The genotoxicity of an aqueous extract of Gyejibokryeong-hwan

Reference standards of amygdalin, coumarin, and cinnamic acid were purchased from Sigma-Aldrich (St. Louis, MO, USA). Albiflorin, cinnamaldehyde, paeoniflorin, and paeonol were obtained from Wako (Osaka, Japan). The purity of the seven reference standards was ≥98.0%. The chemical structures of the seven marker components are shown in Fig. 1a. High-performance liquid chromatography (HPLC)-grade reagents methanol, acetonitrile, and water to obtain the aqueous extract of GJBRHE were obtained from J. T. Baker (Phillipsburg, NJ, USA). Acetic acid was purchased from Merck (Darmstadt, Germany).

The component herbs of Gyejibokryeong-hwan (Poria cocos, Paeonia suffruticosa, Cinnamomum cassia, Paeonia lactiflora, and Prunus persica) were purchased from Kwangmyungdag Medicinal Herbs (Ulsan, Korea) in November 2013 and identified by Prof. Je Hyun Lee (Department of Herbology, College of Oriental Medicine, Dongguk University, Gyeongju, Korea). Voucher specimens (2013–KE31–1 to KE31–5) have been deposited at the K-herb Research Center, Korea Institute of Oriental Medicine (KIOM).

To obtain the Gyejibokryeong-hwan aqueous extract (GJBRHE), the five herbal medicines comprising Gyejibokryeong-hwan were mixed as shown in Table 1 (total weight = 125.0 kg, equivalent to approximately 666.7 single doses), and extracted in a 10-fold mass of water (1250 L) at 100 °C for 2 h. The aqueous extract was filtered by pressing through a filter (10 μm pore size), and then the solution was evaporated and freeze-dried to give a powder. The amount of Gyejibokryeong-hwan decoction powder was 15.7 kg (12.56%). For HPLC analysis, 400 mg of lyophilized Gyejibokryeong-hwan decoction powder was dissolved in 20 mL of distilled water and then extracted by sonication for 10 min on room temperature. The solution was filtered through a 0.2 μm pore-size syringe filter (Woongki Science, Seoul, Korea) before HPLC injection.

The chromatographic analysis for quantitative determination of the seven marker components in Gyejibokryeong-hwan was performed using a Shimadzu Prominence LC-20A system (Shimadzu Co., Kyoto, Japan) consisting of a solvent delivery unit, an on-line degasser, a column oven, a sample autoinjector, and a photodiode array (PDA) detector. The data were acquired and processed using LCsolution software (Version 1.24). The column used for separation of the seven constituents was a Phenomenex Gemini C₁₈ column (250 × 4.6 mm, 5 μm, Torrance, CA, USA) maintained at 40 °C. The mobile phases consisted of water (A) and acetonitrile (B), both containing 1.0% (v/v) acetic acid. The gradient elution of the mobile phase was as follows: 10–60% B for 0–30 min, 60–100% B for 30–40 min, 100% B for 40–45 min, 100–10% B for 45–50 min, and 100% B for 50–60 min. The flow rate and injection volume were 1.0 mL/min and 10 μL, respectively. The detection wavelengths for quantitative analysis of the seven compounds were set according to the maximum absorption wavelengths of each reference compound (amygdalin, albiflorin, paeoniflorin for P. persica, coumarin, cinnamic acid, cinnamaldehyde for C.cassia, paeonol for P.suffruticosa)..

This study was approved by the Institutional Animal Care and Use Committee (IACUC) of Korea Institute of Toxicology (KIT; Daejeon, Korea); it was performed at KIT and was conducted according to the guidelines of the KIT IACUC (Number: G13088), which is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International (1998) under the Good Laboratory Practice Regulations for Nonclinical Laboratory Studies. Specific-pathogen-free Sprague Dawley rats (5 males and 5 females) were purchased from Orient Bio Co. (Seongnam, Korea). The acute oral toxicity study in rats was conducted as described previously [17]. Our preliminary study (N:P13030) showed that a single oral administration of GHBRHE did not induce any toxic effect at dose levels up to 5000 mg/kg/day (data not shown). Based on this study, a dose of 5000 mg/kg/day was selected as the experimental limit dose. GJBRHE was suspended in distilled water. The vehicle control rats received distilled water only. All animals were observed for 14 days and mortality, body weight changes, clinical signs, and gross findings were recorded. This study was conducted in accordance with the guidelines established by the FDA Good Laboratory Practice (GLP) for Nonclinical Laboratory Studies (21 CFR Part 58) and OECD guidelines [18].

