Molecular Psychiatry (1998) 3, 356–361  1998 Stockton Press All rights reserved 1359–4184/98 $12.00 ORIGINAL RESEARCH ARTICLE enriched secondary cultures of O-2A progenitors (A2B5+) express GS (Figure 1c and 1d). In nearly all postmitotic secondary cultures of OL, Oligodendrocyte-type-2 we have investigated GS expression by double immunostaining with the GS antibody and several dif- astrocyte (O-2A) progenitor ferentiation oligodendroglial markers: OL-1,8 2Ј,3Ј-cyc- cells express glutamine lic nucleotide 3Ј-phosphodiesterase (CNP), MBP and myelin/oligodendrocyte glycoprotein (MOG) each spe- synthetase: developmental cific for a particular OL stage of maturation. Figure 2 shows that GS expression in OL secondary cultures and cell type-specific was dependent upon OL differentiation stage. In 4-day OL secondary cultures, 10–15% were A2B5+ O-2A regulation cells. Of these, only 5% expressed GS (data not

1 1 2 shown). Moreover, 10% of pre- or immature OL, D Baas , D Dalenc¸on , C Fressinaud , characterised by the monoclonal rat antibody OL-1 spe- 3 1 L Vitkovic and LL Sarlie`ve cific for the unknown sulfatide, were GS+/OL-1+ and + 1 20% of CNP OL expressed GS. Furthermore, 30% of UPR 416 du CNRS, 5 rue Blaise Pascal, 67084 + + + OL were GS /MBP and about 70% of MOG OL Strasbourg Cedex, France; 2Service de Neurologie B, expressed GS (see also Figure 1). Note that if an OL CHU, Angers, France; 3Division of Neuroscience and was CNP positive it was also OL-1 positive; and if an Behavioral Science, NIMH, NIH, Rockville, USA + + + OL was MOG positive it was OL-1 /CNP /MBP . These data indicated that the number of cells immunostained Keywords: O-2A cells; astrocytes; oligodendrocytes; cul- for GS increased with OL differentiation and matu- ture; glutamine synthetase; mRNA; protein; activity ration (Figure 2). Given that the GS protein was synthesized in each Glutamine synthetase (GS), the enzyme that catalyses macroglial cell type tested (Figure 1), we next com- glutamine synthesis from glutamate and ammonia, pared mRNA levels. GS mRNA expression was ana- plays a central role in the detoxification of brain 9 ammonia.1 In the central nervous system (CNS), GS also lysed by Northern blot using the glial cell line CG-4. subserves additional important functions such as reg- This cell line allowed us to obtain three macroglial cell ulating glutamate, GABA and amino acid metabolism.2 types in vitro. The mRNA sizes were 2.8 kb and 1.4 kb, Oligodendrocytes (OL) form the myelin sheath in the respectively and corresponded to the expected sizes of central nervous system (CNS) and are essential for GS7 and cyclophilin10 mRNA (Figure 3a). GS mRNA efficient propagation of nerve impulses. In culture, OL was expressed in CG-4-derived O-2A progenitors arise from bipotential O-2A progenitor cells. These O- (Figure 3a lane 1), CG-4-derived OL (Figure 3a lane 2) 2A cells give rise to type-2 astrocytes in the presence and CG-4-derived type-2 astrocytes (Figure 3a lane 3). 3–7 of serum. GS is expressed in mature glial cells in vivo Only one GS mRNA species (2.8 kb) was expressed in and in vitro, but it is unknown whether GS is present in CG-4-derived O-2A cells, CG-4-derived type-2 astro- glial progenitors. In addition, a comparison of the GS cytes, and CG-4-derived OL, compared to two GS expression level among the various types of glial cells has never been done in vitro. The current study investi- mRNA species (2.8 kb and 1.4 kb) that were expressed gates in vitro GS expression levels in O-2A progenitors, in type-1 astrocytes. Autoradiogram quantification astrocytes and OL. We demonstrate that the GS gene showed that CG-4-derived O-2A progenitors expressed is expressed in O-2A progenitors and is expressed at 20-fold less GS mRNA than CG-4-derived type-2 astro- different levels in each cultured glial cell type. GS also cytes and 30-fold less than CG-4-derived OL (Figure is stimulated during OL developmental maturation. 