GM1 enhances nerve growth factor (NGF)-stimulated neuritogenesis and prevents apoptotic death of PC12 cells; both may be due to enhancement of TrkA dimerization. In this study, we examined the effect of GM1 on NGF-induced TrkA dimerization in Trk-PC12 (6-24) cells. NGF increased tyrosine phosphorylation of the 140-kDa protein (TrkA monomer), and preincubation with GM1 potentiated this effect. Adding the protein crosslinker bis(sulfosuccinimidyl) suberate with NGF resulted in the appearance of two major bands (220 and 330 kDa) when probed with antibodies against TrkA or phosphotyrosine, and GM1 also enhanced this effect. We interpret the 330-kDa band as being a homodimer of TrkA. The identity of the 220-kDa band is still not certain but may consist of a posttranslationally modified form of TrkA. Our results suggest that GM1 is augmenting the effects of NGF on PC12 cells by enhancing the dimerization and activation of the TrkA receptor. Key Words: Gangliosides-Nerve growth factor-Nerve growth factor-induced receptor dimerization-PCi 2 cells-GM1.

Neutral glycolipids (NGL) were isolated and quantitated in 98 primary human brain tumors; 19 low grade astrocytomas (LGA), 12 anaplastic astrocytomas (AA), 37 high grade astrocytomas (HGA), 18 oligodendroglial tumors, and 12 primitive neuroectodermal tumors (PNET). In 38 of these, the nature of the hexose in the cerebroside was determined using immunothin-layer chromatographic techniques. Galactosylceramide (GalCer) was the major ceramide monohexoside (CMH), and glucosylcerebroside never comprised more than 6% of this fraction in any tumor type. Furthermore, there was no correlation between the proportion of glucosylcerebroside and histological diagnosis. AA had the most characteristic neutral glycolipid pattern, with high levels of total lipid, total neutral glycolipid, CMH, and ceramide dihexoside (CDH) but low water contents. Consistent with this glycolipid composition is the finding that AA usually had neither ceramide trihexoside (CTH) nor globoside. Oligodendrogliomas were somewhat similar to AA in having high levels of CMH and infrequently having CTH or globoside. However, oligodendrogliomas had low water and total lipid contents. PNET had low levels of total lipid, total NGL, and CMH, but frequently contained CTH and globoside. LGA had high water contents but low levels of total lipid and CMH. HGA tended to have intermediate levels of almost all constituents analyzed, probably reflecting the pronounced cellular heterogeneity of these tumors. The frequent presence of GalCer in astrocytomas raises the possibility that some of these contain a population of cells that are related to the oligodendroglial lineage. However, the low amounts of GalCer and infrequent presence of sulfatide in PNET is consistent with their lack of differentiation toward oligodendrocytes. It will be of interest to determine if the neutral glycolipid patterns reported here will correlate with patient survival and be of prognostic significance.

Neutral glycolipid and ganglioside compositions were determined on 11 ependymal tumors, 12 medulloblastomas, 6 other neuronal tumors of the brain, 4 peripheral neuroblastomas, 1 cerebral primitive neuroectodermal tumor (PNET), and 1 PNET of the thoracic wall. Within the group of tumors that can demonstrate neuronal phenotypes, there was an association between the degree of neuronal differentiation usually demonstrated by these tumors and the proportions of both GD1a and 1b-pathway gangliosides. The amount of globoside also correlated with the amount of 1b pathway gangliosides. Patients with medulloblastomas whose 1b gangliosides made up over 15% of the total gangliosides survived longer that those with lower proportions of 1b gangliosides. The only gangliosides in the choroid plexus papilloma were GM3 and GD1a, but other ependymal tumors had significant amounts of GD1b and its metabolic precursors. Ependymoma and anaplastic ependymoma had similar neutral glycolipid compositions, which were different from subependymoma, which lacked ceramide monohexoside and ceramide dihexoside. These differences in glycolipid compositions suggest that there may be fundamental biological differences between these types of ependymal tumors.

Neutral glycolipids (NGL) are promising diagnostic markers of human gliomas, but differences in NGL with age and sex have not been examined. Previous work demonstrated that ceramide dihexosides (CDH) levels in mouse kidney are age- and sex-dependent, probably due to levels of sex hormones. We quantitated CDH in 181 human gliomas and found significant differences with sex and age, particularly menopause and male puberty. This emphasizes the importance of assessing results of studies on glycolipids in disease states with respect to age and sex in order to avoid erroneous conclusions concerning the relationship of glycolipid composition with diagnosis and pathogenesis.

No abstract