During embryonic development, neural crest cells differentiate into a wide variety of cell types including Schwann cells of the peripheral nervous system. In order to establish when neural crest cells first start to express a Schwann cell phenotype immunocytochemical tech niques were used to examine rat premigratory neural crest cell cultures for the presence of Schwann cell markers. Cultures were fixed for immunocytochemistry after culture periods ranging from 1 to 24 days. Neural crest cells were identified by their morphology and any neural tube cells remaining in the cultures were ident ified by their epithelial morphology and immunocyto chemically. As early as 1 to 2 days in culture, approxi mately one third of the neural crest cells stained with m217c, a monoclonal antibody that appears to recognize the same antigen as rat neural antigen-I (RAN-1). A similar proportion of cells were immunoreactive in cultures stained with 192-IgG, a monoclonal antibody that recognizes the rat nerve growth factor receptor. The number of immunoreactive cells increased with time in culture. After 16 days in culture, nests of cells, many of which had a bipolar morphology, were present in the area previously occupied by neural crest cells. The cells in the nests were often associated with neurones and were immunoreactive for m217c, 192-IgG and antibody to S-100 protein and laminin, indicating that the cells were Schwann cells. At all culture periods examined, neural crest cells did not express glial fibrillary acidic protein.

These results demonstrate that cultured premigratory neural crest cells express early Schwann cell markers and that some of these cells differentiate into Schwann cells. These observations suggest that some neural crest cells in vivo may be committed to forming Schwann cells and will do so provided that they then proceed to encounter the correct environmental cues during embry onic development.

During vertebrate development neural crest cells mi grate from the dorsal aspect of the neural tube into many regions of the embryo where they differentiate into a wide variety of different cell types including most of the Schwann cells and neurones of the peripheral nervous system (see review by Le Douarin, 1982). Experiments in which avian and mammalian neural crest cells are cultured alone, in combination with other tissues, on different substrates or in medium differing in composition have demonstrated that the environment in which cells find themselves can determine the type of cell into which they differentiate. This has been shown in the control of skeletogenic (Bee & Thorogood, 1980; Smith & Thorogood, 1983), neurogenic (Le Douarin, 1982; Smith-Thomas et al. 1986), melanogenic (Derby & Newgreen, 1982) and odontoblastic (Lumsden, 1984) differentiation.

Schwann cells derived from the neural crest only fully differentiate in the presence of axons (Bunge et al. 1982, 1983). Holton & Weston (l982a,b) have demonstrated that Schwann cell precursors taken from early dorsal root ganglia will only mature when in contact with neurones. However, it is not known at what stage neural crest cells start to differentiate into Schwann cells and whether they need to be in contact with axons at the time.

The reason why the early events in Schwann cell differentiation have not been investigated to date is due to a lack of suitable markers. However, several anti bodies such as Ran-1 and m217c are now available that allow immature Schwann cells to be identified. Ran-1 recognizes a cell surface antigen found on most rat nervous system cell lines and on cultured Schwann cells (Brockes et al. 1977); m217c recognizes the same cell types as Ran-1, that is it behaves as if it is detecting RAN-1 (Peng et al. 1982; Fields & Dammerman, 1985). Schwann cells also express other markers which are not unique to them but allow one to assess their stage of development. For example, nerve growth factor recep tor is present on Schwann cells deprived of axonal contact in vivo and on Schwann cells in vitro (Zimmer man & Sutter, 1983; Taneuchi 1986; DiStefano & Johnson, 1988); S-100 is an intracellular calcium binding protein that is characteristic to glial cells which has been used by Holton & Weston and others as a marker for mature Schwann cells (Holton & Weston, l982a,b;,Brockes et al. 1979) The intermediate filament protein glial fibrillary acidic protein (GFAP) is also expressed by some non-myelin-forming Schwann cells (Jessen & Mirsky, 1984). Schwann cells in culture produce laminin at all stages of development, whether in contact with axons or not (Cornbrooks et al. 1983). Galactocerebro side (Gale), PO protein and myelin basic protein (MBP) are markers for mature Schwann cells and are only expressed by cells that are actively myelinating axons (Mirsky et al. 1980; Jessen et al. 1987).

