O1+) from embryonic rat cerebrum

Brain Research Protocols 10 (2002) 23–30 www.elsevier.com / locate / bres

Protocol

Culture of oligodendrocyte precursor cells (NG2 1 / O1 2) and oligodendrocytes (NG2 2 / O1 1 ) from embryonic rat cerebrum Kouichi Itoh* Department of Molecular Biodynamics, The Tokyo Metropolitan Institute of Medical Science, Tokyo Metropolitan Organization for Medical Research, Rm. 408, 3 -18 -22 Honkomagome, Bunkyo-ku, Tokyo 113 -8613, Japan Accepted 30 May 2002

Abstract I have established a novel culture technique of oligodendrocyte precursor cells (OPC, NG2 1 / O1 2) and oligodendrocytes (OL, NG2 2 / O1 1 ) from embryonic day 16 (E16) rat cerebrum distinguished by morphological and immunocytochemical analyses. This novel protocol does not require immunopanning techniques using specific antibodies. The OPC were isolated by two passages every 7 days and culturing them on Petri dishes in different culture medium at each step to eliminate neurons and astrocytes. The yield of pure OPC and OL was relatively large compared with immunopanning techniques. In addition, to examine myelination processes in vitro, the OL from different stages were co-cultured with primary neurons from E17 rat cerebrum on a feeder layer of rat cerebrum astrocytes. This co-culture system resulted in successful formation of myelin in the presence or absence of astrocytes. This culture system was useful for studying OL lineage and initiation of myelination.  2002 Elsevier Science B.V. All rights reserved. Theme: Development and regeneration Topic: Glia and other non-neuronal cells Keywords: Cell cultures; Oligodendrocyte precursor cell; Oligodendrocyte; Myelination

1. Type of research • Primary culture of oligodendrocyte precursor cells (OPC) and oligodendrocytes (OL) from embryonic rat brain. • Formation of myelin using OL and neurons and / or astrocytes in culture.

• Step 2, min. • Step 3, min. • Step 4, defined • Step 5, defined

after 7 days, passage of the cultured cells: 30 after 7 days, passage of the cultured cells: 30 after 2 days, changing to serum free-chemical medium with PDGF. after 3 days, changing to serum free-chemical medium with T 3 , T 4 and NT-3.

2. Time required

2.2. Myelination 2.1. Culture of OPC and OL • Step 1, primary culture (including brain dissection and dissociated single cell suspension): 1.5–2 h (excluding preparation of solution). *Tel.: 181-3-3823-2105x5436; fax: 181-3-3823-2130. E-mail address: [email protected] (K. Itoh).

• Primary cultures of neurons from E17 and astrocytes from P0 cerebrum (including brain dissection and dissociated single cell suspension): 1.5–2 h (excluding preparation of solution). • Cultivation of pure astrocytes: up to 2 months. • Cultivation of neurons on astrocytes: up to 14 days. • Cultivation of neurons on astrocytes: up to 14 days.

1385-299X / 02 / $ – see front matter  2002 Elsevier Science B.V. All rights reserved. PII: S1385-299X( 02 )00177-0

24

K. Itoh / Brain Research Protocols 10 (2002) 23–30

• Cultivation of OL with neurons on astrocytes for myelination: up to 14 days.

3. Materials

3.1. Animals Pregnant Wistar rats: embryonic day 16 (E16) for OPC and OL, embryonic day 17 (E17) for neurons and postnatal day 0 (P0) for astrocytes.

3.2. Special equipment • Sterile microdissection kit (two pairs of fine-tipped forceps and small surgical scissors, micro spring scissors) (Roboz, USA). • Dissecting microscope (Nikon, Japan). • Refrigerated centrifuge (Kokusan, H-3R, Japan). • Water-jacketed humidified CO 2 incubator (Napco, Japan). • Sterile 15- and 50-ml centrifuge tubes (Corning, USA). • Pasteur pipettes (Corning). • Sterile 5- and 10-ml plastic pipettes (Falcon, USA). • Culture dishes (Griner, Germany). • Petri dishes (6 and 10 cm diameters) (Sterilin, UK). • Circular coverslips 13 mm diameter for cell culture (Matumani, Japan). • 140 mm pore-sized stainless mesh (JIS, Japan). • Cell strainer (70 mm nylon) (Falcon). • 10 mm pore-sized nylon mesh (NITEX, Switzerland). • Fluorescence microscopy with Zeiss confocal laser scanning LSM510 system with an inverted microscope (Axiophot; Carl Zeiss, Germany). • JEOL 1010 electron microscope (Japan).

