Isolation of hamster brain astroglia by in vitro cultivation and subcutaneous growth, and content of cerebroside, ganglioside, RNA and DNA

The Na+/H+ exchange system in glial cell lines. Properties and activation by an hyperosmotic shock

The properties of the Na+/H+ exchange system in the glial cell lines C6 and NN were studied from 22Na+ uptake experiments and measurements of the internal pH (pHi) using intracellularly trapped biscarboxyethyl-carboxyfluorescein. In both cell types, the Na+/H+ exchanger is the major mechanism by which cells recover their pHi after an intracellular acidification. The exchanger is inhibited by amiloride and its derivatives. The pharmacological profile (ethylisopropylamiloride greater than amiloride greater than benzamil) is identical for the two cell lines. Both Na+ and Li+ can be exchanged for H+. Increasing the external pH increases the activity of the exchanger in the two cell lines. In NN cells the external pH dependence of the exchanger is independent of the pHi. In contrast, in C6 cells, changing the pHi value from 7.0 to 6.5 produces a pH shift of 0.6 pH units in the external pH dependence of the exchanger in the acidic range. Decreasing pHi activates the Na+/H+ exchanger in both cell lines. Increasing the osmolarity of the external medium with mannitol produces an activation of the exchanger in C6 cells, which leads to a cell alkalinization. Mannitol action on 22Na+ uptake and the pHi were not observed in the presence of amiloride derivatives. Mannitol produces a modification of the properties of interaction of the antiport with both internal and external H+. It shifts the pHi dependence of the system to the alkaline range and the external pH (pHo) dependence to the acidic range. It also suppresses the interdependence of pHi and pHo controls of the exchanger's activity. NN cells that possess an Na+/H+ exchange system with different properties do not respond to mannitol by an increased activity of the Na+/H+ exchanger. The action of mannitol on C6 cells is unlikely to be mediated by an activation of protein kinase C.

Specific stimulation of phospholipase activities with phosphatidylethanolamine in transformed cells

Phospholipase A1, A2 and C activities with phosphatidylethanolamine were enhanced in C6 cells relative to primary astrocytic cultures. Enhancement was a function of cell density. Phospholipase activities with phosphatidylcholine were unchanged as a function of cell density, while phospholipase C activity with phosphatidylinositol was reduced. All acid phospholipase activities measured were low or essentially absent in the three transformed cell lines examined. These results suggest that arachidonate release upon confluency is mainly from phosphatidylethanolamine.

Incorporation and metabolism of exogenous fatty acids by cultured normal and tumoral glial cells

We have investigated the transformation of exogenous radioactive free fatty acids by cultured glial cells and their incorporation into complex lipids. The cells were either tumor lines (C6 and NN) or primary cultures from newborn rat hemispheres. The tumor lines could undergo morphological differentiation with dibutyryl cyclic AMP or bromodeoxyuridine. The fatty acid precursors used were palmitic, stearic, oleic, linoleic and linolenic acids. Tumor cells presented a higher incorporation of the precursors in the cell lipid acyl groups than did normal cells. Tumor cells desaturated and/or elongated palmitic, stearic and oleic acid to a higher extent than did normal cells. In contrast, tumor cells transformed linoleic and linolenic acids to their polyunsaturated derivatives to a lower extent than did normal cells. In differentiated tumor cells, these patterns of metabolism were shifted toward the patterns of normal cells. Tumor cells did not exhibit delta 4-desaturase activity, but such activity was restored in the C6 line upon dibutyryl cyclic AMP-induced differentiation. Transformation of linoleic and linolenic acid is likely to proceed through initial delta 6 desaturation. Phospholipids were preferentially labelled with the radioactive fatty acids, and only a little radioactivity was found in the neutral lipid fraction, mainly in diacylglycerols. Each fatty acid precursor label was incorporated in individual phospholipids to a proportion which reflected the typical acyl group composition of glycerophospholipids; we observed high levels of incorporation of palmitic acid and its derivatives into choline glycerophospholipids, and high levels of incorporation of linolenic acid and its derivatives into ethanolamine glycerophospholipids. This pattern was more marked in tumor cells than in normal cells, and the differentiation of tumor cells partially restored the normal pattern, mainly in bromodeoxyuridine-treated NN cells. Both types of differentiation of glial cell lines can be useful as models for the understanding of membrane physiology in normal and tumor cells.

