Evaluation of pro-apoptotic potential of taxifolin against liver cancer

Molecular docking and bioinformatics analysis

Auto Dock Vina (Trott & Olson, 2010) and Chimera (Version 1.13) was used to perform Docking of the compound Taxifolin (PubChem ID 439533) against the proteins Akt (PDB ID: 1GZO) , VEGF (PDB ID: 3QTK) and HIF-1α (PDB ID: 4H6j). Docking was performed through Vina using Opal services in Chimera 1.13 by selecting the appropriate options. The hydrogen bonds between the receptor and ligand were visualized and docking scores were retrieved.

Cell culture and treatment

Human hepatocellular carcinoma cells HepG2 and HuH7 and kidney epithelial cells Vero were obtained from National Control Laboratory for Biologicals (National Institute of Health Islamabad). All cell lines were grown in DMEM (Gibco by Life Technologies) supplemented with 10% Fetal Bovine serum (FBS) (Gibco by Life Technologies) and 1% penicillin-streptomycin with 5% CO₂ at 37 °C. Taxifolin (Sigma Aldrich) dissolved in Dimethyl sulfoxide (0.1% (v/v)) was employed for the treatment of HepG2 and Huh7 cells at a confluency of ∼70%. Cells were treated with taxifolin at 0.1 µM, 0.125 µM, 0.15 µM, 0.20 µM and 0.22 µM for 24 h.

Cell viability assay

MTT Assay was used to study the effect of Taxifolin on the viability of HepG2, HuH7 and Vero cell lines. The cells were plated (1 × 104 cells per well) in one mL of complete cell culture medium, containing 0.1 µM, 0.125 µM, and 0.15 µM concentrations of taxifolin in 96 well plates. Cells were incubated at room temperature for 24 h. After incubation, 100 µL of MTT (5 mg/ml:1XPBS) was added to each well and incubated for 2–3 h in humidified atmosphere. DMSO (100 µL) was then added as solubilization solution and again plate was put overnight in humidified incubator. Afterwards, cell culture plates were centrifuged (1,800 × g for 5 min at 4 °C).The absorbance of 550 nM was recorded on the microplate reader. DMSO treated cells were taken as 100% control and the effect of taxifolin on the inhibition of cell growth was calculated as the % cell viability.

Colony formation assay

The colony formation assay (clonogenic assay) was performed in order to assess the effect of treatment on clonogenic survival of liver cancer cells. HepG2 and HuH7 cells were treated with increasing concentrations (0.1 µM, 0.12 µM, and 0.15 µM) of taxifolin in DMEM medium. Following treatment, the cells were re-plated in triplicates on a 12-well tissue culture plate with 3000 cells/well and cultured in 5% CO₂ at 37 °C for 14 days with the growth media being replaced every four days. The cells were stained with 0.5% crystal violet (in methanol: H₂O; 1:1) and the pictures were taken using a digital camera.

Cell migration assay

HepG2 and HuH7 cells were seeded in 6-well cell culture plates and were allowed to spread, attach and form a monolayer for 24 h. After attachment, the cells were scratched with a sterile pipette tip and then incubated. At 80% confluence they were given a treatment of 0.125 µM and 0.15 µM to examine the effect of taxifolin on the migration of HCC. After the next 24 h, the images of the cell movement were captured at regular intervals through LSM 410 microscope (Zeiss, Germany).

RNA extraction and real-time PCR

Total RNA was isolated from the cells by using trizol solution (Wizbio) according to the manufacturer’s instructions. The concentration of RNA was measured spectrophotometrically and cDNA was synthesized, following the manufacturer protocol (wiz Script cDNA synthesis kit). The reaction mixture for qPCR was prepared using 10 µL of Wiz Pure qPCR master mix (SYBR), 6 µM of reverse and forward primers, and 10 µg cDNA with RNAase free water added to a total volume of 20 µL. The amplification was performed for 40 cycles with the following factors: 95 °C incubation for 10 minutes, then 95 °C for 15 s, amplification at 61 °C for 1 min and real time analysis at 75 °C for 45 s. The data was collected at 61 °C and 7,300 system SDS software was used for the analysis of the data. For normalization, beta-actin primers were used. 2-ΔΔCT method was employed for the gene expression quantification. The primer sequence used were Hif1-α Forward: 5′ CAGATCTCGGCGAAGTAAAG 3′, Hif1-α Reverse 5′ TCACAGAGGCCTTATCAAGAT G 3′; VEGF Forward: 5′ CGAGGGCCTGGAGTGTG 3′, VEGF Reverse: 5′ CCGCATAATCTGCATGGTGAT 3′; Akt Forward 5′ TTCTGCAGCTATGCGCAATGTG 3′, Akt Reverse: 5′ TGGCCAGCATACCATAGTGAGGTT 3′.

