Fucoidan protects hepatocytes from apoptosis and inhibits invasion of hepatocellular carcinoma by up-regulating p42/44 MAPK-dependent NDRG-1/CAP43.

Fucoidan is a traditional Chinese medicine suggested to possess anti-tumor effects. In this study the anti-metastatic effects of fucoidan were investigated in vitro in human hepatocellular carcinoma (HCC) cells (Huh-7 and SNU-761) under normoxic and hypoxic conditions and in vivo using a distant liver metastasis model involving injection of MH134 cells into spleen via the portal vein. Its ability to protect hepatocytes against bile acid (BA)-induced apoptosis was investigated in primary hepatocytes. Fucoidan was found to suppress the invasion of HCC cells through up-regulation of p42/44 MAPK-dependent NDRG-1/CAP43 and partly, under normoxic conditions, through up-regulation of p42/44 MAPK-dependent VMP-1 expression. It also significantly decreased liver metastasis in vivo. As regards its hepatoprotective effect, fucoidan decreased BA-induced hepatocyte apoptosis as shown by the attenuation of caspase-8, and -7 cleavages and suppression of the mobilization of caspase-8 and Fas associated death domain (FADD) into the death-inducing signaling complex. In summary, fucoidan displays inhibitory effects on proliferation of HCC cells and protective effects on hepatocytes. The results suggest fucoidan is a potent suppressor of tumor invasion with hepatoprotective effects.

Introduction

Hepatocellular carcinoma (HCC) is one of the most prevalent cancers worldwide. It is difficult to treat because of its insensitivity to chemotherapy and rapid development of drug resistance. Moreover, as HCC progresses, hydrophobic bile acids (BAs) accumulate during intrahepatic cholestasis and induce apoptosis of hepatocytes which ultimately results in liver cirrhosis or hepatic failure. In fact the poor prognosis of HCC patients is not due to the tumor itself but to the secondary hepatic decompensation reflected in several functional HCC staging systems such as the Barcelona Clinic Liver Cancer System and the Hong Kong Liver Cancer Staging System. For this reason, the development of a potent HCC suppressor with a hepatoprotective effect would represent a marked improvement in the treatment of HCC.

Fucoidan is a family of sulphated polyfucose polysaccharides isolated from brown algae that exhibit anti-tumor8, 9, 10, anti-thrombotic, anti-inflammatory and anti-viral activities. In particular, the anti-cancer effects and low toxicity of fucoidan make it a potentially valuable candidate for cancer chemotherapy. Fucoidan effectively suppresses cancer cell proliferation in vitro, and inhibits metastasis and angiogenesis of B16 melanoma xenografts and Lewis lung adenocarcinomas in vivo9, 16. More recently, it was reported to inhibit the growth of hepatoma Huh-7 cells through down-regulation of the chemokine ligand (CXCL)-12 at 1.0 mg/mL. Nevertheless, the anti-metastatic effects of fucoidan remain to be fully evaluated.

Fucoidan has also been reported to exert a protective effect on hepatocytes showing anti-oxidative effects against acute liver injury and liver fibrosis18, 19. Recently fucoidan was shown to protect against acetaminophen-associated liver injury in vivo. However, the question of whether fucoidan has hepatoprotective effects against BA-induced apoptosis has not been investigated.

In the present study, the effects and mechanism of action of fucoidan in HCC cell invasion in vitro and in vivo and its hepatoprotective effect against BA were evaluated. In particular, given the importance of the hypoxic microenvironment to the survival of HCC cells21, 22, studies were carried out under normoxic and hypoxic conditions.

Materials and methods

Cell lines and cell culture conditions

Three human HCC cell lines were used in this study viz Huh-BAT (a well-differentiated BA-transporter transfected HCC cell line), Huh-7 (a well-differentiated HCC cell line), and SNU-761 (a poorly differentiated HCC cell line)23, 24, 25. Cells were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 100,000 U/L penicillin, and 100 mg/L streptomycin with or without 100 nmol/L insulin. In all experiments, cells were subjected to overnight serum starvation to avoid confounding variables related to serum-induced signaling. Cells were incubated at 37 °C under either normoxic conditions (20% O2, 5% CO2) or hypoxic conditions (1% O2, 5% CO2 and 94% N2). Fucoidan from Fucus vesiculosus was obtained from Sigma-Aldrich Co. LLC. (Seoul, South Korea).