The aim of these experiments was to establish when neural crest cells in vitro first start to express a Schwann cell phenotype. Most antibodies that recognize Schwann cells are specific to rat cells. Thus, in order to establish when neural crest cells first start to express a Schwann cell phenotype (and to enable us subsequently to coculture neural crest cells with other tissues and thus define the developmental cues responsible for Schwann cell differentiation), we developed a technique for culturing rat neural crest cells. This technique is based on the method for preparing avian trunk neural crest cultures (Cohen & Konisberg, 1975). Rat trunk neural crest cell cultures were then examined after different periods of culture time for the presence of a number of the antigenic markers for Schwann cells described above. Schwann cell cultures prepared from sciatic nerve and dorsal root ganglion cultures were used as positive controls.

Rat Schwann cell cultures

Sciatic nerves were removed from 1-to 4-day CFHB or Sprague-Dawley newborn rats and placed into Hanks balanced salt solution (HBSS). After removing associated muscle and connective tissue, the nerves were incubated with 0–1 % collagenase (Sigma no. 9407) in HBSS for 45 min at 37°C. A 0–l % trypsin solution (Sigma no. TO134) was then added to the collagenase solution and the nerves were incubated in this enzyme mixture for 20 min at 37°C. A solution of DNAse (20 μg ml-1) (Sigma no. D5025) was added and the preparation was spun at 100 rev min-1 for 3 min. The supernatant was discarded and the cell pellet was resuspended in 0·25 ml of triturating solution (1 mg BSA, Sigma no. A3912, 2mg DNase, Sigma no. D5025, 50mg trypsin inhibitor, Sigma no. T9003, per 100ml of HBSS). Cells were aspirated through a 23-gauge needle about four times to dissociate the tissue then filtered through 70 μm mesh Nilex gauze attached to the cut-off end of a glass syringe. The cells were counted in a haemocytometer and put in a small volume of medium onto glass coverslips coated with 0 · 01 % poly-o-lysine (Sigma) and collagen (Vitrogen; Collagen Corporation) at a density of 1 to 5 x105 cells per coverslip.

The following day, large numbers of bipolar Schwann cells and some fibroblasts were present. 24 to 48 h after plating, the fibroblasts were killed using antibody to Thy-1 and complement. Cells were subcultured using a light trypsinization when the cultures reached confluence. Secondary cultures of Schwann cells were fixed for antibody staining after different durations of culture. Throughout the culture period the culture medium comprised Dulbecco MEM + 10 % FCS (Flow)+ lOOi.u. penicillin+ 100 μgm1-1 streptomycin+ 10 μg ml-1 glial growth factor (Brockes et al. 1979) + 2 μM forskolin (Sobue et al. 1986).

Dorsal root ganglia cultures

Dorsal root ganglia were dissected from 1-day newborn rats and placed in HBSS. The bundles of sensory axons were cut off and the dorsal root ganglia were cut in half to enable axons and Schwann cells to grow out. The dorsal root ganglias were explanted onto collagen-coated coverslips in 35 mm culture dishes and allowed to attach overnight in a thin film of medium. The culture dishes were placed in a humidified chamber to prevent the medium from drying out. The medium comprised Dulbecco MEM + 10 % FCS + L glutamine +pen/strep+ 100 μg ml-1 nerve growth factor (Sigma). The following day, a small amount of medium was added to the culture.

Rat trunk neural crest cell cultures

Rats were paired in the early evening and checked for plugs the following day. The plug day was taken to be embryonic day 1. E12-timed pregnant rats were anaesthetized with chloroform and killed by dislocation of the neck. Uterine horns were immediately removed and placed into HBSS. Embryos were dissected from decidual swellings and sur rounding extraembryonic membranes. The embryos were transferred to complete medium and the number of somites was counted. 18-to 24-somite embryos were used for the experiments. The neural tubes plus surrounding tissues from the region including the most posterior 6 to 8 somites and some unsegmented tissue were removed using watchmakers’ forceps and electrolytically sharpened tungsten needles (Fig. 1). The medium comprised Eagle’s alpha MEM (Gibco) + 10 % FCS (Flow)+ 10 % of 50 % chick embryo extract (Flow)+ pen/strep. The pieces of neural tube plus surrounding tissue were washed twice with HBSS and incu bated with 0 · 1 % collagenase (Sigma) for 15 min on ice, followed by 10 min at 37°C. The neural tubes were then released from surrounding tissues by gentle pipetting and nudging off associated tissues such as somites, notochord and surface epithelium with tungsten needles. Embryos older than 26 somites are unsuitable, because it is not possible to detach the neural tube without damaging it.