3b). Thus, CG-4-derived type-2 astrocytes expressed Thus, the GS gene is expressed in O-2A cells and is 1.5-fold less GS mRNA than CG-4 derived OL (Figure 3b). regulated in a developmental and macroglial cell type- To determine whether GS was functional and equally specific manner. efficient in all glial cell types, GS activity was determ- In highly enriched cultures of O-2A progenitors, ined and compared in type-1 astrocyte primary cul- type-2 astrocytes, OL or type-1 astrocytes, we investi- tures, O-2A progenitors, type-2 astrocyte and OL sec- gated GS expression by double immunostaining with ondary cultures (Figure 4). GS activity in O-2A cells glial specific markers and a GS antibody. Double and type-1 astrocytes was similar. This level was half immunolabelling with GS (Figure 1a) and glial fibril- of that measured in type-2 astrocytes and OL which are lary acidic protein (GFAP) (Figure 1b) showed that GS two cell types differentiated from bipotential glial O- was expressed in GFAP positive type-1 astrocytes. The 2A progenitors. GS activity in the latter two cell types GS antibody (Figure 1c, 1e and 1g) double immunolab- was similar. Note that type-2 astrocytes expressed 1.5- elled cells with the A2B5 (Figure 1d), GFAP (Figure 1f), fold less GS mRNA, but demonstrated the same GS spe- or myelin basic protein (MBP) (Figure 1h) antibodies. cific activity as OL (Figure 3b). These results indicated that type-2 astrocytes (Figure Previous studies indicated that GS activity was upre- 1e and 1f) as well as mature OL (MBP+) (Figure 1g and gulated by bFGF in type-1 astrocytes but not in OL sec- 1h) express GS. Double immunolabelling with A2B5 ondary cultures.4 Since O-2A progenitors were main- and GS demonstrated, for the first time, that highly tained in culture medium containing bFGF, it was Glutamine synthetase in glial lineage D Baas et al 357

Figure 1 Oligodendrocyte progenitors and all three mature macroglial cell types express glutamine synthetase. Primary type- 1 astrocyte cultures after 26 days in vitro (a, b), O-2A progenitor secondary cultures after 13 days in vitro (c, d), type-2 astrocyte secondary cultures after 13 days in vitro (e, f) and oligodendrocyte secondary cultures after 26 days in vitro (g, h) were double immunolabelled with anti-GS (a, c, e, g), anti-GFAP (b and f), A2B5 (d), and anti-MBP (h) antibodies. Cells were grown as described in Materials and Methods. Each pair of panels represents micrographs of the same field. Note that all cell types express GS. Scale bars = 20 ␮m. Glutamine synthetase in glial lineage D Baas et al 358

Figure 4 Specific activity of glutamine synthetase in mac- Figure 2 Developmental regulation of glutamine synthetase roglial cells is cell type-specific. GS specific activity was mea- in oligodendrocytes. Ratio of double labelled cells (GS and sured in type-1 astrocyte primary cultures after 26 days. O-2A one of the markers of oligodendrocyte maturation, OL-1, CNP, progenitor secondary cultures after 13 days, type-2 astrocyte MBP or MOG) expressed as a percentage of cells labelled with secondary cultures after 13 days and oligodendrocyte second- the particular maturation marker in OL secondary cultures. ary cultures after 26 days. Cells were grown and GS activity Cells were grown as described in Materials and Methods for measured as described in Materials and Methods. Values rep- 4 days in secondary cultures (26 days total time in culture). resent the means ± s.d. from ten separate experiments. Stat- ± Values represent the means s.d. from three independent istical analysis was performed using Student’s t-test. * and ** 3 counts of 10 cells in each of three independent experiments. Significantly different from O-2A progenitor cells at P Ͻ 0.05 and P Ͻ 0.001, respectively.

