Search

Glycolipid synthesis inhibitor, D-PDMP, not only inhibited the production of glycolipid but also inhibited cell growth in human glioma cell line KG-1C in a cell cycle non-dependent manner. The reduction of glycolipid from the cell membrane allowed us to study the biological functions of glycolipids. The ability of cells to adhere to collagen was enhanced by the reduction of glycoli-pids, and random cell migration was also activated by the effect of D-PDMP. The results supported our speculation that glycolipids might function in cell growth, adherence and locomotion.

The effects of cell density on the gangliosides of 4 human glioma cell lines were studied. The cell lines used were KG-1C (GM3-dominant) , A172 and H4 (GM2-dominant), and Hs683 (GM3, GM2-co-dominant) cells. All these cell lines showed higher immunofluorescence with anti-ganglioside antibodies in FACS analysis at sparse density than at confluent density. Steric hindrance from cell surface proteins had been removed by the pretreatment of the cells with trypsin. The chemical content of gangliosides was consistent throughout the time of cell growth. The mechanisms of crypticity of gangliosides at confluent culture were under investigation. We first evaluated the metabolic turnover rate of gang-liosides at different cell densities. The results clearly showed a more rapid turn-over of gangliosides at sparse density from approximately 2 to 4 fold in terms of radioactivities of incorporated tritium into gangliosides. The profiles of labeled gangliosides were also different between the sparse and confluent cultures. We speculate that better accessibilities of antibodies toward gangliosides should be facilitated by the same mechanism which should, in turn, provide easier access of carbohydrate-hydrolysis enzymes to gangliosides at sparse cell density in order to keep an enhanced turnover rate.

The effect of immune cytokines on the composition of neutral and acidic glycolipids was analyzed. Untreated U118MG cells expressed four kinds of neu-tral glycolipid molecules (CMH, CDH, CTH, and Gb?Cer), and two kinds of acidic glycolipid (GM3 and GM2) as its major components. Effect of immune cytokines on the composition of the glycolipid composition was analyzed. U118MG cells were grown subconfiuently before a panel of cytokines (IL-lbeta, 100U/ml ; IL-2, 4000JRU/ml ; IL-4, 100U/ml ; IL-6, 100 ng/ml ; IL-8, 100 ng/ ml ; IFN-alpha, 1000U/ml ; IFN-beta, 1000U/ml ; IFN-gamma, 1000U/ml ; TNF- alpha 10U/ml, and G-CSF 2 ng/ml) were pulsed for 48 hrs. In the neutral glycolipids new bands supposed to be subspecies of CMH and CDH with different fatty acid chains were recognized in the cells treated with IFN-alpha, IL-4, IL-6 and IL-8, although no constitutional change of sugar moieties was noted. In the acidic ones, IFN-gamma and IL-4 treatment brought about remarkable changes in the glycolipid profile enhancing sulfatides, GM1, GD1a, GD1b, and GT1b of the acidc lipids, with or without GM2 and GM3 expression. TNF-alpha also induced GM1. Repeated experiments with the same culture conditions revealed different responses to cytokines in the expression of subspecies of CMH and CDH, and significant changes in the acidic glycolipids were noted in IL-2, G-CSF, TNF-alpha and IL-4 treatment. The 3rd series of repeated experiments using TNF-alpha, IFN-gamma, IL-4 and IL-6 showed only induction of subspecies of CMH by IFN-gamma ; no alteration of acidic glycolipid profile was noted. The 4th series of the same experiments in the same conditions resulted in no response. Raising the dose of cytokines, change of the passage of the cells, addition of DMSO (a potent differentiation factor) did not restore the sensitivity to the cyto-kines. Similar experiments using 6 other glioma cell lines and one melanoma cell line did not induce any glycolipid modulation. We believe that glioma cells have the potential to respond to certain cytokines to modulate their glycolipid expression under undetermined special conditions, and they are biologically unsta-ble in terms of cytokine sensitivity in culture.

We analyzed the glycolipid composition of eight cases with low grade astrocytomas. In all of the cases the composition of neutral glycolipids was notably different from the normal brain. The composition of acidic glycolipids was almost the same as normal. An increase of minor gangliosides as well as some that are usually undetectable in the normal brain was recognized in acidic glycolipids. We classified those glycolipid alterations into three types : a) an increase in short chain gangliosides ; b) novel expression of lacto series gang-liosides ; and c) both a) and b). Glycolipids did not correlate with microscopic findings, proliferation (Ki-index), or prognosis.

We have already found that a human glioma cell line U118MG could modu-late glycosphingolipid (GSL) expression via certain cytokines, though this effect is changeable and very difficult to reproduce even under consistent culture condi-tions. Consecutive analysis of cytokine effect using other 6 glioma (A172, T98G, KG-1C, U87MG, U138MG and U373MG) and one melanoma cell line (Mewo) failed to detect any GSL modulation. A total of ten cytokines were used in this experiment : IL-1beta, IL-2, IL-4, IL-6, IL-8, TNF-alpha, IFN-alpha, IFN-beta, IFN-gamma, and G-CSF, for treating KG-1C mixed glioma cells and 8 cytokines out of those for treating T98G glioblastoma cells. The rest of the cell lines were cultured with IFN-gamma and IL-4 because those two had once shown dramatic effect on GSL expression on U118MG cells. The fact that none of the 6 glioma cell lines responded to cytokines in terms of GSL modulation implies that cytokines do not play a significant role in regulating GSL expression in the glial cell lineage.

In vivo proton magnetic resonance spectroscopy (MRS) provides a noninvasive method of examining a wide variety of cerebral metabolites in both healthy subjects and patients with various brain diseases. Absolute metabolite concentrations have been determined using external and internal standards with known concentrations. When an external standard is placed beside the head, variations in signal amplitudes due to B1 field inhomogeneity and static field inhomogeneity may occur. Hence an internal standard is preferable. The purpose of this study was to quantitatively analyze the metabolite concentrations in normal adult brains and gliomas by in vivo proton MRS using the fully relaxed water signal as an internal standard.Between January 1998 and October 2001, 28 healthy volunteers and 16 patients with gliomas were examined by in vivo proton MRS. Single-voxel spectra were acquired using the point-resolved spectroscopic pulse sequence with a 1.5 T scanner (TR/TE/Ave = 3000 ms/30 ms/64).The calculated concentrations of N-acetyl-asparatate (NAA), creatine (Cre), choline (Cho), and water (H2O) in the normal hemispheric white matter were (23.59 +/- 2.62) mmol/L, (13.06 +/- 1.8) mmol/L, (4.28 +/- 0.8) mmol/L, and (47,280.96 +/- 5414.85) mmol/L, respectively. The metabolite concentrations were not necessarily uniform in different parts of the brain. The concentrations of NAA and Cre decreased in all gliomas (P0.001). The ratios of NAA/Cho and NAA/H2O showed a significant difference between the normal brain and gliomas, and also between the high and low grades (P0.001).Quantitative analysis of in vivo proton MR spectra using the fully relaxed water signal as an internal standard is useful. The concentrations of NAA and the ratios of NAA/H2O and NAA/Cho conduce to discriminating between the glioma and normal brain, and also between the low-grade glioma and high-grade glioma.