High performance liquid chromatography (HPLC)-grade methanol was purchased from Sigma-Aldrich (Merck KGaA, Darmstadt, Germany). Ultrapure water was prepared using a Millipore SAS 67120 system (Merck KGaA). The sulforaphane (98% purity), as the reference standard substance, was purchased from Shanghai Yuanye Biotechnology Co., Ltd., (Shanghai, China). Fetal bovine serum (FBS), penicillin G, streptomycin and amphotericin B were obtained from Gibco (Thermo Fisher Scientific, Inc., Waltham, USA). Dimethyl sulfoxide, ribonuclease (RNase), propidium iodide (PI) and RPMI-1640 were purchased from Sigma-Aldrich (Merck KGaA). Broccoli and green cabbage material were obtained from Infinitus Company Ltd. (Guangzhou, China) and were placed on dry ice and freeze-dried immediately to preserve their freshness.

A Shimadzu LC-20AT HPLC system with an UV detector was used (Shimadzu Corporation, Kyoto, Japan) for quantitative determination. A Phenomenex Luna C18 column (4.6×250 mm, 5 µm; Guangzhou FLM Scientific Instrument Co., Ltd., Guangzhou, China) was used, according to the manufacturer's protocols, and the mobile phase composed of methanol:water (20:80% v/v) at a flow rate of 1.0 ml/min. Furthermore, the detection wavelength was 225 nm and the temperature of the column was set to 30°C. The injection volume was 20 ml. The limit of detection was 0.2 µg/ml. Data acquisition was performed using the LabSolutions CS software version 2.53 (Shimadzu Corporation). There are 6 points in the Standard curve (6 standard samples); additionally, a representative standard sample and tested sample were used. All experiment were repeated twice each one in duplicate.

SYD was prepared according to the procedure described by Rose et al (7). In brief, 100 mg of freeze-dried material, including broccoli and green cabbage at a weight ratio of 1:1, was weighed into a 50-ml polypropylene tube and subjected to two 30 min cycles of ultrasonic disruption (59 kHz) in 70% ethanol (3 ml) at 70°C. The mixture was cooled to room temperature and centrifuged at 3,000 × g for 5 min. Following centrifugation, 1-ml aliquots were removed and condensed in a vacuum to 200 µl. The resulting concentrates were filtered through sterile non-pyrogenic filters (٠.٢ µm; Merck KGaA) and stored at 70°C prior to testing. The extracts for each sample yielded an equivalent concentration of 50 mg/ml. The sulforaphane composition in the extracts was analyzed with a Shimadzu LC-20AT HPLC system with an UV detector (Shimadzu Corporation, Kyoto, Japan).

Following ethical approval by the Institutional Animal Care Committee and the Local Veterinary Office and Ethics Committee of Guangzhou Medical University (Guangzhou, China), the human colon cancer HT-29, LS-174-T and CRL-1790 cell lines were obtained from the Clinic Research Center of Guangzhou Medical University (Guangzhou, China). They were used for the subsequent proliferation, apoptosis, cell cycle and western-blotting assays. The human colon epithelial CRL-1790 cell line was obtained from American Type Culture Collection (Manassas, VA, USA). HT-29, LS-174-T and CRL-1790 cells were grown in Dulbecco's modified Eagle's medium (DMEM; Gibco; Thermo Fisher Scientific, Inc.) with 10% FBS and 1% Penicillin G-Streptomycin and maintained at 37°C with 5% CO₂ atmosphere. All cell studies were approved by the Institutional Animal Care Committee and the Local Veterinary Office and Ethics Committee of Guangzhou Medical University (Guangzhou, China).

Proliferation was assessed through an MTS assay kit (Promega Corporation, Madison, WI, USA), according to the manufacturer's protocols. HT-29, LS-174-T and CRL-1790 cells were inoculated in 96-well plates at the density of 5×10⁴ cells/well. At 0, 24, 48 and 72 h of incubation at 37°C, 20 µl MTS/PMS reagent was added for another 4 h of incubation at room temperature. Dimethyl sulfoxide was used to dissolve the purple formazan. The absorbance at 490 nm was detected by a microplate reader. HT-29 cells were treated by 0, 10, 50, 100 or 200 µg/ml SYD for 48 h at 37°C, and then lactate dehydrogenase (LDH) release was assessed with a LDH activity kit (cat. no. 03002209122; Roche Diagnostics, Indianapolis, IN, USA) and expressed as a percentage [(sample absorbance/lysed cell absorbance-control absorbance) ×100]. Each individual experiment was performed in triplicate.