Protein extraction and Western blot analysis

Protein extraction was performed after the treatment of taxifolin in Huh7 and HepG2 cells. Procedure included the addition of ice-cold lysis buffer (consisting of one mmol/l EDTA, one mmol/l EGTA, 150 mmol/l NaCl, 50 nmol/l Tris–HCl, 20 mmol/l NaF, 1% Triton X-100, 0.5% Nonidet P-40, one mmol/l PMSF and 100 mmol/l Na₃VO₄ with pH 7.4), along with protease inhibitors (Calbiochem, Germany), to the cells and subjected to incubation for 20 min at 0 °C. For immunoblotting, 8–12% poly acrylamide gels were prepared for resolving proteins. Resolved proteins were transferred to nitrocellulose membrane and probed with monoclonal primary antibodies. After incubation with secondary antibodies, proteins of interest were detected through chemiluminescence autoradiography.

DNA ladder assay

DNA was isolated using phenol-chloroform method to estimate the DNA damage. 5 µg of DNA was loaded in 1.5% agarose gel that contained 1.0 µg/ml ethidium bromide (EtBr) including the DNA standards 0.5 µg per well. Electrophoresis was performed for 45 min at 100 volts. The gel was studied under the gel doc system and using digital camera was photographed after observing.

Flow cytometry

Taxifolin-treated HepG2 and Huh7 cells were trypsinized and fixed in Paraformaldehyde:1XPBS (1%) for 60 min. Then, cells were washed with cold PBS twice and centrifuged. Ice chilled 70% ethanol was used to suspend and store over-night the resultant pellet. The cells were again centrifuged (1000 rpm for 5 min) and washing with cold PBS was done to remove ethanol from obtained pellet. Propidium iodide and FITC were used to label the cells through Apo-Direct Kit (BD Pharmagen, CA) as instructed by manufacturer and analyzed through FACScan (Becton Dickinson, NJ). Approximately 10,000 cells per sample were picked and ModFitLT software (Verily Software House, ME) was used to analyze DNA histograms.

Statistical analysis

Statistical analysis was performed using one way ANOVA (multiple comparisons) in graph pad prism and p value less than 0.05 was considered statistically significant.

Table 1:

Docking score and RMSD values of Hif1-α, VEGF and Akt protein docked with ligand taxifolin.

Name of protein	Diagram	Docking score	RMSD upper bond	RMSD lower bond	H bonds all	H bonds ligand atoms	H bonds receptor atoms
Hif1-α		11.6	0.0	0.0	3	3	3
		14.6	1.595	7.169	5	4	4
VEGF		−9.4	0.0	0.0	3	3	2
		−7.9	1.953	6.886	3	2	3
		−7.8	2.219	4.118	2	2	2
Akt		−7.9	0.613	1.429	1	1	1
		−7.2	1.915	2.327	1	1	1
		−7.7	1.497	6.507	1	1	1

DOI: 10.7717/peerj.11276/table-1

Binding interactions of taxifolin with the binding domains of VEGF, Akt and Hif1-α

The hydroxyl group and the oxygen atom on taxifolin resulted in binding specifically with the Cysteine 44 position of VEGF (Fig. 1A). Docking resulted in ten different scores through which the top three hits having maximum number of hydrogen bonds were selected. The score of −9.4, −7.9 and −7.8 was selected. Two hydrogen bonds and high negative docking score depicted that VEGF had the most favorable docking with ligand taxifolin. Akt showed binding affinity to the ligand Taxifolin. This was observed that the hydroxyl group of the ligand showed interaction with the phenylalanine residue of Akt protein (Fig. 1B). Ten possible scores of docking were obtained and the top three hits having maximum hydrogen bonds were selected. The hydroxyl group on the ligand Taxifolin was found to interact with Hif1-α at the TYR325 (tyrosine) by means of hydrogen bond. The ligand docked with the Hif1-α protein with two possible orientations depicting the docking score of 11.6 (RMSD l.b and u.b 0.0) and 14.6 (1.5 and 7.16) respectively. These results showed a favorable binding interaction between our ligand and the protein Hif1-α (Fig. 1C). Table 1 depicts the docking score, RMSD values and the hydrogen bonds that result in the ligand protein interactions values. VEGF, Akt and Hif1-α were shown to exhibit docking scores viz., −9.4, −7.9 and 11.6 respectively.

