Regulatory effects of Spirulina complex polysaccharides on growth of murine RSV-M glioma cells through Toll-like receptor 4

Authors


Correspondence

Keiji Shimizu, Department of Neurosurgery, Kochi Medical School, Kohasu, Okoh-cho, Nankoku, Kochi 783 8505 Japan. Tel/Fax: +81 88 880 2397/+81 88-880 2400; email: kshimizu@kochi-u.ac.jp

Abstract

This study is the first to report that Spirulina complex polysaccharides (CPS) suppress glioma growth by down-regulating angiogenesis via a Toll-like receptor 4 signal. Murine RSV-M glioma cells were implanted s.c. into C3H/HeN mice and TLR4 mutant C3H/HeJ mice. Treatment with either Spirulina CPS or Escherichia coli (E. coli) lipopolysaccharides (LPS) strongly suppressed RSV-M glioma cell growth in C3H/HeN, but not C3H/HeJ, mice. Glioma cells stimulated production of interleukin (IL)-17 in both C3H/HeN and C3H/HeJ tumor-bearing mice. Treatment with E. coli LPS induced much greater IL-17 production in tumor-bearing C3H/HeN mice than in tumor-bearing C3H/HeJ mice. In C3H/HeN mice, treatment with Spirulina CPS suppressed growth of re-transplanted glioma; however, treatment with E. coli LPS did not, suggesting that Spirulina CPS enhance the immune response. Administration of anti-cluster of differentiation (CD)8, anti-CD4, anti-CD8 antibodies, and anti-asialo GM1 antibodies enhanced tumor growth, suggesting that T cells and natural killer cells or macrophages are involved in suppression of tumor growth by Spirulina CPS. Although anti-interferon-γ antibodies had no effect on glioma cell growth, anti-IL-17 antibodies administered four days after tumor transplantation suppressed growth similarly to treatment with Spirulina CPS. Less angiogenesis was observed in gliomas from Spirulina CPS-treated mice than in those from saline- or E. coli LPS-treated mice. These findings suggest that, in C3H/HeN mice, Spirulina CPS antagonize glioma cell growth by down-regulating angiogenesis, and that this down-regulation is mediated in part by regulating IL-17 production.

List of Abbreviations:
CD

cluster of differentiation

CPS

complex polysaccharides

E.coli

Escherichia coli

IFN

interferon

Ig

immunoglobulin

IL

interleukin

LPS

lipopolysaccharides

NF-κB

nuclear factor kappa-light-chain-enhancer of activated B cells

NK

natural killer cells

OD

optical density

PBS-T

PBS, pH7.2, containing 0.05% Tween 20

SEM

standard error of the mean S. platensis, Spirulina platensis

TGF

transforming growth factor

Th

helper T lymphocytes

TLR

Toll-like receptor

TNF

tumor necrosis factor

Treg

immunoregulatory T lymphocytes

Toll-like receptor binding molecules are pathogen-associated factors that induce inflammatory responses [1], [2]. LPS, peptidoglycans, and flagellins are well known ligands for TLR4, TLR2, TLR5, respectively. While analyzing TLR4-activating factors extracted from algae including cyanobacterium, we found that Spirulina CPS are effective in inducing NF-κB in response to either TLR2 or TLR4 [3]. Spirulina is a Gram-negative, oxygenic, photosynthetic, filamentous cyanobacterium (blue-green alga) that Aztecs living near Lake Texcoco in Mexico and residents of Chad used historically as a nutritional and therapeutic supplement [4]. Spirulina CPS (Westphal fraction) are reportedly much less toxic than LPS derived from Salmonella abortus [5], but researchers have not yet extensively investigated their functions, for example as cytokine inducers or anti-tumor reagents. We have found that Spirulina CPS attenuate growth of hepatocellular carcinoma, in part by decreasing production of the inflammatory cytokine, IL-17, which signals through TLR4 (Okuyama H et al., unpublished data, 2010). This conclusion is based on a report that Spirulina CPS suppress tumor growth in C3H/HeN but not in C3H/HeJ mice, the latter being mutant in TLR4 [6].

Three types of Th lymphocytes have been characterized ([7]–[11]). Th1 cells primarily secrete IFN-γ and protect the host by killing tumor cells and intracellular microorganisms; they are also responsible for inducing delayed-type hypersensitivity. Th2 cells mainly secrete IL-4, IL-5, and, IL-13 and induce Ig E production, protecting the host from parasites; they can also induce allergic reactions. Th17 cells mainly secrete IL-17A, IL-17F, and IL-22 and protect the host from infection by bacteria such as Klebsiella pneumoniae and Staphylococcus aureus, and by fungi like Candida albicans. Korn et al. proposed the following three-step model for Th17 cell differentiation: (i) TGF-β and IL-6 induce differentiation, (ii) IL-21 amplifies the precursor frequency of Th17 cells, and (iii) IL-23 terminally differentiates cells and stabilizes the phenotype [11].

