Dulbecco’s Modified Eagle’s Medium (DMEM), fetal bovine serum (FBS), Trypsin-EDTA, and antibiotics such as penicillin (10,000 U/mL)/streptomycin (10,000 $\mu$g/mL)/amphotericin (2500 $\mu$g/mL) reagents for cell culture were obtained from Life Technologies (Paisley, Scotland). MTT reagent, gelatin, agarose, RIPA lysis buffer and other reagents are obtained from Sigma Chemical Co., Ltd. (St. Louis, MO, USA). Silibinin samples were purchased from Sigma Chemical Co. (St. Louis, MO, USA).

HT1080 cell line (ATCC No.CCL-12, Homo sapiens, fibroblast, lung) and IMR90 cell line (ATCC No.CCL-186, Homo sapiens, fibroblast) purchased from ATCC (American Type Culture Collection) were cultured using DMEM containing 10% of FBS and subcultured with trypsin-EDTA. Antibiotics such as penicillin/amphotericin/streptomycin were used to prevent cell culture from bacterial contamination. Mycoplasma testing was performed to authenticate the cell lines used in this study. This was accomplished using the MycoAlert™ Mycoplasma Detection Kit (Lonza, Bend, OR, USA), which detects enzymatic activity associated with viable mycoplasma in cell cultures. Briefly, cells were harvested and lysed, and the resulting lysate was incubated with the MycoAlert™ substrate for 10 minutes at 37 °C. The fluorescence of each sample was then measured to check mycoplasma contamination. The silibinin was freshly dissolved in dimethyl sulfoxide (DMSO) as a solvent before use. Silibinin at 2.5, 5, 10, 15, 20, and 25 $\mu$M was used in all experiments. All doses, including blank and controls, were adjusted at 0.1% DMSO, which is the concentration contained in the highest silibinin dose used, and shown to have no cytotoxic effects.

Silibinin (500 $\mu$L) at 2.5, 5, 10, 15, 20, and 25 $\mu$M was reacted with DPPH (2,2-diphenyl-1-picrylhydrazyl) solution at 0.15 $\mu$M (500 $\mu$L), for 1 h at 25 ℃. Vitamin C at 0.01% was used as a positive control. The optical density (OD) of the product was measured at 532 nm using a visible spectrophotometer (SpectraMax M3, Molecular Devices). The amount of DPPH-generated radicals was represented as % [(OD of silibinin treatment group/OD of the blank group) $\times$100]. Blank means the group containing DPPH spontaneously generated radicals and DMSO without silibinin.

Different doses of silibinin (6 $\mu$L) at 2.5, 5, 10, 15, 20, and 25 $\mu$M, 1% potassium ferricyanide (200 $\mu$L), distilled water (194 $\mu$L), and 200 mM phosphate buffer (200 $\mu$L) at pH6.6 were reacted in a microtube. After incubation for 20 min at 50 ℃, trichloroacetic acid solution at 10% (200 $\mu$L) was added to the reaction product. After centrifugation at 2000 g for 10 min, the supernatant (250 $\mu$L) was mixed with distilled water (250 $\mu$L). Next, after ferric chloride (50 $\mu$L) was added, the optical density was measured at 700 nm using a UV spectrophotometer (SpectraMaxM3, Molecular Devices, San Jose, CA, USA). The blank group contained ferricyanide, FeCl${}_3$, and DMSO without silibinin, and the positive control contained 0.001% of vitamin C. The level of antioxidant activity as a reducing power was displayed as % [(OD of silibinin treatment group/OD of the blank group) $\times$100].

The growth inhibitory effect of silibinin on HT1080 cells was evaluated using 3-(4,5-Dimethyl-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) [5]. The cells at a density of 5 $\times$ 10${}^3$ cells/well were inoculated into 96-well plates. After treatment with silibinin at 2.5, 5, 10, 15, 20, and 25 $\mu$M for 24 h, 20 $\mu$L of MTT (5 mg/mL) were added to each well for 4 h. DMSO (150 $\mu$L) was added to solubilize the formazan salts and measured the OD at 570 nm using a visible spectrophotometer (SpectraMax M3, Molecular Devices). Relative survival of cells was represented as a % compared to the blank group [(OD of the silibinin treatment group/OD of the blank group) $\times$ 100].

