Human /3-¡nterferon(IFN) induced an antiviral state in two fetal brain and six glioma cell lines. The growth-inhibitory effect of IFN was most pronounced on three glioblastoma lines and least on fetal brain and oligodendroglioma cells; IFN growth inhibition of one schwannoma and one anaplastic cell line was intermediate between the two other groups. Thus, the growth-inhibitory effect of IFN generally correlated with the degree of anaplasia of the tissue from which the cells were derived. IFN (1000 units/ml) had to be present for 24 to 48 hr to have a significant inhibitory effect on growth of glioblastoma (12-18) cells. However, growth inhi bition of 12-18 cells exposed to IFN for 3 days persisted for 3 weeks. Both sialic acid-A/-acetylgalactosamine ganglioside and a mixture of normal human brain gangliosides (50 U.M)inhibited growth of fetal brain (CHII) but not glioblastoma 12-18 cells. However, preincubation of cells with either sialic acid-A/-acetylgalactosamine or a mixture of gangliosides did not augment the growth-inhibitory effects of IFN on either CHII or 12-18. These results indicate that gangliosides and IFN may be operating through different mechanisms to cause growth inhibition.

This investigation tested the hypothesis that the growth inhibiting effects of human beta-interferon on cultured human glioma cells involves changes in the ganglioside composition of these cells. Four cell lines derived from human malignant gliomas (12-18, U-251 MG, I29-A, 7-24) and two lines from human fetal brain (CHI, CHII) were cultured in the presence and in the absence of human beta-interferon (HuIFN-beta), 1,000 units per ml medium for three days before harvesting. Human beta-interferon had an inhibitory effect on growth of glioma but not fetal brain cells. Total ganglioside sialic acid for all cell lines ranged between 3.5 and 13.8 micrograms/10(7) cells (0.6-3.9 micrograms/mg protein). No distinct difference in the amount of total ganglioside per cell was observed between neoplastic and non-neoplastic cells, but the latter had more ganglioside per mg total protein. All cell lines displayed different patterns of gangliosides determined by high performance thin layer chromatography, but there was no distinct difference between glioma and fetal brain cells. Human beta-interferon increased the total amount of ganglioside per cell in one fetal brain and two glioma lines, but on a protein basis in only one glioma cell line (I29-A); HuIFN-beta had only minor effects on ganglioside patterns. There was a slight shift towards a greater proportion of structurally simpler gangliosides in both fetal brain and two glioma cell lines exposed to HuIFN-beta, but the reverse occurred in glioma U-251 MG. None of these changes strongly correlated with the degree of growth inhibition due to HuIFN-beta.(ABSTRACT TRUNCATED AT 250 WORDS)

Human beta-interferon (IFN) induced an antiviral state in two fetal brain and six glioma cell lines. The growth-inhibitory effect of IFN was most pronounced on three glioblastoma lines and least on fetal brain and oligodendroglioma cells; IFN growth inhibition of one schwannoma and one anaplastic cell line was intermediate between the two other groups. Thus, the growth-inhibitory effect of IFN generally correlated with the degree of anaplasia of the tissue from which the cells were derived. IFN (1000 units/ml) had to be present for 24 to 48 hr to have a significant inhibitory effect on growth of glioblastoma (12-18) cells. However, growth inhibition of 12-18 cells exposed to IFN for 3 days persisted for 3 weeks. Both sialic acid-N-acetylgalactosamine ganglioside and a mixture of normal human brain gangliosides (50 microM) inhibited growth of fetal brain (CHII) but not glioblastoma 12-18 cells. However, preincubation of cells with either sialic acid-N-acetylgalactosamine or a mixture of gangliosides did not augment the growth-inhibitory effects of IFN on either CHII or 12-18. These results indicate that gangliosides and IFN may be operating through different mechanisms to cause growth inhibition.

Natural killer cells (NK) demonstrate cytotoxicity against a wide range of cultured cells without prior sensitization, and the sensitivity of many target cells to NK attack is altered by exposure to interferon. The structures on the target cells conferring sensitivity or resistance to NK are not known, but glycolipids are suggested to be related to NK susceptibility. To determine possible relationships between target cell glycolipids and NK cytolysis, the effects of human beta-interferon (IFN) on the neutral glycolipid composition and sensitivity to NK cytolysis of cultured cells from four human gliomas and two fetal brains were analyzed. Compared to MOLT-4 and Raji cells all six neural cell lines were quite resistant to NK, and IFN slightly increased this resistance. IFN also caused increases in the amounts of non-hydroxy fatty acid cerebroside, ceramide trihexoside, asialo-GM1, asialo-GM2 and globoside. Increased molar proportions of ceramide tri- and tetra-saccharides occurred in the two glioma lines which had the greatest increases in NK resistance following IFN exposure. It is concluded that neutral glycolipids may play a role in the mechanisms responsible for resistance of some glioma cells to NK cytolysis.