Effect of Genistein on vasculogenic mimicry formation by human uveal melanoma cells

Abstract

Background: Vasculogenic mimicry (VM) was increasingly recognized as a form of aggressive melanoma acquiring blood supply. Genistein had attracted much attention as a potential anticancer agent. Therefore, we examined the effect of Genistein on VM in human uveal melanoma cells.

Methods: VM structure was detected by periodic acid-Schiff (PAS) staining for uveal melanoma C918 cells cultured on the three-dimensional type I collagen gels after exposed to Genistein. We used reverse transcription polymerase chain reaction (RT-PCR) and Western Blot analysis to examine the effect of Genistein on vascular endothelial cadherin (VE-cadherin) mRNA and protein expression. The nude mice models of human uveal melanoma C918 cells were established to assess the number of VM using immunohistochemical and PAS double-staining.

Results: Genistein inhibited the survival of C918 cells in vitro. The ectopic model study showed that VM in tumor tissue sections were significantly reduced by Genistein in vivo. In vitro, the VM structure was found in control, 25 and 50 microM Genistein-treatment groups but not in 100 and 200 microM. RT-PCR and Western Blot showed that 100 and 200 microM concentration of Genistein could significantly decrease VE-cadherin mRNA and protein expression of C918 cells compared with control (P < 0.05). However, the 25 and 50 microM Genistein slightly decreased the VE-cadherin level in vitro (P > 0.05).

Conclusion: Genistein inhibits VM formation of uveal melanoma cells in vivo and in vitro. One possible underlying molecular mechanism by which Genistein could inhibit VM formation of uveal melanoma is related to down-regulation of VE-cadherin.

Figure 1

Comparison of VM channels and the VE-cadherin mRNA level between C918 cells and OCM-1A cells. (A) OCM-1A cells cultured in three-dimensional type I collagen gels do not form the VM network structure. (B) C918 cells have the ability to form the VM. (C) VE-cadherin was expressed by C918 cells but not OCM-1A cells. (A and B, Magnification: × 200)

Figure 2

Effect of Genistein on of human uveal melanoma C918 cells growth. Proliferative activity of C918 cells was determined by the MTT assay after incubation for 48 h with Genistein (0-200 μM). **P < 0.01 vs. control.

Figure 3

The effect of Genistein on the vasculogenic mimicry of human uveal melanoma C918 cells on 3-D collagen I cultures. PAS-stained images of C918 cells cultured on three-dimensional collagen I for 48 h in medium with different concentrations of Genistein. (A) control; (B) 25 μM Genistein; (C) 50 μM Genistein; (D) 100 μM Genistein; (E) 200 μM Genistein. At treatment groups with 25 μM and 50 μM concentrations of Genistein, C918 cells formed fewer VM matrix-association channels than do control. However, the groups treated with higher concentrations of Genistein (100 and 200 μM) did not form the VM channels. (Magnification: × 200)

Figure 4

CD34 and PAS double staining on the C918 human uveal melanoma xenograft sections. (A) Control; (B) 75 mg/kg/day Genistein group; VM channel (arrow) is lined by PAS-positive materials and there are red cells in the center of the channels. (Magnification: × 400)

Figure 5

Effect of Genistein on C918 cells VE-cadherin mRNA expression. (A) The expression of VE-cadherin mRNA in C918 cells was examined by RT-PCR at 48 h after different concentration Genistein pretreatment (0, 25, 50, 100, 200 μM). (B) The results of VE-cadherin mRNA were expressed after normalized by β-actin. Data represent means ± S.E.M from three separate experiments. *P < 0.05, **P < 0.01 vs. control.

Figure 6

Effect of Genistein on C918 cells VE-cadherin protein expression. (A) The expression of VE-cadherin protein in C918 cells was examined by western blot at 48 h after different concentration Genistein pretreatment (0, 25, 50, 100, 200 μM). (B) The results of VE-cadherin protein were expressed after normalized by β-actin. The values were means ± S.E.M. n = 3. *P < 0.05 vs. control.