Recently, Grauer et al. tested TLR ligands, including ligands for TLR2, TLR3, TLR4, TLR5, TLR7 and TLR9, and reported that intratumoral injection of the TLR9 ligand, CpG-oligonucleotide, was the most effective of these at inhibiting growth of glioma GL261 [12]. The anti-tumor effect of CpG-oligonucleotide requires TLR9 expression on nontumor cells. TLR4 mediates, in part, the in vivo anti-tumoral effects of E. coli LPS against glioblastoma multiforme [13]. Treatment with E. coli LPS causes near complete elimination of subcutaneously growing gliomas in wild-type mice, but only produces a 50% reduction in these tumors in TLR4-deficient mice. Although it is well known that E. coli LPS have potent anti-tumor activity, these agents are highly toxic [14], [15]. Evaluation of less toxic forms of LPS, such as Bordetella pertussis LPS, has demonstrated anti-tumor activity [15]. Thus, we were interested in exploring the function of other less toxic CPS, namely those derived from Spirulina, regarding cytokine production and anti-tumor activity.

Here, we assessed Spirulina CPS as a potential immunotherapy reagent against glioma and showed that Spirulina CPS suppresse growth of murine RSV-M glioma cells and angiogenesis in C3H/HeN, but not in TLR4-deficient, C3H/HeJ mice. We also found that administration of anti-IL-17 antibodies suppresses growth of murine RSV-M glioma cells. Spirulina CPS have a unique feature in that they suppresses angiogenesis, in part by regulating the amount of IL-17.

MATERIALS AND METHODS

Animals

C3H/HeN (wild type) mice and C3H/HeJ (tlr4 mutation) mice purchased from Japan SLC, Hamamatsu, Japan were maintained in our animal facility in specific pathogen free conditions. Both strains of mice used in this study were 10-week-old females. All experiments were performed according to the ethical guidelines of Kochi Medical School, Kochi University.

Preparation of Spirulina complex polysaccharides and reagents

Westphal fractions (hot phenol extracts) of Spirulina pacifica (a generous gift from Dr. Genrald Cysewski, Cyanotech, Kailua-Kona, Hawaii and Nobuyuki Miyaji, Toyo Koso Kagaku, Chiba, Japan) were prepared according to described methods [16]. Spirulina pacifica was first selected from a strain of edible Spirulina (Arthrospira) platensis in 1984 (Cyanotech). Briefly, dried cells were washed with acetone, dispersed in distilled water, and then extracted by addition of 90% phenol-water with vigorous agitation at 68°C. The crude preparation was dialyzed to remove phenol and then freeze-dried. The sample was dissolved in water and the remaining solid material eliminated by ultracentrifugation (100,000 g, 6 hrs); the Westphal fraction was obtained from the resultant jelly sediment. According to mass spectral analysis and sodium dodecyl sulfate polyacrylamide gel electrophoresis, the molecular mass of the sample was estimated to be 1000 to 20,000. We call this fraction complex polysaccharide. As a reference, E. coli 055:B5 LPS (Smooth type, Sigma-Aldrich, St. Louis, MO, USA) was used.

Assay of anti-glioma activity

The anti-tumor effect of Spirulina CPS was examined using murine RSV-M glioma cells [17]. RSV-M cells (5,000,000 cells/mouse) were implanted s.c. on each mouse's back. Spirulina CPS (20 μg, 100 μg), E. coli LPS (20 μg) or saline were injected intraperitoneally weekly, starting 6 days after tumor inoculation.

In some experiments, tumors were surgically removed 22 days after implantation and fresh RSV-M glioma cells re-transplanted eight days later. Tumor growth was measured three times a week after implantation. Tumor volume was calculated using the formula volume = width2 × length/2.

Administration of anti-interleukin-17 or anti-interferon-γ antibodies to tumor-bearing mice

To evaluate any anti-tumor effect of anti-IL-17 or anti-IFN-γ antibodies in glioma-bearing mice, RSV-M cells (5,000,000) were implanted subcutaneously on the backs of either C3H/HeN or C3H/HeJ mice. Anti-IL-17 antibodies (R & D Systems, Minneapolis, MN, USA; clone 50104, rat IgG2a, 150 μg/mouse), anti-IFN-γ antibodies (Endogen, Rockford, IL, USA; clone R4–6A2, rat IgG1, 150 μg/mouse), or IgG from rat serum (Sigma-Aldrich 150 μg/mouse) were administered i.p. on day 4 after glioma transplantation and the resultant tumors measured as described above.

Administration of anti-CD4, anti-CD8, and anti-asialo GM1 antibodies to tumor-bearing mice

To explore the mechanisms of antitumor activity, mice were injected i.p. with saline, anti-CD4 monoclonal antibodies (GK1.5, rat IgG2b) or anti-CD8 monoclonal antibodies (53–6.7, rat IgG2a, BioLegend, San Diego, CA, USA) at a dose of 100 μg/100 μL/mouse on days − 1, 0, and + 3 before and after tumor transplantation. In order to immuno-deplete NK cells and macrophages, mice were injected i.p. with control rabbit IgG (1 mg/100 μL/mouse) or rabbit anti-asialo GM1 IgG fraction (1 mg/100 μL/mouse) on days − 1, 0, and + 3 before and after tumor transplantation (Wako Pure Chemical Industries, Osaka, Japan). GK1.5 hybridomas were grown as ascites in BALB/c nu/nu mice, and the IgG fraction prepared by 50% ammonium sulfate precipitation. Other class matched control antibodies were purchased from Sigma-Aldrich.

