Sphingosine 1-phosphate (S1P) is a bioactive lysophospholipid known to be capable of inducing diverse cellular responses through five G protein-coupled receptors, S1P1-5. Previous reports have shown that S1P1, S1P2, and S1P3 are ubiquitously expressed, whereas the expressions of S1P4 and S1P5 are confined largely to lymphoid tissues and central nervous system, respectively. We previously reported that S1P receptor expression profile was closely correlated with the metastatic behavior using the stable B16F10 clone that overexpressed each S1P receptor. Lung metastasis increased in S1P1 or S1P3 overexpressed cells, however decreased in S1P2 overexpressed cells. The functions or relevance of S1P and its receptors in human cancer are not fully understood. At first, we characterized S1P receptor expression profiles on 9 human gastric cancer cell lines using northern blot analysis. Then, we investigated the patterns of S1P induced migration using a Boyden chamber assay and assessed the relation between S1P receptor expressions and migratory responses to S1P. S1P2 was expressed in all gastric cancer cells in a variety of degree. S1P3 was expressed in 4 cancer cells. In contrast, S1P1 expression was very weak in 1 cancer cell and undetectable level in remaining 8 cells. No significant expression of both S1P4 and S1P5 were detected. S1P markedly induced migration of MKN1 cells, which expressed high level of S1P3 and relatively low level of S1P2, in a dose dependent manner up to a concentration of 100 nM. In marked contrast, S1P potently inhibited migration of AZ-521 cells, which expressed high level of S1P2, in a dose dependent manner down to a concentration of 10 nM. HGC-27 cells expressed significant level of these 2 receptors. S1P stimulated migration of HGC-27 cells in a dose dependent manner up to 10 nM. However, at a concentration of 100 nM, HGC-27 cell migration was drastically reduced to the level below control. A S1P2 specific antagonist, JTE-013, dramatically recovered the reduced migration of AZ-521 cells under 1-100 nM of S1P and that of HGC-27 under 100-1000 nM of S1P. This clearly indicates that the S1P-mediated downregulation of the motility of these cell lines operates through S1P2. S1P-mediated migration of MKN1 or HGC-27 was significantly inhibited with 100 ng/ml PTX- and 1 μM Wortmaninn. However, these inhibitors showed marginal effects on that of AZ-521. Our study demonstrates that S1P2 is a preferentially distributed receptor in gastric cancer cells and S1P3 is expressed in several cells concurrently. The balance between S1P2 and S1P3 is critical in determining the migratory response of gastric cancer cell to S1P. Taken together with our previous in vivo study, this suggests a possible association between metastatic behavior of gastric cancer and its expression profile of these two S1P receptors. Therapeutic interventions directed at S1P receptors might be effective to prevent metastasis in gastric cancer.

[Proc Amer Assoc Cancer Res, Volume 46, 2005]