3.3. Chemicals and reagents • Eagle’s-minimal essential medium (E-MEM), Dulbecco’s-MEM (D-MEM), Hank’s balanced salt solution (HBSS), D-PBS, Neurobasal medium, B27 and N2 supplements and GlutaMax (all from Invitrogene, USA). • Heat-inactivated fetal calf serum (FCS) (JRH Biosci, USA). • Poly-L-lysine (PLL, ICN biomedical, USA). • Trypsin solution (0.25% trypsin and 1 mM EDTA) (Invitrogene). • Supplements: insulin, transferrin, BSA, progesterone, putrescine, sodium selenite, N-acetyl-L-cysteine, glutamic acid, platelet-derived growth factors (PDGFAA), forskolin, T 3 (triodothyronin) and T4 (thyroxin), NT-3 (all ingredients from Sigma, USA). • Secondary antibodies: goat anti-mouse IgM-Alexa

Fluor 488 and 594, goat anti-rabbit IgG-Alexa Fluor 488 and 594 (all from Molecular Probes, USA). • Monoclonal antibodies: NG2 (Dr. Stallcup, USA), O4 (ATCC, USA), O1 (ATCC, USA), A2B5 (ATCC, USA), RT97 (DSHB, USA), MBP (SMI94, USA), GFAP (SMI21, USA), Tuj1 (Promega, USA), Rip (DSHB, USA). • Polyclonal antibodies: GalC (Biomakor, USA), NG2 (Dr. Stallcup), PDGFRa (Upstate biotech, USA), PLP (Chemicon, USA). • Vectashield  mounting medium (Vector Labs., USA).

4. Detailed procedure

4.1. OPC and OL culture Step 1, OPC and OL cultures were prepared from an E16 rat cerebrum by a modification of the method of previously described methods [10]. All fetuses from a pregnant rat were pooled for a culture preparation. Step 2, fetal rat brains (8–12 brains) were removed by dissection of the skull using sterile fine-tipped forceps and micro spring scissors in sterile ice-cold HBSS. Cerebral cortices were dissected from the whole brain and meninges and blood vessels were carefully removed. Step 3, the rat embryonic cerebral cortex was mechanically dissociated through 140 mm pore-sized stainless mesh in 10% FCS in E-MEM. Step 4, the cells were dissociated with a fire-polished Pasteur pipette. Trituration involved pipetting the tissue suspension in and out of the pipette |20 times. A 15-ml tube of cells was left still for 5 min to allow the tissue debris to settle down. The supernatant was transferred to a fresh 15-ml tube, and then was centrifuged for 10 min at 1003g at 4 8C. Cells were resuspended in 10% FCS / EMEM and finally sieved through 70 mm pore-sized nylon mesh. This step was repeated twice. The number of viable cells was determined by Trypan blue exclusion in a hemocytometer. Step 5, dispersed cells were seeded on PLL (100 mg / ml)-coated 90 mm diameter culture dishes at a density of 1310 7 cells / dish. Cells were incubated in an atmosphere of 90% air, and 10% CO 2 and |98% humidity at 37 8C. Step 6, after 7 days culture, the cells were passaged with 0.05% trypsin in D-PBS (1st passage). Cells were resuspended in 10% FCS / E-MEM and were finally sieved through 10 mm pore-sized nylon mesh and were cultured for 7 days at a density of 8310 6 cells per non-coated culture dish. Step 7, after 7 days culture, cells were passaged with 0.05% trypsin in D-PBS and were cultured for 2 days in 10% FCS / E-MEM at a density of 3310 6 cells per noncoated Petri dish (2nd passage). Step 8, on the 2nd day of culture, the medium was exchanged to serum-free chemical defined D-MEM

K. Itoh / Brain Research Protocols 10 (2002) 23–30

(CDM) supplemented with 10 mg / ml insulin, 0.5 mg / ml transferrin, 100 mg / ml BSA, 60 ng / ml progesterone, 16 mg / ml putrescine, 40 ng / ml sodium selenite, 60 ng / ml N-acetyl-L-cysteine, 5 mM forskolin and 10 ng / ml PDGFAA [18], and the OPC were cultured for another 2 days. Step 9, to differentiate OPC, the cells were continuously cultured with 30 ng / ml T 3 , 40 ng / ml T 4 , 10 ng / ml NT-3 [2,3] and without PDGF for 3–5 days. These procedures are necessary to eliminate neurons and astrocytes.