A total of 5×10⁴−1×10⁵ control small interfering RNA (siRNA) or c-FLIP siRNA HT-29 cells/well were seeded in 6-well plates. After 24 h at 37°C, the DMEM with 10% FBS was replaced with fresh DMEM with 10% FBS and 0, 100 or 200 µg/ml SYD at 37°C and incubated for 7 days. Cell viability was determined by Typan Blue staining at room temperature for 30 sec and direct cell counting using a hematocytometer (Thorlabs, Inc., Newton, NJ, USA).

The flow cytometric analysis was performed as described previously (12). Subconfluent cultures of HT-29 cells (1.0×10⁵ cells/well) were treated with either PBS or 0, 100, 200 and 400 µg/ml SYD. Subsequent to 24, 48 and 72 h of treatment at room temperature, the cells were harvested by brief trypsinization and centrifugation at room temperature at 3,000 × g for 5 min. The cell pellets were washed twice with ice-cold PBS and 0.5×10⁶ cells were suspended in 500 ml of saponin/PI solution [0.3% (w/v) saponin; 25 µg/ml PI; 0.1 mmol/L EDTA and 10 µg/ml RNase A in PBS] and incubated at ٤°C for 24 h in the dark. Apoptosis was quantified using an annexin V-fluorescein isothiocyanate (FITC) kit (R&D Systems, Inc., Minneapolis, MN, USA), according to the manufacturer's protocols. The stained cells were analyzed using a FACS flow cytometer (BD Biosciences, Franklin Lakes, NJ, USA) and FCS Express V3.0 software (De Novo Software, Glendale, CA, USA). ModFit LT software v3.3 (Verity Software House, Inc., Topsham, ME, USA) was used to analyze the results of the cell cycle.

HT-29 cells (1×10⁶) were seeded into 6 cm dishes and treated with 0, 100, 200 and 400 µg/ml SYD for the indicated times, the activities of caspase-3 (CASP3), CASP8 and CASP9 were measured by cleavage of a 5 µl IETD-ρNA substrate (caspase-8) or DEVD-ρNA substrate (caspase-3, 9). CAS activity was measured using the Colorimetric CaspACE assay system, and the activities of CASP8 (cat. no. AAH-APO-١-٢) and CASP9 (cat. no. 68FL-Casp9-S100) were determined with CASP8 and CASP9 Colorimetric assay kits, according to the manufacturer's protocols (RayBiotech Life, Norcross, GA, USA). To inhibit the tested caspases, the cells were preincubated at 37°C for 2 h with 50 mM Z-DEVD-FMK (cat. no. A1920; CASP3 inhibitor), Z-IETD-FMK (cat. no. B3232; CASP8 inhibitor), Z-LEHD-FMK (cat. no. B3233; CASP9 inhibitor) and Z-VAD-FMK (cat. no. A1902; Pan-caspase inhibitor) (all from ApexBio, Shanghai, China).

Measurement of the release of cytochrome c. The release of cytochrome c from the mitochondria of HT-29 cells into the cytosol was measured using a streptavidin-peroxidase immunohistochemical kit (cat. no. SP0041; OriGene Technologies, Inc., Beijing, China) as described previously (13). Briefly, following incubation at 37°C for 24 h, HT-29 cells (1×10⁶) were seeded into 6 cm dishes and were exposed to either 100 µg/ml or 200 µg/ml SYD or left untreated (control group). Subsequent to an additional ١٢ h incubation at 37°C, the cells were sequentially treated with 3% H₂O₂, blocking buffer (provided in the kit), incubated overnight at 37°C with the primary antibody (1:100; anti-cytochrome c; cat. no. KG22230; Nanjing KeyGen Biotech Co., Ltd., Nanjing, China). All sections were stained using the streptavidin-horseradish peroxidase complex method with goat anti-rabbit antibodies (1:1,000; provided in the kit) as the secondary antibody at 37°C for 30 min, and then 4°C overnight. Brown-yellow granules in the cytoplasm represented a positive staining.

The HT-29 cells incubated with different does of SYD were dyed with rhodamine 123 for 1 h at room temperature. The primary antibody (1:100; anti-cytochrome c; cat. no. KG22230) was incubated for 1h and then incubated with goat anti-rabbit secondary antibodies with APC fluorescence (cat. no. A-11034; Invitrogen; Thermo Fisher Scientific, Inc.) at 37°C for 30 min. Finally, DAPI (cat. no. 40727ES10; Nanjing KeyGen Biotech Co., Ltd.) was directly nucleated for 30 min at room temperature. The release of cytochrome c was observed under a light microscope (Olympus CX31-LV320 Olympus (China) Co., Ltd., Shanghai, China) subsequent to color development (×400).

