InHepG2 cells treated with cisplatin (2.5–20 μM) for 24 and 48 h, (Figure 1A), we found that cisplatin significantly inhibited the cell viability (13% at 24 h and 39% at 48 h, P < 0.05, respectively, at 10 μM cisplatin). In addition, fucoxanthin significantly inhibited the cell proliferation of HepG2 by 17% and 28% after incubation with 10 μM fucoxanthin for 24 h and 48 h, respectively (Figure S1). To investigate whether fucoxanthin increases the sensitivity of cisplatin in HepG2 cells, we pre-incubated HepG2 cells with fucothanxin (1–10 μM) for 24 h followed by incubation with cisplatin (2.5–20 μM) for 24 h. Results reveal that the cell viability of HepG2 cells was significantly and concentration-dependently inhibited (Figure 1B), with an inhibition of 37% at 10 μM fucoxanthin and 10 μM cisplatin, as compared with cisplatin treatment alone. In addition, the combination of fucoxanthin with cisplatin increased early apoptotic cells (PI negative, Annexin V-FITC positive) and late apoptotic cells (PI positive, Annexin V-FITC positive) (Figure 1C). The results indicate that fucoxanthin enhances the anti-proliferative effect of cisplatin in human hepatoma HepG2 cells.

We also evaluated the effect of fucoxanthin on NFκB expression induced by cisplatin in HepG2 cells, as determined by the EMSA and NFκB reporter gene assays. As shown as in Figure 2A, cisplatin μM) most strongly induced NFκB binding activity at 16 h of incubation (by 77%, as compared with untreated cells). However, fucoxanthin concentration-dependently attenuated cisplatin-induced NFκB binding activity, with a 37% inhibition at 5 μM fucoxanthin (Figure 2B). We also showed that fucoxanthin significantly and concentration-dependently attenuated cisplatin-induced NFκB luciferase activity in a similar pattern to that of NFκB binding activity (Figure 2C). In addition, fucoxanthin significantly and concentration-dependently restored cisplatin-inhibited IκB-α-phosphorylation in HepG2 cells at 24 h of incubation, as compared with cisplatin treatment alone (Figure 2D).

Treatment of HepG2 cells with cisplatin (10 μM) for 24 h significantly increased the ratio of Bax/Bcl-2 mRNA expression (by 1.8-fold, P < 0.001, as compared with untreated cells). However, pretreatment of HepG2 cells with fucoxanthin for 24 h followed by incubation with cisplatin for 24 h significantly and concentration-dependently increased the ratio of Bax/Bcl-2 mRNA expression (by 4.3 fold, P < 0.001, as compared with cisplatin treatment alone) (Figure 3A). To further determine whether fucoxanthin in combination with cisplatin enhances the ratio of Bax/Bcl-2 mRNA primarily through NFκB-regulated pathways, we pretreated HepG2 cells with fucoxanthin for 24 h followed by incubation with an NFκB activation inhibitor (NAI) (10 and 20 μM) for 2 h and then by incubation with cisplatin (10 μM) for 24 h. We found that the combination of fucoxanthin, NAI and cisplatin synergistically or additively increased the ratio of Bax/Bcl-2 mRNA expression, as compared with the NFκB activation inhibitor alone (Figure 3B). Thus, the data indicate that fucoxanthin increases the ratio of Bax/Bcl-2 mRNA expression and that this effect is likely associated with inhibition of the NFκB pathway.

Real-time PCR was performed to evaluate the mRNA levels of ERCC1 and TP. As shown in Figure 4, cisplatin (10 μM) treatment alone significantly increased the mRNA expression of ERCC1 and TP in HepG2 cells. However, the induced mRNA expression of ERCC1 and TP in HepG2 cells by cisplatin (10 μM) was attenuated by pretreatment with fucoxanthin (1–10 μM) for 24 h, with a 1.5-fold and a 1.2-fold inhibition, respectively, at 10 μM fucoxanthin, as compared with cisplatin treatment alone.