Cytokine measurements by enzyme-linked immunosorbent assay

Capture antibodies (1 ∼ 5 μg/mL in 0.05 M sodium carbonate, pH 9.6) were added to coat each well of Immuno 96 microwell plates (9018, Corning Costar, Ithaca, NY, USA) and incubated over night at 4°C. After washing the wells twice with 0.15 M PBS-T, blocking buffer (PBS containing 0.05% Tween 20 and 0.5% BSA) was added and the mixture incubated for 30 mins at room temperature. After removing blocking buffer, sera were diluted five times with blocking buffer and wells were incubated for 2 hrs at room temperature or at 37°C. After washing the wells with PBS-T three times, biotin-coupled antibodies (detection antibodies) against each cytokine were added to each well and the resultant mixtures incubated at room temperature for 45 mins. After washing with PBS-T three times, streptavidin-horseradish peroxidase (Invitrogen, Camarillo, CA, USA) was added and incubated at room temperature for 45 min according to the manufacturer's protocol. After washing wells with PBS-T three times, substrate solution containing freshly prepared 0.7 mg/mL of o-phenylenediamine dihydrochloride (Wako Pure Chemical Industries) in citric acid buffer, pH 5.0 containing 0.006% hydrogen peroxide was added and incubated for 10 mins. Reactions were stopped by adding 50 μL of 10% sulfuric acid to each well. Cytokine concentration was estimated by measuring the OD at 490 nm using a micro plate reader (Model 680, Bio-Rad, Hercules, CA, USA) and a standard curve.

Antibodies used in enzyme-linked immunosorbent assay

Capture antibodies included rat anti-mouse IFN-γ antibodies (1 μg/mL), clone R46A2 (Becton Dickinson, Mountain View, CA, USA) and rat anti-IL17 antibodies (2 μg/mL), clone TC11–18H10.1 (Becton Dickinson). Detection antibodies included rat anti-mouse IFN-γ antibodies, clone XMG1.2 (Becton Dickinson) and rat anti-IL17 antibodies, clone TC11–8H4.1 (Becton Dickinson).

Immunohistochemistry

RSV-M gliomas taken from saline- or Spirulina CPS-treated C3H/HeN mice 22 days after tumor implantation were embedded in optimal cutting temperature compound (Sakura Finetek USA, Torrance, CA, USA) and frozen. Frozen tumors were sectioned and stained with anti-CD31 antibodies (rat IgG2a, κ, 390, BioLegend) using Simple stain mouse MAX-PO (F[ab]’2 goat anti-rat Ig and peroxidase coupled to the amino acid polymer) and 3,3’-diaminobenzidine according to the manufacturer's protocol (Nichirei-Biosciences, Tokyo, Japan). Samples were counterstained with hematoxylin.

Statistical analysis

Differences in mean values between groups were calculated using unpaired two-tailed Student's t-tests. P values by Student's t-test are shown unless otherwise indicated.

RESULTS

Anti-tumor activities of Spirulina complex polysaccharides against murine RSV-M glioma cells

To clarify the anti-tumor effect of Spirulina CPS, we administered them i.p. to RSV-M glioma-bearing C3H/HeN and TLR4-mutant C3H/HeJ mice weekly starting six days after inoculation with glioma tumor cells. Treatment of C3H/HeN mice with Spirulina CPS (100 μg) or of C3H/HeN mice with E. coli LPS (20 μg) significantly suppressed RSV-M glioma tumor growth; however, we saw no effect in C3H/HeJ mice (1a). A lower dose of Spirulina CPS (20 μg) also significantly suppressed growth of RSV-M glioma cells, but only in C3H/HeN mice (1b). These results suggest that Spirulina CPS and E. coli LPS exert their effects through TLR4.

Figure 1.

Effect of Spirulina CPS on RSV-M glioma cell growth in C3H/HeN and C3H/HeJ mice. Five million RSV-M glioma cells were inoculated s.c. on the backs of C3H/HeN or C3H/HeJ mice. (a) Spirulina CPS (100 μg), E. coli LPS (20 μg), or saline were injected i.p. every week starting 6 days after tumor inoculation. Data represent means ± SEM of five female 10-week-old mice per group. There is a significant difference between the following groups (P < 0.05; Student's t-test): saline-treated versus Spirulina CPS-treated group on days 17, 19, and 21 after tumor transplantation in C3H/HeN mice. (b) Spirulina CPS (20 μg), E. coli LPS (20 μg), or saline were injected i.p. every week starting 6 days after tumor inoculation. Data represent means ±SEM of six female 10-week-old mice per group. There are significant differences between the following groups (P < 0.01; Student's t-test): saline-treated versus Spirulina CPS-treated group and saline-treated versus E. coli LPS-treated group on days 12, 14, and 19 after tumor transplantation in C3H/HeN mice.

