Exogenously administered gangliosides fail to increase in vivo metastatic frequency or in vitro growth of murine neoplastic cells

Reduced cell migration, tumor growth and experimental metastasis of rat F-11 cells whose expression of GD3-synthase is suppressed

We previously established a rat F-11 cell line whose expression of ganglioside GD3 was inhibited by stable transfection of the anti-sense vector against the GD3-synthase gene, showing that specific inhibition of GD3-synthase expression in tumor cells greatly reduced their growth rate in nude mice. Here, we report that down-regulation of GD3-synthase expression in anti-sense-transfected F-11 cells correlates with reduced cell migration and invasion in vitro and tumor growth and metastasis in vivo. When cultures were denuded of cells in a 1-mm-wide strip, the anti-sense-transfected F-11 cells migrated very slowly into the denuded area. Differences in migration between anti-sense-transfected cells and control parental cells were easily apparent. In vitro invasion assay of F-11 cells revealed a 3-fold decrease in invasion ability from the GD3-synthase-suppressed cells; colony formation in soft agar was not affected. Injection (i.v.) of control sense-transfected and untransfected F-11 cells resulted in multiple, large metastatic nodules in each of the 12 mice, whereas i.v. injection of anti-sense-transfected F-11 cells formed a single, small metastatic nodule in only 2 of the 8 nude mice. In addition, even if metastasis occurred, the anti-sense-induced metastatic nodules were much smaller than the metastatic nodules formed by control F-11 cells. These results demonstrate that suppression of GD3-synthase expression, which results primarily in a marked decrease in the concentration of ganglioside GD3, greatly reduces cell spreading, invasion and both the incidence and growth rate of experimental metastasis of F-11 cells.

Suppression by ganglioside GD1A of migration capability, adhesion to vitronectin and metastatic potential of highly metastatic FBJ-LL cells

Ganglioside GD1a, which is highly expressed in poorly metastatic FBJ-S1 cells, has been shown to inhibit the serum-induced migration capability of highly metastatic FBJ-LL cells. In the present study, the capacity of FBJ-S1 cells to adhere to vitronectin was found to be about half that of FBJ-LL cells. Pre-treatment of FBJ-LL cells with GD1a decreased this capacity by 30% that of the control, whereas GM1-pre-treatment caused only a 10% decrease, indicating that GD1a specifically inhibits FBJ-LL cell adhesion to vitronectin. Since FBJ-LL cells contain almost no GD1a, transfectants capable of expressing GD1a to varying degrees were produced in this study by transfection of FBJ-LL cells with GM2/GD2-synthase cDNA. Decrease in the serum-induced migration capacity of these transfectants was accompanied by an increment in GD1a expression. Adhesion of the transfectants to vitronectin decreased by 30% as compared with mock-transfected cells. Within 4 to 5 weeks after GD1a-expressing transfectant and mock-transfected cells were transplanted into mice, metastatic nodules were observed in liver, lung, kidney and adrenal glands of mock-transplanted mice, but not in those with GD1a-expressing transfectants, indicating that GD1a suppresses the metastasis of FBJ-osteosarcoma cells, possibly by inhibiting cell migration and cell adhesion. The involvement of the ganglioside in the suppression of metastasis is clearly demonstrated in the present study.

Expression of ganglioside GM3 and H-2 antigens in clones with different metastatic and growth potentials isolated from Lewis lung carcinoma (3LL) cell line

In view of the evidence that cell expression of gangliosides in several tumors is positively involved in the metastatic phenotype, Lewis lung carcinoma (3LL) cell line, expressing GM3 as the major ganglioside, was analysed for the cell surface expression of GM3. An indirect immunofluorescence assay, using a M2590 monoclonal antibody recognizing GM3, was used for this purpose. Since the parental 3LL cells consist of heterogenous subpopulations differing in the degrees of GM3 expression, we have developed clones of this cell line with different degrees of metastatic potentials by using an in vitro non-selective procedure in order to investigate whether the expression of GM3 is associated with metastatic potential. The degree of cell surface expression of GM3 among the clones correlated well with their total cellular content of this ganglioside. However, we were unable to confirm the report of increased level of GM3 in high metastatic 3LL clones, nor did a decreased level correlate with weak metastatic ability. In our recent work, an inhibitor of glucosylceramide synthase, D-threo-1-phenyl-2-decanoylamino-3-morpholino-1-propanol (D-PDMP), was found to decrease the levels of all cellular glucosphingolipids and cause the accumulation of the precursors of glucosylceramide. The present study does not, however, rule out the possible involvement of this lipid family in metastatic dissemination, since treatment of 3LL cells with D-PDMP resulted in significant inhibition of their experimental metastatic potential. Clones expressing very low GM3 grew slowly in culture dishes, suggesting that GM3 may have a regulatory role in cell proliferation. The low metastatic clones expressed high levels of H-2Kb antigen, while the expression of the same antigen on the high metastatic clones was relatively low, confirming the previous observation of this tumor system. Moreover, a clone showing the lowest tumorigenic potency revealed both a high cell surface expression of H-2Kb and a high H-2Kb/H-2Db ratio.