The time effect of cisplatin on protein expression of the mitogen-activated protein kinase (MAPK) family (p38, ERK, and JNK) and phosphatidylinositol 3-kinase (PI3K)/AKT in HepG2 cells were determined by Western blotting. Results reveal that cisplatin (10 μM) markedly increased the phosphorylation of ERK, p38 and PI3K/AKT at 6 h of incubation, but it did not affect the phosphorylation of JNK or the protein expression of ERK, p38 and JNK (Figure 5A). We then determined whether pretreatment of HepG2 cells with fucoxanthin (1–10 μM) for 24 h attenuates the induction of MAPK family and PI3K/AKT protein expression by cisplatin (10 μM). We found that fucoxanthin concentration-dependently attenuated cisplatin-induced phosphorylation of ERK, p38 and PI3K/AKT (Figure 5B).

We then determined whether the attenuation of fucoxanthin on ERCC1 and TP mRNA expression induced by cisplatin occur primarily through the inhibition of ERK, p38 and PI3K/AKT pathway. HepG2 cells were pre-incubated with fucoxanthin (5 μM) for 24 h followed by incubation with ERK inhibitor (PD98059; 20 μM), p38 inhibitor (SB203580; 20 μM) or PI3K inhibitor (LY294002; 20 μM) for 1 h and then with cisplatin (10 μM) for 24 h. The concentration of 5 μM fucoxanthin was chosen because it only produced a slight inhibition of mRNA expression for ERCC1 and TP (see Figure 4). We found that the combination of fucoxanthin, cisplatin and the ERK inhibitor or PI3K inhibitor synergistically inhibited ERCC1 mRNA expression but not TP mRNA expression (Figure 6A,B). In contrast, the combination of fucoxanthin with p38 inhibitor enhanced the inhibition of mRNA expression of TP but not ERCC1 (Figure 6C). Thus, the results reveal that fucoxanthin may inhibit ERCC1 mRNA expression through ERK and PI3K/AKT pathway but may inhibit TP mRNA expression through p38 pathway.

Dulbecco’s modified eagle medium (DMEM), fetal bovine serum (FBS), trypsin, penicillin, sodium pyruvate, and non-essential amino acids (NEAA) were purchased from GIBCO/BRL (Maryland, MD, USA). MAPK/extracellular signal-regulated kinase (ERK) 1/2, c-Jun NH2-terminal kinase (JNK)/stress-activated protein kinase and p38 MAPK proteins and phosphorylated proteins, phosphatidylinositol 3-kinase (PI3K)/AKT, ERK inhibitor (PD98059), p38 inhibitor (SB203580) and PI3K inhibitor (LY294002) were purchased from Cell Signaling Technology (Beverly, MA). NFκB activation inhibitor was purchased from Merck Millipore (Billerica, MA, USA). Fucoxanthin was extracted from Undaria pinnatifida and purified, as we reported previously [62]. The purified fucoxanthin was dissolved in ethanol to a final concentration of 10 mM as the stock solution. Before the experiment, fucoxanthin solutions were prepared freshly in a mixture of ethanol and FBS (1:9), as adopted from the preparation of lycopene solution [63].

The human hepatoblastoma HepG2 cell line was obtained from Food Industry Research and Development Institute (FIRDI, Hsinchu, Taiwan) and maintained in DMEM supplemented with 10% fetal bovine serum without antibiotics under 5% CO₂ at 37 °C.

Cell viability was evaluated using the modified acid-phosphatase (ACP) assay, with p-nitrophenyl phosphate (PNPP) disodium salt as a substrate. The cell culture media were aspirated, and the cells were washed with phosphate-buffered saline (PBS). Following the wash, 100 μL of the ACP reagent (0.1 M sodium acetate (pH 5.5), 0.1% Triton X-100, and 10 mM PNPP) was added. After 1 h of incubation at 37 °C, the enzyme activity was stopped by adding 10 μL of 1 N NaOH, and the enzyme activity was determined photometrically at a wavelength of 405 nm [64].