The effect of Spirulina complex polysaccharides on cytokine production in the sera of RSV-M glioma-bearing mice

We first monitored serum concentrations of IFN-γ and IL-17 in C3H/HeN and C3H/HeJ mice 11 days after tumor inoculation. Glioma cells weakly stimulated production of IL-17 in both C3H/HeN and C3H/HeJ tumor-bearing mice (Table 1). E. coli LPS induced IFN-γ production in C3H/HeN mice, but not in C3H/HeJ mice, which are TLR4 mutants (Table 1). E. coli LPS also induced much greater IL-17 production in C3H/HeN mice than in C3H/HeJ mice (Table 1). These results confirmed that TLR4 mediates E. coli LPS-induced enhancement of production of these cytokines. In contrast, neither dose of Spirulina CPS induced or marginally induced production of IFN-γ in either C3H/HeJ or C3H/HeN mice (Table 1). Neither dose of Spirulina CPS significantly increased IL-17 concentrations in either C3H/HeJ or C3H/HeN mice compared to that in the saline-treated group (Table 1). We also compared serum concentrations of IL-17 in mice administered 20 μg of Spirulina CPS and mice administered 20 μg E. coli LPS in the experiment shown in Figure 1b (Table 2). At a dose of 20 μg, Spirulina CPS-treated mice had lower concentrations of IL-17 than did the saline-treated group, suggesting that Spirulina CPS down-regulate production of IL-17 induced by glioma inoculation. In contrast, E. coli LPS-treated mice had higher concentrations of IL-17 than did saline-treated mice. However, at a higher dose (100 μg), Spirulina CPS marginally enhanced the serum concentration of IL-17; this enhancement was not significant (Table 1). Taken together, these results suggest that Spirulina CPS weakly suppress or do not significantly increase the serum concentration of IL-17 compared with E. coli LPS and induce a different balance between IL-17 and IFN-γ than that induced by E. coli LPS.

Table 1. Serum cytokine concentrations in glioma RSV-M bearing mice treated with Spirulina CPS or E. coli LPS
Mouse Strain CytokineC3H/HeJC3H/HeN
IFN-γIL-17IFN-γIL-17
  1. On day 11 after tumor inoculation, IFN-γ and IL-17 concentrations were measured as described in Materials and Methods. Detection limit of IFN-γ is 10 pg/mL. Results represent mean values of five mice (pg/mL ± SEM). Spi CPS, Spirulina pacifica complex polysaccharides. *, P < 0.01, compared with saline-treated group, (Student's t-test).
No tumor< 102114 ± 176< 102121 ± 101
Glioma + Saline< 102933 ± 326< 102962 ± 482
Glioma + E. coli LPS< 104872 ± 343*474 ± 9910,226 ± 1,842*
Glioma + Spi CPS 20 μg< 103565 ± 270< 102272 ± 116
Glioma + Spi CPS 100 μg< 102958 ± 31827 ± 153876 ± 664
Table 2. Serum IL-17 concentrations in glioma RSV-M bearing mice treated with low doses of Spirulina CPS or E. coli LPS
Mouse strainC3H/HeJC3H/HeN
  1. On day 13 after tumor inoculation in the experiment of Figure 1b, mice serum IL-17 concentrations were measured as described in Materials and Methods. Results are shown as mean values of six mice (pg/mL ± SEM). Spi CPS, Spirulina pacifica complex polysaccharides. *, P < 0.01, compared with no tumor group; **, P < 0.01, compared with saline-treated group; ***, P < 0.01, compared with saline-treated group (all Student's t-test).
No tumor2105 ± 1532100 ± 130
Glioma + saline3184 ± 343*2864 ± 151*
Glioma + E. coli LPS 20 μg2965 ± 4314069 ± 380**  
Glioma + Spi CPS 20 μg3131 ± 3881882 ± 117***

Anti-tumor activity of anti-interleukin-17 antibodies against glioma

To clarify the effects of IL-17 and IFN-γ on growth of RSV-M glioma cells, we administered antibodies against each cytokine to C3H/HeJ and C3H/HeN mice four days after glioma transplantation. Spirulina CPS suppressed tumor growth in C3H/HeN but not in C3H/HeJ mice, as shown in Figures 1 and 2. Treatment with anti-IFN-γ antibodies alone had no effect on glioma cell growth in either mouse strain (Fig. 2). By contrast, anti-IL-17 antibodies alone suppressed growth of glioma cells in both strains of mice. Treatment with anti-IL-17 antibodies did not increase the degree of suppression by Spirulina CPS in C3H/HeN mice, indicating that Spirulina CPS and anti-IL-17 antibodies do not have a synergistic suppressive effect on glioma growth (2a). Overall, in both strains of mice, IL-17 apparently supports glioma growth. In accordance with this, transplantation of glioma cells increased concentrations of IL-17 in the sera of both C3H/HeJ and C3H/HeN mice (2b).

