Genistein induces apoptosis of colon cancer cells by reversal of epithelial-to-mesenchymal via a Notch1/NF-κB/slug/E-cadherin pathway

HT-29 (ATCC number: HTB-38) colon cancer cells (ATCC (American Type Culture Collection), Manassas, VA) were cultured in RPMI-1640 medium (GIBCO) containing 10% FBS (Gibco), 100 U/mL penicillin and 100 U/mL streptomycin, at 37 °C and 5% CO2.

To examine the effects of GEN on proliferation, cells were loaded on 96-well plates for overnight and then changed to medium contained with 25–400 μmol/L GEN (LC Laboratories, Woburn, MA) respectively for another 48 h. To examine the effects of GEN on EMT, overnight monolayers were treated with medium added by GEN (200 μmol/L) and TNF-α (10 ng/mL) (Sigma-Aldrich) respectively for another 48 h. During the treatment, cells were placed in serum-free and antibiotic-free medium.

An inhibitory effect of GEN on proliferation of colon cancer cell lines was evaluated by the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5- diphenyl tetrazolium) assay. HT-29 cells were plated in 96-well plates (5000 cells per well). After incubation for 24 h, various concentrations of GEN were added into each well and each concentration was repeated in five wells. After 48 h incubation, the medium was aspirated and 0.5 μg/mL MTT was added. Cells were incubated at 37 °C for another 4 h and the formazan product was solubilized with dimethylsulfoxide (DMSO). The optical density (OD) of each well was then measured at 570 nm on an enzyme linked immunosorbent assay (ELISA) microplate reader (Multiskan EX, Labsystems, Helsinki, Finland). Each test was performed in triplicate experiments.

The levels of nuclear condensation and fragmentation were observed by means of nucleic acid staining with DAPI (4′,6-diamidino-2-phenylindole) (Solarbio, Beijing, China). Briefly, HT-29 cells were plated in 6-well plates (10⁵ cells per well). After treatment, the cells were washed twice with PBS, and were fixed with methanol (MeOH), acetic acid (HAc) (3:1, v/v) for 10 min at 4 °C. Cells were stained with DAPI (10 mg/mL) for 20 min in the dark, and were then observed under a fluorescence microscope (Olympus BX41, Japan) in less than 15 min.

Acridine orange and ethidium bromide (AO/EB) staining (Solarbio, Beijing, China) was carried out to further prove the cell apoptosis. Briefly, HT-29 cells were plated in 6-well plates (10⁵ cells per well). After treatment, Cells were washed with PBS for three times and then stained with the staining solution containing 100 μg/mL acridine orange and 100 μg/mL ethidium bromide for 20 min at room temperature in the dark. Cells were observed under an inverted fluorescence microscope (Olympus BX41, Japan) after the staining.

Cell invasion assays were performed using a Transwell chamber (8.0 μm, Polycarbonate, Corning, USA). Transwell chambers were precoated with Matrigel (1: 8; BD, Bedford, MA, USA) and exposed to ultraviolet light for 2 h following air-drying at 4 °C. Transwell chambers were then inserted into a 24-well plate containing culture medium with 20% FBS in lower chamber. Cells were starved overnight and then seeded on the upper chamber (1 × 10⁵ cells per well in 0.4% FBS culture medium). After incubation for 24 h, the filter inserts were removed from the wells and the cells on the upper side of the filter were removed using cotton swabs. Cells invaded to the underside of the filter were first fixed with methanol (15 min), and then stained with 2% ethanol containing 0.1% crystal violet powder (15 min). After being dried, the stained cells were enumerated under a microscope (Olympus BX41, Japan).