Western blot analyses were performed as described previously (13) using the following antibodies: Major histocompatibility complex, Class II, DR Beta 4 (DR4; cat. no., 42533S; 1:1,000), major histocompatibility complex, Class II, DR Beta 5 (DR5; cat. no., 3696S; 1:1,000), Fas cell surface death receptor (Fas; cat. no., 8023S; 1:1,000), CASP8 (cat. no., 4927S; 1:1,000), CASP9 (cat. no., 9502S; 1:1,000), CASP3 (cat. no., 9664S; 1:1,000) poly (ADP-ribose) polymerase (PARP; cat. no., 9532S; 1:1,000), β-actin (cat. no., 4970S; 1:1,000), TNF receptor associated factor 2 (TRAF2; cat. no., 4724S; 1:1,000), TRAIL (cat. no., 3219S; 1:1,000), TNF-α (cat. no., 34; 1:500), X-linked inhibitor of apoptosis (XIAP; cat. no., 14334S; 1:1,000), BCL2, apoptosis regulator (Bcl-2; cat. no., 15071; 1:1,000), BCL2 like 1 (Bcl-xl; cat. no., 2762S; 1:500), MCL1, BCL2 family apoptosis regulator (Mcl-1; cat. no., 94296S; 1:1,000), BCL2 associated X, apoptosis regulator (Bax; cat. no., 2774S; 1:1,000) and truncated (t)-BH3 interacting domain death agonist (BID; cat. no., 8762S; 1:500). All aforementioned antibodies were purchased from Cell Signaling Technology, Inc., (Danvers, MA, USA). In addition, cellular-FLICE-like inhibitory protein (c-FLIP; cat. no., sc-8346, 1:500) was purchased from Santa Cruz Biotechnology, Inc., (Dallas, TX Santa Cruz Biotechnology). Total proteins were extracted from cells using lysis buffer containing phenylmethyl sulfonylfluoride (Beyotime Institute of Biotechnology, Haimen, China) at 25°C for 30 min and protein concentration was determined with a BCA Protein Assay kit (Beyotime Institute of Biotechnology). A total of 20 µg protein was separated by SDS-PAGE (١٠٪ gel) and transferred onto a polyvinylidene fluoride membrane. Membranes were blocked with 5% non-fat milk at room temperature for 2 h and incubated with primary antibodies aforementioned at 4°C for 12 h. Anti-β-actin (cat. no., 4970S; 1:1,000; Cell Signaling Technology, Inc.) was used as a loading control. Membranes were then washed and incubated with horseradish peroxidase-conjugated secondary antibody (cat. nos. A21020 and A٢١٠١٠; ١:٨,٠٠٠; Abbkine Scientific Co., Ltd., Wuhan, China) at room temperature for 2 h. Immunoreactive bands were visualized using a chemiluminescence solution with an enhanced chemiluminescent kit (Beyotime Institute of Biotechnology). Quantity One (v4.6.8; Bio-Rad Laboratories, Inc., Hercules, CA, USA) was used as the software for quantification.

For the transient knockdown of c-FLIP, the HT-29 cells were transfected with 150 pmol Stealth RNAi siRNA directed against c-FLIP (si-c-FLIP) or non-targeting control siRNA (Scr) (both from Invitrogen; Thermo Fisher Scientific, Inc.) using the TransMessenger transfection reagent kit (Qiagen GmbH, Hilden, Germany). Cells were incubated with the transfection complexes for 4 h under their normal growth conditions. Subsequently, the complexes were removed from the cells, the cells were washed once with PBS, and then 500 µl fresh medium containing FBS and antibiotics (1% penicillin G-streptomycin; cat. no. 15140122; Thermo Fisher Scientific, Inc.) was added to the cells, according to the manufacturer's protocols. c-FLIP, sense, 5′-CGGACTATAGAGTGCTGATGG-3′ and antisense 5′-GATTATCAGGCAGATTCCTAG-3′.