4.2. Neuron cultures Neuron cultures were prepared from E17 rat brain [12]. Briefly, the rat cerebral cortex was mechanically dissociated through 140 mm pore-sized stainless mesh in 10% FCS in high glucose D-MEM. The dissociated cells were finally sieved through 70 mm pore-sized nylon mesh and then were centrifuged for 10 min, 1003g at 4 8C. The cells were resuspended with Neurobasal medium supplemented with B27 and N2 to purify neurons. Before experiments, neurons were plated on a feeder layer of astrocytes and plated on PLL (100 mg / ml)-coated glass coverslips.

4.3. Astrocyte cultures Astrocyte cultures were prepared from P0 rat brain. Dispersed cells were suspended in 10% FCS / D-MEM and were seeded on PLL (100 mg / ml)-coated 90 mm diameter culture dishes at a density of 1310 6 cells. After 7 days culture, cells were passaged with 0.25% trypsin in D-PBS to eliminate other cells. After being resuspended in 10% FCS / D-MEM, astrocytes were seeded and cultured for 7 days. After 3–5 passages, cells were used for feeder layer. Before experiments, astrocytes were plated on PLL (100 mg / ml)-coated glass coverslips.

4.4. Myelination in culture Neurons (1–5310 4 cells / ml) were co-cultured on a feeder layer of cultured astrocytes on PLL-coated glass coverslips for 4–7 days to differentiate neurons in 10% FCS / D-MEM. OL were plated on the neurons / astrocytes co-cultures in the serum-free CDM. The plating density of OL was greater than that of neurons with ratio varying between 3:1 and 5:1.

4.5. Immunofluorescence staining Step 1, for cell surface staining with A2B5 [1], O4 [19], O1 [20], anti-galactocerebroside (GalC) antibodies, OPC and OL growing on a Petri dish were fixed with 2% paraformaldehyde solution for 15 min at RT.

25

Step 2, for staining for NG2 [17], immunolabeling of cell surface was carried out on living culture for 15 min, and cells were fixed in 4% paraformaldehyde for 15 min at RT. Step 3, for anti-MBP anti-PLP antibodies [5] it was done by 15 min of 4% paraformaldehyde incubation followed by 10 min of 0.1% Triton X-100 incubation on ice. Step 4, in myelination cultures, RT97 antibody against phosphorylated 200 kDa neurofilament [14] was used to identify matured axons and MBP immunolabeling allowed visualization of OL. Step 5, for intracellular staining of neurofilament (RT97), b tubulin III (Tuj1) [7] and GFAP [4], cells were fixed in 4% paraformaldehyde for 15 min at RT and for 5 min in acidic alcohol (95% ethanol and 5% glacial acetic acid) at 220 8C. Step 6, following washing, cells were incubated with primary antibodies for 1 h at RT or overnight at 4 8C (except for NG2). After washing, cells were incubated with Alexa Fluor-conjugated anti-rabbit IgG, Alexa Fluorconjugated anti-mouse IgG or Alexa Fluor-conjugated antimouse IgM for 1 h at RT. Step 7, for double immunolabeling, the same procedures were repeatedly performed. Stained cells were mounted on slide-glasses or Petri dish using Vectashield  mounting medium. Fluorescence microscopy was performed with a Zeiss confocal laser scanning LSM510 system with an inverted microscope (Axiophot; Carl Zeiss) and Olympus confocal laser scanning FV300 system with an inverted microscope (IX70, Olympus, Japan).

4.6. Identification of cell types Step 1, OPC were positively identified by immunostaining for PDGFRa, A2B5 and NG2. Step 2, OL were positively identified by immunostaining for O4, O1 / GalC, Rip, MBP and PLP. Step 3, neurons were positively identified by immunostaining for A2B5, Tuj1 and RT97. Step 4, astrocytes were positively identified by immunostaining for GFAP and Ran2. Step 5, cells were double-labeled with antibodies to show that the antibodies specifically stained.

