The effects of oral administration of IMWE and HMWF on murine splenic NK cell populations are shown in Figure 3. Oral administration of IMWF (4.6 ± 0.4%) and HMWF (5.2 ± 0.5%) increased the size of the NK cell population in splenic tissue compared with that observed in the control group (4.1 ± 0.2%). There was a significant difference between the HMWF group and the control group.

Some reports have indicated that fucoidan stimulates NK cell activity in vivo [16,22,23]. Previous reports have also shown that fucoidan extracted from Cladosiphon okamuranus increases the size of NK cell populations. In this study, the population of NK cells was increased significantly in the HMWF group compared to that observed in the control group. Shimizu et al. reported that high-molecular-weight fucoidan extracted from Cladosiphon okamuranus (200–300 kDa) promotes greater increases in the proportion of murine cytotoxic T cells than middle- (2–3 kDa) or low- (0.5–1 kDa) molecular-weight fucoidan [18]. They administered fucoidan by containing 5% fucoidan in laboratory chew for 70 days. In the present study, we also administered the same amount of fucoidan for 70 days. Our results indicate feedings fucoidan also increase the population of the NK cells in mouse spleen. NK cells have the ability to recognize and destroy a wide range of abnormal cells, including tumor cells, without damaging healthy and normal self cells [24]. In vivo and in vitro studies have shown that NK cells can eliminate tumor cells [25]. Our results indicate that one of the anti-tumor mechanisms of fucoidan extracted from Cladosiphon okamuranus might be to mediate increases in NK cell activity. Our results also indicate that differences in the molecular weight of fucoidan might affect NK cell activity in vivo. Further studies must be conducted in order to understand the relationships between the molecular weight of fucoidan and NK cell activity in vivo.

To reveal the relationships between the anti-tumor effects of fucoidan and innate immunity, we examined the anti-tumor effects of fucoidan using myeloid differentiation primary response gene-88 (Myd-88) knockout mice. The results are shown in Table 2. No significant changes in tumor weight were observed in the HMWF/Myd-88 group compared to that seen in the control/Myd-88 group. No change was observed in the serum IL-2 level between the HMWF/Myd-88 and control/Myd-88 groups. The serum IL-10 level in the HMWF/Myd-88 group was increased slightly compared to that observed in the control/Myd-88 group.

Previous reports have indicated that the bioactivity of fucoidan activates innate immunity [4,16,17]. The activation of innate immunity is essential for the activation of adaptive immunity [26]. In particular, toll-like receptors (TLR) on the surface of intracellular organelles recognize the specific structures of bacteria, viruses and fungi [27]. Adapter molecules such as Myd-88 and TIR-domain-containing adapter-inducing interferon-β (TRIF) play important roles in inducing the production of cytokines via TLRs [28,29]. In our results, fucoidan exhibited no anti-tumor effects in Myd-88 knockout mice. This result indicates that the anti-tumor effects of fucoidan arise from the Myd-88 pathway in vivo.

The serum IL-2 level in the HMWF/Myd-88 group was not changed compared to that observed in the control/Myd-88 group. The serum IL-10 level in the HMWF/Myd-88 group was increased compared to that observed in the control/Myd-88 group. TLR-4 stimulates cytokine production via both Myd-88 and TRIF signaling pathways [29]. Our results indicate that increased serum cytokine levels might result from the TRIF-dependent pathway, not the Myd-88-dependent pathway. To our knowledge, there are no reports investigating the relationships between TLRs and fucoidan. Our results indicate that fucoidan stimulates TLR-4 in tumor mouse models. To understand the mechanisms underlying the anti-tumor effects of fucoidan, further studies focusing on TLRs must be conducted both in vivo and in vitro.

Some reports have indicated that the bioactivity of fucoidan depends on fucoidan’s molecular weight [18,30,31]. Yang et al. described the anti-cancer activities of fucoidan extracted from Undaria pinnatifida. Their results indicated that 490 kDa of fucoidan exhibits more effective anti-cancer activities than natural fucoidan (5100 kDa) or low-molecular-weight fucoidan (260 kDa) [30]. Shimizu et al. reported that high-molecular-weight fucoidan promotes greater increases in the proportion of murine cytotoxic T cells than middle- or low- molecular-weight fucoidan [18]. Park et al. reported that high-molecular-weight fucoidan enhances inflammation, while low-molecular-weight fucoidan shows anti-inflammatory activities through the suppression of Th1-mediated immune reactions [31]. Our results demonstrated the presence of HMWF anti-tumor activities following activation of innate immunity in vivo. To understand the mechanisms of the anti-tumor activities exhibited by IMWF and LMWF, further studies must be conducted.