Fig. 1.

Photograph of a 25-somite rat embryo showing the position from which the neural tube is removed for a neural crest culture. Bar, 320 μm.

Fig. 1.

Photograph of a 25-somite rat embryo showing the position from which the neural tube is removed for a neural crest culture. Bar, 320 μm.

The neural tubes were washed twice with medium and plated onto glass coverslips coated with 0 · 01 % polylysine (Sigma) and 25μ gm1-1 fibronectin (Sigma) and preincubated for 1 h in medium. Two to three neural tubes were plated onto each coverslip and left to attach for 1 to 2 h. After the neural tubes had attached 1·5 ml of medium was addded to each 35 mm culture dish. On the second or third day of culture, at which time a sizeable neural crest cell outgrowth was present, the neural tube was removed from the culture using a tungsten needle and the cultures were fed with either 1-to 2-day or 7-to 14-day neural-tube-conditioned medium (NTCM). The cultures were subsequently fed every third day with either fresh medium or NTCM.

Cultures were maintained for periods up to 4 weeks and were fixed for immunocytochemistry after different durations of culture time.

In order to study melanocyte differentiation, a few trunk neural crest cultures were prepared from embryos of hooded lister rats. These cultures were grown in medium containing 0·2 μg ml-1 alpha melanocyte stimulating hormone (alpha MSH) for up to 4 weeks with a complete medium change every 3 or 4 days. After 12 to 15 days, cultures were examined for the presence of melanocyte precursors using the DOPA reaction (see Ito & Takeuchi, 1984 for method).

Ventral neural tube cultures

The dorsal half was dissected off 18-to 24-somite neural tubes and the remaining half of the neural tube was cultured in the same conditions as neural crest cultures. Cultures were fixed after 2 days for immunocytochemistry.

Neural-tube-conditioned medium

Neural crest cultures were prepared as described above. After 1 to 2 days, the medium was removed from the culture and collected (1-to 2-day NTCM). Medium was also collected from neural crest cultures (from which the neural tube had been removed on day 2 or 3) on days 7 to 14 of the culture period (7-to 14-day NTCM).

Staining procedures

m217c

Cultures were washed twice with PBS/1 % BSA and then fixed for 15 min in 4 % paraformaldehyde pH 7·3. Cultures were then washed three times with PBS/1 % BSA, blocked with PBS/5 % goat serum for 30 min and then incubated in m217c supernatant diluted 1:80 in PBS/1 % goat serum for 1 h at room temperature (RT). Cultures were then washed three times in PBS/1 % BSA followed by biotinylated goat anti mouse (1:50) (Gibco BRL) for 30 min at RT. The cultures were then washed twice in PBS and incubated with FITC Streptavidin (1:100) (Serotec) for 20– 30 min at RT, washed three times in PBS and mounted in glycerol/PBS (1:1). The m217c supernatant was a gift from J. de Vellis.

192-IgG

This monoclonal antibody which recognizes the rat nerve growth factor receptor (Chandler et al. 1984) was a gift from Dr E. Johnson. The procedure was the same as for m217c. The antibody was used at a concentration of 2– 3 μg m1-1.

151-IgG

This monoclonal antibody generated against PC12 cell epider mal growth factor receptor (Parsons Chandler et al. 1985) was a gift from Dr E. Johnson and was used at a concentration of 3 μgm1-1 The procedure was the same as for m217c and 192-IgG.

S-100

Cultures were washed three times in PBS/1 % BSA, fixed in 4 % paraformaldehyde for 15 min at RT, washed three times in PBS/1 % BSA, blocked in PBS/0· 2 % Triton/5 % goat serum for 30min and then incubated in antibody to S-100 (1:200) (Dako) in PBSμriton/1 % goat serum for lh. The cultures were then washed twice in PBS and then incubated in FITC- or RITC-conjugated goat anti-rabbit (Tago 1: 50) for 30min at RT.

Laminin

The procedure was the same as for S-100 but omitted the permeabilization steps. Anti-laminin (Collaborative Re search) was used at 1:200.