4.7. Electron microscopic analysis After washing with PBS, cultured cells were fixed in 2.5% glutaraldehyde for 1 h at 4 8C and then postfixed in 1% OsO 4 . After dehydration in a graded series of ethanol solutions, cultures were embedded in Epon–Araldite mixture. These plastic embedded ultrathin preparations were examined in a JEOL 1010 electron microscope operated at 80 kV.

K. Itoh / Brain Research Protocols 10 (2002) 23–30

26

5. Results

5.1. OL can be differentiated from OPC in culture ( Fig. 1) To eliminate differentiated neurons, 7-day-old primary

cultured cells (first phase, Ph. 1) were passaged using 0.05% trypsin and trituration. The dispersed cells were plated on non-coated cultured dishes (second phase, Ph. 2). After 7 days the same procedure was repeated (Fig. 1A). After a second passage, dispersed cells in 10% FCS / EMEM were plated on a Petri dish (third phase, Ph. 3). The cells had to be cultured for 2 days on the Petri dish (Fig. 1B). After this process, serum-free CDM with PDGF was replaced to proliferate OPC for 3 days (Fig. 1C). During this culture period, the number of cells with bipolar and tripolar processes like OPC was dramatically increased by PDGF. To allow these OPC-like cells to differentiate into OL, T 3 , T 4 and NT-3 in serum-free CDM were replaced with PDGF (Fig. 1D). Most of the OPC-like cells differentiated into cells with many processes and the larger cell bodies (.10 mm) resembling OL. When the cells were cultured on a culture dish, instead of a Petri dish, during this step, it was difficult to eliminate neurons and astrocytes.

5.2. Identification of OPC in culture OPC in this culture system was determined by morphological and immunocytochemical analyses using various antibodies. OPC defined by NG2 positive ( 1 ) / O4 negative ( 2) and O1 2 immunoreactivities formed a few bipolar processes (Fig. 2, Table 1). NG2 1 OPC also were recognized by A2B5 antibody, but not Tuj1 as premature neuronal marker and GFAP as astrocyte marker. Thus, OPC could be obtained (.98%) from rat cerebrum of embryonic day 16 without immunopanning techniques (Table 1).

5.3. Identification of OL in culture After 3 days in culture, T 3 , T 4 and NT-3 were added into medium in the absence of PDGF to enhance differentiation of OPC for 4–5 days. OL were defined by O4 1 , O1 1 , Rip 1 and MBP 1 immunoreactivities, but NG2 2 and A2B5 2 (Fig. 3, Table 1). The results for each combination of antibodies analyzed in the OPC and OL are shown in Table 1. Thus, it was shown that most of OPC differentiated into OL in culture. Yield of OL was 10 to 20 dishes of 5310 5 –2310 6 per 6 cm diameter Petri dish. This amount of OL was enough to perform biochemical and molecular biological analyses.

5.4. Myelination in culture Fig. 1. Preparation of OPC and OL in culture. Phase contrast micrographs of each step during the procedure. (A) In the second week (second phase, Ph2–7 DIV), just before 2nd passage. (B) In the third week (third phase, Ph3), 1 day (1 DIV) after replacement of medium with PDGF. (C) 3-day culture (3 DIV) in the presence of PDGF before replacement of medium. (D) 4-day culture (7 DIV) in the presence of T 3 , T 4 and NT3. DIV, day in vitro. Bar520 mm.

Myelination quantified at 7 DIV MBP 1 OL were distinguished by morphological differences. OL designated as ‘premyelinating OL’ had many thin processes (Fig. 4A). OL designated ‘myelinating OL’ had thick processes, and a reduced number of processes compared with premyeli-

K. Itoh / Brain Research Protocols 10 (2002) 23–30

27

Fig. 2. Double immunofluorescence staining of OPC cultures with anti-NG2, anti-A2B5, anti-O1, anti-Tuj1 and anti-GFAP antibodies. Most of the cells were positively stained with NG2 (A–D). The NG2 1 cells were A2B5 1 (E), but not O1 (F), Tuj1 (G) for neurons and GFAP (H) for astrocytes. Corresponding phase contrast micrographs to fluorescence, I for A and E, J for B and F, K for C and G, L for D and H. Scale bar520 mm.