Animal experiments

In vivo studies were carried out in C3H mice (Orient Bio Inc). The in vivo study protocol was approved by the Institutional Animal Care and Use Committee (IACUC) and Ethics Committee of Seoul National University Hospital, Central South University and all experiments were carried out in strict accordance with the recommendations of the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. Surgical procedures were performed under anesthesia with sodium pentobarbital unless otherwise stated and all efforts were made to minimize animal stress and suffering.

Cell proliferation analysis (MTS assay)

Cell proliferation was determined using the Cell Titer 96 Aqueous One Solution cell proliferation assay (Promega, Madison, WI, USA). This measures the conversion of the colorimetric reagent 3,4-(5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium salt (MTS) into soluble formazan by dehydrogenase enzymes found in metabolically proliferating cell. Following each treatment, 20 μL dye solution was added to each well of a 96-well plate and incubated for 2 h. Subsequently, absorbance was recorded at 490 nm using an enzyme-linked immunosorbent assay (ELISA) plate reader (Molecular Devices, Sunnyvale, CA, USA). The half maximal inhibitory concentration (IC50) for each cell line was calculated using the Softmax Pro plate reader program.

Cell invasion assay

Invasion of HCC cells (Huh-7 and SNU-761) under both normoxic and hypoxic conditions was measured in 24-well chambers. Inserts were transferred into wells and coated with Matrigel (BD Biosciences, Billerica, MA, USA) for 30 min at 37 °C. Cells were then suspended in serum-free medium and implanted on the Matrigel-coated upper chambers (5×104 cells/chamber); DMEM containing 10% fetal bovine serum (FBS) was added to the lower chambers. Cells were incubated for 6 h at 37 °C and then incubated with or without fucoidan for 24 h before being stained with 4 μg/mL calcein AM (BD Biosciences, Billerica, MA, USA) in Hank's balanced salt solution (HBSS) at 37 °C for 90 min. Invasion capacity was determined as relative fluorescence units measured at excitation and emission wavelengths of 494 and 517 nm respectively using a multifunctional plate reader (EnVision Multilabel Reader; PerkinElmer Inc., Waltham, MA, USA).

Immunoblot analysis

HCC cells were lysed on ice for 20 min using lysis buffer and centrifuged at 14,000×g for 10 min at 4 °C. To carry out sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), samples were transferred to nitrocellulose membranes, blotted with appropriate primary antibodies at a dilution of 1:1000, and treated with peroxidase-conjugated secondary antibodies (Biosource International, Camarillo, CA, USA). Bound antibodies were visualized using chemiluminescent substrate (ECL; Amersham, Arlington Heights, IL, USA) and exposed to Kodak X-OMAT film (Kodak, New Haven, CT, USA). Primary antibodies included: rabbit anti-phospho-p42/44 mitogen-activated protein kinase (MAPK), rabbit anti-caspase 9, 6, 3, and anti-caspase 7 (cleaved) (all from Cell Signaling Technology, Danvers, MA, USA); anti-caspase 8 (BD Biosciences, San Jose, CA, USA); rabbit anti-N-myc downstream-regulated gene (NDRG)-1/CAP43 (Invitrogen Corporation, Camarillo, CA, USA). Goat anti-actin antibody was from Santa Cruz Biotechnology Inc. (Santa Cruz, CA, USA). Densitometric analysis was performed with Image J software (National Institutes of Health, Bethesda, MD, USA).

