromethane extract. When compared with a normal zebrafish heart (Fig. 2a), pericardial edema, a misshaped atrium and ventricle as well as reduced number of endothelial cells and cardiomyocytes were seen in the zebrafish treated with dichloromethane extract (Fig. 2b). The overall ranking of cardiovascular toxicity of RST active fractions was dichloromethane extract>diethyl ether extract>n-butyl ethanol extract.

Table 2.

RST active fractions-induced cardiovascular toxicity (n=10)

RST active fraction Pericardial edema (% of control) Slow circulation (% of control)
 n-Butyl ethanol extract 10% at MNLC & LC10 40% at MNLC
 Diethyl ether extract 80% at MNLC; 90% at LC10 100% at MNLC & LC10
 Dichloromethane extract 30% at MNLC 100% at MNLC
Open in a new tab

Fig. 1.

Fig. 1

Open in a new tab

Cardiovascular toxicity induced by RST active fractions in zebrafish

(a) Pericardial edema: circled area is a zebrafish heart or pericardial edema of untreated control zebrafish and zebrafish treated with n-butyl ethanol extract, dichloromethane extract, and diethyl ether extract. (b) Zebrafish atrial and ventricle heart rate changes as compared with the untreated control zebrafish. All the values represent the means±SE (n=10). * P<0.05 and *** P<0.001 vs. control

Fig. 2.

Fig. 2

Open in a new tab

Histopathological analysis of zebrafish hearts

(a) Untreated larval zebrafish exhibited a normal heart morphology and histology. (b) Larval zebrafish treated with dichloromethane extract showed pericardial edema, misshaped atrium and ventricle as well as reduced number of endothelial cells and cardiomyocytes. A: atrium; V: ventricle

3.3. Hepatotoxicity

In the acute toxicity and target organ toxicity study above, we found that RSTE and four of the RST active fractions (dealkalized water extract, ethanol sedimentation extract, dichloromethane extract, and n-butyl ethanol extract) induced zebrafish hepatotoxicity. Here we systematically assessed and characterized the liver toxicity of these RSTEs. Untreated zebrafish exhibited clear liver tissue and were perfused with circulating blood cells. After treatment with RSTE or one of these RST active fractions, the zebrafish liver lost transparency and became dark or brown and liver blood flow was no longer observable. All tested RST active fractions except the ethanol sedimentation extract also induced zebrafish yolk retention and showed an obvious dosage–toxicity relationship (Fig. 3).

Fig. 3.

Fig. 3

Open in a new tab

Visual phenotype of hepatotoxicity in zebrafish at 5 dpf after exposure to RSTE or RST active fractions for 2 d

Untreated larval zebrafish exhibited a clear, healthy liver (a). Larval zebrafish treated with RSTE (b), RST dealkalized water extract (c), ethanol sedimentation extract (d), n-butyl ethanol extract (e), and dichloromethane extract (f) exhibited liver tissue degeneration, hepatatrophia, and yolk sac retention. L: liver; Y: yolk sac

