Arenobufagin was purchased from Baoji Herbest Bio-Tech Co., Ltd. (Baoji, China). XTT, propidium iodide (PI), proteinase K, Hoechst 33258 and ribonuclease A (RNaseA) were purchased from Merck KGaA (Darmstadt, Germany). Dulbecco's modified Eagle's medium (DMEM), phenazine methosulfate, agarose X and 25% glutaraldehyde solution were purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). Crystal violet (C.I. 42555) Certistain^® was purchased from Merck KGaA. SB203580, a specific inhibitor of p38 MAPK, and its negative control SB202474 were purchased from Merck KGaA. Fetal bovine serum (FBS) was obtained from Nichirei Biosciences, Inc. (Tokyo, Japan). High performance liquid chromatography (HPLC)-grade acetonitrile and methanol were purchased from Thermo Fisher Scientific, Inc. (Waltham, MA, USA). Mass spectrometry (MS)-grade formic acid was obtained from ROE Scientific, Inc. (Newark, DE, USA).

Dried skin of Bufo bufo gargarizans Cantor was purchased from Anhui Jinchan Biochemistry Co., Ltd. (Huaibei, China), and further identified by Professor Hongjie Wang (Institute of Chinese Materia Medica, China Academy of Chinese Medical Sciences, Beijing, China). The dried toad skin (10 kg) was cut into pieces, and then extracted under reflux with 95% ethanol into 20 liters. The extracting solution was dried with rotary evaporation at 45°C under reduced pressure (vacuum drying) to yield ~150 g residue. Following separation through a silica gel (2,000 g; 160-200 mesh; Qingdao Haiyang Chemical Co., Ltd., Qingdao, China) column chromatography with chloroform-methanol solution (50:1-1:1) with gradient elution, a total of eight fractions were obtained (Fr. 1-8). Fr. 4 (8 g) was further separated by C18 (320 ml; SP-120-50-ODS-RPS; DAISO Co., Ltd., Osaka, Japan) column chromatography using 30% (500 ml, Fr. 4.1-4.5), 40% (500 ml, Fr. 4.6-4.10), 50% (500 ml, Fr. 4.11-4.15) and 60% (400 ml, Fr. 4.16-4.19) methanol. Hellebrigenin was purified from Fr. 4.10-4.16 by preparative HPLC. It was obtained as a white powder with molecular formula of C₂₄H₃₂O₆ based on high-resolution electrospray ionization MS (HR-ESI-MS). The compound was identified as hellebrigenin with >96% purity according to previously reported values (28).

U-87, a human glioblastoma cell line, was obtained from the American Type Culture Collection (Manassas, VA, USA) and cultured in DMEM supplemented with 10% heat-inactivated FBS, 100 U/ml of penicillin and 100 µg/ml of streptomycin (both from Wako Pure Chemical Industries, Ltd.) in a humidified atmosphere with 5% CO₂ at 37°C. Primary mouse glial cells were prepared according to a method previously reported with modifications (29). Briefly, the cerebral cortex of 4-5 newborn C57BL/6J mice [2-4 days old (sex unconfirmed), body weight 2-3 g, housed at 23±1°C in a room with a constant humidity of 55±5% and a regular 12-h light/12-h dark cycle together with dams] was removed and digested by trituration in PBS (Wako Pure Chemical Industries, Inc.) containing 0.25% trypsin. Dissociated glial cells were cultured in 75 cm² flasks (BD Biosciences, Franklin Lakes, NJ, USA) in DMEM containing 10% FBS in a humidified atmosphere with 5% CO₂ at 37°C.

After 2 days of cultivation, the medium was changed to remove unattached cells. Glial cells were allowed to form a confluent monolayer for 7-9 days and harvested from the flask by treatment with 0.125% trypsin. Subsequently, cells were seeded at a density of 5×10⁵ cells/ml in 96-well plates and allowed to grow to ~80% confluency. The glial cell preparation was highly enriched in astrocytes (>90%) according our previous report (29). The U-87 and primary glial cells were treated with SB203580, a specific inhibitor of p38 MAPK, and its negative control SB202474 at various concentrations (1, 5 or 10 µM) for 30 min prior to treatment with arenobufagin and hellebrigenin [20 ng/ml, almost equal to their half maximal inhibitory concentration (IC₅₀) values at 48 h], in the presence or absence of these reagents. The present study was approved by the Committee of Animal Care and Welfare of Tokyo University of Pharmacy and Life Sciences (Tokyo, Japan; registration no. P17-60).