Real-time polymerase chain reaction (PCR) analysis

Total ribonucleic acids (RNAs) were extracted from Huh-7 and SNU-761 cells using Trizol reagent (Invitrogen, Carlsbad, CA, USA). Complementary deoxyribonucleic acid (cDNA) templates were prepared using oligo(dT) random primers and Moloney murine leukemia virus (MoMLV) reverse transcriptase. After the reverse transcription reaction, the cDNA template was amplified by PCR using Taq polymerase (Invitrogen). VMP-1 was determined by real-time PCR (LightCycler; Roche Molecular Biochemicals, Mannheim, Germany) using SYBR green as the fluorophore (Molecular Probes, Eugene, OR, USA). After electrophoresis in 1% agarose gel, the portion of gel containing the expected PCR product of vacuole membrane protein 1 (VMP-1) was excised (242 bp) and the product eluted into Tris–HCl buffer using a DNA gel extraction kit (Qiagen, Valencia, CA, USA). Forward and reverse primers were 5′-GTGGCTTTCATTGGTGCTGTCC-3′ and 5′-GAGTTCAACCGCTGCTGGA TTC-3′, respectively. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene expression was used as a control. The level of VMP-1 mRNA expression was calculated as the relative intensity of the PCR product bands compared with that from the GAPDH gene using the 2–ΔΔCt method. All PCR experiments were performed in triplicate.

SiRNA transfection

Cells were seeded in a 6-well culture plate (2×105 cells/well) in 2-mL antibiotic-free medium supplemented with 10% FBS. At 60%–80% confluence, cells were transfected with small interfering RNA (siRNA) using the siRNA transfection reagent (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA) according to the manufacturer's instructions. The cells were treated with siRNA for 6 h at 37 °C and then growth medium containing 20% FBS and antibiotics was added. After 18 h, the medium was replaced with fresh medium containing 10% FBS and antibiotics and after 24 h, the cells were used in subsequent experiments.

Distant metastasis model (splenic injection)

Five-week-old male C3H mice were anesthetized with ether and subjected to splenic injection as previously described. After exposing the spleen by an upper quadrant incision, 1×105 MH134 cells (a mouse hepatocellular carcinoma cell line) were injected into its lower pole using a 31-gauge needle. Mice were then treated with intraperitoneal (i.p.) injections of normal saline (control group, n=4) or 10 mg/kg fucoidan (treatment group, n=4) every two days. After eight days, the number of macroscopic liver metastases and the sum of the maximal diameter for each nodule were determined. The i.p. dose was determined based on a previous study of the antitumor effects of fucoidan in vivo. Tumor specimens were fixed in 4% formaldehyde and embedded in paraffin. For immunohistochemical (IHC) analysis, specimens were cut into 4 μm sections and subjected to IHC staining using the anti-mouse CD31 antibody (Vector Laboratories, Inc., Burlingame, USA) at a 1:300 dilution or anti-glypican (GP)-3 antibody (Novus Biologicals, Littleton, USA) at a 1:200 dilution.

Primary hepatocyte isolation and culture

Primary hepatocytes from five-week-old C3H mice were isolated and cultured as previously described. The mice were housed and handled under specific pathogen-free conditions. Hanks' solution I consisted of basal Hanks' solution containing 2.5 mmol/L EGTA, 0.1% glucose and penicillin/streptomycin at a dilution of 1:100. Hanks' solution II consisted of basal Hanks' solution containing 0.3 mg/mL collagenase CLSII and 5 mmol/L CaCl2. Hanks' solutions were prewarmed to 42 °C and collagenase added immediately prior to liver perfusion. After shaving the abdomen, the abdominal cavity was opened and the portal vein cannulated with a 25-gauge catheter to which a silicon tube was connected. Hanks' solution I was perfused using a peristaltic infusion pump at a flow rate of 8 mL/min. After infusion was initiated, the inferior vena cava was incised to permit sufficient outflow. The liver was perfused with Hanks' solution I for 2 min followed by Hanks′ solution II for 5 min. Efficient perfusion via the portal vein was evidenced by the color change in liver from dark red-brown to light brown. Following perfusion, the gallbladder and liver capsule were removed using a pincette and livers transferred to sterile Petri dishes.