necessary to determine whether GS activity was depen- dent on bFGF. Therefore, we measured GS activity in secondary cultures highly enriched with O-2A progeni- tors maintained with or without bFGF (2 ng ml−1). GS specific activity was similar in O-2A cells grown, either in the presence of bFGF (0.10 ± 0.03 ␮mol mg protein−1 min−1) or absence of bFGF (0.09 ± 0.04 ␮mol mg protein−1 min−1) (average ± s.d. from 10 separate experiments). These data suggested that bFGF had no effect on GS expression in O-2A progenitor cells. The CG-4 cell line proved to be interesting for this study because it expressed OL differentiation markers. Indeed, in culture, the CG-4 cell line yielded a high percent of each cell type studied. Ninety eight percent were A2B5+/galactocerebroside (GC)− O-2A cells in serum-free medium. Upon growth factor omission, most cells (98%) acquired a typical differentiated OL (GC+) phenotype or, in the presence of 20% FCS, a type-2 astrocytes phenotype.9 Primary and secondary culture of O-2A progenitors, type-2 astrocytes, OL or type-1 astrocytes, contained also a high percent of each studied cell type and less than 1% of microglia as pre- viously described.3 The present results showed that functional GS was expressed in vitro, not only in one macroglial cell type, but in all including O-2A cells. Previously, it was thought that GS was expressed only in mature macro- glial cells and not in O-2A progenitors. Thus, in these Figure 3 Northern blot analysis of glutamine synthetase mRNA in CG-4 cell line. (a) PolyA+ mRNA prepared from the studies, we report the novel finding that GS was CG-4 cell line maintained as O-2A progenitor cells (lane 1), expressed and was highly efficient in O-2A cells. The differentiated into OL (lane 2) or type-2 astrocytes (lane 3) original GS protein expression study reported that it were hybridized consecutively with 32P-labelled GS and was only found in astrocytes.11 Later studies,3,4,12,13 cyclophilin cDNA specific probes. (b) Blots were analysed by using GS immunostaining or enzyme activity, found phosphorimager (from Molecular Dynamics, Sunnyvale, CA, that in vivo and in vitro GS was expressed in OL as USA) and the data were normalized to the cyclophilin signal. well as astrocytes. This GS protein likely also plays an Values represent averages of two independent experiments. important role in OL development and maturation because the protein is already enzymatically active in Glutamine synthetase in glial lineage D Baas et al 359 O-2A progenitors. The enrichment of GS mRNA in glu- methods. After 11 days, primary mixed glial cultures20 tamatergic areas in the CNS, and the role of the GS were enriched with O-2A progenitor cells dispersed enzyme in neurotransmitter metabolism and brain over the astroglial layer. To obtain O-2A progenitor cell ammonia detoxification have already been described.7 secondary cultures, these glial precursors (A2B5+) were ° Moreover, the discovery of glutamate receptors in O- dislodged and incubated at 37 C in a 5% CO2 humidi- 2A cells14 is consistent with our results which show fied atmosphere for 1 h to allow astrocyte attachment. GS mRNA and active protein in O-2A progenitors. O-2A progenitors were dislodged again, (ie more than The results presented here indicate that GS 80% O-2A progenitors, A2B5+)19 centrifuged and resus- expression was differentially regulated in cultured pended in Dulbecco’s modified Eagle’s medium macroglial cell types. In O-2A progenitors, GS activity (DMEM, Gibco, Ref. 07401600) supplemented with was not increased by bFGF which indicated that a dif- 