Cysteine sulfinic acid uptake in cultured neuronal and glial cells

The presence of an efficient high affinity uptake system for L-CSA has been demonstrated in cultured neuronal and glial cells of various types. In neurons and most glial cells L-CSA uptake is inhibited by acidic amino acids, L-glutamate and L-aspartate and requires sodium ions; however the sodium dependence varies from one cell type to the other. The characteristics of the uptake system change during cell maturation, especially in astroblasts. The predominance of CSA uptake in glial cells as compared to neurons, the similarity of the kinetic parameters and of the structural specificity of L-glutamate uptake suggest that both excitatory amino acids are transported by a common system. In view of accumulating evidence, the present results are in agreement with a role of CSA as a neurotransmitter and as a precursor for taurine biosynthesis in the central nervous system.

Inhibitory action of serum from a Laron dwarf on normal cellular function

Glucose uptake and O2 consumption of confluent glial cells grown in culture were measured in the presence of serum-free buffer and compared with those measured in the presence of serum from a normal volunteer, from an hGH-deficient dwarf and from a Laron dwarf. Cellular glucose uptake and respiration in the absence or presence of insulin or hGH are inhibited by Laron serum.

Effect of hydrocortisone and thyroxine on ATPase activities of neuronal and glial cell lines in culture

The effect of hydrocortisone and thyroxine, on the activities of Ca2+- and Mg2+- ATPase was studied in cultured neuronal (clone M1) and glial (clones NN and C6) cell lines. For M1 and NN cells an increase in Ca2+- and Mg2+-ecto-ATPase activity was found when the cells were cultured during 4-6 days in presence of hydrocortisone or together with thyroxine. In the same conditions, a decrease in Ca2+- and Mg2+-ecto-ATPase activity was found for the C6 cells. In C6 cells the effect of hormones was more pronounced for the Mg2+- than for the Ca2+-ecto-ATPase activity. The observed decrease may be related to the tumoral origin of the C6 cells. The activity of (Na+, K+)-ATPase in all three cell lines increased in presence of hydrocortisone or together with thyroxine when the cells were cultured during 4-6 days, in presence of the hormones, whereas the total Mg2+- ATPase activity increased only after 6 days of treatment. Thyroxine alone has very few effect either on Ca2+- and Mg2+-ecto-ATPase, or on (Na+, K+)- and total Mg2+-ATPase activity. These observations are interpreted to indicate that hormones may modulate or induce enzymatic activities involved in active transport phenomena in nervous tissue.

Effect of ethanol on superoxide dismutase activity in cultured neural cells

Superoxide dismutase (EC 1.15.1.1) activity was investigated in several types of neural cells cultivated in the presence of 100 mM ethanol. Superoxide dismutase was inhibited by acute treatment with ethanol. Chronic treatment with ethanol specifically inhibited superoxide dismutase in glial cells. In all instances withdrawal of ethanol produced a quick return to control values. Inhibition of superoxide dismutase by ethanol may increase toxic oxygen radicals in nervous tissue.

Ethanol and neuronal metabolism

The effect of ethanol on membrane enzymes (Na+, K+ and Mg2+ ATPases, 5'-nucleotidase, adenylate cyclase) alcohol dehydrogenase, aldehyde dehydrogenase and superoxide dismutase were studied in nerve cells (established cell lines, primary cultures of chick and rat brain) cultured in the presence of 100 mM ethanol, and in total rat brain, following various ethanol treatments of the rats (20% ethanol as the sole liquid source, intraperitoneal injection). The results show a difference between neuronal and glial cells. Most of the observed changes in enzymatic activities returned rapidly to control values when ethanol was withdrawn from the culture medium or from the diet. Alcohol dehydrogenase was more stimulated by ethanol than aldehyde dehydrogenase; therefore acetaldehyde may be accumulated. The inhibition of superoxide dismutase activity may allow an accumulation of cytotoxic O2- radicals in nervous tissue and may explain the polymorphism of lesions brought about by alcohol intoxication.