After disintegrating livers by gentle shaking, suspensions were filtered through a 100 mm cell strainer and transferred to 50 mL Falcon tubes. Cells were washed twice with William's medium E (WME) at room temperature and 37.5 °C before being centrifuged for 2 min and resuspended in WME. Hepatocytes were then plated on collagen-coated culture dishes with fetal calf serum (FCS) cell culture medium and kept in a humidified cell culture incubator at 37 °C under 5% CO2. After 4 h during which hepatocytes attached to the collagen-coated dish, FCS medium was removed and replaced with serum-free cell culture medium. After a further 14–20 h, cells were washed three times with starvation medium and incubated in starvation medium for the indicated time periods.

Death-inducing signaling complex immunoprecipitation (DISC IP)

Cell pellets were treated with lysis buffer (100 mmol/L Tris–HCl, 5 μmol/L EDTA, and 1% NP-40) containing protease inhibitor and phosphatase inhibitor cocktails. Total protein (500 μg) was incubated with rat anti-death receptor (DR) 5 antibody (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA), mouse anti-Fas-associated death domain (FADD) antibody (BD Biosciences, San Jose, CA, USA), mouse anti-cellular FLICE inhibitory protein (cFLIP) antibody (Enzo Life Science, Farmingdale, NY, USA), or mouse anti-caspase 8 antibody (Upstate, Lake Placid, NY, USA) at 4 °C for 1 h. Protein A/G beads (Santa Cruz, San Diego, CA, USA) were added to the lysate and incubated overnight with gentle rocking at 4 °C. The beads were washed five times by repeated suspension in 500 μL lysis buffer followed by centrifugation at 10,000×g for 1 min. After the last wash, 35 μL Laemmli buffer (4% SDS, 20% glycerol, 10% 2-mercaptoethanol, 0.004% bromophenol blue, 0.125 mol/L Tris HCl, pH 6.8) was added to the pellet and heated at 100 °C for 5 min. Centrifugation was performed again to collect supernatant and finally samples were subjected to immunoblot analysis as described above.

Statistical analysis

Statistical analysis was performed using PASW version 18.0 (SPSS Inc., Chicago, IL, USA). All experimental results are based on three independent experiments using cells from three separate isolations and are presented as mean±standard deviation (SD). For comparisons between groups, data were analyzed by the Mann–Whitney U test or one-way ANOVA. For all tests, P<0.05 was regarded as statistically significant.

Results

The effect of 0–2 mg/mL fucoidan on Huh-7 and SNU-761 cell growth was assessed in vitro using the MTS assay. Cell proliferation following fucoidan treatment decreased in a dose-dependent manner under both normoxic and hypoxic conditions (Fig. 1A). Moreover, 1 mg/mL fucoidan enhanced apoptosis of Huh-7 cells as compared to control as shown by the expression of bax, bid, caspase-8, -7, and -9 cleavages, under both normoxic and hypoxic conditions (Fig. 1B). The IC50 values in Huh-7 cells were estimated to be 0.80 mg/mL under normoxic conditions and 1.05 mg/mL under hypoxic conditions. Corresponding values in SNU-761 cells were 0.28 and 0.6 mg/mL, respectively.

Figure 1

Effect of fucoidan on proliferation and invasion of HCC cells. (A) Huh-7 and SNU-761 cell growth following treatment with fucoidan (0, 0.1, 0.25, 0.5, 1, and 2 mg/mL). *Concentration used in the invasion assay; (B) Fucoidan (1 mg/mL) enhances apoptosis of Huh-7 cells under both normoxic and hypoxic conditions; (C) Huh-7 cell invasion through a Matrigel-coated filter following treatment with fucoidan 0.1 mg/mL under normoxic conditions and 0.5 mg/mL under hypoxic conditions; (D) SNU-761 cell invasion following treatment with fucoidan 0.1 mg/mL under both normoxic and hypoxic conditions.