Figure 2.

Effects of anti-IL-17 and anti-IFN-γ antibodies on glioma cell growth. (a) Five million RSV-M glioma cells were inoculated s.c. on day 0 as described in Materials and Methods. Each antibody (0.1 mL of 1.5 mg/mL antibodies in saline) was administered i.p. to mice four days after tumor inoculation. Left panel: Anti-IL-17 antibodies suppressed growth of RSV-M glioma cells in C3H/HeJ mice from days 8 to 12 after tumor inoculation. Results represent means ± SEM of five female mice. There is a significant difference between the following groups (P < 0.05; Student's t-test): anti-IL-17 antibodies-treated and other groups on days 8, 10, and 12 after tumor inoculation. Right panel: Anti-IL-17 antibodies, but not anti-IFN-γ antibodies, suppressed tumor growth in C3H/HeN mice. There are significant differences between the following groups (P < 0.05; Student's t-test): rat-IgG-treated versus anti-IL-17 antibodies-treated group, Spirulina CPS (100 μg) + rat-IgG group, and Spirulina CPS (100 μg) + anti-IL-17 antibodies group on days 10 and 12 after tumor inoculation. Results represent means ± SEM of five female mice. (b) Serum IL-17 concentrations three weeks after tumor inoculation in glioma RSV-M-bearing C3H/HeJ and C3H/HeN mice that had received rat IgG, anti-IFN-γ antibodies, or Spirulina CPS + rat IgG. Results represent means of five female mice (pg/mL ± SEM). Serum IL-17 concentrations in the groups that received anti-IL-17 antibodies are not shown because the presence of anti-IL-17 antibodies prevented accurate measurement. *, P < 0.05: no tumor versus rat IgG-treated group; Student's t-test. **, P < 0.01: rat IgG-treated versus Spirulina CPS + rat IgG-treated group; Student's t-test.

Three weeks after tumor transplantation, we found that glioma-bearing C3H/HeN mice treated with 100 μg of Spirulina CPS plus rat IgG had lower concentrations of IL-17 in their sera than did those treated with rat IgG alone (2b). We did not observe a similar suppressive effect of Spirulina CPS on IL-17 production in C3H/HeJ mice, confirming the importance of signaling through TLR4.

Spirulina complex polysaccharides enhance immunity against RSV-M glioma

To examine whether Spirulina CPS induce immunity against RSV-M glioma, we resected implanted primary tumors from C3H/HeN mice and re-inoculated the same RSV-M glioma cells into the mice eight days later. Before re-implantation, we had treated these mice with E. coli LPS, Spirulina CPS or saline as described in Materials and Methods. Spirulina CPS severely suppressed growth of re-transplanted RSV-M cells in C3H/HeN mice (Fig. 3). Saline minimally but significantly suppressed growth of re-transplanted glioma cells. Glioma growth in E. coli LPS-treated mice was as robust as in normal mice that had not received primary tumors. This result suggests that Spirulina CPS enhance the ability to generate acquired immunity, while E. coli LPS do not, and that Spirulina CPS could serve as an adjuvant to increase the T cell immunity.

Figure 3.

Spirulina CPS-treated mice mount a strong immune response against re-implanted tumors. Five million RSV-M glioma cells were inoculated s.c. on the backs of C3H/HeN mice and these tumor-bearing mice were treated with either E. coli LPS or Spirulina CPS as in Figure 1. Twenty-two days later, the tumors were surgically removed from E.coli LPS (20 μg) or Spirulina CPS (100 μg)-treated tumor-bearing C3H/HeN mice and the mice re-transplanted with RSV-M glioma cells on day 30. Normal: RSV-M glioma cells were inoculated into normal C3H/HeN mice. Tumor growth was measured as described in Materials and Methods. Data represent means ± SEM of five female 10-weeks-old mice per group. There is a significant difference between the following groups (P < 0.05; Student's t-test): E. coli LPS-treated versus Spirulina CPS-treated group on days 8 and 10 after tumor re-transplantation; saline-treated versus Spirulina CPS-treated group on day 10 after tumor re-transplantation.

Spirulina complex polysaccharides suppresses angiogenesis in RSV-M gliomas

Because IL-17 reportedly promotes angiogenesis and tumor growth [18] and angiogenesis reportedly a key event in the progression of malignant gliomas [19], we examined the effect of Spirulina CPS on angiogenesis by staining gliomas from Spirulina CPS-treated C3H/HeN mice in the experiment shown in Figure 1 with antibodies against CD31. CD31 is highly expressed on endothelial cells and concentrates at the junctions between them. We found fewer CD31 stained cells in gliomas from Spirulina CPS-treated mice than in those from saline-treated or E. coli LPS-treated mice (Fig. 4a, b). This result suggests that decreased angiogenesis mediates the regulatory effect of Spirulina CPS on growth of murine RSV-M glioma cells, in part due to decreased IL-17 production. The degree of angiogenesis did not always correlate with the serum concentration of IL-17 (Table 1 and Fig. 4).