Briefly, the cell suspension (1 × 10⁵/mL) was inoculated on cover slips which were partitioned previously into the wells of a 6-well plate. After 24 h, HT-29 cells were treated with 200 μmol/mL GEN and 10 ng/mL TNF-α respectively for 48 h. Cells were fixed with 3% formaldehyde in phosphate buffered saline (PBS, pH 7.4) for 20 min, and washed thrice with PBS. Washed cells were permeabilized using 0.2% Triton X-100 and blocked in 2% BSA in PBS. Then cells were washed thrice with PBS, and incubated with the antibody E-cadherin (dilution 1:200) with 2% BSA in PBS at 37 °C for 1 h. The resulting cells were washed thrice with PBS and incubated with fluorescein FITC- labeled polyclonal goat anti-mouse IgG antibody (dilution 1:200) at 37 °C for 1 h. Cells were stained with propidium iodide (DAPI) (Sigma) and scanned by LSCM. All images were acquired using the same intensity and photodetector gain.

Experimental monolayers were washed with serum free media, and then total and fractionated proteins were extracted by cell lysis buffer (Cell Signaling Technology, Danvers, MA). The lysates were centrifuged at 12,000×g for 20 min at 4 °C. Equal amounts of protein, after concentration was determined by the Bradford assay (Bio-Rad, Hercules, CA), were loaded on SDS-PAGE and transferred to nitrocellulose membranes (Bio-Rad). After blocking, specific antibodies such as Bax, caspase-3, caspase-8, Bcl-2, PI3K, Notch1, p-NF-κB, NF-κB, E-cadherin, N-cadherin and β-actin from AB clonal Biotechnology Co., Ltd. (Wuhan, China) were used to perform detection. Finally each protein was detected using an enhanced chemilumi-nescence system (GE Healthcare, USA). Blot images were digitized (Chemidoc, Bio-Rad, Milan, Italy) and the area of each band was quantified using the computerized imaging system (QuantityOne, Bio-Rad). Relative optical density (arbitrary units) was normalized for control bands in each series and for protein loading (as probed by anti-actin blots). Each test was performed in triplicate experiments.

Total cellular RNA was extracted using the trizol reagent (TransGen Biotech, Beijing, China) according to manufacturer’s instructions. One microgram of total RNA was reverse transcribed at 42 °C for 50 min using a TransScript First-Strand cDNA Synthesis SuperMix according to manufacturer’s instructions (TransGen Biotech, Beijing, China). PCR was then performed using Taq (TaKaRa, Shiga, Japan) polymerase. Each amplification was performed for 35 cycles, one cycle profile consisted of denaturation at 94 °C for 30 s, annealing at 55 °C for 30 s and extension at 72 °C for 120 s. PCR products were visualized by eletrophoresis through 1.2% agarose gels and quantifed with Glyko Bandscan gel analyzing software (Glyko, Novato, CA, USA). Parallel reactions were run using human GAPDH as a control for RT–PCR. The primer sequences that used for RT-PCR of slug, twist1, zeb1, zeb2, foxc1, foxc2 [24‐27] and GAPDH were shown in Table 1.

HT-29 cells were cultured with the indicated concentrations of GEN for 48 h, and cell viability was determined by MTT assay. The result showed that GEN inhibited the growth of HT-29 cells in a dose-dependent manner, with the best inhibition at 48 h in a concentration of 200 μmol/L (Fig. 1a). And the dose-dependent increase of inhibition ratio were suppressed when the concentration of GEN were over 200 μmol/L. The DAPI and AO/EB staining suggested that cell apoptosis can be significantly induced by GEN at 48 h with a concentration of 200 μmol/L (Fig. 1b). The cell-cycle phase distribution and the ratios of apoptotic cells were further determined by flow cytometry with PI staining. The percentage of cells in G1, S and G2/M phase was evaluated using Multi-cycle software, respectively. The results showed a significant increase of GEN treated cells in the G0/G1 phase from 44.60 ± 3.32% to 58.51 ± 9.20 (p < 0.05) compared with control, and the apoptotic rate increased significantly (p < 0.05) from 2.49 ± 0.16% to 21.50 ± 8.50% (Fig. 1c). These data indicated that GEN can induce apoptosis of HT-29 cells significantly at 200 μmol/L for 48 h. Thus, all the treatments of cells in the following experiment were carried out with 48 h of 200 μmol/L GEN.