For the measurement of mRNA expression levels in HT-29 cells with SDY treatment, the mRNA expression levels of DR3, DR4, DR5, Fas and TNF-α receptor were detected by RT-qPCR following treatment with 0, 10, 50, 100, 150 or 200 µg/ml SYD for 24 h at room temperature. The housekeeper gene GAPDH was used as a control. The RNA of cells was extracted using TRIzol^® reagent (cat. no. 15596018; Invitrogen; Thermo Fisher Scientific, Inc.). cDNA templates from HT-29 cell lines were prepared using a TIANScript RT kit (Tiangen Biotech, Co., Ltd., Beijing, China), according to the manufacture's protocol. RT-qPCR cycling was performed in 96-well plates on a LightCycler 480 Real-Time PCR system (Roche Applied Science, Penzberg, Germany). The reaction was performed in a 20 µl total volume containing ١٠ µl SYBR^® Select Master mix (cat. no. 4472908; Thermo Fisher Scientific, Inc.), 1 µl each primer (١٠ µM) and ٢ µl template cDNA. The primer sequences used for PCR are presented in Table I. The amplification protocol consisted of an initial denaturation step at 95°C for 5 min, followed by two-step PCR for 40 cycles at 95°C for 30 sec and 60°C for 30 sec. The mRNA expression levels of each target were determined based on the cycle threshold (Cq) value for the reference and each target and calculated as 2-∆Cq (5). Three independent experiments were performed.

Data were analyzed using SPSS software (version 19.0; IBM Corp., Armonk, NY, USA). The data are presented as the mean ± standard deviation. The data were analyzed using a two-tailed Student's t-test. For multiple comparisons, a one-way analysis of variance was used with Scheffe as the post-hoc test. P<0.05 or P<0.001 was considered to indicate a statistically significant difference. The calibration was assessed using Hosmer-Lemeshow goodness-of-fit test and Pearson's correlation coefficient was used to calculate the coefficient of correlation.

For the quality control of SYD, a HPLC analysis was applied in the present study. The chromatogram of a reference standard and a typical chromatogram obtained from the analyses of SYD are indicated Fig. 1. The sulforaphane peak (9.6 min) in the chromatogram of SYD was identified by comparing the retention time with that of its reference compound (sulforaphane). The linearity of the calibration curve for sulforaphane was assessed at six concentration levels ranging from 0.0281 to 1.4060 mg/ml, and triplicate injections were applied for each concentration. A calibration curve was constructed by plotting the integrated chromatographic peak areas (Y) vs. the corresponding concentration of the injected standard solutions (X). A least squares regression analysis was employed and a regression equation (Y=2,623,732.13X+7,853.33), and coefficient of correlation (r² was 0.99990) were obtained over relatively wide concentration ranges for all the analyses. The sulforaphane concentration in the sample was indicated to equal 0.0624 mg/ml.

The anti-proliferative effect of SYD in human colorectal cancer cells was tested and it was revealed that SYD inhibited the viability of colon cancer cells in a dose- and time-dependent manner. No significant difference was revealed at the 24 h time point, but significant differences were noted between 0, 10, 50, 100, 200 and 400 µg/ml SYD following 48 and 72 h of treatment (P=0.001). The viability of HT-29 and LS-174-T cells decreased with increasing doses of SYD at 48 and 72 h, respectively (Fig. 2A and B). At a concentration of 400 µg/ml, SYD inhibited the viability of HT-29 and LS-174-T cells. A concentration of 400 µg/ml SYD yielded ~88/80 and 95/92% inhibition of HT-29/LS-174-T cell viability compared with the control group (no treatment) at 48 and 72 h, respectively, and the corresponding half maximal inhibitory concentration (IC₅₀) values were 103.89/132.05 and 94.23/93.15 µg/ml. After 48 h of treatment with 0–200 µg/ml SYD, LDH increased in a dose-dependent manner in the HT29 cell line (Fig. 2C). As a control, the indicated dose of SYD did not significantly decrease the viability of CRL-1790 (normal human colon epithelial) cells, indicating that SYD exhibits selective cytotoxicity in colon cancer cells (Fig. 2D).

The effects of SYD on the cell cycle progression of HT-29 cells are indicated in Fig. 3. SYD treatment resulted in a significant dose-dependent accumulation of cells in the G2-M phase at 24 h (P<0.001, compared with the control; Fig. 3A and B). The population of G2 cells was increased by 6.2, 8.4 and 18.9% following treatment with 100, 200 and 400 µg/ml SYD, respectively, compared with the control. However, marked G1 arrest was observed with 200 and 400 µg/ml SYD following 48 and 72 h treatment (Fig. 3A, C and D). The greatest differences of 41 and 18% were obtained at 48 and 72 h, respectively, between the 0 and 400 µg/ml SYD groups, indicating that SYD decreased the population of cells in the S phase.