nated OL (Fig. 4B and C). In the culture, myelinating OL (168 cells) comprised 72% of MBP 1 OL (232 cells) and there were only 64 premyelinating OL (28%). Electron microscopic analysis was performed in cultures. In this system, structure of compacted myelin appeared 1 day after addition of OL. The number of Table 1 Developmental profile of OPC and OL expressing lineage-specific markers in vitro Markers (antibodies)

Positive cells (%) OPC

OL

NG2 A2B5 1 O4 1 O1 1 / GalC 1 MBP 1

98.865.2 97.962.7 1.561.4 1.262.8 ND

2.663.5 1.569.4 96.161.2 97.268.5 59.5666.9

NG2 1 /A2B5 1 NG2 1 / GC 2 NG2 1 / GC 1 O4 1 / GC 1 MBP 1 / GC 1

97.962.7 96.264.2 3.563.4 1.262.8 ND

1.569.4 2.262.5 0.461.3 96.161.2 59.566.9

1

OPC and OL were double-labeled by anti-A2B5, NG2, O4, O1 (GalC), MBP antibodies. Figures as the ratio of positive cells of each lineagespecific marker against total number of cells (1000 cells) represent mean6S.E. of three independent cultures. ND, not detectable.

compacted myelin was gradually increased until 7 days in culture (data not shown).

6. Discussion In this study, I describe for the first time the method required for culture of OPC and OL from rat embryonic cerebrum, not using immunopanning techniques with specific antibodies. I have successfully used this method to study the OL lineage using biochemical and immunocytochemical techniques [10,16]. This culture system used a Petri dish to purify OPC after second passage instead of immunoselection. The OPC were recognized not only with NG2, but also with A2B5 and antibodies against GD 3 and PDGFR-a. Most of OPC at 1 day were bipolar in shape, but a few of the cells already had a multipolar morphology like OL. These cells with multipolar processes included not only O1 1 OL, but also OPC. Bipolar OPC were also not completely synchronized. However, as shown in Table 1, more than 98% of 1-day-old cultured cells were NG2 1 OPC and 97% of 7-day-old cultured cells were O1 1 OL. Thus, these OPC were freshly isolated as OPC population at second passage. The other small population of cells was already OPC after first passage. If analysis of OL lineage

28

K. Itoh / Brain Research Protocols 10 (2002) 23–30

Fig. 3. Double immunofluorescence staining of OL cultures with anti-O4, anti-O1, anti-MBP and anti-Rip antibodies. O4 1 (A) and O1 1 (D), O1 1 (B) and MBP 1 (E), and Rip 1 (C) in OL. Corresponding phase contrast micrographs to fluorescence, G for A and D, H for B and E, F for C. Scale bar520 mm.

cell population is required, development between OPC and OL is observed during 7 days in serum-free CDM. Other methods are available to yield and purify OPC and OL. These cells have been purified on a Percoll gradient [9]. In this method, many animals have to be utilized, although the procedure does not require a long time. In another method green fluorescent protein (GFP) transgenic animals are used to isolate OPC and OL. Using fluorescence-

activated cell sorting (FACS), the developmental stagespecific promoter driven GFP 1 cells are isolated and enriched to purity [6]. However, this method requires GFP transgenic animals, FACS instrument and molecular biological techniques to purify these cells. On the other hand, the present protocol was simple and easy to isolate OPC and differentiate into OL when a particular brand of Petri dish was used in culture.

K. Itoh / Brain Research Protocols 10 (2002) 23–30

Fig. 4. Immunofluorescence staining of premyelinating (A) and myelinating OL (B) in culture with anti-MBP antibody at 4 days. Premyelinating OL (28% of MBP 1 ) had thin and many processes (A) and myelinated OL (72% of MBP 1 ) had thick processes, and decreased number of processes compared with premyelinated OL. (C) Double immunofluorescence staining of myelinating OL in culture with RT-97 (green) and anti-MBP (red) antibodies. Scale bar510 mm.

