Dulbecco's modified Eagle's medium (DMEM), fetal bovine serum (FBS) and the other cell culture reagents were purchased from Gibco (Thermo Fisher Scientific, Inc., Waltham, MA, USA). Protease inhibitor cocktail, dihydroethidium (DHE), RIPA lysis buffer, sulforhodamine B (SRB) and a caspase 3 activity assay kit were obtained from Sigma-Aldrich (Merck KGaA, Darmstadt, Germany). JC-1 was purchased from Cayman Chemical Company (Cayman Chemical Company, Ann Arbor, MI, USA). The primary antibodies against p21^Cip/WAF1 (cat. no. 2947), cyclin D1 (cat. no. 2978), β-actin (cat. no. 4970), and anti-rabbit IgG horseradish peroxidase (HRP)-linked antibody (cat. no. 7074) antibody were purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA). iScript reverse transcription Supermix for reverse transcription-quantitative polymerase chain reaction (RT-qPCR) and SsoFast EvaGreen Supermix were supplied by Bio-Rad Laboratories, Inc. (Hercules, CA, USA).

Edible leaves of CF were collected from Udon Thani Province, Thailand, in May 2014. Identification was performed by the Pharmaceutical Laboratory Service Center, Faculty of Pharmaceutical Science, Prince of Songkla University (Hat Yai, Thailand; specimen no. SKP083030601) was deposited at the Prince of Songkla University Herbarium. Dried leaves were extracted using 50% ethanol, filtered, evaporated and lyophilized to obtain the dry extract. The yield was 12.25% of the starting dry weight of the leaves. The CF leaf extract was maintained at −20°C until use.

The human HepG2 liver cancer cell line was obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA) and maintained according to the recommendations of the ATCC at 37°C and 5% CO₂ in complete DMEM, supplemented with 100 U/ml penicillin G, 100 µg/ml streptomycin and 10% FBS. Subsequent to reaching confluence, the HepG2 cancer cells were detached using 0.25% trypsin-EDTA and 1×10⁶ cells were seeded into the same complete medium. The DMEM medium was replaced every 3 days.

The SRB assay was used to determine the effect of CF on the viability of HepG2 cancer cells, as previously described (20). Briefly, cells were cultured in a 96-well plate for 24 h and fresh medium containing various concentrations of CF (0–500 µg/ml) were added. Subsequent to 24–48 h, cells were fixed with ice-cold 10% trichloroacetic acid at 4°C, stained with 0.4% SRB for 30 min at room temperature, and dissolved with 10 mM Tris base solution. Absorbance was measured at a filter wavelength of 540 nm using a spectrophotometer (Opsys MR™ Microplate Reader; Dynex Technologies, Chantilly, VA, USA).

The colony formation assay was used to determine the effect of CF on the cell regrowth of HepG2 cancer cells, as previously described (20). Briefly, 500 viable HepG2 cancer cells were seeded in 6-well plates (500 cells/well) for 24 h, treated with various concentrations of CF (0–200 µg/ml) for 24 h, washed once with phosphate-buffered saline (PBS) and resuspended in fresh medium. HepG2 cells were grown for another 24 days. Subsequently, the DMEM medium was discarded, the cells were washed with PBS buffer three times, fixed with 100% methanol at −20°C, stained with 0.5% crystal violet in 100% methanol for 1 h at room temperature, washed with tap water, and the colonies were viewed and captured using a digital camera (Nikon D3100). Colonies containing >50 individual cells were counted using Image-Pro Plus software (Media Cybernetics, L.P., Silver Spring, MA, USA).

Cell migration was assessed using a wound healing assay, as previously described (20). Briefly, HepG2 cancer cells were seeded into 24-well plates for 24 h. Cells were scratched using a sterile 0.2-ml pipette tip, certain cells were untreated and others were treated with different concentrations of CF (0–100 µg/ml). Images were obtained from 0 to 48 h. The closing of the scratched wound was determined by image capture of the uncovered area along the scratch. The wound distance was calculated by dividing the area by the length of the scratch, and this was compared with the untreated control group. Cell migration was monitored by phase contrast microscopy (Nikon Eclipse TS100 inverted microscope; magnification, ×10).