The used experimental methods were based on the report of Maron and Ames [19], with minor modifications. The Ames test was conducted as described previously [17]. Distilled water was selected as the vehicle based on the results of the solubility test and aseptic test. Various concentrations of GJBRHE were incubated with tester strains in the presence or absence of metabolic activation with the S9 mix along with vehicle. Each concentration of GJBRHE was assayed in triplicate. The present study was conducted in accordance with OECD guidelines for the Testing of Chemicals Section 4 Health Effects Test No. 471 about Bacterial Reverse Mutation Test (21 July 1997).

Chromosome aberration tests were conducted as described previously [17] with minor modifications as described by Ishidate [20] and by Dean and Danford [21]. Chinese hamster lung (CHL) cells were selected as the test system because they are sensitive to mutagens, their low chromosome number facilitates scoring and the same cell line can be used repeatedly. Furthermore, because CHL cells are a widely used test system for in vitro mutagenicity studies, a broad historical database exists. The modal chromosome number of this cell line is 25 with a doubling time of ~15–17 h. The cells are thawed in culture medium and then grown for more than 7 days as a monolayer. Sterility was checked by inverted microscopy to determine any gross mycoplasma contamination and confirmed by polymerase chain reaction amplification. Cells were cultured in reconstituted modified Eagle’s medium (Gibco-Invitrogen, Carlsbad, CA, USA) supplemented with Na₂HCO₃ (2.2 g), l-glutamine (292 mg), streptomycin sulfate (100 μg), penicillin G·Na (10⁵ units) and 10% (v/v) fetal bovine serum (Gibco-Invitrogen) per liter. The cultures were incubated at 37 °C under humidified 1.5% CO₂ in air. Based on the results, the dose range for the present study was designed considering the solubility and cytotoxicity of SST. Ethyl methanesulfonate (EMS) was used as a positive control chemical without metabolic activation using the S9 mix, and cyclophosphamide (CPA) was used as a positive control with metabolic activation. The present study was conducted in accordance with OECD guidelines for the Testing of Chemicals Section 4 Health Effects Test No. 473 about In vitro Mammalian Chromosome Aberration Test (21 July 1997).

The MN test was performed as described by Muangphra and Gooneratne [22] with slight modifications. Bone marrow preparations were made according to the method described by Schmid [23]. MN tests using ICR male mice were conducted as described previously [17, 24]. The preliminary study showed that oral administration of GJBRHE did not induce any toxic effect at 2000 mg/kg (the dose limit for treatment up to 14 days according to OECD guidelines). This was selected as the maximum dose. Specific-pathogen-free male ICR strain mice weighing 27.2–30.0 g were obtained from Orient Co., Ltd. (Seongnam, Korea) at 6 weeks of age. According to Schmid [23] method, ICR mice were euthanized by CO₂ gas after the final administration of GJBRHE and bone marrow preparations were made. Small round or oval bodies, measuring approximately 1/5 to 1/20 the diameter of a polychromatic erythrocyte (PCE), were counted as micronuclei. A total of 2000 PCEs were scored per animal to determine the frequency of MNPCEs and PCE/(PCE + NCE) ratio was calculated by counting 500 cells. Differences in the numbers of MNPCEs between treatment and control groups were analyzed using the Kruskal–Wallis H test and Dunn’s rank sum test. The Mann–Whitney nonparametric U test was used to compare the PCE/(PCE + NCE) ratios of treatment and control groups. Significance for all tests was accepted at P < 0.05.

This study was approved by the Institutional Animal Care and Use Committee (IACUC) of Korea Institute of Toxicology (KIT; Daejeon, Korea); it was performed at KIT and was conducted according to the guidelines of the KIT IACUC (Number: N14005), which is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International (1998) under the Good Laboratory Practice Regulations for Nonclinical Laboratory Studies.