We quantified the liver size, liver degeneration, and yolk sac retention in zebrafish treated with RSTE and RST active fractions using an image-based morphometric analysis. As indicated in Table 3 and Fig. 4, the relative liver areas of zebrafish treated with RSTE, dealkalized water extract, ethanol sedimentation extract, dichloromethane extract, and n-butyl ethanol extract were 95.8%–96.8%, 98.1%–103.9%, 97.7%–104.8%, 97.1%–100.8%, and 89.9%–102.4% of the untreated control zebrafish, respectively, and a statistically significant reduced liver size (P<0.01) was observed only in the zebrafish treated with RSTE at a tested concentration of LC10. The relative liver brightness of zebrafish treated with RSTE, dealkalized water extract, ethanol sedimentation extract, dichloromethane extract, and n-butyl ethanol extract was 7.8%–30.1%, 3.4%–17.8%, 2.0%–16.9%, 1.7%–25.2%, and −1.2%–20.8% of the untreated control zebrafish, respectively; liver degeneration was demonstrated in zebrafish treated with RSTE, ethanol sedimentation extract, and n-butyl ethanol extract at 1/3 MNLC and higher concentrations, and with dealkalized water extract and dichloromethane extract at MNLC and LC10 concentrations (P<0.05, P<0.01, or P<0.001). The relative yolk sac areas were 44%–538%, 80%–345%, 96%–196%, 127%–834%, and 124%–689% of the untreated control zebrafish, respectively, in zebrafish treated with RSTE, dealkalized water extract, ethanol sedimentation extract, dichloromethane extract, and n-butyl ethanol extract. Statistically significant delayed yolk sac absorption was found in zebrafish treated with n-butyl ethanol extract at MNLC (P<0.001), and with RSTE, dealkalized water extract, and dichloromethane extract at MNLC and LC10 (Fig. 4). Based on the liver degeneration at MNLC, the hepatotoxic severity was RSTE>dichloromethane extract>n-butyl ethanol extract>ethanol sedimentation extract>dealkalized water extract.

Table 3.

RSTE and RST active fractions-induced liver toxicity (n=10)

RST active fraction Liver size change (% of control) Liver degeneration (% of control) Yolk sac retention (% of control)
RSTE 88.6% at LC10 9.4% at 1/3 MNLC 24.3% at MNLC 30.1% at LC10 516% at MNLC 538% at LC10
Dealkalized water extract No 9.5% at MNLC 17.8% at LC10 210% at MNLC 345% at LC10
Ethanol sedimentation extract No 8.0% at 1/3 MNLC 11.2% at MNLC 16.9% at LC10 No
Dichloromethane extract No 23.8% at MNLC 25.2% at LC10 573% at MNLC 834% at LC10
n-Butyl ethanol extract No 11.7% at 1/3 MNLC 20.8% at MNLC 689% at MNLC
Open in a new tab

Fig. 4.

Fig. 4

Open in a new tab

Zebrafish liver size change, liver degeneration, and yolk sac retention after exposure to RSTE and RST active fractions

(a) Liver size was significantly reduced in zebrafish treated with RSTE (*** P<0.001 vs. control). (b, c) Liver degeneration and delayed yolk sac absorption were found in zebrafish treated with RSTE and four RST active fractions (* P<0.05, ** P<0.01, *** P<0.001 vs. control). All the values are presented as mean±SE (n=10)

To confirm the hepatotoxicity in zebrafish induced by RST dichloromethane extract but not liver damage in zebrafish treated with RST diethyl ether extract, we performed further experiments in mice by analyzing liver functions and liver pathology. Various serum biochemical biomarkers for liver functions in control mice and mice administered either dichloromethane extract or diethyl ether extract are represented in Table 4. Compared with the untreated control mice, ALT activity was significantly increased (P<0.01), whereas TBiL level was statistically decreased in mice administered dichloromethane extract. No liver function changes were found in mice administered diethyl ether extract (P>0.05). The histological observations consistently supported the results obtained from serum enzyme assays. Gross necrosis, massive fatty degeneration, broad lymphocyte, and Kupffer cell infiltration around the central vein, and cellular boundary loss were observed in mice administered dichloromethane extract. No abnormal histopathology was found in the diethyl ether extract group (Fig. 5).

Table 4.

Liver functions of mice administered the RST active fractions (n=12)

 Group ALT (IU/L) AST (IU/L) TBiL (μmol/L) ALP (IU/L) TBA (μmol/L)
Control 22.5±5.27 102.8±21.78 1.4±0.47 529.1±203.10 2.1±1.23
Diethyl ether extract 27.1±11.28 102.1±24.46 1.3±0.39 525.3±141.25 1.7±1.04
Dichloromethane extract 42.8±45.66* 105.8±23.85 1.0±0.32* 505.6±143.04 1.7±0.91
Open in a new tab

All values are expressed as mean±SE (n=12).