Intracellular ROS generation was measured using DHE, the cell-permeable fluorescent probe. Briefly, HepG2 cancer cells were seeded and cultured in black 96-well plates for 24 h. The cells were treated with CF (0–250 µg/ml) plus 25 µM DHE in serum-free medium and maintained at 37°C for 90 min in a 5% CO₂ incubator in the dark. The fluorescence intensity was measured on a fluorescence microplate reader at 518 nm (excitation) and 605 nm (emission). Data were expressed as the percentage of ROS relative to untreated control groups.

Caspase 3 activity was measured using caspase 3 fluorimetric assay kits according to the manufacturer's instructions. Briefly, HepG2 cancer cells were treated with CF (0–100 µg/ml) for 24 h, lysed, and the protein concentrations were measured using Bradford's reagent (Bio-Rad Laboratories, Inc.). The caspase 3 activity reactions were composed of cell lysates and buffer containing the caspase 3 substrate, acetyl Asp-Glu-Val-Asp 7-amido-4-methylcoumarin (AMC; Santa Cruz Biotechnology, Inc., Dallas, TX, USA). These mixtures were incubated at 37°C in the dark for 90 min. The AMC fluorescence intensity was read using a fluorescence plate reader at 360 nm (excitation) and 460 nm (emission). AMC served as a standard to calculate the caspase 3 activity.

JC-1, the lipophilic cationic fluorescent dye, was used to measure changes in ΔΨm. Briefly, HepG2 cancer cells were seeded into black 96-well plates for 24 h, treated with CF (0–100 µg/ml) for 24 h, loaded with JC-1, and incubated for 30 min at 37°C in the dark. Subsequently, the cells were rinsed with PBS buffer and added 200 µl JC1 assay buffer and then read the fluorescent intensity. The ΔΨm was measured using a fluorescence plate reader at 485 nm (excitation) and 535 nm (emission). In normal cells, JC-1 transforms to J-aggregate and in dead cells, JC-1 exists in monomeric form. The fluorescence intensity ratio of J-aggregates to JC-1 monomers served as an indicator of the depolarization of ΔΨm.

The expression levels of matrix metalloproteinase-9 (the MMP-9) in conditioned medium were detected by gelatin zymography. Briefly, HepG2 cancer cells were seeded in 24-well culture plates for 24 h. Subsequently, HepG2 cancer cells were cultured in complete DMEM medium containing different concentrations of CF (0–100 µg/ml) at 37°C for 48 h. The culture medium was then collected, centrifuged at 400 × g for 5 min at 4°C, and the protein concentration was measured using Bradford's reagent. The protein samples were combined with 2X non-reducing sample buffer without heating, loaded onto a 10% SDS-polyacrylamide gel (Bio-Rad Laboratories, Inc.) containing 0.1% (w/v) gelatin and subjected to electrophoresis at 120 V for 1.5 h. Following electrophoresis, the gel was washed three times with 2.5% Triton X-100 to remove the SDS and incubated with developing buffer [50 mM Tris-HCl buffer (pH 7.45) and 10 mM CaCl₂] at 37°C for 12 h. The gels were stained with 0.1% Coomassie Brilliant Blue R-250 for 1 h at room temperature and washed with destaining solution until clear bands were observable against an intensely stained background.

Briefly, the HepG2 cancer cells were seeded in 6-well plates for 24 h and treated with CF for 24 h. RNA was isolated and cDNA was prepared. PCR amplification was performed using specific primers for ras-related C3 botulinum toxin substrate 1 (rho family, small GTP binding protein Rac1) (RAC1) and cyclin dependent kinase 6 (CDK6), and ACTB served as an internal control. The PCR primer sequences are presented in Table I.

RT-qPCR was performed in a final reaction volume of 20 µl containing SYBR-Green PCR Master Mix, the target gene and the internal control, ACTB under the following conditions: Denaturation at 95°C for 3 min; amplification (40 cycles) at 95°C for 15 sec and 60°C for 30 sec. Expression of each gene was monitored using an Applied Biosystems^® StepOne™ real-time PCR system (Thermo Fisher Scientific, Inc.). Differences in gene expression levels were calculated using the 2^−ΔΔCq method for relative quantification, and expressed as the fold change relative to the untreated control (21).