Figure 4.

Spirulina CPS suppresses angiogenesis in RSV-M gliomas. Frozen sections of gliomas prepared from tumor-bearing C3H/HeN mice in Fig. 1 were stained with anti-CD31 antibodies as described in Materials and Methods. (a) Representative photomicrographs from each group of mice are shown. Glioma bearing mice were treated with saline, E.coli LPS (20 μg) and Spirulina CPS (20 μg, 100 μg) as in Figure 1. In the case of Spirulina CPS, only one group (100 μg) section is shown. Bar, 50 μm. (b) Each group consisted of ten randomly chosen 300 μm × 400 μm fields from tumor sections for each treatment group. Cell numbers stained with anti-CD31 antibodies are shown as mean values ± SEM. There are significant differences between the following groups (P <0 .01; Student's t-test): saline versus either dose of Spirulina CPS; E. coli LPS versus either dose of Spirulina CPS.

CD8+ T and asialo GM1+ cells mediate suppression of growth of RSV-M glioma by Spirulina complex polysaccharides

To identify which cells suppress glioma growth, we depleted a subpopulation of effector cells by injecting antibodies against T lymphocyte markers, or a marker for NK cells and macrophages, before administering Spirulina CPS (Fig. 5a, b). Anti-CD8 antibodies enhanced glioma growth, suggesting that CD8+ T cells are involved in suppression of tumor growth (5a). Although anti-CD4 antibodies alone had no effect on tumor growth, they enhanced it when administered with anti-CD8 antibodies (5a). This suggests that CD4+ T cells are also involved in tumor immunity against this glioma. This conclusion is confirmed by the fact that treating tumor bearing mice with anti-CD8 and anti-CD4 antibodies enhanced angiogenesis compared with that seen in mice treated with anti-CD8 antibodies alone (Fig. 5c, d). These results suggest that both CD4+ T and CD8+ T cells mediate the anti-tumor effect of Spirulina CPS.

Figure 5.

Both T and asialo GM1 positive cells are involved in the glioma-suppressing activity of Spirulina CPS. Five million RSV-M glioma cells were inoculated s.c. on the backs of C3H/HeN mice. Spirulina CPS (100 μg) or saline was injected i.p. every week starting 6 days after tumor inoculation. Data represent means ± SEM of five female 10-week-old mice per group. Antibodies were administered as described in Materials and Methods. (a) Effects of anti-CD4 and anti-CD8 antibodies on the suppressive effect of Spirulina CPS (Spi CPS) on the growth of glioma cells in C3H/HeN mice. There is a significant difference between the following groups at the indicated periods (P < 0.05; Student's t-test): saline versus Spi CPS; Spi CPS + anti-CD8 antibodies versus Spi CPS + rat IgG from days 13 to 21; Spi CPS + anti-CD8 antibodies versus Spi CPS + anti-CD4 + anti-CD8 antibodies from days 13 to 18. (b) Effect of anti-asialo GM1 antibodies on the suppressive effect of Spirulina CPS on the growth of glioma cells in C3H/HeN mice. There is a significant difference between the following groups at the indicated periods (P < 0.05; Student's t-test): Spi CPS + anti-asialo GM1 antibodies versus Spi CPS + rabbit IgG from days 11 to 21. (c) Frozen sections of gliomas prepared from tumor-bearing C3H/HeN mice in (a) and (b) were stained with anti-CD31 antibodies as described in Materials and Methods. Representative photomicrographs from each group of mice are shown. Bar, 50 μm. (d) CD31 positive cells were compared among groups treated with anti-CD4 antibodies, anti-CD8 antibodies, anti-CD4 and anti-CD8 antibodies, and anti-asialo GM1 antibodies. CD31 positive cells in ten randomly chosen 300 μm × 400 μm fields of tumor sections for each treatment were counted. The cell numbers stained with anti-CD31 antibodies are shown as means ± SEM. Asterisks indicate significant differences between the groups (P < 0.01; Student's t-test).

Anti-asialo GM1 antibodies enhanced tumor growth as efficiently as did anti-CD4 and anti-CD8 antibodies, suggesting that NK cells or macrophages are involved in suppression of the growth of glioma cells by Spirulina CPS (5b). We observed equivalent angiogenesis in mice treated with anti-CD4 and anti-CD8 antibodies and those treated with anti-asialo GM1 antibodies (5d). Taken together, both T cells and NK cells or macrophages contribute equally to suppression of glioma growth by Spirulina CPS.