The activities of MMP-2 and MMP-9 were examined using gelatin zymography according to a previous study [6]. HT1080 cells were cultured in the presence of silibinin at 2.5, 5, 10, 15, 20, and 25 $\mu$M for 1 h. Then, phorbol myristate acetate (PMA) at 1 ng/mL was added for 3 days to induce the expression of MMP. The conditioned medium collected, cleared by centrifugation, and used for the analysis of gelatin zymography. The bands representing MMPs activity were detected as clear zones, and the degree of the bands’ intensity was measured with Davinch-ChemiTM. The MMP enzyme activity was expressed as % in comparison to the blank group [(Silibinin treatment group / blank group $\times$ 100)].

Western blot analysis was carried out according to standard procedures. HT1080 cells were exposed to silibinin at 2.5, 5, 10, 15, 20, and 25 $\mu$M for 1 h. Then, they were stimulated using 1 ng/mL PMA for 24 h in the presence of silibinin. Next, cell lysis was performed with RIPA lysis buffer. The proteins from cell lysates were transferred from a 10% polyacrylamide gel to a nitrocellulose membrane. Thereafter, the membrane was treated with 5% of skim milk. Next, after the treatment of primary antibodies against the target protein such as MMP-2, TIMP-1, p-JNK, ERK-1/2, IL-1$\beta$ and $\beta$-actin, secondary antibody treatment was performed. Target proteins were determined using a chemiluminescent ECL kit (Amersham Pharmacia Biotech, Piscataway, NJ, USA). The degree of band intensity was analyzed with a LAS3000 ® image analyzer (Fujifilm Life Science, Tokyo, Japan).

HT1080 cells were cultured in a slide chamber at 37 °C for 24 hours. After treatment with each concentration of silibinin, for 1 h, PMA at 1 ng/mL was added and incubated for 24 h. After fixing with 10% formalin for 15 min, the cells were permeabilized with phosphate-buffered saline (PBS) containing 0.5% of Tween 20 (PBS T-20) for 30 min and washed 3 times with 0.1% PBS T-20. After blocking with 5% of donkey normal serum, primary antibodies (MMP-2) were added for 2 h. Then, after washing 3 times for 5 min each with 0.1% PBS T-20, secondary antibodies (donkey anti-goat conjugated CY3, donkey anti-mouse conjugated CY3, donkey anti-rabbit conjugated FITC) were added for 1 h. Then, after washing them, the slides were exposed to DAPI reagent for nuclei staining and observed with the iRiS Digital Cell Imaging System (Logos Biosystems, Gyeonggi-do, Korea). All reagents were purchased from Sigma Chemical Co. (St. Louis, MO, USA).

The invasion of HT1080 cells was carried out in accordance with the Chemicon® methodology. The invasion chamber from the Cell Invasion Assay Kit (ECM550) consists of a 24-well tissue culture plate and 12 cell culture inserts containing polycarbonate membrane (8.0 $\mu$m pore size), over which a thin layer of ECMatrix solution is applied. HT1080 cells in 300 $\mu$L of serum-free media were added to each insert and 500 $\mu$L of media containing 10% fetal bovine serum (chemoattractant) was added to the lower chamber. After the cells adhered to insert, they were treated with each concentration of silibinin and stimulated with 1 ng/mL PMA. Chambers were then incubated in 5% CO${}_2$ and at 37 °C for 72 hours. Non-invading cells as well as the ECM gel layer were removed using a cotton-tipped swab and washed with PBS. On the other side of the filter, invasive cells on the lower surface of the membrane were stained by dipping inserts in the staining solution for 20 min. After washing them with PBS, the stained cells were dissolved in 10% acetic acid. Then, the optical density was measured at 560 nm using a UV spectrophotometer (SpectraMaxM3, Molecular Devices). Relative invasion of cells was represented as a % compared to the blank group [(OD of the silibinin treatment group/OD of the blank group) $\times$ 100].