P0 protein

Adult rat sciatic nerves were fixed in 5 % acetic acid in ethanol, embedded in Tissue Tek, sectioned and stained according to the procedure for S-100 and laminin. Dorsal root ganglia and neural crest cultures were fixed in 5 % acetic acid in ethanol and then stained according to the procedure for S-100 and laminin. PO antiserum (gift from J. Brockes) was used at 1:200.

GalC

The procedure was the same as for m217c. GalC supernatant was used 1:20.

Anti-neurofilament protein

(i) Antiserum R-39

Cultures were fixed in 4 % paraformal dehyde for 15 min, washed three times in PBS/BSA, blocked and permeabilized in PBS/BSAμriton for 30 min, and then incubated in R-39 (1: 200) for 1 h. They were washed three times and then incubated in FITC goat anti-rabbit (TAGO) (1:40) for 30min. Antiserum R-39 was a gift from D. Dahl.

(ii) Monoclonal no. 32 (Sternberger-Meyer)

This recognizes all neurofilament subclasses. Fixation and permeabilization were as for R-39. The primary was used at 1: 2000. The secondary was biotinylated goat anti-mouse (BRL) followed by FITC streptavidin.

(iii) Monoclonal 3AJO

This recognizes 68×103Mr neurofila ments. The fixation and staining procedure were the same as for monoclonal no. 32. The primary antibody was used at 1: 20 and was a gift from T. Jessell.

GFAP

The procedure was the same as for anti-neurofilament R-39. The anti-GFAP antiserum (Dako) was used at 1: 100 or Boehringer GFAP monoclonal at 1: 20. The secondary for the polyclonal was Caltag FITC goat anti-rabbit at 1: 50 and the secondary for the monoclonal was biotinylated goat anti mouse human adsorbed (Caltag (1:50) followed by FITC streptavidin (Serotec 1:100).

F16.4.4

F16.4.4 is a monoclonal antibody that recognizes Class 1 antigen of the rat major histocompatibility complex (Hart & Fabre, 1981).

Sections of adult rat liver (fixed overnight in 4 % parafor maldehyde) were incubated with F16.4.4 overnight at 4°C. The staining procedure was subsequently the same as for m217c. Neural crest cultures were fixed and stained according to the procedure used for m217c. The antibody was a gift from J. Butcher.

Rat Schwann cell culturess (prepared from sciatic nerve based on the method of Brockes et al. 1979)

Young sciatic nerve cultures contained small bipolar Schwann cells, and large flat fibroblasts. In time, many of the Schwann cells attained a more flattened mor phology. Both m217c and 192-IgG stained the bipolar cells strongly, with a typical speckled cell surface pattern (Fig. 2A and B). In the older cultures, many cells with flattened cell bodies stained with these antibodies as well as those cells that still had a bipolar morphology. The cells in older cultures (even after 7 weeks culture) stained as brightly as cells in younger cultures. In cultures stained for laminin or S-100 pro tein, a similar population of cells was stained (Fig. 2C). With all of these antibodies, we often observed nests of stained cells surrounded by unstained fibroblasts. Large flattened cells with typical fibroblastic morphology stained with anti-Thy 1, but not any of the other antibodies. Our anti-GFAP antibodies did not stain any cells in our cultures (in agreement with observations of Brockes et al. 1979), although they did stain cultured neonatal astrocytes.

Fig. 2.

(A) Schwann cells cultured for 18 days and then stained with mAb m217c. Schwann cells are immunoreactive and the fibroblasts are unstained. (B) Schwann cells cultured for 11 days and then stained with mAb 192-IgG. Schwann cells are immunoreactive and the fibroblasts are unstained. (C) Schwann cells cultured for 18 days and then stained with antibody to S-100 protein. Schwann cells are immunoreactive and the fibroblasts are unstained. Bars A-C,40 μm.

Fig. 2.

(A) Schwann cells cultured for 18 days and then stained with mAb m217c. Schwann cells are immunoreactive and the fibroblasts are unstained. (B) Schwann cells cultured for 11 days and then stained with mAb 192-IgG. Schwann cells are immunoreactive and the fibroblasts are unstained. (C) Schwann cells cultured for 18 days and then stained with antibody to S-100 protein. Schwann cells are immunoreactive and the fibroblasts are unstained. Bars A-C,40 μm.