10% donor calf serum. Cells (105 cells per cm2) were ferential regulation of GS activity by bFGF existed replated in Falcon Petri dishes (Becton Dickinson and among various glial cell types. Indeed, Fressinaud et Co, Oxnard, CA, USA) precoated with poly-l-lysine al4 showed that GS activity was also not regulated by (60 000 Da, Sigma Chemicals, St Louis, MO, USA, bFGF in OL secondary cultures, but was upregulated in 20 ␮gml−1 in 0.1 M sodium borate-NaOH buffer, pH type-1 astrocytes. Thus, GS gene expression was likely 8.4).20 After 30 min, the calf serum-containing medium regulated by bFGF in a macroglial cell type-specific was removed and replaced with a chemically defined manner. medium,20 to which 2 ng ml−1 of basic fibroblast growth Furthermore, in OL secondary cultures which con- factor (bFGF) was added. After 2 days in this medium tained cells at all stages of differentiation and develop- (13 days, total time in culture) 80–90% of cells were ment, we demonstrated that the number of cells A2B5+. expressing GS protein increased during OL maturation. Type-2 astrocyte secondary cultures were prepared These cultures showed that 70% of more mature by modification of the Deloulme et al18 method and (MOG+) cells expressed GS. However, only 10% of OL- using high density O-2A progenitor cell cultures pre- 1+ cells which comprise immature cells, expressed GS. pared as previously described. These cells were grown Thus, during OL maturation in vitro, GS was expressed for 2 days in the chemically defined medium described at a stage preceding MOG expression. We noted that above with 2 ng ml−1 bFGF added. Then, DMEM was another protein, the growth-associated protein-43 removed and DMEM supplemented with 15% fetal calf (GAP-43), an abundant membrane phosphoprotein of serum (FCS) was added. Eighty to eighty-five percent of the CNS, was also expressed in type-1 astrocytes, type- O-2A precursor cells differentiated into glial fibrillary 2 astrocytes, OL and O-2A progenitors, and its acidic protein (GFAP+)-containing type-2 astrocytes expression was developmentally and cell-specifically within 3 days (16 days, total time in culture). regulated.15,16 In addition, the developmental time- To obtain OL secondary cultures, OL were dislodged course of mRNA encoding GS in vitro in type-1 astro- from 22-day primary mixed glial cell cultures18,21 and cytes17 was similar to the one described here for OL. plated for 2 h as described for O-2A progenitor cells. In conclusion, our data strongly indicated that GS After 2 h, the serum-containing medium was removed was expressed and was functional not only in type-1 and the chemically defined medium added. OL second- astrocytes, but also in type-2 astrocytes, OL, and O-2A ary cultures (105 cells per cm2) were grown in the progenitors. However, GS gene expression and GS gene chemically defined medium which was changed every regulation were cell type-specific and developmentally 3 days. The OL secondary culture system used in this regulated suggesting multiple roles or functions in the study was highly selective for the OL lineage. After 4 various macroglial cell types. The finding that the GS days in chemically defined medium (26 days, total time gene was subject to cell type-specific regulation broad- in culture), these subcultures contained at least 85% ens the number of potential biological metabolic path- GC- or CNP-positive OL.20,22 In addition, about 70% ways that this enzyme could subserve in vivo. This and 24% of these cells expressed MBP and MOG, might be expected, since the regulation of ammonia respectively. Both