Data were analyzed using ANOVA and post hoc (Duncan) test as means of values $\pm$ SD from three independent experiments (*, p 0.05, **, p 0.01 and ***, p 0.001).

The antioxidant activity of silibinin was examined using the DPPH radical scavenging assay and the reducing power assay. Vitamin C (Vit. C) at 100 $\mu$g/mL, used as a positive control, decreased by 65% of the radicals generated by DPPH, whereas silibinin showed no radical scavenging activity (Fig. 1A). The difference between blank group and silibinin treatment groups above 10 $\mu$M was significantly observed (Fig. 1B). When looking at the reducing power (Fig. 1B), vitamin C at 10 $\mu$g/mL (the highest dose of 100 $\mu$g/mL gave results out of scale) showed a 300% increase over the blank, and silibinin at doses of 10 $\mu$M and higher showed a modest but significant activity, which reached a 49% increase at 25 $\mu$M. These results indicate that silibinin has no radical scavenging activity, although being endowed with some reducing power. Therefore, its efficacy as an antioxidant is quite low.

Fig. 1.

Antioxidant activity of silibinin. (A) The scavenging effect of silibinin on DPPH radicals is shown in this experiment. Vitamin C was used as positive control at 100 $\mu$g/mL. (B) Reducing power of silibinin. Vitamin C was used as a positive control at 10 $\mu$g/mL. Data are shown as mean $\pm$ SD from three independent experiments, all run in triplicates. The level of significance between blank and silibinin treatment was evaluated statistically (*, p 0.05; ***, p 0.001) using ANOVA and post hoc (Duncan) test.

The effect of silibinin on cell viability was examined in HT1080 and IMR-90 cells. In HT1080, silibinin at low dose (2.5 $\mu$M and 5 $\mu$M) resulted in a 37% and 38% increase in cell growth, respectively. On the contrary, at the higher doses (20 $\mu$M and 25 $\mu$M), silibinin induced a 26% and 29% reduction in cell viability, respectively (Fig. 2A). Silibinin effects on IMR-90 cells resulted in a milder growth inhibition (around 20%) at all doses, with no evident dose-effect. Moreover, the growth inhibitory effect of silibinin above 20 $\mu$M on HT1080 cells was significantly higher than that on IMR-90 cells, indicating some specificity of the effect, likely due to a different growth control in tumor versus normal cells.

Fig. 2.

Effect of silibinin on cell viability. The effects of silibinin on cell viability were examined in HT1080 cells (A) and IMR-90 cells (B), respectively. The cells were treated with silibinin at 2.5, 5, 10, 15, 20, and 25 $\mu$M. Cell viability was examined by MTT assay after 24 h treatment. Data show mean $\pm$ SD from three independent experiments, each one run in triplicate. The significance level between blank and silibinin treatment was evaluated statistically (**, p 0.01; ***, p 0.001) using ANOVA and post hoc (Duncan) test.

The effect of silibinin on MMPs activity was examined in HT1080 cells. Silibinin at low doses had no inhibitory effect, and only at the two higher doses of 20 $\mu$M and 25 $\mu$M reduced MMP-9 activity by 188% and 943% compared to the PMA group, respectively (Fig. 3). Silibinin effects on MMP2 were quite different, because there was a stimulation of its activity at the lower doses (5-10-15 $\mu$M), whilst at only the highest concentration of 25 $\mu$M it reduced MMP-2 activity by 147%, compared to the PMA group.

Fig. 3.

Effects of silibinin on MMPs activation. The inhibitory effects of silibinin on the inhibitory activities of MMP-9 and MMP-2 was evaluated in PMA-stimulated HT1080 cells to induce MMPs expression. The silibinin at 2.5, 5, 10, 15, 20, and 25 $\mu$M was added under serum-free conditions for 72 h. MMP-9 and MMP-2 activities were analyzed using gelatin zymography assay. Data display means values $\pm$ SD from triplicate experiments. The significance level between PMA groups and silibinin treatment groups was determined statistically (${}^{\mathrm{\#}\mathrm{\#}\mathrm{\#}}$, p 0.001 for the activity increase of MMPs, ***, p 0.001 for the activity decrease of MMPs) using ANOVA and post hoc (Duncan) test.