OPC in the present method were obtained from E16 rat cerebrum. When E18 cerebra were used, it was difficult to obtain OPC. When cells were cultured from E14–15 rat cerebrum, OPC were purified after second passage, how-

29

ever the yield was 10 times lower than E16. Thus, it was important to dissociate E16 rat cerebrum to purify OPC on a Petri dish. Several cell attachment substrates and Petri dishes were tested for their purification of OPC. OPC could not be successfully purified on any substrates and Petri dishes, except Petri dishes from one company. The reason for the ability of this particular brand of Petri dishes to enable successful OPC purification is not clear. In some cases, the use of non-coated coverslips was successful, although larger numbers of astrocytes were contaminated compared with the Petri dish. In addition, the quality of FCS had an effect on the purification of OPC and OL in this system. In this study, embryonic rat brains were used to purify OPC and OL. OL (O4 1 / O1 2 and O4 2 / O1 1 ) were isolated from postnatal mouse brain using O4 and O1 mAbs immunoselection [8]. However, immunopanning techniques cannot be used to purify OPC from mouse brain, because a good antibody to select OPC has not been available so far. Thus, my method has the potential advantage of using a Petri dish for culturing OPC and OL in mice. I have tried to purify OPC and OL from mouse. A small number of mouse OL could be purified by the modification of this protocol (age, culturing period etc.). However, OPC and large numbers of OL could not be obtained yet as in the case of rat cultures. The OPC and OL purification from cerebrum required a careful choice of the age and strain of mice. To settle this issue, I am continuing to determine optimum culture conditions. In the peripheral nervous system, signal molecules for myelination are known to interact with axons and Schwann cells [21,22]. However, little is known about the information of myelination at the initial stage in brain. Zalc et al. have reported that myelination occurs 2 weeks in vitro after primary cultures from E15 mouse cerebrum [15]. In this culture system, OL and myelination could not be controlled. In the present system, developmentally different stages of OL were added to neurons / astrocytes cultures and processes of myelination were observed. The OL in my method had myelinogenic potential in vitro. The OL were maintained in serum-free CDM in the presence of T 3 , T 4 and NT-3 for at least 4 days and after that the OL gradually started dying. However, the OL cultured with neurons and astrocytes were maintained in serum-free CDM for 2 more weeks. This observation indicated that the survival of OL was promoted by interaction with neurons and astrocytes. To determine interaction between OL and axons at the initial stage of myelination, this myelination technique is useful. Immunocytochemical and electron microscopic analyses are performed to aim at blocking an important cell surface molecule by IgG [F(ab9) 2 ] fragment of the antibody or reagents were added to the culture medium on neurons and astrocytes before or after adding OL and myelination quantified at 1–14 DIV. Myelinating MBP 1 OL were distinguished by morphological differences. Thus, it is demonstrated whether or not the

K. Itoh / Brain Research Protocols 10 (2002) 23–30

30

IgG [F(ab9) 2 ] fragment of the antibody or reagents influence myelination at the initial stage.

[7] [8]

7. Trouble shooting OPC could be successfully purified on this particular brand of Petri dishes. In addition, Petri dish selection of OPC required careful choice of FCS quality. When batch of FCS is tested in your laboratory, FCS should be chosen by growth inhibition of astrocytes as selection index. A good FCS for OPC and OL is a bad one for astrocytes.

8. Essential literature references Oligodendrocyte cultures: [3,10,16,18]. Cerebrum neuron and astrocyte cultures: [11]. Immunocytochemistry: [13,16].

[9]

[10]

[11]

[12]

Acknowledgements

[13]

This work was supported by a grant-in-aid for Scientific Research from JSPS (K.I.). The author thanks Dr. A. Nishiyama (University of Connecticut, USA) for critical reading of the manuscript, Dr. B. Stallcup (The Burnham Institute, La Jolla Cancer Research Center, CA, USA) for kindly providing NG2 antibodies and the members of the Department of Molecular Biodynamics (Dr. Umeda’s lab.) for discussion and support.

[14]

References [1] E.R. Abney, B.P. Williams, M.C. Raff, Tracing the development of oligodendrocytes from precursor cells using monoclonal antibodies, fluorescence-activated cell sorting, and cell culture, Dev. Biol. 100 (1983) 166–171. [2] B.A. Barres, M.A. Lazar, M.C. Raff, A novel role for thyroid hormone, glucocorticoids and retinoic acid in timing oligodendrocyte development, Development 120 (1994) 1097–1108. [3] B.A. Barres, M.C. Raff, F. Gaese, I. Bartke, G. Dechant, Y.A. Barde, A crucial role for neurotrophin-3 in oligodendrocyte development, Nature 367 (1994) 371–375. [4] A. Bignami, L.F. Eng, D. Dahl, C.T. Uyeda, Localization of the glial fibrillary acidic protein in astrocytes by immunofluorescence, Brain Res. 43 (1972) 429–435. [5] A.T. Campagnoni, Molecular biology of myelin proteins from the central nervous system, J. Neurochem. 51 (1988) 1–14. [6] K.J. Chandross, R.I. Cohen, P. Paras Jr., M. Gravel, P.E. Braun, L.D. Hudson, Identification and characterization of early glial progenitors