Rat dorsal root ganglion cultures

We examined 2-day-old cultures which contained many axons and Schwann cells and a few fibroblasts. m217c stained Schwann cells, recognized by their bipolar morphology, brightly and there was staining on most axons (Fig. 3A). When cultures were stained with antibodies to nerve growth factor receptor, axons and adjacent Schwann cells were brightly stained (Fig. 3B). Fibroblasts were unstained.

Fig. 3.

(A) Neonatal dorsal root ganglion cultured for 2 days and then stained with m217c. Axons (large arrows) and Schwann cells (small arrows) are stained. (B) Neonatal dorsal root ganglion cultured for 2 days and then stained with 192-lgG. Axons (large arrows) and Schwann cells (small arrows) are stained. Bars, 40 μm.

Fig. 3.

(A) Neonatal dorsal root ganglion cultured for 2 days and then stained with m217c. Axons (large arrows) and Schwann cells (small arrows) are stained. (B) Neonatal dorsal root ganglion cultured for 2 days and then stained with 192-lgG. Axons (large arrows) and Schwann cells (small arrows) are stained. Bars, 40 μm.

Rat trunk neural crest cultures

Phase-contrast and time-lapse observations

2–4 h after explantation of neural tubes, a number of cell processes could be observed extending from the neural tube. By 24 h the explant was surrounded by cells, radiating out about 400 μm. The cells migrating from the tubes were mainly of two clearly distinct morphologies. (1) Cells of mesenchymal appearance with spaces in between them, predominantly seen migrating from the dorsal aspect of the neural tube, and morphologically very similar to cells in avian neural crest cultures (Fig. 4A). These cells were not seen in cultures of ventral neural tube. We assume these cells to be neural crest cells. (2) Cells of flattened epithelioid appearance, confluent and migrating largely from the ventral aspect of the neural tube, and present also in cultures derived from ventral neural tube. We assume these to be neural-tube-derived cells. The proportions of these cells varied from culture to culture; 60–70 % of cultures contained cells of mesenchymal morphology, presumed to be neural-crest cells.

Fig. 4.

(A) Neural crest outgrowth from a neural tube cultured for 1 day. nc, neural crest; nt, neural tube. (B) Neural crest cultured for 7 days. There are some patches of closely associated cells. (C) DOPA-positive cells (identified by dark reaction product) on epithelial sheet in a 12-day neural crest culture. Bars A-C, 80 μm.

Fig. 4.

(A) Neural crest outgrowth from a neural tube cultured for 1 day. nc, neural crest; nt, neural tube. (B) Neural crest cultured for 7 days. There are some patches of closely associated cells. (C) DOPA-positive cells (identified by dark reaction product) on epithelial sheet in a 12-day neural crest culture. Bars A-C, 80 μm.

Cultures were observed intermittently for up to 4 weeks, and by continuous time-lapse videorecording for up to 3 days. Videorecording showed frequent cell divisions in the neural crest cell population, and con siderable mobility of cells at the margin of the culture. Cells continued to separate from the neural tube, generally migrating a small distance before settling down. Some of these late migrating cells may have been neuronal, since from 3 days on we started to see bipolar cells of neuronal morphology mixed with the neural crest cells, mostly near the neural tube explant. Also, at around 3 days we usually saw some degree of cell lysis, mostly in the neural tube cell population. The neural tube was removed on the 2nd or 3rd day of culture and the amount of Iysis was much reduced by addition of conditioned medium to the culture at this time (see later). Most cultures continued to grow after this, sometimes increasing in size until they covered a 22 mm coverslip. Most of the cells in this outgrowth were mesenchymal in morphology; very similar to those seen in avian neural crest cultures (Fig. 4B). At early stages, these cells were rather homogeneous in shape and size, but later there was usually considerable variation in cell size and density; older cultures had areas of large loosely associated cells, surrounding smaller closely packed regions. In the absence of a marker that specifically recognizes rat neural crest cells, it is not possible to judge the proportion of neural-tube-derived cells in these areas of predominantly mesenchymal cells. There was always an area of neural-tube-derived cells of epithelial morphology; many of these cells could eventually be stained for CNS immunohistochemical markers (see later).

