The involvement of gangliosides in growth and migration of human glioblastoma cells

Abstract

Human glioblastoma is a primary brain tumour with a bad prognosis. One major contributing factor is that the tumour grows in an infiltrative way. Single tumour cells, undetectable with today's techniques, invade the brain and escape therapy. There is a need to find specific target molecules on the tumour cells to utilise in new treatment strategies. One group of potentially useful molecules is the gangliosides, sialic acid-containing glycosphingolipids. They are expressed on the cell surface and they have been seen to be altered in tumour cells compared to normal brain cells. In glioblastoma, so-called glioblastoma-associated gangliosides comprise overexpression of normal gangliosides (e.g. GD3 and GM2) and expression of gangliosides not detected in normal adult brain (the lacto series). The overall aim of this work was to elucidate the role of the glioblastoma-associated gangliosides in the uncontrolled growth and migration of the glioblastoma cells by using different cell culture systems, a rat model and monoclonal antibodies (MAb) directed against specific gangliosides. To investigate the role of gangliosides in tumour growth, glioblastoma cell lines in monolayer cultures were exposed to MAb directed against two of the glioblastoma-associated gangliosides (GM2 and GD3). The glioblastoma cell proliferation was measured with 3H-thymidine and Brd-U and the results showed a significant and specific inhibitory effect of anti-GD3 antibodies (proliferation decreased by 70-90%, Paper I). These results support the suggestion that ganglioside GD3 might be involved in the growth of glioblastomas. To elucidate a selective expression of gangliosides in glioma cells in vivo, a new intracranial glioblastoma rat model was used. Biopsies of human glioblastoma, briefly cultured, were implanted into nude rat brain (human xenograft). This animal glioblastoma model, in contrast to previously used ones, shows the characteristic infiltrative growth of the human tumour. We used MAb directed against three glioblastoma-associated gangliosides (GD3, GM2 and the lacto series ganglioside 3'-isoLM1) in immunohistochemical analyses on sections of the human xenografts. The results showed ganglioside GD3 expression in areas of tumour growth, while 3'-isoLM1 was expressed in the areas of invading tumour cells, in the periphery of the tumour. Both gangliosides were expressed by the tumour cells, confirmed by double labelling with an anti-human cell antibody (3B4 MAb) (Paper II). These results lent further support to the hypothesis that GD3 is involved in tumour cell growth and that 3'-isoLM1 is involved in the migration of the tumour cells.To further explore the involvement of gangliosides in the migration of glioblastoma cells, the influence of laminin on the ganglioside biosynthesis was investigated. Laminin is a protein in the extracellular matrix surrounding glioblastomas and has been shown to increase the migration of glioblastoma cells in vitro. We used spheroid cultures of glioblastoma cells and 3H-galactose and 14C-serin to label the ganglioside synthesis during presence or absence of laminin. The results showed that laminin influenced ganglioside biosynthesis, possibly including the lacto series gangliosides 3'-isoLM1 and/or 3'-LM1, which possibly promote tumour cell migration.In conclusion, this work has produced new results supporting the hypothesis that ganglioside GD3 is involved in glioblastoma growth and that ganglioside 3'-isoLM1 is involved in the migration of the glioblastoma cells. As both gangliosides are overexpressed in malignant glioblastoma in vivo, they are both suitable target molecules to be utilised in glioblastoma therapy.