SYD induced dose-dependent apoptosis of HT-29 and LS-174-T cells. The apoptosis significantly increased in the 10, 50, 100, 150 and 200 µg/ml SYD groups, compared with the 0 µg/ml SYD group (P<0.05; Fig. 4A). The activation of caspases by SYD treatment was subsequently analyzed. A western blot analysis was performed to verify the rapid caspase activation observed with the FACScan assay. The results revealed an accumulation of cleaved CASP8 and CASP3 expression in addition to an increase in the expression level of cleaved CASP9 (P<0.001, compared with the 0 µg/ml SYD group; Fig. 4B). SYD-induced apoptosis was significantly decreased by inhibition of CASP8 (IETD-FMK) compared with control (P<0.05), whereas the inhibition of CASP9 (LEHD-FMK) had a minimal effect on SYD-triggered HT-29 cell apoptosis (Fig. 4C). Furthermore, the treatment of HT-29 cells with a pan-caspase inhibitor (Z-VAD-FMK) significantly delayed SYD-induced apoptosis compared with control (P<0.05; Fig. 4C).

To determine which apoptosis pathway serves a role in SYD-induced apoptosis, apoptosis-associated genes were first assessed by western blot analysis, which revealed that DR (DR3, DR4 and DR5), tumor necrosis factor-α receptor and the corresponding Fas exhibited minimal alterations in SYD-treated HT-29 cells (data not shown). Furthermore, a RT-qPCR analysis revealed that the mRNA levels were unchanged (data not shown). In contrast, as the concentration of SYD increased, the expression levels of TRAIL and the Fas ligand (FasLG) also increased. Importantly, the c-FLIP expression levels significantly decreased in a concentration-dependent manner (P<0.05 for 10–100 µg/ml SYD and P<0.001 for 150–200 µg/ml SYD, compared with 0 µg/ml SYD; Fig. 5A).

The activation of CASP8 by SYD encouraged the focus on nuclear factor κβ (NF-κβ)-regulated proteins that might regulate CASP8 or DR-mediated apoptosis. c-FLIP modulates CASP8 activity by competing with CASP8 for binding at the death domain, which cleaves pro-CASP8 (14). c-FLIP has two alternatively spliced isoforms, short (c-FLIP-S) and long (c-FLIP-L). Although TRAF2 was activated by different doses of SYD, its intensity was weaker than that of CASP8 in HT-29 cells (P<0.05, compared with the 0 µg/ml SYD group; Fig. 5A). The CASP8-processed N-terminal fragment of c-FLIPL is more efficient than c-FLIPL at recruiting TRAF2 and RIP1 (15), as was confirmed by the results of the present study.

To verify the potential significance of c-FLIP downregulation in SYD cytotoxicity, c-FLIP siRNA was transfected into HT-29 colon cancer cells, which decreased the c-FLIP protein expression levels compared with Src-transfected cells (P<0.05, compared with si-c-FLIP; Fig. 5B). The viability of HT-29 cells was examined with and without c-FLIP knockdown and it was indicated that the knockdown of c-FLIP in HT-29 cells decreased sensitivity to SYD, increased viability and prolonged survival compared with transfection with control siRNA 0, 100 and 200 µg/ml (P<0.05, compared with cells transfected with control siRNA and treated with 100 µg/ml or 200 µg/ml SYD; Fig. 5C). Comparative proliferation assays of cells with and without c-FLIP knockdown were performed and no significant differences were indicated, suggesting that the observed changes in viability are not an artifact of the altered proliferation rate (Fig. 5C and D).

To further confirm the SYD-induced apoptosis pathway, the expression levels of other NF-iβ-regulated proteins were examined. The results indicated that SYD did not affect the total expression level of the anti-apoptotic Bcl-2 family members, including Bcl-2, Bcl-xl, Mcl-1 and Bax (data not shown). XIAP was indicated to serve as a potent inhibitor of the downstream effectors CASP3, caspase-7 and CASP9, and XIAP cleavage increased in a concentration-dependent manner. The Pan-caspase inhibitor Z-VAD-FMK partially blocked XIAP cleavage, indicating that the inhibition of viability by SYD in colon cancer is partly due to caspase-dependent apoptosis induction (P<0.05, compared with the 0, 50 or 100 µg/ml SYD groups; Fig. 6A).

Although the total expression level of pro-apoptotic Bax was not altered, the cytosolic expression level of Bax was decreased and the mitochondrial levels were significantly increased following treatment with increasing concentrations of SYD (P<0.05). Inversely, increasing concentrations of SYD resulted in significant increases in the levels of cytochrome c in the cytosol and therefore significant decreases in the mitochondrial levels, as determined by western blotting and a cytochrome c assay (P<0.05, compared with the 0 µg/ml SYD group; Fig. 6B and C). Cytochrome c increased with higher doses of SYD (Fig. 6D). Subsequently, a concentration-dependent accumulation of tBid was further revealed. (Fig. 6A). Z-VAD-FMK had no effect on Bid cleavage, indicating a potential caspase-independent underlying mechanism.