*

P<0.01, compared with untreated control mice

Fig. 5.

Fig. 5

Open in a new tab

Histopathology of liver tissue section

(a) The central vein surrounded by a hepatic cord of cells from the untreated control mice; (b) Patches of liver cell necrosis with inflammatory cell infiltration around the central vein and focal necrosis with sinusoidal dilatation in mice treated with the dichloromethane extract group; (c) A normal liver histology from mice treated with diethyl ether extract

3.4. Fingerprint analyses of RST dichloromethane extract and diethyl ether extract

The fingerprint of the chemical constituents from active fractions of dichloromethane and diethyl ether was determined by HPLC. Under optimum chromatographic conditions, the major components of each active fraction were identified and a well-resolved baseline separation was obtained. The main peaks existing in all of the sample profiles of the same fraction were selected as the common peaks. Quantitative HPLC fingerprint was calculated based on the relative retention time and relative peak area with the reference peak of these common peaks. The similarities among samples were analyzed using the Similarity Evaluation System for Chromatographic Fingerprint of TCM (Version 2004A). The results are shown in Fig. 6, where there are four representative chromatograms for each fraction, of which S1, S2, and S3 represent the profiles of three batches of active fractions, respectively, and R shows the profile of the reference. Our results demonstrated a more than 0.96 similarity among these three batches of samples, which means the fingerprint derived from the HPLC assay was a reproducible and reliable method and could be used as a quality control for preparation of active fractions. The main characteristics of the chromatograms obtained from the diethyl ether fraction and dichloromethane fraction were different, implying that the chromatograms of the dichloromethane fraction and the diethyl ether fraction were different.

Fig. 6.

Fig. 6

Open in a new tab

HPLC fingerprint chromatograms of dichloromethane extract (a) and diethyl ether extract (b) recorded at a wavelength of 210 nm

A Kromasil C18 Column, a linear gradient of acetonitrile (5%–35%) in water containing 0.1% trifluoroacetic acid, and a flow rate of 1.0 ml/min were used for quantitative determination

4. Discussion

In this study, we prepared and obtained RSTE and five RST active fractions by extracting RST using dealkalized water, ethanol, n-butyl ethanol, dichloromethane, and diethyl ether. The toxicity of the various active fractions extracted by different solvents from RST was assessed and compared in a zebrafish model. We found that diethyl ether extract induced cardiovascular toxicity, and RSTE, dealkalized water extract and ethanol sedimentation extract induced hepatotoxicity, n-butyl ethanol extract and dichloromethane extract induced both cardiovascular and liver injuries. Overall ranking of cardiovascular toxicity of RST active fractions was dichloromethane extract>diethyl ether extract>n-butyl ethanol extract. Severity of hepatotoxicity was RSTE>dichloromethane extract>n-butyl ethanol extract>ethanol sedimentation extract>dealkalized water extract. Hepatotoxicity in zebrafish induced by dichloromethane extract as well as the absence of liver damage in zebrafish treated with diethyl ether extract was further confirmed in a mouse study. Our findings in this study demonstrated that the toxicity from RST active fractions can be different from the toxicity from the whole RSTE and active fraction technology could be a strategy, or part of a strategy, for reducing or even eliminating RST toxicity.

It appears that dealkalized water extract and ethanol sedimentation extract could not remove whole RSTE-induced hepatotoxicity; and n-butyl ethanol extract and dichloromethane extract actually induced additional organ/cardiovascular toxicity in addition to liver toxicity. Hepatotoxicity was not found in diethyl ether extract, but unfortunately, cardiovascular toxicity occurred. We postulate that due to higher concentrations/amounts or the toxic ability of some chemical components in extracts, compared with RSTE, some active fractions could induce an additional cardiovascular toxicity.