antigens were expressed after GC levels is of great importance. Additionally, in the brain, during OL maturation in vitro.23,24 OL secondary cul- GS may play a major role in glutamate metabolism. tures also contained O-2A progenitor cells (10–15% Myelination requires that myelin components, purines, A2B5+ cells) and a small proportion of astrocytes and pyrimidines, and amino acids be synthesized. This can microglia (Ͻ1.5% each) as previously described.18,21 be done in part by the GS pathway. Thus, these data Primary type-1 astrocyte cultures were prepared by indicate that GS is not only involved in later myelin- modification of the Booher and Sensenbrenner ation, but that GS is also likely an early OL differen- method.25 Cells were plated at a density of 5 × 103 cells tiation marker. per cm2 in Falcon Petri dishes. Culture medium was changed after 5 days and then twice a week. After 22 days in culture, 95% of the cells were positively Methods stained for GFAP. Astrocyte cultures were maintained Cell culture in the chemically defined medium. After 24 h in this Secondary O-2A progenitor cell cultures were prepared defined medium, the medium was changed and the from newborn Sprague–Dawley rat brains by modifi- cells were maintained for 3 additional days (ie, 23–26 cation of the Deloulme et al18 and Laeng et al19 days, total time). Glutamine synthetase in glial lineage D Baas et al 360 The CG-4 cell line was routinely cultured in DMEM Acknowledgments supplemented with N1 biotin as described by Louis et DB is supported by an MRE grant. We are grateful to al.9 Nearly all cells were maintained as self-renewing Dr JP Delaunoy, Dr MS Ghandour, Dr C Linnington, Dr O-2A progenitors (98% are A2B5+/GC−) in serum-free NM Neskovic and Dr B Ranscht for generous supplies medium by concurrent treatment with platelet-derived of GS antibody, rat mAb OL-1, mouse mAb against growth factor (PDGF) (5 ng ml−1) and bFGF (5 ng ml−1) MOG, CNP antibody and mouse mAb against galacto- as described.26 Upon PDGF and bFGF removal, CG-4 cerebroside, respectively. We also wish to thank Prof J cells differentiated readily into OL. After 2 days, most Kempf, Prof G Tholey, Prof G Vincendon, Dr A Reeber, cells (98%) acquired a typical OL differentiated pheno- Dr M Schmitt for stimulating discussions and for criti- type (GC+). Forty-eight hours after growth factor with- cal reading of the manuscript. We thank Dr K Bumsted drawal, 2% FCS was added to promote oligodendro- for helping in the preparation of the manuscript. We cyte survival and maturation. To induce O-2A also thank Dr JC Louis for the CG-4 cell line. progenitor cells to differentiate into type-2 astrocytes (A2B5+/GFAP+ cells), 20% FCS was added to the medium after growth factor withdrawal.9 References 1 Elliott WH. Glutamine synthesis. In: Colowick SP, Kaplan NO (eds). Methods in Enzymology. Vol. 2. Academic Press: New York, Immunocytochemistry 1955, pp 337–342. Immunostaining for the O-2A progenitors character- 2 Van den Berg CJ. Glutamate and glutamine. In: Lajtha A (ed). Hand- istic for A2B5 mouse monoclonal antibody (mAb), the book of Neurochemistry. Vol. 3. Plenum Press: New York, 1969, unknown sulfatide OL-1 characteristic for immature pp 355–379. 