As mice are too small to successfully collect clear cerebrospinal fluid without blood contamination, and there is no clear difference in the structure and function of the blood-brain barrier (BBB) in rats and mice, rats were used instead of mice for the qualitative assessment. Nine male Sprague Dawley rats (220-240 g) were purchased from Shanghai Shibei Biotechnology Co., Ltd. (Shanghai, China), and housed at 23±1°C in a room with a constant humidity of 55±5% and a regular 12/12-h light/dark cycle for one week prior to experimentation. Throughout the experiment, the rats had free access to food and water. After an overnight fasting with water not limited, all animals were randomly divided into three groups (n=3) and administered the following treatments: Vehicle control (saline), arenobufagin (50 mg/kg) and hellebrigenin (50 mg/kg). All animals received a single oral dose of 50 mg arenobufagin/hellebrigenin or the same volume of saline/kg of body weight. At 30 min after the oral administration, cerebrospinal fluids were collected according to the method previously described with slight modifications (30). Briefly, the fur on the head and neck regions of rats were carefully removed, and animals were placed in an induction chamber and anesthetized. Following disinfection with ethanol, an ~2-cm skin incision was made along the midline. The spinous process and lamina were clamped with hemostatic forceps in order to make the arachnoid membrane more visible. Then, the cerebrospinal fluid was carefully extracted from the subarachnoid space using a 1-ml syringe with a 23G needle without blood contamination. Following centrifugation at 9,730 × g for 15 min at 4°C, the cerebrospinal fluid supernatants were collected and stored at −20°C for later analysis.

The analysis of cerebrospinal fluids was performed using the ultimate 3000 hyperbaric LC system coupled with a Velos Pro LTQ Orbitrap Mass Spectrometer via an ESI interface from Thermo Fisher Scientific, Inc. Eclipse XDB-C18 (4.6×150 mm, 3.5 µm; Agilent Technologies, Inc., Santa Clara, CA, USA) was used as the separation column in the ultra HPLC. The mobile phase consisted of 0.1% formic acid in water (solvent A) and acetonitrile (solvent B), with the gradient elution as follows: 0-5 min, 20-40% B; 5-20 min, 40-50% B; 20-21 min, 50-95% B; 21-26 min, held at 95% B; 26-33 min, balanced to 20% B. The flow rate was 0.2 ml/min and the temperature-controlled column was maintained at 30°C. Column effluents were monitored at 296 nm following the injection of 4 µl cerebrospinal fluid obtained from the control and treatment groups. MS detection was performed in the positive ion mode. The ESI source parameters were as follows: Ion spray voltage, +5.0 kV; sheath gas flow rate, 40 arb; aux gas flow rate, 10 arb; capillary temperature, 350°C; and S-lens RF level, 60%. The mass resolution of Fourier transform was 30,000 with a full scan in the range of m/z 200-800. The tandem MS (MS/MS) experiments were set as a data-dependent scan. The experimental procedures complied with the Animal Ethics Committee Guidelines of Beijing Animals Science Biology Technology Co., Ltd. (Beijing, China; registration no. 170703002).