We analyzed the fingerprint chromatograms of the chemical constituents of dichloromethane extract and diethyl ether extract taken as the representatives. Based on the toxicity study findings and HPLC fingerprint profiles, we consider that there was a correlation between the different active fractions and toxic responses. Toxic target organs and toxic levels induced by different active fractions of RST were probably related to the various substances extracted by different solvents. The literature has indicated that the major chemical components are polysaccharide in ethanol extract, flavonoid in n-butyl ethanol extract, alkaloid in dichloromethane extract, and organic acid in diethyl ether extract (Wang et al., 2004; Qian et al., 2015; Zhou et al., 2015). Further study to investigate the chemical components that are responsible for toxicity in RST active fractions is in progress.

Zebrafish share physiological, morphological, and histological similarities with mammals and have been recognized as an inexpensive and rapid alternative to rodents for evaluating drug toxicity and safety liabilities (Zhang et al., 2003; Hill et al., 2005; 2012; Zon and Peterson, 2005; Jones et al., 2008; McGrath and Li, 2008; Rekha et al., 2008; Yang et al., 2010; Hill, 2011; Selderslaghs et al., 2012; Zhang and Ding, 2013; Qian et al., 2015). As a whole organism, zebrafish are apparently able to capture toxicity associated with toxic metabolites, which are unlikely to be found in vitro (Hill et al., 2008). Very recently, several studies regarding the use of zebrafish for hepatotoxicity and cardiovascular assessment were published from our group (He J.H. et al., 2013; Zhu et al., 2014; Shen et al., 2015). These and other studies strongly support the use of zebrafish for toxicity studies, including assessing and comparing the toxicity of Chinese herbs and their active fractions.

5. Conclusions

In summary, this study assessed and compared the toxicity of RSTE and five RST active fractions in a larval zebrafish model. We found that diethyl ether extract induced cardiovascular toxicity, and RSTE, dealkalized water extract, and ethanol sedimentation extract induced hepatotoxicity, whereas n-butyl ethanol extract and dichloromethane extract induced both cardiovascular and liver injuries. The effects of dichloromethane extract and diethyl ether extract on the zebrafish liver were confirmed in mice. Our findings suggest that the toxicity from RST active fractions could be different from the whole RSTE and active fraction technology could be a strategy for reducing or eliminating RST toxicity.

Footnotes

*

Project supported by the National Science and Technology Major Project of China (Nos. 2012ZX09505001-002 and 2015ZX09501004-002-002) and the Zhejiang Provincial Science and Technology Planning Project of China (No. 2014C03009)

Compliance with ethics guidelines: Hong-cui LIU, Xiao-yu ZHU, Jiang-hua CHEN, Sheng-ya GUO, Chun-qi LI, and Zhong-ping DENG declare that they have no conflict of interest.

All institutional and national guidelines for the care and use of laboratory animals were followed.