3 Cammer W. Glutamine synthetase in the central nervous system is OL rat mAb, the mature OL specific MOG mouse mAb, not confined to astrocytes. J Neuroimmunol 1990; 26: 173–178. the rabbit anti-GS polyclonal antibody, the mature OL 4 Fressinaud C, Weinrauder H, Delaunoy JP, Tholey G, Labourdette specific SMI 99 mAb anti-MBP or mouse mAb anti- G, Sarlie`ve LL. Glutamine synthetase expression in rat oligodendro- GFAP was performed as previously described.26,27 Dou- cytes in culture: regulation by hormones and growth factors. J Cell Physiol 1991; 149: 459–468. ble immunolabelling with polyclonal guinea pig anti- 5 Juurlink BHJ, Hertz L. Establishment of highly enriched type-2 rat brain myelin CNP antibody prepared by Dr NM Nes- astrocyte cultures and quantitative determination of intense gluta- kovic in our laboratory (1/50 dilution, 1 h at RT) was mine synthetase activity in these cells. J Neurosci Res 1991; 30: done with anti-guinea pig IgG (1/100 dilution, Biosys 531–539. SA, France) in the same manner as MBP double immu- 6 Martinez-Hernandez A, Bell KP, Norenberg MD. Glutamine syn- thetase: glial localization in brain. Science 1977; 195: 1356–1358. nolabelling. After further washing in PBS, coverslips 7 Mearow KM, Mill JF, Vitkovic L. The ontogeny and localization of were mounted in 58% glycerol in PBS, viewed and glutamine synthetase gene expression in rat brain. Mol Brain Res photographed using an epifluorescence illumination. 1989; 6: 223–232. 8 Ghandour MS, Nussbaum JL. Oligodendrocyte cell surface recog- nized by a novel monoclonal antibody specific to sulfatide. Neuro- Other analyses report 1990; 1: 13–16. Glutamine synthetase activity was determined in crude 9 Louis JC, Magal E, Muir D, Manthorpe M, Varon S. CG-4, a new 28 bipotential glial cell line from rat brain, is capable of differentiating cell extracts according to Perraud et al. To analyze in vitro into either mature oligodendrocytes or type-2 astrocytes. J the expression of mRNAs coding for GS, a 1300-bp Neurosci Res 1992; 31: 193–204. EcoRI/PstI cDNA fragment inserted into pBR3227 was 10 Danielson PE, Forss-Petter S, Brow MA, Calavetta L, Douglass J, radiolabelled by random priming with 32P-dCTP (NEN, Milner R et al. p1B15: a cDNA clone of the rat mRNA encoding 3000 Ci mmol−1, Les Ulis, France) using a random cyclophilin. DNA 1988; 7: 261–267. + 11 Hallermayer K, Hamprecht B. Cellular heterogeneity in primary priming kit (BRL, Cergy Pontoise, France). Poly(A) cultures of brain cells revealed by immunochemical localization of mRNA was prepared as described by Hartmann et al.29 glutamine synthetase. Brain Res 1984; 295: 1–11. Five micrograms of mRNA were electrophoresed on a 12 D’Amelio F, Eng LF, Gibbs MA. Glutamine synthetase immuno- 1% agarose gel (20 mM MOPS, 5 mM sodium acetate, reactivity is present in oligodendroglia of various regions of the central nervous system. Glia 1990; 3: 335–341. 1 mM EDTA, pH 7.0, 0.66 M formaldehyde). The RNAs 13 Tansey FA, Farooq M, Cammer W. Glutamine synthetase in oligo- were transferred onto nylon and blots were hybridized dendrocytes and astrocytes: new biochemical and immunocyto- at 42°C overnight with 2 × 106 cpm ml−1 of 32P-labelled chemical evidence. J Neurochem 1991; 56: 266–272. 14 Gallo V, Zhou JM, McBain CJ, Wright P, Knutson PL, Armstrong cDNA in 50% deionized formamide, 50 mM NaH2PO4, RC. Oligodendrocyte progenitor cell proliferation and lineage pro- pH 7.4, 1 mM EDTA, 0.75 M NaCl, 1% Denhardt’s, gression are regulated by glutamate receptor-mediated K+ channel 0.1% SDS. Blots were washed twice at room tempera- block. J Neurosci 1996; 16: 2659–2670. ture in 1× sodium salt citrate (SSC), then once with 1× 15 da Cunda A, Vitkovic L. Regulation of immunoreactive GAP-43 SSC, 1% SDS for 20 min at 42°C and twice with 1× expression in rat cortical macroglia is cell type specific. J Cell Biol SSC, 1% SDS for 20 min at 50°C. Blots were autoradio- 1990; 111: 209–215. − ° 16 Vitkovic L. GAP-43 expression in macroglial cells: potential func- graphed on Kodak XAR5 film at 70 C. Relative mRNA tional significance. Perspect Develop Neurobiol 1992; 1: 39–43. levels were quantitated on autoradiograms by densito- 17 da Cunda A, Aloyo VJ, Vitkovic L. Developmental regulation of metry. The recovery of mRNAs was determined by GAP-43, glutamine synthetase and ␤-actin mRNA in rat cortical hybridization with a probe for cyclophilin, an unre- astrocytes. Dev Brain Res 1991; 64: 212–215. 