Invasion assays were performed using a concentration of fucoidan that did not inhibit proliferation of HCC cells (Huh-7 and SNU-761) under normoxic and hypoxic conditions. Huh-7 cell invasion through a Matrigel-coated filter was evaluated after treatment with fucoidan 0.1 mg/mL under normoxic conditions and 0.5 mg/mL under hypoxic conditions. SNU-761 cell invasion was evaluated after treatment with fucoidan 0.1 mg/mL under both normoxic and hypoxic conditions. At 24 h, fucoidan significantly suppressed Huh-7 and SNU-761 cell invasion compared to controls under both normoxic and hypoxic conditions (Fig. 1C and D).

Enhanced p42/44 MAPK pathway dependent NDRG/CAP43 by fucoidan

The expressions of NDRG-1/CAP43 (a stress-responsive protein involved in cell growth that acts as a tumor suppressor in many cell types) and mesenchymal markers (CK19, MMP-2, MMP-9, LOX, cathepsin D, vimentin, E-cadherin, and fibronectin) were evaluated by immunoblotting. The expression of NDRG-1/CAP43 was higher in Huh-7 and SNU-761 cells treated with fucoidan than in controls at 48 h (Fig. 2A and B). The up-regulation of NDRG-1/CAP43 expression in fucoidan-treated SNU-761 cells was slightly less prominent which may be due to the different cell lines. Increased expression of NDRG-1/CAP43 at 48 h due to fucoidan was also found by densitometric analysis (Fig. 2C). In contrast, fucoidan treatment was found to decrease the expressions of vimentin, E-cadherin, and fibronectin in Huh-7 cells as compared to controls (Fig. 2A).

Figure 2

Fucoidan enhances the expression of NDRG-1/CAP43 and mesenchymal markers. Expression of NDRG-1/CAP43 and mesenchymal markers (CK19, MMP-2, LOX, cathepsin D, vimentin, e-cadherin and fibronectin) were evaluated by immunoblotting in (A) Huh-7 and (B) SNU-761 cells and by (C) densitometric analyses (data are expressed as the relative ratio of NDRG-1/CAP43 to actin).

In order to determine the pathway activated by fucoidan, the expressions of total p42/44, phosphorylated-p42/44 (p-p42/44), total Akt, and phosphorylated-Akt (p-Akt) were evaluated in HCC cells by immunoblot assay. At early time points, the expression of p-p42/44 was higher in Huh-7 cells compared to control cells (Fig. 3A) and expression was also higher in SNU-761 cells (Fig. 3B). However, changes in the expression of total Akt and p-Akt were not consistent in the two cell lines.

Figure 3

Fucoidan enhances the expression of NDRG-1/CAP43 by the p42/44 MAPK pathway. Immunoblotting reveals that fucoidan enhances expression of phosphorylated-p42/44 (p-p42/44) in (A) Huh-7 and (B) SNU-761 cells and (C) that expression of NDRG-1/CAP43 in Huh-7 and SNU-761 cells is attenuated by treatment with U0126 (MAPK inhibitor). (D) The outcome of densitometric analysis data are expressed as the relative ratio of NDRG-1/CAP43 to actin.

The effect of fucoidan on the expression of p42/44 MAPK pathway-dependent NDRG-1/CAP43 in Huh-7 and SNU-761 cells was also evaluated by immunoblotting and found to be attenuated by U0126 (MAPK inhibitor) treatment (Fig. 3C). Densitometric analysis also confirmed that fucoidan activates the expression of NDRG-1/CAP43 in a p42/44 MAPK pathway-dependent manner (Fig. 3D).

An invasion assay was also carried out to evaluate whether NDRG-1/CAP43 siRNA increases invasion of fucoidan-treated HCC cells. First, the specific knockdown effect of siRNA targeting NDRG-1/CAP43 was examined by immunoblot assay. As shown in Fig. 4A, NDRG-1/CAP43 siRNA significantly decreased the expression of NDRG-1/CAP43 in Huh-7 cells. In addition, fucoidan suppressed HCC cell invasion via NDRG-1/CAP43 under both normoxic and hypoxic conditions (Fig. 4B and C).

Figure 4

Invasion of HCC cells following fucoidan treatment with NDRG-1/CAP43 siRNA. (A) Immunoblot analysis reveals that NDRG-1/CAP43 siRNA significantly decreases the expression of NDRG-1/CAP43 in Huh-7 cells and increased invasion of fucoidan-treated HCC cells. Fucoidan suppresses invasion of (B) Huh-7 and (C) SNU-761 cells through NDRG-1/CAP43 activity.