References

  • 1.Chai NL, Fu Q, Shi H, et al. Oxymatrine liposome attenuates hepatic fibrosis via targeting hepatic stellate cells. World J Gastroenterol. 2012;18(31):4199–4206. doi: 10.3748/wjg.v18.i31.4199. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Cho CH, Chuang CY, Chen CF. Study of the antipyretic activity of matrine. A lupin alkaloid isolated from Sophora subprostrata . Planta Med. 1986;52(5):343–345. doi: 10.1055/s-2007-969179. [DOI] [PubMed] [Google Scholar]
  • 3.Chui CH, Lau FY, Tang JC, et al. Activities of fresh juice of Scutellaria barbata and warmed water extract of Radix Sophorae Tonkinensis on anti-proliferation and apoptosis of human cancer cell lines. Int J Mol Med. 2005;16(2):337–341. doi: 10.3892/ijmm.16.2.337. [DOI] [PubMed] [Google Scholar]
  • 4.CPC (Chinese Pharmacopoeia Commission) Pharmacopoeia of the People’s Republic of China (Part I) China Medical Science Press, Beijing, China; 2015. pp. 25–26. (in Chinese) [Google Scholar]
  • 5.Ding PL, Chen DF. Isoprenylated flavonoids from the roots and rhizomes of Sophora tonkinensis . Helv Chim Acta. 2006;89(1):103–110. doi: 10.1002/hlca.200690000. [DOI] [Google Scholar]
  • 6.Ding PL, Huang H, Zhou P, et al. Quinolizidine alkaloids with anti-HBV activity from Sophora tonkinensis . Planta Med. 2006;72(9):854–856. doi: 10.1055/s-2006-946639. [DOI] [PubMed] [Google Scholar]
  • 7.He CM, Cheng ZH, Chen DF. Qualitative and quantitative analysis of flavonoids in Sophora tonkinensis by LC/MS and HPLC. Chin J Nat Med. 2013;11(6):690–698. doi: 10.1016/S1875-5364(13)60081-3. [DOI] [PubMed] [Google Scholar]
  • 8.He JH, Guo SY, Zhu F, et al. A zebrafish phenotypic assay for assessing drug-induced hepatotoxicity. J Pharmacol Toxicol Methods. 2013;67(1):25–32. doi: 10.1016/j.vascn.2012.10.003. [DOI] [PubMed] [Google Scholar]
  • 9.Hill A. Hepatotoxicity testing in larval zebrafish. In: McGrath P, editor. Zebrafish: Methods for Assessing Drug Safety and Toxicity. John Wiley & Sons, Inc., Hoboken, NJ, USA; 2011. pp. 89–102. [DOI] [Google Scholar]
  • 10.Hill AJ, Teraoka H, Heideman W, et al. Zebrafish as a model vertebrate for investigating chemical toxicity. Toxicol Sci. 2005;86(1):6–19. doi: 10.1093/toxsci/kfi110. [DOI] [PubMed] [Google Scholar]
  • 11.Hill A, Ball J, Jones M, et al. Implementation of zebrafish toxicity testing between in vitro and in vivo models to advance candidate selection. The 29th Annual Meeting of the American College of Toxicology, Tucson, AZ, USA; 2008. pp. 9–12. [Google Scholar]
  • 12.Hill A, Mesens N, Steemans M, et al. Comparisons between in vitro whole cell imaging and in vivo zebrafish-based approaches for identifying potential human hepatotoxicants earlier in pharmaceutical development. Drug Metab Rev. 2012;44(1):127–140. doi: 10.3109/03602532.2011.645578. [DOI] [PubMed] [Google Scholar]
  • 13.Jones KS, Alimov AP, Rilo HL, et al. A high throughput live transparent animal bioassay to identify non-toxic small molecules or genes that regulate vertebrate fat metabolism for obesity drug development. Nutr Metab (Lond) 2008;5:23. doi: 10.1186/1743-7075-5-23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Kimmel CB, Ballard WW, Kimmel SR, et al. Stages of embryonic development of the zebrafish. Dev Dynam. 1995;203(3):253–310. doi: 10.1002/aja.1002030302. [DOI] [PubMed] [Google Scholar]