Inhibitory effect of Taxifolin ongrowth and viability of hepatic carcinoma cell lines

HepG2 and Huh7 cell lines were grown and maintained in DMEM media. The normal kidney cells (Vero cells) were also grown in DMEM media and were seeded continuously for cytotoxicity experiments. HepG2 cells were given the treatment with taxifolin at different concentrations of 0.10 µM, 0.125 µM, and 0.150 µM (Fig. 2). The IC50 for HepG2 was achieved at a concentration of 0.15 µM, where 50% of cancer cells growth was inhibited. Huh7 cells were treated with taxifolin at different concentrations of 0.200 µM and 0.220 µM. The IC50 of taxifolin for Huh7 cells was 0.220 µM. Cell counting was done by trypan blue assay, performed before each cell culture experiment.

In three independent experiments, cell cytotoxicitywas measured as a percentage of corresponding control (value of untreated cells). The toxicity for taxifolin was estimated on Vero cells; MTT assay for Vero was performed with respect to different taxifolin concentrations i.e., 0.1 µM, 0.125 µM, 0.15 µM, 0.17 µM and 0.20 µM for 24 h. A distinctive pattern for IC50 values at the three tested cell lines Vero, Huh7 and HepG2 was observed. Treatment of Vero cell lines with taxifolin did not pose any cytotoxic effect (Fig. 3A).

Since taxifolin is known to exhibit anti-cancer activities, it was investigated whether taxifolin extract would be more efficacious in the inhibition of growth and viability of liver cancer cells. We performed MTT assay against liver cancer cells (Huh7 and HepG2) for the investigation of the anti-proliferative potential of taxifolin and checked the percentage of cancer cells relative to the control cells. It was observed that taxifolin treatment (0.1 µM, 0.125 µM and 0.15 µM, 0.17 µM and 0.20 µM for 24 h) to liver cancer cells caused cell growth inhibition in a dose-dependent manner revealing that both cell lines responded to taxifolin treatment within 24 h. As depicted in Figs. 3B and 3C, the IC50 value of taxifolin treated Huh7 and HepG2 cells was 0.22 µM and 0.150 µM (24 h), respectively. At half maximal inhibitory concentration (IC50) it was observed that the growth of cancer cells was inhibited by 50%. In viable cells, mitochondrial dehydrogenase enzyme resulted in the cellular reduction of MTT; soluble yellow tetrazolium salt to purple colored formazan product formation. Significant decrease in viability of HepG2 was observed in a dose-dependent manner. Each experiment was performed in triplicates and the results of three independent experiments are reported. The percent cell viability of the HepG2 control cells (control; DMSO) was 100%, with a considerable reduction of 25% in the cells treated with a dose of 0.125 µM and 50% in cells at 0.15 µM. Similar trend was observed in Huh7 cells after treatment with varying concentrations of taxifolin, indicating the decline in viability of liver cancer cells upon increasing taxifolin dosage.

Reduction of clonogenic survival and migratory potential of hepatic cancercells after taxifolin treatment

Both cell lines were treated with increasing concentrations of taxifolin (0.1 µM, 0.125 µM, and 0.15 µM). After incubation and treatment of HepG2 and Huh7 cells for 14 days, treated cells exhibited a dose-dependent inhibition in colony formation relative to controls (untreated). Treatment of cell lines with increasing concentration of taxifolin resulted in a decline in cell survival and their reproductive ability to form clone (a large colony), with respect to the control. Figure 4 shows affected clonogenicity of Huh7 and HepG2 cells upon taxifolin treatment in comparison to control. Clonogenicity of liver cancer cells was highest in control. The surviving fraction declined as the concentration was increased to 0.1 µM and even further at 0.125 µM. Finally at the concentration of 0.150 µM for HepG2 and 0.220 µM for Huh7, taxifolin minimized the surviving fraction affirming its cytotoxic role for liver cancer cells (Clonogenicity 0.1 µM > 0.125 µM > 0.150 µM).

Cell scratch assay exhibited that after culturing for 24 h, taxifolin significantly suppressed the migration of HepG2 and Huh7cells. Control cells exhibited a significant potential to migrate having no hindrance in cell motility (Figs. 5A and 5D). In both treated cell lines, taxifolin caused inhibition in the migration of liver cancer cells and reduced the speed of wound healing in a concentration-dependent manner. The results indicated that treatment of HepG2 cells with dose 0.125 µM and 0.15 µM and Huh7 cells with dose 0.20 µM and 0.22 µM slowed the motility and the wound remained unfilled at 24 h. The migration distance decreased significantly upon increasing concentration of taxifolin which depicts that taxifolin reduced the migratory capability of hepatic cells in vitro. The migration ability of HepG2 cells at 0.125 µM concentration of taxifolin was notably decreased (Fig. 5B) whereas further minimized at 0.15 µM (Fig. 5C). Similarly, taxifolin treatment also brought about significant reduction in migration potential of Huh7 cells at 0.20 µM (Fig. 5E) and its increased concentration (0.22 µM) further decreased their migration (Fig. 5F).