Enhanced expression of VMP-1 by fucoidan under normoxic conditions

The effect of fucoidan treatment on VMP-1 (a novel molecule that inhibits metastasis and proliferation in HCC) was investigated using quantitative real-time PCR. VMP-1 mRNA expression was significantly increased in SNU-761 cells (Fig. 5A), VMP-1 siRNA significantly decreased VMP-1 mRNA expression in both Huh-7 and SNU-761 cells (Fig. 5B, P<0.05) and VMP-1 siRNA increased invasion of SNU-761 cells (Fig. 5C), all under normoxic conditions. Thus VMP-1 appears to be involved in HCC cell invasion but only under normoxic conditions. The increased expression of VMP-1 in Huh-7 cells was attenuated by U0126 (MAPK inhibitor) under normoxic conditions (Fig. 5D, P<0.05).

Figure 5

Fucoidan enhances expression of VMP-1 under normoxic conditions. (A) RT-PCR reveals that fucoidan significantly increases VMP-1 mRNA expression; (B) VMP-1 siRNA significantly decreases VMP-1 mRNA expression in both Huh-7 and SNU-761 cells; (C) VMP-1 siRNA increases invasion of fucoidan-treated HCC cells; (D) increased expression of VMP-1 in Huh-7 cells is attenuated by U0126 (MAPK inhibitor).

Fucoidan inhibited liver metastasis in an intrahepatic portal vein metastasis model in vivo

Fucoidan was anti-metastatic in the distant metastasis model as shown by the fact that the number of intrahepatic metastases in the fucoidan treatment group was significantly lower than in the control group (Fig. 6A, P<0.05). In addition, the sum of the maximal diameter of each nodule in the fucoidan treatment group was significantly lower than that in the control group (Fig. 6B, P<0.05). Fucoidan attenuated the nuclear grade of MH134 cells as determined by hematoxylin and eosin (H&E) staining and attenuated angiogenesis of MH134 cells as shown by decreased CD31 expression on IHC staining (Fig. 6C). The expression of GP-3 was also attenuated by fucoidan treatment when compared to control.

Figure 6

Fucoidan inhibits liver metastasis in the distant metastasis model. (A) The number of intrahepatic metastases in the fucoidan treatment group and control group; (B) the sum of the maximal diameter of each nodule in the two groups; (C) histological analysis following fucoidan treatment. The nuclear grade of MH134 cells was determined by H&E-staining and CD31 and GP-3 expression by IHC staining (200×).

Fucoidan suppressed BA-induced apoptosis

To evaluate whether fucoidan decreased BA-induced hepatocyte apoptosis, the hepatoprotective effect of fucoidan on Huh-BAT cells stably transfected with a BA transporter was evaluated as previously described30, 31, 32. Fucoidan significantly decreased BA-induced Huh-BAT cell apoptosis as shown by attenuation of caspase-8, -9, -3, and -7 cleavages, p-JNK, and bid as compared to BA-treated cells (Fig. 7A). DISC-IP was used to evaluate changes in DISC assembly and revealed that fucoidan suppressed the mobilization of caspase-8, cFLIP, DR-5 and FADD into DISC as compared to BA-treated cells (Fig. 7B). Decreased recruitment to DISC leads to suppression of autocatalytic processes of the procaspases within DISC. Finally, following BA treatment of primary mouse hepatocytes in culture, immunoblotting showed fucoidan significantly decreased BA-induced hepatocyte apoptosis as shown by the attenuation of caspase-8, and -7 cleavages compared to BA-treated cells (Fig. 7C).

Figure 7

Fucoidan suppresses BA-induced hepatocyte apoptosis. (A) Suppression of BA-induced Huh-BAT cell apoptosis is shown by the attenuation of caspase-8, -9, -3, and -7 cleavages, p-JNK, and bid as compared to control; (B) Suppression of the mobilization of caspase-8, cFLIP, and FADD into DISC is shown by DISC-IP; (C) the decrease in BA-induced primary cultured-hepatocyte apoptosis as compared to BA-treated cells.