  • 15.Lee JW, Lee JH, Lee C, et al. Inhibitory constituents of Sophora tonkinensis on nitric oxide production in RAW 264.7 macrophages. Bioorg Med Chem Lett. 2015;25(4):960–962. doi: 10.1016/j.bmcl.2014.12.012. [DOI] [PubMed] [Google Scholar]
  • 16.Li SJ, Yang J, Qian XL, et al. Experimental study on chronic toxicity in rats caused by water extract components of radix et Rhizoma Sophorae tonkinensise . Chin J Pharmacovigil. 2011;8(2):89–92. doi: 10.3969/j.issn.1672-8629.2011.02.007. (in Chinese) [DOI] [Google Scholar]
  • 17.Li XN, Sha N, Yan HX, et al. Isoprenylated flavonoids from the roots of Sophora tonkinensis . Phytochem Lett. 2008;1(3):163–167. doi: 10.1016/j.phytol.2008.08.001. [DOI] [Google Scholar]
  • 18.Li XN, Lu ZQ, Chen GT, et al. NMR spectral assignments of isoprenylated flavanones from Sophora tonkinensis . Magn Reson Chem. 2008;46(9):898–902. doi: 10.1002/mrc.2274. [DOI] [PubMed] [Google Scholar]
  • 19.Li X, Luan Y, Li X. Study on anti-inflammatory efficacy accompanied by side effects of different components of Sophorae tonkinensis radix et Rhizoma . China J Chin Mater Med. 2012;37(15):2232–2237. doi: 10.4268/cjcmm20121510. (in Chinese) [DOI] [PubMed] [Google Scholar]
  • 20.Liu XS, Jiang J, Jiao XY, et al. Matrine-induced apoptosis in leukemia U937 cells: involvement of caspases activation and MAPK-independent pathways. Planta Med. 2006;72(6):501–506. doi: 10.1055/s-2006-931534. [DOI] [PubMed] [Google Scholar]
  • 21.Long Y, Lin XT, Zeng KL, et al. Efficacy of intramuscular matrine in the treatment of chronic hepatitis B. Hepatobiliary Pancreat Dis Int. 2004;3(1):69–72. [PubMed] [Google Scholar]
  • 22.McGrath P, Li CQ. Zebrafish: a predictive model for assessing drug-induced toxicity. Drug Discov Today. 2008;13(9-10):394–401. doi: 10.1016/j.drudis.2008.03.002. [DOI] [PubMed] [Google Scholar]
  • 23.Pan QM, Li YH, Hua J, et al. Antiviral matrine-type alkaloids from the rhizomes of Sophora tonkinensis . J Nat Prod. 2015;78(7):1683–1688. doi: 10.1021/acs.jnatprod.5b00325. [DOI] [PubMed] [Google Scholar]
  • 24.Qian H, Zhao BT, Chan B, et al. Relationship between the content of polysaccharides, flavonoids and polyphenols from the sporocarp of Phellinus linteus and the antioxidant activity. Sci Technol Food Ind. 2015;36(12):104–108. doi: 10.13386/j.issn1002-0306.2015.12.014. (in Chinese) [DOI] [Google Scholar]
  • 25.Rekha RD, Amali AA, Her GM, et al. Thioacetamide accelerates steatohepatitis, cirrhosis and HCC by expressing HCV core protein in transgenic zebrafish Danio rerio . Toxicology. 2008;243(1-2):11–22. doi: 10.1016/j.tox.2007.09.007. [DOI] [PubMed] [Google Scholar]
  • 26.Selderslaghs IWT, Blust R, Witters HE. Feasibility study of the zebrafish assay as an alternative method to screen for developmental toxicity and embryotoxicity using a training set of 27 compounds. Reprod Toxicol. 2012;33(2):142–154. doi: 10.1016/j.reprotox.2011.08.003. [DOI] [PubMed] [Google Scholar]
  • 27.Shen B, Liu HC, Ou WB, et al. Toxicity induced by Basic Violet 14, Direct Red 28 and Acid Red 26 in zebrafish larvae. J Appl Toxicol. 2015;35(12):1473–1480. doi: 10.1002/jat.3134. [DOI] [PubMed] [Google Scholar]
  • 28.Sun R, Yang Q, Zhao Y. Comparative study on acute toxicity of different components of radix et Rhizome Sophorae Tonkinensis in mice. Chin J Pharmacovigil. 2010;7(5):257–262. doi: 10.3969/j.issn.1672-8629.2010.05.001. (in Chinese) [DOI] [Google Scholar]