Membrane lipids in bromodeoxyuridine-differentiated astroglial cells in culture

Embryonic hamster astroblasts (NN strain) grown in continuous line were cultivated in the presence of bromodeoxyuridine (BrdU). A decrease in the growth rate of the cells and striking changes in their morphology were observed, the morphology of the cells resembling that of mature astrocytes. Membrane lipids of BrdU-differentiated and standard cells were compared. No modification of the lipid/protein ratio was observed. Phospholipids and cholesterol were increased in the same proportions in the cells, and no modification of the phospholipid distribution was observed. Ganglioside sialic acid remained at the same level, but the ganglioside distribution was highly modified. Complex gangliosides appeared (GM1 and GD1a), while the proportion of simple gangliosides (GM3 and GD3) decreased. However, neither GT1 nor GQ1 were detected in differentiated cells. The distribution of phosphoglyceride acyl groups was highly modified, the proportion of arachidonic and docosapentaenoic acids being 2 to 3 times higher in BrdU-treated cells than in proliferating ones. These results were compared to those obtained with another clonal line of glial cells (C6) which exhibited no morphological differentiation in the presence of BrdU; the lipids of these cells were not modified by such a treatment.

High-affinity uptake of gamma-aminobutyric acid in cultured glial and neuronal cells

Both glial and neuronal cells maintained in primary culture were found to accumulate [3H]GABA by an efficient "high-affinity" uptake system (apparent Km = 9 muM, Vmax = 0.018 and 0.584 nmol/mg/min, respectively) which required sodium ions and was inhibited by 1 mM ouabain. Strychnine and parachloromercuriphenylsulfonate (pCS) (both at 1mM) also strongly inhibited uptake of [3H]GABA, but metabolic inhibitors (2,4-dinitrophenol, potassium cyanide, and malonate) were without effect. Only three structural analogs of GABA (nipecotate, beta-alanine, and 2,4-diaminobutyrate) inhibited uptake of [3H]GABA, while several other compounds with structural similarities to GABA (e.g. glycine, L-proline, and taurine) did not interact with the system. The kinetic studies indicated presence of a second uptake (Km = 92 muM, Vmax = 0.124 nmol/mg/min) in the primary cultures containing predominantly glioblasts. On the other hand, only one of the neuronal cell lines transformed by simian virus SV40 appeared to accumulate [3H]GABA against a concentration gradient. Apparent Km of this uptake was relatively high (819 muM), and it was only weakly inhibited by 1 mM ouabain and 1 mM pCS. The structural specificity also differed from that of the uptake observed in the primary cultures. Significantly, non of the nontransformed continuous cell lines of either tumoral (glioma, C6; neuroblastoma, M1; M1NN) or normal (NN;I6) origin actively accumulated [3H]GABA. It is suggested that for the neurochemical studies related to GABA and requiring homogeneous cell populations, the primary cultures offer a better experimental model than the continuous cell lines.

Effects of amino acids on calcium uptake by glial and neuroblastoma cells

The uptake of [45Ca] has been studied in clonal glial and neuronal cells. It was somewhat more efficient in the neuroblastoma clone M1 compared to glial clones. In all cases [45Ca] uptake was shown to depend on the phosphate concentration in the incubation medium. It was decreased by the ionophore A 23187 at 200 microM concentration in both neuronal and glial clones. The influence of amino acids some of which are putative neurotransmitters was investigated; the interactions between [45Ca] uptake and these amino acids were related to their concentration and the type of cells used (neuronal or glial). L-aspartate and taurine for example had two opposite effects on [45Ca] uptake by the glial clone NN at two different concentrations; they could therefore play a role in the control of calcium level in the synaptic cleft.