18 Deloulme JC, Laeng P, Janet T, Sensenbrenner M, Baudier J. lated gene constitutively expressed in all animal Expression of neuromodulin (GAP43) and its regulation by basic 10 cells. All values were normalized to the cyclophilin fibroblast growth factor during the differentiation of O-2A progeni- signal. tor cells. J Neurosci Res 1993; 36: 147–162. Glutamine synthetase in glial lineage D Baas et al 361 19 Laeng P, De´cimo D, Pettmann B, Janet T, Labourdette G. Retinoic 26 Baas D, Fressinaud C, Ittel ME, Reeber A, Dalenc¸on D, Puymirat J acid regulates the development of oligodendrocyte precursor cells et al. Expression of thyroid hormone receptor isoforms in rat oligo- in vitro. J Neurosci Res 1994; 39: 613–633. dendrocyte cultures. Effect of 3,5,3Ј-triiodo-l-thyronine. Neurosci 20 Besnard F, Perraud F, Sensenbrenner M, Labourdette G. Effects of Lett 1994; 176: 47–51. acidic and basic fibroblast growth factors on proliferation and 27 Baas D, Bourbeau D, Carre´ JL, Sarlie`ve LL, Dussault JH, Puymirat maturation of cultured rat oligodendrocytes. Int J Devl Neurosci J. Expression of ␣ and ␤ thyroid receptors during oligodendrocyte 1989; 7: 401–409. differentiation. NeuroReport 1994; 5: 1805–1808. 21 Labourdette G, Roussel G, Nussbaum JL. Oligodendroglia content 28 Perraud F, Besnard F, Pettmann B, Senbrenner M, Labourdette G. of glial cell primary cultures from newborn rat brain hemispheres, Effects of acidic and basic fibroblast growth factors (aFGF and depends on the initial plating density. Neurosci Lett 1980; 18: bFGF) on the proliferation and the glutamine synthetase expression 203–209. of rat astroblasts in culture. Glia 1988; 1: 124–131. 22 Fressinaud C, Vallat JM, Rigaud M, Cassagne C, Labourdette G, Sar- 29 Hartmann KP, Papa V, Brown EJ, Doerries U, Rosenthal S, Goldfin lie`ve LL. Investigation of myelination in vitro: polar lipid content I. A rapid and simple one step method for isolation of polyA RNA and fatty acid composition of myelinating oligodendrocytes in rat from cells in monolayer. Endocrinology 1990; 127: 2038–2040. oligodendrocyte cultures. Neurochem Int 1990; 16: 27–39. 23 Duboid-Dalcq M, Behar T, Hudson L, Lazzarini RA. Emergence of Correspondence to Dr D Baas at her present address: Department of three myelin proteins in oligodendrocytes cultured without neu- Ophthalmology and Visual Science, Yale University School of Medi- rons. J Cell Biol 1986; 102: 384–392. cine, 330 Cedar Street, PO 208061, New Haven, CT 06520-8061, USA. 24 Matthieu JM, Amiguet P. Myelin/oligodendrocyte glycoprotein E-mail: dominique.baasȰyale.edu expression during development in normal and myelin-deficient Reprint requests: Dr LL Sarlie`ve, UPR 416 du CNRS, Centre de Neuro- mice. Dev Neurosci 1990; 6: 293–302. chimie 5, rue Blaise Pascal, 67084 Strasbourg Cedex, France. E-mail: 25 Booher J, Sensenbrenner M. Growth and cultivation of dissociated sarlieveȰneurochem.u-strasbg.fr neurons and glial cells from embryonic chick, rat and human brain Received 15 September 1997; revised 2 January 1998; accepted 5 Janu- in flask cultures. Neurobiology 1972; 2: 97–105. ary 1998