Melanocytes

In order to detect melanocytes, neural crest cells from hooded lister rats were cultured in medium containing 0-2 μg m1-1 alpha MSH. Under these culture con ditions, even after 4 weeks culture, fully differentiated melanocytes never appeared. However, if the DOPA reaction was performed on the cultures to detect melanocyte precursors, a few positive cells were ob served in all of the cultures examined (after a culture period of 12 to 17 days). These melanocyte precursors were detected by a dark reaction product, were always in dose proximity to each other and were always on the epithelial sheet and not observed elsewhere in the culture (as reported previously by Ito & Takeuchi, 1984 in mouse neural crest cultures) (Fig. 4C). The absence of melanocytes in rat neural crest cultures contrasts with avian neural crest cultures in which melanocytes differentiate readily. In avian neural crest cultures, melanocyte differentiation is favoured by the presence of chick embryo extract and certain batches of fetal calf serum (Derby & Newgreen, 1982) both of which were present in our culture medium. Thus, melanogenesis in rat neural crest in vitro must require an environmental stimulus that is not required by cultured avian neural crest cells. To our knowledge, fully differentiated melanocytes have only been detected in mouse neural crest cultures cultured in the presence of alpha MSH (Ito & Takeuchi, 1984).

Removal of the neural tube and effects of neural-tube conditioned medium

Experiments in which the neural tube was removed at various stages of the culture period and in which the medium was replaced with either fresh, 1-to 2-day or 7-to 14-day NTCM after removal of the neural tube, demonstrated the following. If the neural tube is removed before day 2, and the medium is replaced with fresh medium, most cells in most cultures lyse. If the neural tube is removed on day 3 and the medium is completely replaced with fresh medium, many cells in a large proportion of the cultures lyse. However, if, after removal of the neural tube at days 2 or 3, the medium is replaced with either medium conditioned by neural crest cultures over the first 2 days of the culture period, or by medium from days 7 to 14 of the culture period, cell lysis is rarely observed, and the cells continue to multiply. These observations suggest that the presence of diffusible factors produced by the neural tube and older neural crest cultures are important for neural crest cell survival in the early stages of culture (days 1 to 7).

Addition of 5 % whole (unfractionated) rat embryo extract (prepared from E17 to E19 embryos based on the procedure for chick embryo extract (Howard & Bronner-Fraser, 1985)) to the culture medium did not reduce the incidence or amount of cell lysis or increase the proliferation of the neural crest cells.

Immunostaining for Schwann cell markers (antigen recognized by m217c, nerve growth factor receptor recognized by mAb 192-IgG S-100 protein, laminin, GFAP, GalCal PO protein)

(a) Neural crest cells

Essentially identical results were observed with m217c and 192-IgG at all ages of culture. The results are summarized in Table l. As early as .1 to 2 days after explantation of the neural tubes, approximately one third of the neural crest cells were immunoreactive; the staining was cell surface and heterogeneous in intensity (Fig. 5A and B). The staining was less bright than after longer periods of culture, the immunoreactive cells were intermixed with unstained cells and there was no difference in the morphology of stained and unstained cells. Large numbers of brightly stained neural crest cells were adjacent to or near the predominantly unstained epithelial sheet. By day 3 one third to one half of the neural crest cells were immunoreactive for m217c.

7 to 8 days after explantation a similar proportion of neural crest cells (identified by their mesenchymal morphology) as in the younger cultures were immuno reactive for m217c and 192-IgG (Fig. 5C and D). The staining was heterogeneous and there was no difference in morphology between stained and unstained cells. Large numbers of neural crest cells near the predomi nantly unstained epithelial sheet were brightly stained.

9 to 15 days after explantation, large numbers of neural crest cells stained brightly for m217c. As ob served previously, if any epithelial sheet was present in the culture, there was a very distinct demarcation between the brightly stained neural crest cells next to the predominantly unstained neural tube cells of the epithelial sheet. In long-term cultures (16 to 24 days) stained with m217c or 192-IgG and antiserum R-39 (anti-neurofilament), the majority of stained cells were in nests which were observed throughout the culture. Many of the cells stained with m217c or 192-IgG had a bipolar morphology (Fig. 6A) but some had the typical mesenchymal morphology seen in younger cultures. These cells did not express neurofilament protein but round neuronal cell bodies and some fibres (identified by antiserum R-39) were present in most of these nests (Fig. 6B). These nests are therefore strikingly reminis cent of dorsal root ganglia, being roughly spherical structures, containing both neurones and Schwann cells. There was no obvious difference between the morphology or staining of the m217c and 192-IgG immunoreactive cells in nests with or without mature neurones.