Taxifolin induces DNA damage as well as apoptosis in hepatic cancer cells

Taxifolin influence on DNA damage and apoptosis was also assessed. DNA ladder assay showed that DNA damage was present in control as well as the DMSO treated group. Gel electrophoresis was used to examine DNA fragmentation. Control HepG2 cells did not present any DNA fragmentation signifying a high molecular weight DNA (intact DNA) (Fig. 6A). In contrast, the cells treated with taxifolin at 0.125 µM and 0.15 µM for 24 h resulted in fragmented bands (DNA breakage) forming a ladder pattern. This laddering of DNA depicts that DNA damage is a characteristic feature of apoptosis. The DNA fragmentation pattern in the cells treated at 0.125 µM was significantly less prominent than the cells treated at 0.15 µM concentration of taxifolin.

Next, the number of cells going through apoptosis after taxifolin treatment was assessed through flow cytometric analysis. A significant increase in the apoptotic population of cells in both cell lines was observed. In Figs. 6B and 6C, x-axis (FL1-H) refers to cells labeled with Annexin-V while y-axis refers to the cells labeled with propidium iodide (PI). PI-positive cells have heavily injured plasma membrane and are probably dead (shown in upper left quadrant). Annexin-V-positive cells represent cell population in initial phase of apoptosis (shown in lower right quadrant). Cells in late apoptotic phase binds both PI and Annexin-V (shown in upper right quadrant) while healthy cells bind with neither (shown in lower left quadrant). Dose-dependent increase in the ratio of dead cells to live cells was also recorded (Table S1), suggesting a positive correlation between taxifolin concentration and induction of cellular apoptosis.

Taxifolin influence on the expression of apoptotic markers (caspase-3, PCNA and PARP) was also assessed through western blotting (Fig. 6D). Immunoblot analysis revealed that PCNA expression in taxifolin-treated Huh7 and HepG2 cells was remarkably reduced while taxifolin induced PARP cleavage and caspase-3 activation was also seen. Furthermore, taxifolin pro-apoptotic influence was concentration-dependent and maximum reduction in protein levels of truncated-PARP and PCNA and increased activated caspase-3 was observed at 0.150 µM in HepG2 and 0.220 µM in Huh7 cells.

Taxifolin causes downregulation of Hif1-α, VEGF and Aktexpression in hepatic cancer cells

Taxifolin influence on transcription and translation of Hif1- α, VEGF and Akt in HepG2 and Huh7 cells was determined by qPCR and immunoblotting. Both hepatic carcinoma cell lines were treated with taxifolin for 24 h and then assessed for taxifolin effect on Hif1-α, VEGF and Akt mRNA and protein expression level. mRNA expression of Hif1-α was significantly decreased (p < 0.001) in taxifolin treated cells in comparison to control cells. Figure 7A shows the mean fold change normalized to β-actin plotted against the two concentrations of taxifolin. Taxifolin resulted in a 0.6 fold decrease in Hif1-α at 0.125 µM and 1 fold decrease at 0.15 µM concentration. These results confirmed that taxifolin is an effective inhibitor of Hif1-α at post transcriptional level (Fig. 6A). The relative gene expression of VEGF was significantly reduced (p < 0.0187) in taxifolin treated cells as compared to control cells. Taxifolin caused a 0.5 fold decrease in VEGF expression at 0.125 µM and 0.6 folds decrease at 0.15 µM. A significant decrease (p < 0.029) in Akt levels was observed with taxifolin treatment as compared to control. Akt was downregulated by 1 fold at both treatment concentrations of taxifolin.

Immunoblotting results also depicted the reduction of VEGF, Akt and Hif1-α expression at protein level. In cancer cell, signal transduction begins with VEGF binding with its receptor that leads to activation of PI3K/Akt pathway. Akt activation consequently promotes activation and stability of Hif1- α. HepG2 and Huh7 cells treated with taxifolin showed remarkable decrease in the protein concentrations of VEGF, Akt and Hif1- α in dose-dependent manner. Maximum decrease in HepG2 cells was observed at 0.150 µM concentration while in Huh7 cells, maximum reduction was recorded at 0.220 µM (Fig. 7B). This effect of taxifolin is significant for the suppression of cell survival signals and the induction of apoptosis. Our results depicted attenuation in mRNA and protein expression levels of Hif1- α, VEGF and Akt upon taxifolin treatment in HepG2 and Huh7 cells, suggesting taxifolin can be a viable option for curbing cell survival and angiogenesis in hepatic carcinoma.