Following treatment with various concentrations (10, 20, 30, 40, 100 and 150 ng/ml) of arenobufagin and hellebrigenin for 48 h, cell viability was measured using the XTT assay as described previously (31). Relative cell viability was expressed as the ratio of the absorbance at 450 nm of each treatment group against those of the corresponding untreated control group. The IC₅₀ values of each drug were calculated using GraphPad Prism^® 6.0 software (GraphPad Software, Inc., La Jolla, CA, USA). With respect to the morphological alterations of U-87 cells, the cells were imaged using an inverted microscope (CKX53; Olympus Corporation, Tokyo, Japan) fitted with a digital camera following treatment with various concentrations (20, 40, 100, 150 and 200 ng/ml) of arenobufagin and hellebrigenin for 48 h. Untreated cells were used as the control. Clonogenic survival assays were performed according to a method previously described, with slight modifications (14). Briefly, U-87 cells were seeded at a density of 5×10³ cells/well in 6-well plates, and treated with various concentrations (5, 10, 20, 30 and 40 ng/ml) of arenobufagin and hellebrigenin for 24 h. Untreated cells were used as the control. The medium was then replaced with fresh DMEM (supplemented with 10% heat-inactivated FBS, 100 U/ml penicillin and 100 µg/ml streptomycin) and the cells were allowed to grow for 7-10 days in a humidified 5% CO₂ atmosphere at 37°C. Following washing gently with PBS twice, the cells were fixed with 0.25% glutaraldehyde/PBS for 15 min prior to staining with 0.2% crystal violet/PBS for 10 min at room temperature. Following a washout of extra crystal violet with water to get an adequate staining pattern, the images of crystal violet-stained cells were scanned into a computer, followed by dissolution of the violet-stained cells in 1% SDS. The absorbance of cell lysates was determined at 550 nm. The relative colony formation rate was expressed as the ratio of the absorbance at 550 nm of each treatment group against those of the corresponding untreated control group.

Following treatment with various concentrations (10, 20, 30 and 40 ng/ml) of arenobufagin or hellebrigenin for 48 h, and 20 ng/ml arenobufagin or hellebrigenin for 24, 48 or 72 h, cell cycle analysis was performed using a FACSCanto™ flow cytometer (BD Biosciences), according to a previous method (14,21). Briefly, cells were washed twice with cold PBS, fixed with 1% paraformaldehyde/PBS on ice for 30 min, washed twice again with cold PBS, permeabilized in 70% (v/v) cold ethanol at −20°C for at least 4 h. Cell pellets were then washed twice with cold PBS following centrifugation (430 × g for 5 min at 4°C) and incubated with 0.25% Triton X-100 for 5 min on ice. Following centrifugation (430 × g for 5 min at 4°C) and washing with PBS, cells were resuspended in 500 µl PI/RNase A/PBS (5 µg/ml PI and 0.1% RNase A in PBS) and incubated for 30 min in the dark at room temperature. A total of 10,000 events were recorded, and FACSDiva™ software (v6.0; BD Biosciences) and ModFit LT™ v3.0 (Verity Software House, Inc., Topsham, ME, USA) were used to calculate the number of cells in each G₀/G₁, S and G₂/M phase fraction.

For preparation of the protein samples, cell pellets (1-2×10⁶ per 110 µl buffer) were suspended in Laemmli buffer containing 100 mM DTT, 2 µg/ml leupeptin, 2 µg/ml aprotinin, 1 µg/ml pepstatin, 1 mM PMSF. Cell suspensions were sonicated (Qsonica, LLC, Newtown, CT, USA) with 10 short bursts of 2 sec followed by intervals of 2 sec for cooling. The suspensions were then kept in an ice bath. Sonicated cells were heated in 95°C for 5 min, and then centrifuged at 13,000 × g for 15 min at 4°C. Protein concentrations of the supernatant were determined according to Bradford's method using the Protein Assay Dye Reagent Concentrate (Bio-Rad Laboratories, Inc.), according to the manufacturer's instructions, using BSA as the standard. Western blot analysis was carried out according to methods previously described (31). Briefly, after separation of proteins (10-20 µg protein/lane) via SDS-PAGE (8, 10 or 12.5% according to the molecular weight of the target protein), followed by transference to a polyvinylidene difluoride (PVDF) membrane, which was then blocked with 5% skim milk/PBST (PBS containing 0.5% Tween-20) for 1 h at room temperature. Protein bands were detected using the following primary antibodies: Mouse anti-human β-actin (1:5,000 dilution; cat. no. A-5441; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany), rabbit anti-human Cdc25c (1:1,000 dilution; cat. no. 4688), rabbit anti-human p27 (1:1,000 dilution; cat. no. 2552), mouse anti-human Cyclin B1 (1:2,000 dilution; cat. no. 4135), mouse anti-human survivin (1:1,000 dilution; cat. no. 2802), rabbit anti-human phospho-Akt (Ser473) (1:2,000 dilution; cat. no. 4060) and Akt (1:1,000 dilution; cat. no. 4691), rabbit anti-human phospho-p38 (Thr180/Tyr182, 1:1,000 dilution; cat. no. 9211) and p38 (1:1,000 dilution; cat. no. 9212), rabbit anti-human phospho-MK2 (Thr334) (1:1,000 dilution; cat. no. 3041), rabbit anti-human phospho-ERK (Thr202/Tyr2042) (1:2,000 dilution; cat. no. 4370) and ERK (1:1,000 dilution; cat. no. 4695; all from Cell Signaling Technology, Inc., Danvers, MA, USA). PVDF membranes containing blotted protein bands were incubated overnight with the respective primary antibody at 4°C, followed by the appropriate horseradish peroxidase-conjugated secondary antibody (anti-mouse IgG, 1:3,000 dilution, cat. no. A5906; anti-rabbit IgG, 1:3,000 dilution, cat. no. A0545; both from Sigma-Aldrich; Merck KGaA) for 1 h at room temperature, and then detected with an enhanced chemiluminescence (ECL) analysis system (Amersham; GE Healthcare Life Sciences, Chalfont, UK).