  • 29.Tang L, Dong LN, Peng XJ, et al. Pharmacokinetic characterization of oxymatrine and matrine in rats after oral administration of radix Sophorae tonkinensis extract and oxymatrine by sensitive and robust UPLC-MS/MS method. J Pharm Biomed Anal. 2013;83:179–185. doi: 10.1016/j.jpba.2013.05.003. [DOI] [PubMed] [Google Scholar]
  • 30.Wang SH, Li TF, Ran BD, et al. Analysis on contents of organic acids and volatile components in tabacco leaves of Yunnan Province. Chin Tob Sci. 2004;(2):35–37. doi: 10.13496/j.issn.1007-5119.2004.02.010. (in Chinese) [DOI] [Google Scholar]
  • 31.Westerfield M. The Zebrafish Book: A Guide for the Laboratory Use of Zebrafish (Danio rerio) University of Oregon Press, Eugene, OR, USA; 1995. [Google Scholar]
  • 32.Xiao P, Kubo H, Komiya H, et al. (−)-14β-Acetoxymatrine and (+)-14α-acetoxymatrine, two new matrine-type lupin alkaloids from the leaves of Sophora tonkinensis . Chem Pharm Bull (Tokyo) 1999;47(3):448–450. doi: 10.1248/cpb.47.448. [DOI] [Google Scholar]
  • 33.Yang Q, Zheng LN, Xie YZ, et al. Study on the “dosage-time-toxicity” relationship of hepatotoxicity induced by different components of radix et Rhizoma Sophorae Tonkinensis in mice. Chin J Pharmacovigil. 2010;7(7):385–389. doi: 10.3969/j.issn.1672-8629.2010.07.001. (in Chinese) [DOI] [Google Scholar]
  • 34.Yoo H, Chae HS, Kim YM, et al. Flavonoids and arylbenzofurans from the rhizomes and roots of Sophora tonkinensis with IL-6 production inhibitory activity. Bioorg Med Chem Lett. 2014;24(24):5644–5647. doi: 10.1016/j.bmcl.2014.10.077. [DOI] [PubMed] [Google Scholar]
  • 35.Zhang C, Willett C, Fremgen T. Zebrafish: an animal model for toxicological studies. Curr Protoc Toxicol. 2003;17:1.7.1–1718. doi: 10.1002/0471140856.tx0107s17. [DOI] [PubMed] [Google Scholar]
  • 36.Zhang HY, Ding TH. Survey of clinical application and toxic reaction of shandougen. Western J Tradit Chin Med. 2013;26(3):121–124. doi: 10.3969/j.issn.1004-6852.2013.03.052. (in Chinese) [DOI] [Google Scholar]
  • 37.Zhou HJ, Ma HL, Guo DZ, et al. Physicochemical properties and antioxidant activity of intracellular polysaccharides from Phellinus igniarius precipitated by different ethanol concentrations. Food Sci. 2015;36(19):34–38. doi: 10.7506/spkx1002-6630-201519006. (in Chinese) [DOI] [Google Scholar]
  • 38.Zhou J, Guo SY, Zhang Y, et al. Human prokinetic drugs promote gastrointestinal motility in zebrafish. Neurogastroenterol Motil. 2014;26(4):589–595. doi: 10.1111/nmo.12306. [DOI] [PubMed] [Google Scholar]
  • 39.Zhu JJ, Xu YQ, He JH, et al. Human cardiotoxic drugs delivered by soaking and microinjection induce cardiovascular toxicity in zebrafish. J Appl Toxicol. 2014;34(2):139–148. doi: 10.1002/jat.2843. [DOI] [PubMed] [Google Scholar]
  • 40.Zhu XY, Liu HC, Guo SY, et al. A zebrafish thrombosis model for assessing antithrombotic drugs. Zebrafish. 2016;13(4):335–344. doi: 10.1089/zeb.2016.1263. [DOI] [PubMed] [Google Scholar]
  • 41.Zon LI, Peterson RT. In vivo drug discovery in the zebrafish. Nat Rev Drug Discov. 2005;4(1):35–44. doi: 10.1038/nrd1606. [DOI] [PubMed] [Google Scholar]

Articles from Journal of Zhejiang University. Science. B are provided here courtesy of Zhejiang University Press

RESOURCES