The expressions of MAPK and other mediators regulating MMPs were examined to elucidate how silibinin influences the regulation of MMP-2 and MMP-9 expression and activity. The expression of ERK-1/2, p-p38, p-JNK (Fig. 4A), MMP-2, TIMP-1, and IL-1$\beta$ (Fig. 4B) were analyzed in PMA-stimulated cells with or without silibinin presence, by using western blot. The expression of ERK-1/2 was diminished by 27% only in the presence of the highest dose (25 $\mu$M) of silibinin, whereas p-p38 expression was inhibited at all doses tested, with values between 23% and 50%. In particular, the expression level of p-JNK was significantly higher than in the PMA-treated control group in the presence of silibinin at 10 $\mu$M, with values of 28%. Silibinin at all doses progressively inhibited the expression of MMP2, TIMP1 and IL-1$\beta$ and at the highest dose of 25 $\mu$M reduced the expression level of MMP-2 by 64%, TIMP-1 by 149% and IL-1$\beta$ by 91% compared to the PMA-stimulated group.

Fig. 4.

The effect of silibinin on the expression of proteins associated with invasion and metastasis in HT1080 cells. (A) Effects of silibinin on the expressions of ERK-1/2, p-p38, and p-JNK. (B) Effects of silibinin on the expressions of MMP-2, TIMP-1, and IL-1$\beta$. The level of protein expression in cell lysates was determined by western blot analysis using the indicated antibodies. Target proteins were normalized using the expression of $\beta$-actin as housekeeping reference. Data represent means $\pm$ SD from triplicate experiments. The level of significance between PMA groups and silibinin treatment groups was identified statistically (**p 0.01; ***, p 0.001) using ANOVA and post hoc (Duncan) test.

To investigate the effect of silibinin on the expression of metastasis-related proteins, immunofluorescence staining of MMP-2 was performed in PMA-stimulated HT1080 cells, with or without silibinin treatment. The cell’s nuclei were observed in blue color after being stained with DAPI. MMP-2 were stained with CY3 and displayed in red color. Silibinin treatment at 25 $\mu$M decreased the degrees of MMP-2 compared to the PMA-stimulated group (Fig. 5), thus confirming that silibinin at 25 $\mu$M could reduce MMP2 expression.

Fig. 5.

The immunofluorescence staining analysis of MMP-2. HT1080 cells were treated with silibinin at 2.5, 5, 10, 15, 20, and 25 $\mu$M in the presence of PMA. Cells were detected with specific antibodies and counterstained with DAPI. The cells were observed at 200$\times$ of magnification using the iRiS™ Digital Cell Imaging System.

Tumor cells degrade collagen in the extracellular matrix to obtain more nutrients and make space to move into other tissues through blood vessels. Therefore, in this study, an invasion assay was performed using HT1080 cells stimulated with PMA, in order to evaluate the efficacy of silibinin in the inhibition of cell invasion. Paradoxically, though in line with data on cell growth and effects on MMP2, silibinin treatment at low doses promoted invasion, while at concentrations above 10 $\mu$M progressively diminished cell invasion (Fig. 6). Silibinin at 25 $\mu$M inhibited cell invasion by 73% compared to PMA-stimulated cells, thus supporting the effect of silibinin as inhibitor of cancer metastasis.

Fig. 6.

Effects of silibinin on HT1080 cell invasion. Cell penetration into the ECM layer through the polycarbonate membrane was examined in the presence of silibinin. Using a 24-well tissue culture plate with an insert and a polycarbonate membrane with an 8 $\mu$m pore size, an invasion assay was carried out in the invasion chamber. To treat HT1080 cells, silibinin was used at concentrations of 2.5, 5, 10, 15, and 25 M. Data display means values $\pm$ SD from triplicate experiments. The level of significance between PMA groups and silibinin treatment groups was identified (***, p 0.001) using ANOVA and post hoc (Duncan) test.