Total RNA in cell cultures was extracted with REzol reagent (PROtech Technologies, Inc., Placentia, CA, USA), and 1 μg of total RNA was reverse-transcribed by using oligo-dT as a primer in 20 μL reverse-transcription solutions containing 1 μL reverse transcriptase (Promega, Sunnyvale, CA, USA). Real-time PCR performed with a Corbett instrument (Applied Biosystems, Carlsbad, CA, USA) using SYBR Green Master Mix (ProTech, Placentia, CA, USA) according to the manufacturer’s instructions. In all real-time PCR experiments, both a non-template control (NTC) and a standard curve were amplified, as well. The RNA abundance was normalized to β-actin RNA in each sample. The primers used in this study were as follows: ERCC1 forward 5′-CCCTGGGAATTTGGCGACGTAA-3′, reverse 5′-CTCCAGGTACCGCCCAGCTTCC-3′; TP forward 5′-AGCTGGAGTCTATTCCTGGATT-3′, reverse 5′-GGCTGCATATAGGATTCCGTC-3′; β-actin forward 5′-GTGGGGCGCCCCAGGCACCA-3′, reverse 3′-CACCCCGCGGGGTCCGTGGT-5′.

Protein expression of MAPK family (ERK, p38, JNK, p-JNK, p-ERK and p-p38) and PI3K/AKT (PI3K, AKT and p-AKT) was measured by Western blotting. In cell culture experiments, the medium was removed and cells were rinsed with PBS twice. After the addition of 0.5 mL of cold RIPA buffer and protease inhibitors, cells were scraped and followed by a mild vortexing at 0 °C for 20 min. The cell lysates were then subjected to a centrifugation of 10,000 rpm for 30 min at 4 °C. Total protein (50 μg) from the supernatant was resolved on SDS-PAGE and transferred onto a PVDF membrane. After blocking with TBS buffer (20 mmol/L Tris-HCl, 150 mmol/L NaCl, pH 7.4) containing 5% nonfat milk, the membrane was incubated with monoclonal antibody followed by incubation with horseradish peroxidase-conjugated anti-goat IgG, and then visualized using an ECL chemiluminescent detection kit (Amersham, Sweden). The relative density of the immunoreactive bands was quantitated by densitometry (Gel Pro Analyzer TM, version 3.0, Media Cybernetics, Rockville, MD, USA).

Nuclear protein extracts (5 μg) were prepared according to the modified method of a previous study [65]. Binding activities of transcription factors including NFκB were analyzed by EMSA. Electrophoretic mobility shift assay (EMSA) was performed with LightShift Chemiluminescent EMSA Kit (Pierce Biotechnology, Rockford, IL, Country), as described previously [66]. The NFκB consensus oligonucleotide probe (5′-AGTTGAGGGGACTTTCCCAGGC-3′) was end-labeled with biotin (Sangon, Shanghai, China). Briefly, nuclear extract (5 μg) was incubated with 10 ng NFκB (p65) probe. For the cold probe assay, 40 ng of unlabeled (cold) NFκB probe was mixed with sample 5 min before adding 10 ng biotin-labeled NFκB probe. Protein-DNA complexes were then resolved by non-denaturing polyacrylamide gel electrophoresis (PAGE). After blocking, avidin-HRP was applied and detected by enhanced chemiluminescence (ECL, Amersham). The relative NFκB levels were quantitated by Matrox Inspector 2.1 software.

HepG2 cells (1.8 × 10⁴ cells/well) were plated in 96-microwell-white-plates (Nalge Nunc, Rochester, New York, NY, USA) before transfection. The NFκB plasmid vector (pGL4.32 (luc2P/NF-κB-RE/Hygro)) contains five copies of an NFκB response element was purchased from Promega (Sunnyvale, CA, USA). Transfection of NFκB plasmid vector (0.15 μg) into HepG2 cells was performed using TransIL-LT1 Transfection Reagent (Mirus, Madison, WI, USA), and in all experiments, the pRL-TK Renilla reporter vector (0.02 μg) (Promega, Sunnyvale, CA, USA) was used as an internal control. The cells were then incubated with cisplatin (10 μM) for 16 h after pretreatment with fucoxanthin (1–10 μM) for 24 h. Renilla and firefly luciferase activities were measured using the Dual-Luciferase Reporter Assay System (Promega, Sunnyvale, CA, USA).

All experiments were repeated at least thrice. Values are expressed as means ± SD and analyzed using one way ANOVA followed by LSD test for comparisons of group means, when the F ratios were significant. All statistical analyses were performed using SPSS for Windows, version 10 (SPSS, Inc., Armonk, NY, USA); a P value < 0.05 is considered statistically significant.