br Materials and methods br Results br Discussion Many studi

Materials and methods
Results
Discussion Many studies investigate the role of 12/15-LOX in cancer cell,23, 24, 25 however we here first found that host 12/15-LOX also plays an important role in metastasis progression. We have demonstrated that 12(S)-HETE increased melanoma cell adhesion to the host epithelial myc inhibitor and various ECM substrates in vitro and lung metastasis in vivo. Lung is one of the most common sites for melanoma cell dissemination. Here we found that tumour growth was reduced in 12/15-LOX null mice when the B16F10 melanoma cells were subcutaneously injected. Furthermore, the lung nodule formation was decreased in 12/15-LOX null mice when the melanoma cells were intravenously injected. Co-treatment of 12(S)-HETE with melanoma cells also increased lung metastasis. It has been reported that angiogenesis inhibition can decrease lung metastasis and 15(S)-HETE, a 12/15-LOX metabolite, is a well-known angiogenic factor. 15(S)-HETE increased vessel density in chick CAM, induced sprouting in rat aortic rings and increased endothelial cell–cell contact and the formation of tubular network-like structures in human umbilical vein endothelial cells (HUVECs). Similarly, 12-HETE has been shown to contribute to tumour angiogenesis via a vascular endothelial growth factor (VEGF)-dependent pathway and to stimulate endothelial cell mitogenesis.29, 30 Although there are several emerging experimental studies indicated that 12/15-LOX enhanced cancer cell growth via angiogenesis, we here found that 12(S)-HETE and 15(S)-HETE but not LXA4 or LXB4 could directly increase the absorbance of MTT reaction of B16F10 melanoma cells in vitro. Since HETEs and lipoxins, the downstream products of 12/15-LOX from AA, may alter cellular proliferation and apoptosis,31, 32 and possibly explain the increase of tumour progression and metastasis. However, we here found that HETEs did not affect the proliferation of melanoma in vitro. This may partially be due to the shorter duration and lower dose of HETEs used here than the previous studies. It was found that HETEs increased the number of adherent cells on collagen- or fibronectin-coated plates. The ability of tumour to invade new tissues requires the complex interaction of various cell surface-associated elements. It has also been reported that HETEs may increase the adhesion of B16F10 cells to endothelial cells or extracellular matrix.33, 34, 35, 36 We also demonstrated that co-injection with 12(S)-HETE into C57BL/6 mice increased the cell number of melanoma cells to lung tissue. Furthermore, it was found that 12(S)-HETE treatment enhanced the adherence of melanoma cells onto the lung epithelia cells isolated from 12/15-LOX null mice. Therefore, the HETEs-induced adhesion of melanoma cells plays an important role in lung metastasis. The defective adhesion might explain the diminished metastasis in 12/15-LOX null mice. Cell adhesion makes an important contribution to the promotion of cell migration and the transduction of information across the plasma membrane. Adhesion signalling pathways are very complex and are triggered by various transmembrane receptors and kinases such as Akt, ERK, p38, Src and FAK. We here found that FAK and ERK but not p38 or PI3K inhibitors reduced the 12(S)-HETE-induced adhesion of B16F10 cells to type I collagen-coated plate. On the other hand, p38 and PI3K inhibitor failed to show any significant effect on the 12(S)-HETE-induced cell adhesion. Furthermore, we also found that treatment of adhering B16F10 cells with 12(S)-HETE increased the phosphorylation of FAK and ERK. FAK is a non-receptor protein tyrosine kinase that localises at focal adhesion or focal contact. The recruitment of FAK to focal adhesions is essential for its regulation by integrin-dependent adhesion to the extracellular matrix. Furthermore, ERK phosphorylates FAK both in vitro and in vivo, leading to the paxillin–FAK interaction, one of the mechanisms that regulate adhesion formation.41, 42, 43 We also showed that 12(S)-HETE increased the phosphorylation of FAK and ERK, which was antagonised by the ERK inhibitor. Therefore, ERK/FAK crosstalk was involved in 12(S)-HETE-induced increase of B16F10 melanoma cells adhesion to ECM or epithelia cells. The ERK phosphorylation in response to 12(S)-HETE was transient, which may explain the reason that the proliferation was not affected by 12(S)-HETE application. In addition, 12(S)-HETE also enhanced the adhesion of human melanoma cell lines of A375, A2058 and C32 onto type I collagen-coated plate. The in vivo lung adhesion of C32 human melanoma was also enhanced by 12(S)-HETE. Furthermore, overexpression of 12/15-LOX in PC3 or A431 cells enhances the surface expression of αvβ3 and αvβ5 integrins in PC3 cells and αvβ5 in A431 cells. Therefore, the increased expression and/or activation of integrins may be involved in the action of 12(S)-HETE. These results indicated that 12(S)-HETE potentiated the metastasis potential of melanoma. More melanoma cells were sequestered in the lung tissue of 12/15-LOX knockout mice 24h after intravenous injection. In vivo attachment of cells on endothelial cells or lung epithelial cells needs further investigation.