Briefly, HepG2 cells were treated with 100 µg/ml CF for 24 h, lysed with RIPA lysis buffer and centrifuged at 10,000 × g at 4°C for 30 min. The supernatant was collected and the protein concentration was determined. A sample of 20 µg total protein was then separated by 12% SDS-polyacrylamide gel electrophoresis (120 V for 1.5 h) and transferred to PVDF membranes (Immobilon^®; EMD Millipore, Billerica, MA, USA). The membranes were incubated with each primary antibody (p21^Cip/WAF1, cyclin D1 and ACTB), at a dilution of 1:2,500, overnight and with the HRP-conjugated secondary antibody (dilution, 1:5,000) for 2 h. Bands were detected using an enhanced Clarity™ Western ECL Substrate (Bio-Rad Laboratories, Inc.).

Statistical comparison of the control and CF groups was performed using Student's t-test and one-way analysis of variance, followed by Tukey's post-hoc test. The analyses were conducted using SigmaStat software version 3.5 (Systat Software Inc., San Jose, CA, USA) and values are expressed as the mean ± standard error of the mean of three determinations. P<0.05 was considered to indicate a statistically significant difference.

To assess the cytotoxicity of CF on human HepG2 liver cells, the cells were cultured with various concentrations of CF for 24–48 h and the cytotoxicity was determined using the SRB assay (Fig. 1A). The results indicate that cell growth was strongly inhibited in a dose- and time-dependent manner with half maximal inhibitory concentration (IC₅₀) values of 219.03±9.96 µg/ml at 24 h and 124.90±6.86 µg/ml at 48 h (Fig. 1A).

To determine the effect of CF on the replicative potential and longer term viability of HepG2 cells, a colony formation assay was used. CF caused a dose-dependent decline in the colony forming ability of HepG2 cells with IC₅₀ values of 20.06±1.52 µg/ml (Fig. 1B and C). CF inhibited cell regrowth at a lower concentration when compared with the CF concentration that induces cancer cell death. CF exhibited cytotoxic and antiproliferative effects against the HepG2 liver cancer cells.

To determine whether CF inhibits cancer cell migration, a wound healing assay was performed. The results demonstrate that CF significantly inhibited cancer cell migration in a dose-dependent manner. At a concentration of 100 µg/ml, CF inhibited HepG2 cancer cell migration by ~40% when compared with the untreated control group (Fig. 2A and B).

The effects of CF on the expression of invasion-linked matrix metalloproteinase-9 (MMP-9) were then assessed. The expression level of MMP-9 was relatively high in the HepG2 cancer cells in the untreated control group. CF treatment suppressed MMP-9 in a dose-dependent manner, and significantly decreased MMP-9 at a concentration of 100 µg/ml when compared with the untreated control group (Fig. 2C and D).

To establish the mechanism by which CF causes cancer cell death, the intracellular accumulation of ROS was examined using a DHE-enhanced fluorescent probe. Upon treatment of the HepG2 cancer cells with CF, ROS were released in a dose-dependent manner (Fig. 3A). CF induced the production of ROS in a dose-dependent manner when compared with the untreated control group (P<0.05).

In addition, the involvement of mitochondria in CF-induced cytotoxicity was determined. The function of mitochondria was investigated by measuring caspase 3 activities, and mitochondrial dysfunction using the fluorescent dye JC-1. The results indicated that CF treatment activates caspase 3 activities in a dose-dependent manner (Fig. 3B). Furthermore, CF depolarized the ΔΨm, as shown by the decline in the JC-1 aggregates/JC-1 monomers ratio. The lowest concentration at which the decrease in ΔΨm was significant was 50 µg/ml (Fig. 3C).

Whether CF inhibits mRNA expression of the cell cycle regulator, RAC1 and the downstream gene, CDK6 was investigated. CF did not alter the expression levels of either of these genes in the HepG2 cells (Fig. 4A and B). Proteins, p21 and cyclin D1, which are associated with cell survival, were examined by western blot analysis to determine whether CF inhibits HepG2 cell proliferation. The results indicated that CF significantly induces p21 expression to inhibit the cell cycle, which is correlated with a reduction in cyclin D1 protein expression in HepG2 cells (Fig. 4C-E). The present results found that CF causes HepG2 cancer cell death by stimulating the cell cycle protein inhibitor, p21, and reducing expression levels of the cell cycle protein, cyclin D1.