DISCUSSION

Here, we report that Spirulina CPS suppresses growth of RSV-M glioma cells in C3H/HeN, but not in TLR4-mutant C3H/HeJ, mice (Fig. 1). Anti-IL-17 antibodies suppressed RSV-M gliomas in both C3H/HeN and C3H/HeJ mice, suggesting that IL-17 supports RSV-M glioma growth (Fig. 2). The fact that Spirulina CPS and anti-IL-17 antibodies did not work synergistically to suppress RSV-M glioma cell growth in C3H/HeN mice (2a) supports this observation. Serum concentrations of IL-17 in mice administered Spirulina CPS plus IgG were significantly lower than those in mice administered IgG alone, suggesting that Spirulina CPS down-regulate production of IL-17. We observed the suppressive effect of Spirulina CPS on serum IL-17 concentrations in C3H/HeN, but not in C3H/HeJ, mice (2b). Because anti-IL-17 antibodies neutralize the function of IL-17 and Spirulina CPS reduce its serum concentration, they do not synergize in suppressing growth of gliomas. We speculate that treatment of glioma-bearing C3H/HeN mice with either anti-IL-17 antibodies or Spirulina CPS results in smaller amounts of active IL-17. We conclude, therefore, that the suppressive effect of Spirulina CPS on RSV-M glioma growth is due, in part, to the regulation of the ability to produce IL-17.

Tartour et al. reported that IL-17 promotes tumorigenicity of human cervical tumors in nude mice [20]. Benchetrit et al. reported that tumor-specific T lymphocytes activated by IL-17 produce significantly more suppression of growth of IL-17 DNA-transfected plasmacytoma J558L and mastocytoma P815 cells than of vector-transfected tumors [21]. They concluded that IL-17 is a pleiotropic cytokine with potential pro- or anti-tumor effects, depending on the immunogenicity of the tumor models investigated. We have provided here direct evidence for IL-17 involvement in tumor growth by demonstrating that administration of anti-IL-17 antibodies suppresses growth of RSV-M gliomas (Fig. 2).

Our data suggest that E. coli LPS suppress glioma growth by inducing IL-17 and IFN-γ and that Spirulina CPS suppresses it without strongly inducing either of these cytokines (Fig. 1 and Tables 1, 2). These results remind us that although Spirulina CPS suppress glioma growth by decreasing angiogenesis, E. coli LPS suppress it without doing so (Figs. 1, 4).

Martin-Orozco et al. reported that Th17 cells promote cytotoxic T cell activation as part of the anti-tumor immune response [22]. They found that IL-17A-deficient mice are more susceptible to development of pulmonary melanoma metastases and that adoptive transfer of tumor-specific Th17 cells prevents tumor development more effectively than do Th1 cells. They also reported that Th17 cells promote dendritic cell recruitment to tumor tissues and that the draining lymph nodes contain increased numbers of CD8α+ dendritic cells carrying tumor-derived materials. By contrast, IL-23, a Th17 maturation and maintenance factor, reportedly promotes tumor growth [23]. IL-23 may have additional functions besides maintaining Th17 cells, or Th17 cells may have pro- or anti-tumor activities depending on the magnitude of cytotoxic T lymphocyte response and the tumor microenvironment.

Lipopolysaccharides from E. coli and B. pertussis reportedly have anti-tumor activity [14], [15]. E. coli LPS exert their effects by forming a complex with TLR4-MD2 [24] and induce NF-κB much more efficiently via TLR4 than via TLR2 [3]. By contrast, Spirulina CPS induce NF-κB weakly through either TLR2 or TLR4 [3]. We observed that pre-incubating each of these polysaccharides with the antibacterial drug polymyxin B (data not shown) completely abolishes the ability of either E. coli LPS or Spirulina CPS to induce NF-κB, suggesting a similarity in their physicochemical nature. It is possible that Spirulina CPS have a moiety that resembles lipid A, since polymyxin B reportedly binds to the lipid A portion of bacterial LPS [25]. Porphyromonas gingivalis LPS reportedly contain multiple lipid A species that interact functionally with both TLR2 and TLR4 [26]. In the course of analyzing TLR4-responsive materials extracted from algae including cyanobacterium, we found that both Spirulina CPS and Petalonia binghamiae polysaccharides activate TLR4 and TLR2 equally and by consequence up-regulate NF-κB, suppressing growth of MH134 hepatocellular carcinoma [3].

Studies using Spirulina polysaccharide preparations free of LPS have demonstrated anti-tumor activity. A sulfated polysaccharide from Spirulina platensis that chelates calcium suppressed murine melanoma metastasis [27]. Balachandran et al. reported that a high molecular weight polysaccharide fraction from S. platensis induces IL-1β and TNF-α mRNAs in human monocyte THP-1 cells by activating NF-κB and that this induction is suppressed by antibodies to CD14 and TLR2, but not by antibodies to TLR4 [28]. Akao et al. reported that an orally administered water-soluble extract of S. platensis enhances tumoricidal NK activation through the MyD88 pathway [29]. They found that IFN-γ-mediated activation of NK, but not CD8 T cells, suppress B16 melanoma growth. By contrast, Spirulina CPS do not activate NK cells, although spleen cells from C3H/HeN mice injected with E. coli LPS show remarkable killing of NK-sensitive YAC-1 cells (Okuyama H et al., unpublished data, 2010). This finding is reminiscent of findings reported here that Spirulina CPS can induce immunity against RSV-M glioma, whereas E. coli LPS do not induce immunity against RSV-M glioma in a tumor re-challenge assay (Fig. 3).