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

using a transgenic selection strategy, J. Neurosci. 19 (1999) 759– 774. S.S. Easter Jr., L.S. Ross, A. Frankfurter, Initial tract formation in the mouse brain, J. Neurosci. 13 (1993) 285–299. A.L. Gard, S.E. Pfeiffer, W.C. Williams II, Immunopanning and developmental stage-specific primary culture of oligodendrocyte progenitors (O4 1 GalC 2) directly from postnatal rodent cerebrum, Neuroprotocols: A Companion Methods Neurosci. 2 (1983) 209– 218. M. Hirayama, D.H. Silberberg, R.P. Lisak, D. Pleasure, Long-term culture of oligodendrocytes isolated from rat corpus callosum by Percoll density gradient. Lysis by polyclonal antigalactocerebroside serum, J. Neuropathol. Exp. Neurol. 42 (1983) 16–28. K. Itoh, Y. Sakurai, H. Asou, M. Umeda, Differential expression of alternatively spliced neural cell adhesion molecule L1 isoforms during oligodendrocyte maturation, J. Neurosci. Res. 60 (2000) 579–586. K. Itoh, Y. Ikarashi, Y. Maruyama, Morphological changes and neural cell adhesion molecule expression in mouse cerebrum primary cultures following long-term exposure to phorbol ester, Neurosci. Res. 6 (1989) 350–357. K. Itoh, H. Kawamura, H. Asou, A novel monoclonal antibody against carbohydrates of L1 cell adhesion molecule causes an influx of calcium in cultured cortical neurons, Brain Res. 580 (1992) 233–240. K. Itoh, B. Stevens, M. Schachner, R.D. Fields, Regulated expression of the neural cell adhesion molecule L1 by specific patterns of neural impulses, Science 270 (1995) 1369–1372. S.N. Lawson, A.A. Harper, E.I. Harper, J.A. Garson, B.H. Anderton, A monoclonal antibody against neurofilament protein specifically labels a subpopulation of rat sensory neurones, J. Comp. Neurol. 228 (1984) 263–272. C. Lubetzki, C. Demerens, P. Anglade, H. Villarroya, A. Frankfurter, V.M. Lee, B. Zalc, Even in culture, oligodendrocytes myelinate solely axons, Proc. Natl. Acad. Sci. USA 90 (1993) 6820–6824. Y. Nakai, Y. Sakurai, A. Yamaji, M. Umeda, K. Uyemura, K. Itoh, Lysenin-sphingomyelin binding at the surface of oligodendrocyte lineage cells increases during differentiation in vitro, J. Neurosci. Res. 62 (2000) 521–529. A. Nishiyama, X.H. Lin, N. Giese, C.H. Heldin, W.B. Stallcup, Co-localization of NG2 proteoglycan and PDGF alpha-receptor on O2A progenitor cells in the developing rat brain, J. Neurosci. Res. 43 (1996) 299–314. M.C. Raff, R.H. Miller, M. Noble, A glial progenitor cell that develops in vitro into an astrocyte or an oligodendrocyte depending on culture medium, Nature 303 (1983) 390–396. I. Sommer, M. Schachner, Cells that are O4 antigen-positive and O1 antigen-negative differentiate into O1 antigen-positive oligodendrocytes, Neurosci. Lett. 29 (1982) 183–188. I. Sommer, M. Schachner, Monoclonal antibodies (O1 to O4) to oligodendrocyte cell surfaces: an immunocytological study in the central nervous system, Dev. Biol. 83 (1981) 311–327. P. Wood, F. Moya, C. Eldridge, G. Owens, B. Ranscht, M. Schachner, M. Bunge, R. Bunge, Studies of the initiation of myelination by Schwann cells, Ann. NY Acad. Sci. 605 (1990) 1–14. P.M. Wood, M. Schachner, R.P. Bunge, Inhibition of Schwann cell myelination in vitro by antibody to the L1 adhesion molecule, J. Neurosci. 10 (1990) 3635–3645.