Following treatment with various concentrations (10, 20, 40, 100, 150 and 200 ng/ml) of arenobufagin or hellebrigenin for 48 h as described in the previous section of the cell viability assay, LDH leakage from U-87 cells was measured using a LDH cytotoxicity detection kit (Wako Pure Chemical Industries, Ltd.) according to the method previously described with slight modifications (31). Briefly, culture medium served as the negative control (NC). Culture supernatants (S) were collected by centrifugation at 450 × g for 5 min at 4°C and stored at −80°C until use. Cultured cells without treatment were lysed in the culture medium containing 0.2% Tween-20, and the cell lysate was used as the non-damaged positive control (PC) following centrifugation at 12,000 × g for 5 min at 4°C. In order to avoid the influence of Tween-20, culture medium containing 0.2% Tween-20 served as the negative control for PC and was referred to as NCT. Samples were diluted 16-fold with PBS and 50 µl of the diluted solution was transferred into wells of a 96-well plate. LDH activities were determined by adding 50 µl reaction reagent from the kit, followed by incubation at room temperature for 30 min. The reaction was stopped by the addition of 100 µl stopping solution provided with the kit at room temperature, and the absorbance at 560 nm was measured with a microplate reader (EMax^® Plus; Molecular Devices, LLC, Sunnyvale, CA, USA). Cell damage was calculated as a percentage of LDH leakage from damaged cells using the following formula: (S-NC)/(PC-NCT) × 100.

DNA fragmentation analysis was performed according to a method described previously (32). Briefly, between 5×10⁵ and 1×10⁶ cells were used for the DNA preparation. The cells were suspended in 500 µl lysis buffer (50 mM Tris-HCl, pH 7.8, 10 mM EDTA-2Na, 0.5% sodium-N-lauroyl sarcosinate). The suspension was incubated in an ice bath for 30 min following the addition of 50 µl of 10% SDS with gentle agitation every 5 min. The lysates were incubated successively at 50°C for 90 min with proteinase K (1 mg/ml), at 50°C for 30 min with RNase A (1 mg/ml) and at 60°C for 15 min following the addition of the same volume of NaI solution [6.0 M NaI in 26 mM Tris-HCl, pH 8.0, 13 mM EDTA-2Na, 10 mg/ml of bovine glycogen (Nacalai Tesque, Inc., Kyoto, Japan)]. DNA was precipitated by addition of an equal volume of 100% isopropanol. DNA pellets were successively washed with 50% isopropanol, 100% isopropanol and 70% ethanol. Extracted DNA was dissolved in TE buffer (10 mM Tris-HCl, pH 7.8, 1 mM EDTA), and the DNA concentrations were determined by staining with Hoechst 33258 as described (33). The DNA samples (~20 µg DNA in 20 µl TE buffer) and 100 bp DNA ladder (Invitrogen; Thermo Fisher Scientific, Inc.) as a DNA size marker were electrophoresed on a 2% agarose X gel using TBE buffer (89 mM Tris, 89 mM boric acid, 2 mM EDTA). Gels were visualized using 0.5 µg/ml ethidium bromide staining (at room temperature, 30 min), followed by viewing under a Luminograph II chemiluminescent imaging system (ATTO Corporation, Tokyo, Japan).