In this study, we used fucoidan which was not changed the composition and sulfation degree in the process of hydrothermal treatment. Bilan et al. reported even a single algal species might contain, at least in minor amounts, several sulfated polysaccharides differing in molecular structure [32]. This result indicates fractionation of crude “fucoidans” may give non-fucoidan polysaccharides. Some results indicate the bioactivities of fucoidan are depended on not only their molecular weight but also their derived seaweeds and compositions [5,6]. When we compare the bioactivities of polysaccharides such as fucoidan, not only the molecular weight but also the derived seaweeds and compositions must be considered.

BALB/c mice (4- or 5-week-old, female) were purchased from CREA Japan (Osaka, Japan). Myd-88 knockout mice (BALB/c background, 4-week-old, female) were purchased from Oriental BioService, Inc., (Kyoto, Japan). All mice were maintained under conventional conditions. The use of the mice and the procedures used in this study were approved by the Animal Research Committee of Tottori University.

Fucoidan (HMWF group; M_w 300–330 kDa, lot No.1010) from Cladosiphon okamuranus was provided by Marine Products Kimuraya Inc. (Tottori, Japan). Teruya et al. reported that fucoidan from Cladosiphon okamuranus consisted of L-fucose, D-xylose, D-gulucuronic acid in the ratio of 4.0:0.03:1.0 [33]. Degree of sulfation (DS) was evaluated by the elemental analysis method to be 0.33–0.34. Degree of acetylation (DA) was determined by the ¹H NMR analysis method to be 0.2. Fucoidan samples of various molecular weights were prepared by the application of hydrothermal depolymerization method [34]. These values are given per mole of fucose residues. The molecular weights (MW) of fucoidan were calculated by the gel permeation chromatography method (columns: TSKgel PW_XL (Tosoh CO., Tokyo, Japan); mobile phase: 0.1 M NaNO₃; flow rate: 0.5 mL/min; column temperature: 40 °C; detector: RI; calibration: pullulan standard). Hydrothermal treatment of aqueous solution of fucoidan at 140 °C for 15–60 min gave the low molecular weight fucoidan group (LMWF group; M_w 6.5–40 kDa). Similar treatment at 140 °C for 5–8 min gave intermediate molecular weight fucoidan group (IMWF group; M_w 110–138 kDa). Under these hydrothermal conditions, all samples maintained their DS and DA values.

The murine colon cancer cell line (colon 26) was provided in 2-mm tumor tissue blocks by Cancer Institute Hospital (Tokyo, Japan). The blocks were dipped into cell banker^® (ZENOAQ, Inc., Fukushima, Japan) and stored at −80 °C until use.

The mice were anesthetized with inhalation of 3%–5% isoflurane (Intervet, Inc., Tokyo, Japan). The tumor tissue blocks were thawed at room temperature and washed three to four times with saline. The tumor tissue blocks were minced to 1-mm blocks, and pieces of the blocks were transplanted subcutaneously into the dorsal regions of the mice.

When the transplanted tumors grew to 10 mm in size, the mice were euthanized with inhalation of 5% isoflurane followed by cervical dislocation. The tumor tissues were extirpated, and the coats, microvessels and necrotic tissues were removed. After being washed with saline, the tumor tissues were minced to 1-mm blocks and transplanted subcutaneously into the dorsal regions of the mice. We used tumor tissues with passage numbers greater than two in our experiments.

Twenty-five BALB/c mice were randomized into four groups: the control group (control, n = 10), the LMWF group (LMWF, n = 5), the IMWF group (IMWF, n = 5) and the HMWF group (HMWF, n = 5).

Except for the control group, each fucoidan group was fed 5% (w/w) fucoidan with a normal powdered diet (CE-2; CREA Japan, Osaka, Japan). All group mice fed fucoidan were fed about 5 g/kg·day of fucoidan. The control group was fed a normal powdered diet. Before transplantation of the tumor tissues, the mice were bred for 28 days by being fed a powdered diet with or without each weight of fucoidan. After being bred for 28 days, each mouse underwent subcutaneous transplantation of one piece of tumor tissue into the dorsal region. Each mouse had been fed a powdered diet with or without fucoidan for 14 days. Then, all mice were euthanized with inhalation of 5% isoflurane followed by cervical dislocation. The tumor tissues were removed and weighed (g). After being weighed, the tumor tissues were fixed in 10% buffered formalin.