Effect of insulin-induced epileptic seizures on neutral glycolipids of rabbit cerebral cortex

The neutral glycolipids of rabbit central cortex were analyzed during epileptic seizures produced by insulin or pentetrazol injection. The two agents gave similar results. A decrease of glycolipid content occurred in the cortex and in the neuronal fraction during seizures. The normal glycolipid level was restored during the recovery phase.

Essential fatty acid metabolism in cultured astroblasts

The fetal calf serum on which two astroblast cell lines were grown was shown to be deficient in essential fatty acids. The fatty acid profiles of lipids of these two cell lines showed very low amounts of polyunsaturated fatty acids. In order to know if this low level of unsaturation was due to the lack of essential fatty acids in serum, or to a lack of desaturase activities in the cells, we have investigated the modifications of the cell lipid fatty acid patterns when serum was enriched in essential fatty acids. Linoleic acid was incorporated in rather high amounts in the cell lipids, while linolenic acid was very poorly incorporated. These two essential fatty acids were converted into polyunsaturated fatty acids only when they were added alone to the serum. Both cell clones exhibited a lack in the delta 4 desaturase activity. No morphological changes of the cells occured after nine days of culture with modified serum.

Ganglioside content and pattern in human gliomas in culture. Correlation of morphological changes with altered gangliosides

The ganglioside level and pattern of human gliomas in monolayer cultures were examined. These gliomas revealed morphological variations that correlated with several features of ganglioside analysis. Glioblastoma lines TC 178 and TC 501 that morphologically had changed during extended subculture revealed reduced amounts and a simplified pattern of gangliosides with almost total loss of the characteristic brain complex gangliosides. In contrast, two glioblastoma lines TC 526 and TC 593, as well as the oligodendroglioma line TC 620 showed brain-like gangliosides and the cells in these cultures had maintained their characteristic morphology observed during early subcultures. The possibility that altered ganglioside levels occur in conjunction with morphological changes after propagation in vitro is discussed.

Neuroblasts-glia interaction in tissue culture as evidenced by the study of ectoenzymes. Ecto-ATPase activity of mouse neuroblastoma cells

Ecto-ATPase and ecto-5'-nucleotidase activities of hamster astroblasts, clonal line NN, and mouse neuroblasts, clonal line Ml in coculture, have been studied. The originally low ecto-ATPase activity in both cell lines increased many fold when these cell lines were cocultured. An increase of ecto-ATPase activity was also found in coculture of neuroblastoma cells with chick fibroblasts. Neuroblastoma Ml cells were separated from coculture with hamster astroblasts after 7 days and 2 months. Reisolated M1 cell lines exhibited higher ecto-ATPase activity than the original M1 cell line. The M1 cell line separated after 2 months of coculture had higher ecto-ATPase activity than the M1 cell line separated after 7 days of coculture with hamster astroblasts. This higher ecto-ATPase activity continued for more than 20 replications after separation of the M1 cells from the glial cells.

Biochemical evidence on the role of gangliosides in nerve-endings

From the biochemical and developmental data in the literature, there appears to be some specific relationship of gangliosides to synapses. Whether the concentration of polysialo-gangliosides is located in the immediately pre-synaptic part of the neuronal plasma membrane is not known. There are many unanswered questions about the gangliosides, which we have attempted to pose in this article. There are as many hypotheses about roles for gangliosides, and virtually no data to back them up. In fact there are not even real indications of possible functions. We believe that it will be very difficult to define the functions of gangliosides until more of the basic questions are answered-even if their synaptic localization is a constant temptation to speculate.