Experiments were independently repeated three times, and the results are presented as the means ± standard deviation of three assays. Statistical analysis was performed using GraphPad Prism^® 6 software. The Student's t-test was used to compare sample means between two groups, and one-way analysis of variance followed by Dunnett's post hoc test was used to compare sample means among three or more groups. P<0.05 was considered to indicate a statistically significant difference.

First, whether the two active bufadienolides were able to cross the BBB was investigated. The [M+H]⁺ ion of arenobufagin at m/z 417.2264 (C₂₄H₃₃O₆, Cal.417.2272, error −1.88 ppm) with a retention time of 8.97 min was only detected in the cerebrospinal fluid of rats who received a single oral dose of arenobufagin, instead of saline (vehicle control) (Fig. 1A). However, the [M+H]⁺ ion of hellebrigenin at m/z 417.2277 (C₂₄H₃₃O₆, Cal.417.2272, error 1.26 ppm) with a retention time of 8.91 min was hardly detected in the cerebrospinal fluid of rats who received a single oral dose of hellebrigenin due to its very low signal intensity. The further identification of arenobufagin in cerebrospinal fluids was performed using MS/MS experiments (Fig. 1B). These results indicated that arenobufagin, but not hellebrigenin, were able to cross the BBB.

A significant decrease in cell viability was observed in a dose-dependent manner in U-87 cells following treatment with various concentrations of arenobufagin and hellebrigenin for 48 h. The IC₅₀ values of arenobufagin and hellebrigenin were 24.9±2.8 ng/ml and 23.5±2.4 ng/ml, respectively (Fig. 2A and B). In comparison, under the same treatment conditions, no cytocidal effect was observed in mouse primary astrocytes when treated with arenobufagin (Fig. 2C), and a slight increase in viability was observed following treatment with hellebrigenin at concentrations >10 ng/ml (Fig. 2D). Morphological alterations of U-87 cells were also observed using an inverted microscope. As shown in Fig. 2E and F, treatment with arenobufagin and hellebrigenin at concentrations >100 ng/ml for 48 h, caused a large number of U-87 cells to detach from the plate. In contrast, concentrations <40 ng/ml of each drug exhibited little effect on cell morphology, although an evident decrease in the number of cells/field was observed, indicating an inhibition of cell viability. In comparison, almost no morphological alterations were observed in the primary astrocytes following treatment with the two bufadienolides at concentrations ≤150 ng/ml (Fig. 2G).

To explore whether exposure to arenobufagin and hellebrigenin suppressed the surviving fraction of U-87 cells, a colony formation assay was performed. As shown in Fig. 3, significant suppression of the colony numbers of U-87 was observed following long-term treatment with arenobufagin or hellebrigenin, at concentrations ≥5 ng/ml, indicating their potency against the survival of the cells.

To explore whether cell cycle arrest is involved in the cytotoxic effects of arenobufagin and hellebrigenin, cell cycle analyses were performed using flow cytometry (Figs. 4 and 5). In comparison with control group (untreated cells), a significant increase in the number of cells in the G₂/M phase was observed (Figs. 4A and C, and 5A and C) following treatment with various concentrations of the two drugs for 48 h. Concomitantly, a significant decrease in the number of cells in the G₀/G₁ and S phase was also observed (Figs. 4A and C, and 5A and C). Furthermore, following treatment with arenobufagin or hellebrigenin at 20 ng/ml, which was almost equal to their IC₅₀ values, for 24, 48 and 72 h, G₂/M cell cycle arrest-inducing activity of each drug was observed at 24-h post-treatment, which reached the maximum at 48-h post-treatment, and was sustained for up to 72 h (Figs. 4B and D, and 5B and D). A significant decrease in the number of cells in the G₀/G₁ and S phase was also concomitantly observed (Figs. 4B and D, and 5B and D). As shown in Fig. 6, compared with the control group, the expression of Cdc25C and Cyclin B1 was remark-ably downregulated following treatment with arenobufagin or hellebrigenin in a dose-dependent manner, although almost no evident alteration was observed in the expression levels of p27. Similarly, a dose-dependent downregulation in the expression of survivin was also observed.

