Trifolirhizin standard preparation (purity >98%) were purchased from Chinese National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China). All other chemicals used were provided from Sigma-Aldrich (St. Louis, MQ, USA). Trifolirhizin was dissolved in DMSO. Cells were treated with increasing concentrations of trifolirhizin for designated hours. Control groups were exposed to the same volumes of DMSO under the same conditions.

MKN45, L02, HEK293 cells were purchased from the Cell Bank of Chinese Academy of Sciences (Shanghai, China). All cells were maintained in RPMI-1640 medium (Gibco, Carlsbad, CA, USA) with 10% fetal calf serum (Gibco) containing antibiotics (100 IU/ml penicillin and 100 IU/ml streptomycin) at 37°C in an atmosphere containing 5% CO₂. The 3 types of cell lines were treated with various concentrations of trifolirhizin for suitable times. Cell viability of control and experiment cells were investigated using the 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyl-2H-tetrazolium bromide (MTT) assay. The absorbance at 570 nm was assessed using a microplate reader (Perkin-Elmer Inc., Waltham, MA, USA).

MKN45 cells (10⁵) were cultured on cover-slide and treated as described above. For Hoechst 33342 staining, cells were added with Hoechst 33342 (5 µg/ml) at 4°C for 10 min in the dark (Beyotime, Haimeng, China). For terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick end labeling (TUNEL) staining (Roche, Basel, Switzerland), TUNEL kit was used to analyze cell DNA fragmentation in apoptotic cells according to the manufacturer's instructions. Trifolirhizin-treated cells after fixing was added by moderate TUNEL and DAPI (Beyotime). Finally, apoptotic cells were observed using an inverted fluorescence microscope (D5100, Nicon, Tokyo, Japan).

MKN45 cell (2–5×10⁵) apoptosis quantification was analyzed according to the Annexin V&PI kit manufacturer's instructions (Selleck Chemicals, Houston, TX, USA). A minimum of 10⁵ trifolirhizin-treated cells were counted for each detection and immediately assessed by FACSCalibur flow cytometry (BD Biosciences, San Jose, CA, USA). Final sample data were reported as percentage of apoptotic cells (sum of early-phase apoptosis at right down quadrant and late-apoptosis at right upper quadrant).

The JC-10 assay kit (AAT Bioquest, CA, USA) was used to assess the change tendency in mitochondrial membrane potential (MMP). The kit was applied according to the manufacturer's instructions. MKN45 cells were incubated in 96-well plate and exposed to trifolirhizin under desired concentrations. JC-10 working solutions (50 µl) were added into each well. After incubation and re-suspending in buffer B, the treated and control MKN45 cell suspensions were measured with an excitation filter of 490 nm while emission filter of 525 nm (green) and another excitation filter of 490 nm while emission filter of 590 nm (red) by a microplate reader (Perkin-Elmer).

PI was employed for cycle analysis (Beyotime). After trifolirhizin treatment at 48 h, MKN45 carcinoma cells were harvested from 6-cm culture plates and subsequently fixed with 70% pre-cooled ethanol. After washing and centrifufation, cell pellets were re-suspended in staining solutions containing RNase A (10 µg/ml) with PI (5 µg/ml) under darkness. Cells treated with trifolirhizin and control were incubated for 30 min and then tested with a FACSCalibur cytometer system (BD Biosciences). Cell cycles were measured using ModFit LT3.2 software.

Western blotting was used to investigate the levels of apoptotic and related signaling pathway proteins. Concisely, cells treated with trifolirhizin and control were lysed with lysis buffer (Beyotime) and centrifuged. Cell proteins were fractionated by 8–12% SDS-PAGE. The proteins were electro-transferred onto the nitrocellulose membranes, using the BCA kit to adjust protein levels (Beyotime). The proteins levels were assessed using primary antibody, including Bcl-2, Bax, PARP, c-Myc, caspase-3, caspase-9, p-JNK, p-ERK, p-P38, JNK, P38, ERK, P53, β-actin (CST, Beverly, MA, USA), followed by incubation with anti-mouse/rat HRP-labeled secondary antibodies. The β-actin antibody was performed as control. The signals were invested by ECL chemiluminescence (BD Biosciences). The tendency of increasing or decreasing of proteins levels were analyzed comparing to control.

The animal study was licensed by the Institutional Animal Care and Use Committee of Fudan University and performed strictly accordance with the Declaration of Helsinki and the Guide for the Care and the Use of Laboratory Animals. The 6-week-old BALB/C nude mice were obtained from Shanghai Slac Laboratory Animal Corporation and maintained in a sterile humidified environment with average temperature at 22°C. The mice were divided into 4 groups: 1 control group and 3 treatment groups (N=6) with increasing trifolirhizin. MKN45 cells were harvested, diluted to desired concentrations (5×10⁶) and injected intraperitoneally into the right flank of each mouse. Then, trifolirhizin was injected into the BALB/C nude mice when tumor volume achieved a mean group size of 100 mm³. The saline was injected into control group. All mouse groups were injected every 3 days. BALB-nu mice were sacrificed at 24 h when tumor volume achieved average group size of 1,000 mm³. MKN45 tumor xenografts were evaluated and fixed for further experiments. Tumor volumes were evaluated as (W² × L)/2 per 3 days using micrometer calipers, and tumor weights were calculated.