The chronic and acute effects of ethanol on adenosine triphosphatase activity in cultured astroblast and neuroblastoma cells

Mg2+ATPase and (Na+ + K+)ATPase activities were measured in clonal line NN hamster astroblasts and in clonal lines M1 and N1E-115 mouse neuroblastoma cells after the cells had been subjected to the acute and chronic actions of 100 mM ethanol. Exposure of the astroblasts to ethanol for periods as long as 68 days produced an increase in total cellular Mg2+ ATPase activity, as measured in cell homogenates; however, activity reverted to control levels upon withdrawal of ethanol. Chronic exposure of clonal line N1E-115 neuroblastoma cells to ethanol produced an increase in Mg2+ATPase and (Na+ +K+)ATPase activities. In contrast, the activities of both ATPases of clonal line M1 neuroblasts were unaffected by chronic exposure to ethanol. Acute exposure of cell homogenates to 100 mM ethanol inhibited Mg2+ ATPase and (Na+ + K+)ATPase of astroblasts but not that of neublastoma cells. These findings suggest that neural cells in culture may serve as useful models for studying the effects of ethanol on specific cell types.

The relative utilization of the acyl dihydroxyacetone phosphate and glycerol phosphate pathways for synthesis of glycerolipids in various tumors and normal tissues

Rates of phosphatidate synthesis from dihydroxyacetone phosphate via acyl dihydroxyacetone phosphate or glycerol phosphate are compared in homogenates of 13 tissues, most of which are deficient in glycerol phosphate dehydrogenase (EC 1.1.1.8). In all tissues examined, dihydroxyacetone phosphate entered phosphatidate more rapidly via acyl dihydroxyacetone phosphate than via glycerol phosphate. Tissues with a relatively low rate of phosphatidate synthesis via glycerol phosphate, showed no compensating increase in the rate of synthesis via acyl dihydroxyacetone phosphate. The rates at which tissue homogenates synthesize phosphatidate from dihydroxyacetone phosphate via glycerol phosphate increase as glycerol phosphate dehydrongenase increase. Both glycerol phosphate dehydrogenase and glycerol phosphate: acyl CoA acyltransferase (EC 2.3.1.15) are more active than dihydroxyacetone phosphate : acyl CoA acyltransferase (EC 2.3.1.42). Thus, all the tissue homogenates possessed an apparently greater capability to synthesize phosphatidate via glycerol phosphate than via acyl dihydroxyacetone phosphate, but did not express this potential. This result is discussed in relation to in vivo substrate limitations.

The lipid composition of isolated brain cells and axons

The current status of the published work on the lipid composition of isolated brain cells is reviewed and some new work on the sphingolipids of these cells is presented. In spite of considerable differences in isolation techniques between different groups, the lipid analyses of different cell preparations are similar enough to permit several generalizations. This fact is an encouraging sign that cell separation methods have considerable usefulness in defining the composition of normal brain cells. It is a general finding that astrocytes have more lipid than neuronal perikarya but that the gross lipid composition of these two cell types is surprisingly similar. Oligodendroglial lipids are quite different from those of the other two cell types and are characterized by a high galactolipid content. Although such a lipid pattern might be expected in oligodendroglia, which are myelin-forming cells, axonal lipids have an even higher galactolipid content. In an effort to find more cell-specific patterns, the glycosphingolipids were examined in more detail. Differences were seen in the distribution and fatty-acid patterns of these minor lipids in neurons and astrocytes, although it may be premature to conclude that these differences will prove to be cell-specific. All of the isolated cells were found to contain galactosylceramide, sulfatide, glucosylceramide, dihexosylceramide, and gangliosides. The distribution of these lipids in the normal cells was found to differ considerably from that reported in cultured neuroblastoma cells or astrocytoma cells. Not only were gangliosides present in all cells but the ganglioside patterns of neurons and astrocytes were nearly identical. The fatty-acid patterns of the neuronal and astroglial sphingolipids generally do not resemble each other, and both are quite different from those found in oligodendroglia and axons. However, the fatty-acid composition of the sphingolipids from bovine oligodendroglia and from axons are similar and resemble those of myelin lipids. The fatty acids of glucosylceramide and dihexosylceramide are similar in all three cell types. They have rather large amounts of 16:0 and acids longer than C18; thus they are considerably different from the ganglioside fatty acids (which have mostly 15:0) isolated from the same fractions.