The release of LDH provides an accurate measure of the cell membrane integrity and cell viability (13,14,31). Following treatment with various concentrations of arenobufagin and hellebrigenin for 48 h, LDH leakage analysis was performed to examine whether the treatments affected cell membrane integrity. A dose-dependent increased in LDH leakage was observed in U-87 cells treated with arenobufagin or hellebrigenin (Fig. 7). Notably, no significant difference was observed between cells treated with each drug at 20 ng/ml and their respective control group (0 ng/ml), indicating that cell membrane damage was more likely due to relatively high concentrations, but not relatively low concentrations, of each drug.

To explore whether apoptosis is involved in the cytotoxic effects of arenobufagin and hellebrigenin, DNA fragmentation analysis was preformed using DNA gel electrophoresis. As shown in Fig. 8, no DNA fragmentation was observed in U-87 cells when treated with arenobufagin or hellebrigenin, even at the concentrations >100 ng/ml, which significantly induced a large number of U-87 cells to lose their adhesive properties (Fig. 2E and F) and LDH release (Fig. 7). These results indicated that apoptosis induction was less likely to contribute to the cytotoxicity caused by arenobufagin or hellebrigenin regardless of the drug concentration. These results were also supported by the cell cycle analysis, demonstrating almost no accumulation of cells in the sub-G1 phase, which typically represents apoptotic cells (Figs. 4 and 5).

To explore whether the PI3K/Akt and/or MAPK signaling pathway are involved in the cytocidal effects of arenobufagin and hellebrigenin, the activation of Akt, p38, Erk and JNK was determined in U-87 cells following treatment with various concentrations of arenobufagin and hellebrigenin for 48 h. As shown in Fig. 9, a substantial increase in the expression levels of phospho-p38 MAPK over the endogenous levels was detected in the cells when treated with arenobufagin or hellebrigenin, although almost no alterations in the expression levels of total p38 MAPK expression was observed. Furthermore, similar phenomena were observed in the expression levels of phospho-MAPKAPK2 (MK2), a direct downstream target of p38 MAPK. In contrast, the expression level of phospho-Akt was slightly increased along with a modest decrease in the total Akt expression. Similarly, a modest increase in the expression level of phospho-Erk and total Erk was also observed in the cells, although the highest concentrations of hellebrigenin (40 ng/ml) slightly downregulated the expression of phospho-Erk. These results indicated that the activation of p38 MAPK was involved in the cytocidal effects of arenobufagin and hellebrigenin, although the partial participation of Akt and Erk may not be completely excluded. In addition, the expression level of phospho-JNK was not detected regardless of the presence or absence of each drug, indicating that there was no association between JNK and the cytocidal effects of the two drugs.

To provide further confirm the direct involvement of p38 MAPK in the cytocidal effects of arenobufagin and hellebrigenin, alterations to cell viability were determined in U-87 cells following treatment for 48 h with 20 ng/ml arenobufagin or hellebri-genin in the presence or absence of various concentrations of SB203580, a specific inhibitor of p38 MAPK. Consistent with the IC₅₀ values of each drug (Fig. 2), treatment with 20 ng/ml of arenobufagin or hellebrigenin alone significantly reduced cell viability by ~50% (Fig. 10A–C). Compared with the groups treated with arenobufagin or hellebrigenin alone, combining various concentrations of SB203580 (1, 5 and 10 µM) with each drug significantly enhanced the cytotoxicities of the two drugs (Fig. 10A–C). Although significant reductions in the viability following treatment with each drug in combination with SB202474, a negative control for SB203580, were also observed compared with single drug treatment, the efficacy of each drug was increased more by the addition of SB203580 in comparison with SB202474, indicating the critical role of p38 MAPK. Notably, a dose-dependent decease in the cell viability was observed in cells treated with various concentrations of SB202580 (1, 5 and 10 µM) alone, but not SB202474 alone. In order to evaluate whether there was an association between the activation of the p38 MAPK pathway and other cellular responses, alterations to cell cycle arrest, LDH leakage and DNA fragmentation were further determined in cells. No alteration in LDH leakage (Fig. 10D), G₂/M cell cycle arrest (Fig. 11) or DNA fragmentation (Fig. 8B and C) was observed in the cells treated with a combination of each drug and SB203580 when compared with treatment with each drug alone, indicating no association between the activation of p38 MAPK signaling pathway and these cellular responses.