Acute toxicity study, a one-way analysis of variance was performed at α = 0.05 in cases in which Bartlett’s test indicated no significant deviations from variance homogeneity. When significance was noted, a multiple-comparison test (Dunnett’s test) was used to determine which pairs of groups were significantly different. Statistical analyses were performed using the PATH/TOX system. Chromosome aberration assay, the statistical analyses were based on the methods used in the published reports [25] using Statistical Analysis System (SAS) software. Each metaphase was classified as a normal metaphase or an aberrant metaphase with one or more aberrations, and the frequency of aberrant metaphases was analyzed statistically. The χ² test and Fisher’s exact test [26] were performed to compare the vehicle control and GJBRHE-treated groups. The in vivo MN results were evaluated as described previously [27] using the method of Lovell et al. [28] with minor modifications. Data with heterogeneous variances were analyzed using Kruskal–Wallis analysis of variance followed by multiple comparisons using Dunnett’s test [29]. The significance was accepted when all of the PCE/(PCE + NCE) ratios were > 0.1. The result was judged as positive when there was a significant and dose-related increase or a reproducible increase in the frequency of MNPCEs or aberrant metaphases at one or more dose levels. Differences were regarded as significant at P < 0.05.

The optimized HPLC–PDA method was used for quantitative determination of the seven marker compounds in the Gyejibokryeong-hwan sample. Using optimized chromatography conditions, all analytes separated in under 30 min; a typical HPLC chromatogram is shown in Fig. 1b. The wavelength of the PDA ranged from 190 to 400 nm and the detection wavelengths for quantitative analysis were 262 nm (amygdalin), 230 nm (albiflorin and paeoniflorin), and 280 nm (coumarin, cinnamic acid, cinnamaldehyde, and paeonol). The correlation coefficient (r²) of the five compounds showed good linearity of ≥ 0.9999. The retention times of the seven components, coumarin, amygdalin, paeoniflorin, albiflorin, cinnamic acid, cinnamaldehyde, and paeonol were 9.60, 14.80, 15.55, 19.56, 22.08, 24.53, and 26.37 min, respectively. The amounts of the seven marker compounds were 0.52–45.07 mg/g (Table 2). Among the compounds, paeoniflorin, the marker compound of P. lactiflora, was detected as the main component of GJBRHE at 45.07 mg/g. This established HPLC–PDA method will be useful for improvement of the quality control of Gyejibokryeong-hwan.

During the experimental period, no clinical symptoms or no mortality of toxicity were observed in rats of either sex in any of the GJBRHE-treated groups during the 14-day observation period. The changes in body weight are shown in Fig. 2. For both sexes, the changes in body weight did not differ significantly between the treated rats with 5000 mg/kg/day of GJBRHE and the control group. At the time of the scheduled autopsy, there were no abnormal pathological observations for the lung, thymus, heart, liver, stomach, adrenals, and spleen in the male or female rats given 5000 mg/kg/day of GJBRHE.

No positive mutagenic response when compared with concurrent vehicle control groups was observed in any of the S. typhimurium TA1535, TA1573, TA100, and TA98 or E. coli WP2 uvrA strains at any dose level of GJBRHE regardless of the presence (Fig. 3a) or absence (Fig. 3b) of the S9 fraction at up to 5000 μg/plate. There was a significant increase in all positive control groups (Fig. 3a and b).

There were significant increases the frequency of metaphase cells with structural aberrations in cells treated for 6 h after S9 fraction with 4200 μg/mL GJBRHE in the presence of the S9 fraction and in those treated for 22 h S9 fraction with 800 μg/mL GJBRHE without the S9 fraction (Table 3). The positive controls showed significant increases in the frequency of metaphases (Table 3).

No clinical signs and no deaths of toxicity were observed in any of the GJBRHE treatment groups. As shown in Table 4, there was no significant change in body weight compared with the vehicle control between GJBRHE at 500, 1000, or 2000 mg/kg. The cytotoxicity indices, i.e. the PCE/(PCE + NCE) ratios, were 0.59, 0.53, 0.55, and 0.48 for the vehicle control (0), 500, 1000, and 2000 mg/kg/day groups, respectively. The frequencies of micronucleated PCEs among 2000 PCEs were 1.67, 0.67, 0.33, and 0.67 for the vehicle control (0), 500, 1000, and 2000 mg/kg/day groups, respectively (Table 5). There was no significant increase in the frequencies of micronucleated PCEs at any dose level of GJBRHE. The positive control substance induced a significant increase in the ratio, to 54.33 compared with that of the vehicle controls (P < 0.01).