The xenografted MKN45 tumor tissue samples were removed and cut to desired small pieces. samples (5 µm) were fixed and stained by cleaved-caspase-3 and Ki67 antibodies, respectively, for immunohistochemistry. We observed the visual images using an inverted fluorescence microscope (D5100, Nicon). The immunohistochemistry positive cells were counted and analyzed using ImageJ Pro 6.0 (National Institutes of Health).

We performed triplicate experiments using independent samples. Results were expressed as mean values ± standard deviation (mean ± SD). Statistical significance of our present results was determined using the one-way ANOVA. All statistical analyses were performed with commercial SPSS software (SPSS19.0, Chicago, IL, USA).

Published reports have manifested that trifolirhizin is cytotoxic on cancer cells. MTT assay was used to assess the numbers of viable MKN45 cells. Exposure under trifolirhizin results in a concentration- and time-dependent suppression of MKN45 cell viability (Fig. 1B). The trifolirhizin exposure time was 48 h in the present examination. The IC₅₀ value of trifolirhizin on MKN45 cells was 33.27±2.06 µg/ml. We also conducted MTT assay to evaluate the efficacy of trifolirhizin on a human normal liver cell line and human normal kidney cell line in vitro. To our satisfaction, >60% tested cells were viable when treated with increasing trifolirhizin dose ≤250 µg/ml. The apparently higher IC₅₀ value of trifolirhizin on two normal human cell lines exhibited more safety and further implication possibility (Fig. 1C).

MKN45 cells were treated with desired concentrations of trifolirhizin as above at 48 h. Hoechst staining exhibited visual images with cell shrinkage, nuclear fragmentation and chromatin compaction of apoptosis (Fig. 4A). Images indicated that trifolirhizin lead to increasing population of apoptotic cells. Consistently, TUNEL staining showed a significant difference between trifolirhizin treated cells vs. the vehicle control, the ratio of green staining (apoptotic) vs blue staining (normal) arise with increasing trifolirhizin (20–40 µg/ml, shown in Fig. 4B). Annexin V/PI assay was performed to evaluate the apoptotic population quantitation after trifolirhizin exposure. In this study, trifolirhizin significantly triggered apoptosis. The apoptotic population in trifolirhizin (20–40 µg/ml) induced MKN45 cells increased (Fig. 2). The JC-10 assay was used to indicate the mitochondrial outer membrane permeabilization after trifolirhizin treatment in early phase. Fig. 1D demonstrated that red fluorescence (normal cells) was decreased while green fluorescence (apoptotic cells) increased with trifolirhizin treatment (20–40 µg/ml) at 24 h, so the upgrade ratio of green/red fluorescence reflected that the MMP (∆Ψm) in MKN45 after trifolirhizin treatment (20–40 µg/ml) was decreased.

The expression levels of the epidermal growth factor receptor (EGFR) signaling pathways and its downstream mitogen activated protein kinase (MAPK) signaling pathway were shared involving multiple anticancer agent molecular activities. The EGFR levels, containing RAS/RAF/MEK/ERK are downregulated with the increasing trifolirhizin, Fig. 4A indicates that trifolirhizin (20–40 µg/ml) triggered the increase of p38 and elicited no impact on p-JNK. Bcl-2 and Bax family members are the critical regulators in mitochondrial apoptosis pathways. Trifolirhizin leads to disorder of expression ratio of Bcl-2 and Bax. The induction of apoptosis by trifolirhizin was subsequently mediated through activities of c-Myc, caspase-9, and caspase-3.

PI was employed to determine the cell cycle distribution under trifolirhizin exposure. Fig. 3 shows that trifolirhizin exposure (20–40 µg/ml) resulted in increasing accumulation in G₂/M phase in MKN45 cells. Furthermore, trifolirhizin exhibited impact on the cell cycle through regulation of cell cycle related proteins. There are an appreciable downregulation of cyclin B as well as Cdc2 in trifolirhizin treatment (20–40 µg/ml). The P53 protein, which is located upstream of cyclin complex and downstream of EGFR signaling pathways, is a key point between signaling pathway and cellular response in cell cycle arrest. Trifolirhizin treatment (20–40 µg/ml) elicited P53 activation (Fig. 5A).

Efficacy of trifolirhizin was also confirmed because there were few changes in the mouse body weight in the experiment (Fig. 1E). These results certified the relative safety of trifolirhizin in vivo. After 21 administration days, xenograft tumors treated with trifolirhizin (1–3 mg/kg) were smaller than control treated with saline (Table I and Fig. 1F). The significantly decreasing levels of Ki67-positive cells were found in the study, which hinted the anti-proliferation effect of trifolirhizin (1–3 mg/kg) in vivo. Cleaved-caspase-3 positive cells were increased which reflected the trifolirhizin apoptotic effect (Fig. 4D). These results also manifested the antitumor effects of trifolirhizin.