We have analyzed the anti-tumor effect of Spirulina CPS against melanoma B16-F10 growth in C57BL/6 and C57BL/6-IFN-γ KO mice, C57BL/6-Ifngtm1Ts [30]. In C57BL/6 mice, neither E. coli LPS nor Spirulina CPS show significant anti-tumor activity against B16-F10 melanoma [3]. However, in IFN-γ KO mice, Spirulina CPS, but not E. coli LPS, suppress growth of this melanoma line, suggesting that Spirulina CPS exert an anti-tumor effect in the absence of IFN-γ [3]. Indeed, Spirulina CPS did not induce greater IFN-γ production in glioma-bearing mice than did E. coli LPS (Table 1). Spirulina CPS may enhance tumor immunity without inducing IFN-γ. Shankaran et al. proposed the concept “cancer immunoediting” to account for why tumors that develop in the presence of an intact immune system that includes lymphocytes and IFN-γ are less immunogenic than tumors that develop in immunodeficient hosts [31]. It may be easier to clarify the antitumor effect of Spirulina CPS in IFN-γ deficient subjects where glioma cells are not edited.

Anti-asialo GM1 antibodies also abolish the anti-tumor activity of Spirulina CPS (5b). Since we observed no difference between Spirulina CPS-treated and control mice (data not shown) in the number of NK cells stained with anti-CD49b antibodies (clone DX5), asialo GM1 positive macrophages may be involved in the anti-tumor activity of Spirulina CPS. Indeed, we observed that Spirulina CPS induce F4/80 positive macrophages in gliomas and that there are fewer of these after treatment with either anti-CD4 antibodies and anti-CD8 antibodies, or anti-asialo GM1 antibodies (data not shown).

These results suggest that asialo GM1 positive macrophages are important in the anti-tumor activity that occurs following administration of Spirulina CPS to mice. We found that Spirulina CPS induce smaller amounts of IL-6 and TGF-β in antigen-presenting cells such as macrophages and dendritic cells than do E. coli LPS (Okuyama H et al., unpublished data, 2010). This may explain why Spirulina CPS does not amplify IL-17 production. On the other hand, the efficacy of anti-tumor responses of TLR-activating factors may depend on the tumor microenvironment.

We also suggest that both CD4+ T and CD8+ T cells are important in the antitumor response against this glioma. Treatment with anti-CD8 antibodies significantly enhanced glioma growth in Spirulina CPS-treated mice and addition if anti-CD4 antibodies further amplified this enhancement (5a). These findings are in accordance with the fact that Spirulina CPS helped C3H/HeN mice to induce acquired immunity against glioma (Fig. 3). Angiogenesis assessed with anti-CD31 antibodies also confirmed that both CD4+ T and CD8+ T cells are involved in antitumor activity of Spirulina CPS against gliomas (Fig. 5c, d).

Our data suggest that Spirulina CPS suppresses RSV-M glioma cell growth by regulating IL-17 production. Numasaki et al. reported that IL-17 promotes angiogenesis not only via stimulation of vascular endothelial cell migration but also via elaboration of proangiogenic factors, which creates an imbalance between angiogenesis activators and inhibitors in the vasculature [18]. It is thought that angiogenesis is necessary for both tumor growth and migration of leukocytes into tumors to suppress tumor growth. In the case of RSV-M glioma cells, reducing IL-17 concentrations suppresses tumor growth, as revealed by the inhibitory effects of anti-IL-17 antibodies (Fig. 2). There were fewer endothelial cells expressing CD31 in RSV-M gliomas from Spirulina CPS-treated mice than in those from control mice (Fig. 4). There was a definite decrease in numbers of endothelial cells in gliomas from Spirulina CPS-treated mice but not in those from E. coli LPS-treated mice. The lower concentrations of IL-17 induced by Spirulina CPS compared with E. coli LPS may lead to less angiogenesis. Although the serum IL-17 concentration in the Spirulina CPS (100 μg)-treated group was slightly higher than that in the saline-treated group (Table 1), the degree of expression of CD31 in the former group was much less than that in the control group (4b). This suggests that not only down-regulation of IL-17, but also other factors, mediate the suppression of angiogenesis by Spirulina CPS.

In conclusion, Spirulina CPS suppress glioma growth in a TLR4 dependent way by reducing angiogenesis, in part by regulating the concentration of IL-17. Activating both T cells and macrophages against tumors whose growth is highly dependent on angiogenesis is also useful.

ACKNOWLEDGMENTS

We are grateful to Dr. G. Cysewski (Cyanotech) and Mr. N. Miyaji (Toyo Koso Kagaku) for supplying Spirulina pacifica. This work was partly supported by a Japan Society for the Promotion of Science Grant-in-Aid for Scientific Research (KAKENHI 21550168 and 24791508).

DISCLOSURE

The authors have no financial conflicts of interest.

Ancillary