Discussion

The present study clearly demonstrates that fucoidan, a sulphated polysaccharide purified from brown algae, possesses anti-metastatic effects on HCC both in vitro (Huh-7 and SNU-761 cells) and in vivo (C3H mouse) and exerts a hepatoprotective effect by decreasing BA-induced hepatocyte apoptosis. Fucoidan inhibited HCC cell invasion through up-regulation of p42/44 MAPK-dependent NDRG-1/CAP43 and partly, under normoxic conditions, through up-regulation of p42/44 MAPK-dependent VMP-1. Fucoidan also decreased the migratory potential of Huh-7 cells by suppressing mesenchymal markers. In vivo, fucoidan decreased the expression of the angiogenesis marker CD31. It also reduced the expression of GP-3 which correlates with poorly-differentiated HCC and poor prognosis33, 34.

NDRG-1/CAP43 (also known as Drg-1, RTP, Rit42 and PROXY-1) regulates tumor progression in various cancers and is involved in cell differentiation. It is mainly expressed in non-neoplastic tissue where it is regulated by hypoxia, oxidative stress, histone deacetylase-targeting drug, lysophosphatidylcholine, oncogene, and tumor suppressor gene36, 37. Although some studies have reported conflicting results, NDRG-1/CAP43 appears to be a tumor suppressor or metastasis suppressor in many cancer cells including those of breast, prostate, cervical, ovarian and pancreatic cancers38, 39, 40, 41, 42. In this study, knockdown of NDRG-1/CAP43 significantly enhanced the invasion of HCC cells and its expression was modulated by the p42/44 MAPK pathway.

As previously reported, VMP-1 is an inhibitor of cell proliferation and secretory membrane transport. Moreover, it is necessary for autophagy and its expression induces the formation of autophagosomes. Interestingly, VMP-1 was significantly up-regulated under normoxic conditions in this study. VMP-1 was recently identified as a membrane protein putatively associated with suppression of HCC growth and pulmonary metastases. Based on our results, fucoidan appears to exert anti-metastatic effects by up-regulation of NDRG-1/CAP43 and VMP-1 in a p42/44 MAPK pathway-dependent manner.

The MAPK pathway usually controls proliferation of normal and tumor cells by activation of RAS, Raf and p42/44 but there have been some conflicting reports. Very high signal intensity of p42/44 interrupts the cell cycle by inducing cell cycle inhibitors including p21Cip/Waf and p27KIP. This suggests that high signal intensity of p42/44 MAPK exerts a paradoxical effect on tumors. In the present study, fucoidan was shown to exert anti-metastatic effects on HCC cells by activating the p42/44 MAPK pathway.

Surprisingly, fucoidan decreased BA-induced apoptosis of hepatocytes while enhancing apoptosis of HCC cells. This dual ability to preserve hepatic function may represent an ideal strategy to treat HCC. The cytotoxic effects of hydrophobic BAs on normal hepatocytes have been well documented and previous studies have shown that mitochondrial dysfunction due to oxidative stress and the non-mitochondrial nitric oxide-dependent pathway result in hepatocyte apoptosis47, 48. Procaspases are recruited to DISC via homotypic interactions between death effector domains (DED) of FADD and initiator caspases and, once recruited, undergo autocatalytic processing. Both the recruitment and processing is regulated by cFLIP. A potential mechanism by which toxic BA promote death receptor cytotoxic signaling is by modulating the composition of DISC. The results of our DISC IP studies suggest fucoidan suppresses these cascades.

Conclusions

We have demonstrated that fucoidan displays anti-metastatic effects on HCC through up-regulation of p42/44 MAPK-dependent NDRG-1/CAP43 and protects against BA-induced apoptosis of hepatocytes. This ability to suppress tumor invasion and exert hepatoprotective effects suggests fucoidan may represent an ideal therapeutic agent for the treatment of patients with HCC.