GEN was confirmed in this paper inhibit proliferation and induce apoptosis of colon cancer cells in vitro. We next characterize the effect of GEN on cell invasion in HT-29 cells by transwell chamber assay with TNF-α treatment as a positive control, since TNF-α has been proved by several research can induce EMT of kinds of cancer cells [7, 12, 28]. The results showed that few cells moved into the lower chamber of the control group, and there was fewer cells moved into the lower chamber of the GEN group, and there was significant decrease compared with control group (n = 3, p < 0.05), while the number of cells that moved into the lower chamber of the TNF-α was significantly higher than that of the control group and GEN group (p < 0.01). These results indicated that the invasion ability of the HT-29 cells in GEN group was significantly reduced than in the control and TNF-α group (Fig. 2a and b).

To further characterize the reversal of EMT induced by GEN, we analyzed the effect of GEN on EMT-related markers, E-cadherin and N-cadherin, using immunofluorescence staining and western blot assay. The Immunofluorescence results showed that the intensity of E-cadherin signal treated by GEN was obviously stronger than that of the TNF-α and control group (Fig. 3a). And the percentage of E-cadherin positive cells was 79.41 ± 12.59% in GEN group, significantly higher (p < 0.05) than 19.37 ± 2.94% in control group and 7.56 ± 2.50% in TNF-α group (Fig. 3b). Western-blot results further showed that TNF-α significantly reduced the E-cadherin expression (p < 0.05) and increased the expression of N-cadherin (p < 0.05) which suggested a positive effect of TNF-α on EMT (Fig. 3c). However, in parallel with the marked increase in the E-cadherin expression (p < 0.01), GEN significantly decreased the expression of N-cadherin (p < 0.01) within 48 h (Fig. 3c). These data suggested that GEN can reverse the EMT of HT-29 cells.

In addition to the changes of EMT markers, the mRNA expressions of invasion-related genes in the cells were also evaluated using RT–PCR assay. The results showed that GEN significantly decreased the mRNA expression of slug, zeb1, zeb2, foxc-1, foxc-2 and twist1 in HT-29 cells (p < 0.05) (Fig. 4). While TNF-α significantly increased the mRNA expression of zeb1. Marked increases of the mRNA expression of slug, zeb2 and twist1 didn’t found, and fox-1 and fox-2 mRNA expression were even lower than control group. But all these mRNA expressions were significantly higher than GEN group (p < 0.05) (Fig. 4). These data demonstrates that GEN can significantly inhibit mRNA expression of invasion-related genes in HT-29 cells.

NF-κB has been found represses E-cadherin expression and enhances EMT of several kinds of cancer cells [29‐31]. TNF-α can induce the EMT via the NF-κB pathway [32]. We found in this paper, GEN significantly down-regulated the expression of both NF-κB p65 and p-NF-κB p65 (p < 0.05) (Fig. 5). However, exposure to TNF-α resulted in remarkable increase of NF-κB p65 and p-NF-κB p65 (p < 0.05) (Fig. 5). These results suggested that GEN can reverse EMT through NF-κB pathway in HT-29 cells.

In addition, we found that GEN significantly inhibited the expression of both notch-1 (p < 0.05) (Fig. 6). TNF-α significantly reduced the level of notch-1 expression, however, the level was significantly higher when compared with GEN treatment (p < 0.05) (Fig. 6). It has been confirmed that the genes such as anti-apoptotic (B-cell lymphoma-2, Bcl-2) and pro-apoptotic (Bax) are important regulators of apoptosis in colon cancer cell lines [33‐35]. And Caspases play a central role in apoptosis-induction [36]. Here our results showed that GEN significantly increase the expression of all the proteins including Bcl-2/Bax, Caspases-8 and Caspases-3 (P < 0.05) (Fig. 6). TNF-α was found also increase the expression of these proteins (p < 0.05), while the levels were lower than GEN conditions except Caspases-3 which no significant difference was found between the two treatments (Fig. 6).