In the control and IMWF groups, thin sections (5 μm) were made from each sample for histopathological observation after hematoxylin-eosin (HE) staining. Each section was examined microscopically, and the number of cell divisions in each tumor was calculated. The calculation of cell divisions was performed in 10 fields at 400× magnification using five mice in each group. The mean score of 50 fields was considered to be the number of cell divisions in each group.

We also examined the apoptosis rate of the tumor tissues using terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphatate-biotin nick end labeling (TUNEL) staining. Each section was examined microscopically and the number of cell divisions in each tumor was calculated. The calculation of cell divisions was performed in 10 fields at 400× magnification using four mice in each group. The mean score of 40 fields was considered to be the number of cell divisions in each group.

Twenty-four BALB/c mice were randomized into four groups: the control group (control, n = 4), the LMWF group (LMWF, n = 4), the IMWF group (IMWF, n = 4) and the HMWF group (HMWF, n = 4).

Except for the control group, each fucoidan group was fed 5% (w/w) fucoidan with a normal powdered diet (CE-2; CREA Japan, Osaka, Japan). The control group was fed a normal powdered diet. Before transplantation of the tumor tissues, the mice were bred for 28 days by being fed a powdered diet with or without each weight of fucoidan. After being bred for 28 days, each mouse underwent subcutaneous transplantation of one piece of tumor tissue into the dorsal region. After the tumor tissues were transplanted (day 0), the number of days until the mice died was measured. Thereafter, a survival curve was created and the median survival time (days) was calculated.

Nine BALB/c mice were randomized into three groups: the control group (control, n = 3) fed a normal powdered diet (CE-2, CREA Japan, Osaka, Japan) for 70 days, the IMWF group (IMWF, n = 3) fed a normal powdered diet with 5% (w/w) IMWF for 70 days and the HMWF group (HMWF, n = 3) fed a normal powdered diet with 5% (w/w) HMWF for 70 days.

After breeding, all mice were euthanized with inhalation of 5% isoflurane followed by cervical dislocation. Spleens were obtained from each mouse and splenocytes were collected. Erythrocytes were removed by mixing with a lysis buffer (150 mM NH₄Cl, 10 nM KHCO₃ and 100 μM EDTA) for 10 min at room temperature.

FITC-conjugated anti-mouse NKG2A/C/E monoclonal antibodies (BD Bioscience, San Jose, CA) and PE-conjugated anti-mouse CD8a monoclonal antibodies (BD bioscience, San Jose, CA) were used. FITC- and PE-conjugated IgG2a isotype-matched antibodies were used as controls for background staining. Leukocytes were then stained at 4 °C using predetermined optimal concentrations of each antibody for 30 min. Antibody binding was analyzed on a FACSAria flow cytometer (BD Biosciences) by gating the cells with the forward and side light scatter properties of lymphocytes. NKG2A/C/E-positive and CD8a-negative lymphocytes were deemed to be NK cells. We calculated the percentage of NK cells in the splenic lymphocytes.

Ten Myd-88 knockout mice were divided into two groups (n = 5 each): the control/Myd-88 (control, n = 5) and the HMWF/Myd-88 groups. The HMWF/Myd-88 group was fed 5% (w/w) fucoidan with a normal powdered diet (CE-2; CREA Japan, Osaka, Japan). The control/Myd-88 group was fed a normal powdered diet. Before transplantation of the tumor tissues, the mice were bred for 28 days by being fed a powdered diet with or without fucoidan. After being bred for 28 days, each mouse underwent subcutaneous transplantation of one piece of tumor tissue into the dorsal region. Each mouse had been fed a powdered diet with or without fucoidan for 14 days. Then, all mice were euthanized with inhalation of 5% isoflurane followed by cervical dislocation. Tumor tissues and sera were obtained. The tumor tissues were weighed.

The serum interleukin (IL)-2 and IL-10 levels were quantified using a sandwich enzyme-linked immunosorbent assay (ELISA). The serum IL-2 and IL-10 levels were measured using murine IL-2 and IL-10 ELISA kits (Thermo SCIENTIFIC, Rockford, IL, USA) according to the manufacturer’s protocols.

The data are expressed as the mean ± S.D. The statistical analyses were performed using one-way ANOVA followed by Tukey-Kramer’s test or Welch’s t-test. We performed the log-rank test to evaluate the survival times of the tumor mice. A p-value < 0.05 was considered statistically significant.