Fibrinogen conjugated with Alexa Fluor 488 and IMDM culture medium was purchased from Invitrogen (OR, USA), and fibrinogen was purchased from Sigma-Aldrich (St. Louis, MO, USA). Blocking antibodies of mouse β₁ (Biolegend, CA, USA, HMb1-1) and mouse β₃ integrins (Santa Cruz Biotechnology, Santa Cruz, CA, USA, 2C9.G2), human α₅β₁ (Chemicon, Millipore, Billerica, MA, USA, HA5) and human α_vβ₃ integrins (Santa Cruz Biotechnology, HM2034) were prepared according to the manufacturer's instructions. G. lucidum polysaccharides were purchased from Johnsun Mushroom (Hangzhou, China), and dissolved in phosphate-buffered saline (PBS).

Melanoma cells B16 and A375 were obtained from the Shanghai Cell Bank (Chinese Academy of Science, Shanghai, China). Cells were cultured in IMDM medium and 10% fetal bovine serum (FBS) was added in a humidified cell incubator (37˚C, 5% CO₂). These cells were harvested with trypsin and EDTA. Subsequently, PBS-washed cells were suspended in culture medium for further use. Six- to eight-week-old C57Bl/6J healthy male mice (specific pathogen free) were obtained from the Jilin University Animal Center (Changchun, China). All of the protocols applied in the animal experiments were approved by the Animal Care and Use Committee of Northeast Dianli University (Jilin, China).

Melanoma cells were harvested and resuspended in IMDM. For the adhesion assay, Alexa Fluor 488 conjugated fibrinogen (green fluorescence) was used. Each aliquot (0.1 ml) of tumor cells was mixed with fibrinogen (1 mg/ml) and incubated at 37˚C for 30 min. After washing twice, cells were suspended in 0.5 ml of PBS for further analysis by a flow cytometer (FACScan, Beckman-Coulter, Miami, FL, USA).

Human NK cells were isolated from human peripheral blood mononuclear cells (PBMCs) with a negative magnetic bead NK cell isolation kit (Miltenyi Biotec, Bergisch Gladbach, Germany). In brief, B cells, T cells, stem cells, monocytes, dendritic cells, granulocytes and erythroid cells were labeled by a cocktail of magnetically biotin-conjugated antibodies. Highly pure NK cells were separated by depletion of magnetically labeled cells in the MicroBead. Mouse NK cells were separated from spleen cells with a similar kit (Miltenyi Biotec).

An NK cytotoxic assay was performed in the presence or absence of fibrinogen. In brief, Na⁵¹CrO₄ pre-labeled tumor cells in 0.05 ml (1×10⁴) culture medium were pre-incubated with 0.05 ml of various concentrations of fibrinogen in the presence (0.1 mg/ml) or absence of G. lucidum polysaccharides for 30 min at 37˚C. Subsequently, 100:1 of NK cells (0.05 ml) were added. Following 4 h of co-incubation at 37˚C, cell culture plates were centrifuged at 1500 rpm for 10 min, and supernatants (0.05 ml) were collected for the radioactivity measurement. The cytotoxicity was measured using the formula % = (A − B)/(C − B) × 100, where A is radioactivity in the test well, B is spontaneous radioactivity from the well target cells without effect cells, and C is the total target released radioactivity.

Data were presented as the mean ± SD. The significance of differences between the means was calculated by the analysis of variance. When the difference of the means was shown to be significant, multiple comparisons by pairs were calculated by the T-test. Probability values of P<0.001, P<0.01 or P<0.05 were considered to indicate a statistically significant difference.

Fibrinogen is known to adhere to α_vβ₃ and α_vβ₁ integrins on the surface of tumor cells. To confirm the adhesion of fibrinogen to the tumor cells we used, fibrinogen conjugated with Alexa Fluor 488 was incubated with B16 or A375 cells at 37˚C for 30 min with or without sufficient blocking antibodies. The results showed that fibrinogen strongly bound to the A375 (~94%) and B16 (~98%) cell surface. Both α_vβ₃ and α₅β₁ integrins bind to fibrinogen. In A375 and B16 cells, total eradication of the adhesion appeared when the functions of the two integrins were blocked (Fig. 1). Our results revealed that these two types of integrins mediated the adhesion of melanoma cells to fibrinogen.

Fibrinogen may be coated on tumor cells and may protect these cells from NK cytotoxicity. Moreover, the adhesion of fibrinogen to tumor cells is essential. Melanoma cells strongly adhere to fibrinogen (Fig. 1). Therefore, we examined the ability of fibrinogen to protect melanoma cells from NK cytotoxicity. The results revealed that fibrinogen was capable of protecting tumor cells from NK cytotoxicity in a concentration-dependent manner (Fig. 2).

To observe the effects of G. lucidum polysaccharides, we added them into the flow cytometric adhesion system. The results revealed that G. lucidum polysaccharides inhibited adhesion to the level of the blocking antibodies, and the combination of G. lucidum polysaccharides and antibodies (GLP+Ab) had no further blocking effect (Fig. 3). The results suggest that G. lucidum polysaccharides eradicate melanoma cell-fibrinogen adhesion by blocking α_vβ₃ and α₅β₁ integrins.

Fibrinogen coats are capable of blocking NK cytotoxicity against melanoma cells, and fibrinogen adhesion mediated by α_vβ₃ and α₅β₁ integrins are eliminated by G. lucidum polysaccharides. Therefore, we examined the effect of G. lucidum polysaccharides on NK cytotoxicity affected by fibrinogen. The results showed that G. lucidum polysaccharides (FBG+GLP) reduced the blocking effect to almost the level of the control (Fig. 4). The results were also similar to the samples with antibodies added (FBG+Ab), and a combination of G. lucidum polysaccharides and antibodies (FBG+Ab+GLP) revealed no further improvement. Our results suggested that G. lucidum polysaccharides are able to eliminate the blocking effect of fibrinogen on NK cytotoxicity to melanoma cells.

NK cytotoxicity is an important immune defense for tumor metastasis, which may be blocked by fibrinogen, and G. lucidum polysaccharides may eliminate the blocking effect of fibrinogen on NK cytotoxicity against tumor cells. To detect the effect of G. lucidum polysaccharides on the metastasis of melanoma cells in a mouse model, we injected G. lucidum polysaccharides or function-blocking antibodies 5 min prior to injection of B16 tumor cells through the lateral tail vein. Our results, as shown in Fig. 5, shown that G. lucidum polysaccharides (GLP) and antibodies (Ab) decreased metastasis efficiently. The inhibitive rate of G. lucidum polysaccharides and antibodies was almost identical, and the combination of G. lucidum polysaccharides and antibodies (GLP+Ab) provided no further inhibition.