Fig. 5.

(A) Neural crest cultured for 2 days and stained with m217c. A subpopulation of the cells (many of which have rounded up during fixation) are immunoreactive. (B) Neural crest cultured for 2 days and stained with 192-IgG. A subpopulation of the cells are immunoreactive. (C) Neural crest cultured for 7 days and stained with m217c. A subpopulation of cells are stained. The staining is heterogeneous and stained cells have a similar morphology to unstained cells. (D) Neural crest cultured for 8 days and stained with 192-IgG. A subpopulation of cells are stained. The staining is heterogeneous. (E) Immunoreactive cells on epithelial sheet in a 7-day neural crest culture stained with m217c. (F) Immunoreactive cells on epithelial sheet in an 8-day neural crest culture stained with 192-IgG. Bars A-F, 40 μm.

Fig. 5.

(A) Neural crest cultured for 2 days and stained with m217c. A subpopulation of the cells (many of which have rounded up during fixation) are immunoreactive. (B) Neural crest cultured for 2 days and stained with 192-IgG. A subpopulation of the cells are immunoreactive. (C) Neural crest cultured for 7 days and stained with m217c. A subpopulation of cells are stained. The staining is heterogeneous and stained cells have a similar morphology to unstained cells. (D) Neural crest cultured for 8 days and stained with 192-IgG. A subpopulation of cells are stained. The staining is heterogeneous. (E) Immunoreactive cells on epithelial sheet in a 7-day neural crest culture stained with m217c. (F) Immunoreactive cells on epithelial sheet in an 8-day neural crest culture stained with 192-IgG. Bars A-F, 40 μm.

Fig. 6.

(A) Nest of 192-IgG immunoreactive cells in a 16-day neural crest culture. Large numbers of cells have a bipolar morphology (arrows). (B) Neurofilament protein immunoreactive neurones in the centre of a nest of 192-IgG immunoreactive cells. (C) Nest of S-100 immunoreactive cells in a 24-day neural crest culture. Large numbers of cells have a bipolar morphology (arrows). Bars A-C, 40 μm.

Fig. 6.

(A) Nest of 192-IgG immunoreactive cells in a 16-day neural crest culture. Large numbers of cells have a bipolar morphology (arrows). (B) Neurofilament protein immunoreactive neurones in the centre of a nest of 192-IgG immunoreactive cells. (C) Nest of S-100 immunoreactive cells in a 24-day neural crest culture. Large numbers of cells have a bipolar morphology (arrows). Bars A-C, 40 μm.

Cultures were examined for S-100 protein and GFAP immunoreactivity at 3 days, 7 to 8 days, 9 to 12 days and 16 to 24 days. Cultures were also examined for laminin, GalC and PO protein. The results obtained with S-100 protein and laminin are summarized in Table 1. GF AP, GalC and PO protein were not observed in the early stages of culture or in the ‘nests’ of cells in 16-to 24-day cultures and the results are therefore not included in the table.

Table 1.

Schwann cell markers in cultured rat neural crest cells

Schwann cell markers in cultured rat neural crest cells
Schwann cell markers in cultured rat neural crest cells

In the early stages of culture neither laminin, GFAP nor S-100 protein were observed in the neural crest cells. However, 16 days after explantation large num bers of nests of cells were observed in the cultures; the cells in these nests contained S-100 protein, were associated with small amounts of laminin and often had a bipolar morphology (Fig. 6C). In double-labelling experiments, neurones, identified with antibodies to neurofilament protein, were observed to be present in most of the nests of S-100-positive cells. These nests of cells did not contain GalC or PO protein. In addition to these cells, bipolar cells containing S-100 protein were also observed on the epithelial sheet.

(b) Epithelial sheet

2 days after explantation, the majority of neural tube cells were unstained with m217c. Many axons and some cells on the epithelial sheet were immunoreactive for 192-IgG. 7 to 8 days after explantation, the neural tube cells which constituted the epithelial sheet were pre dominantly unstained with m217c and 192-IgG. How ever, on the epithelial sheet which tended to constitute the central part of the culture, a few cells were stained. These cells tended to have flattened cell bodies and often had long cell processes (Fig. 5E). A very similar pattern of staining was observed with 192-IgG (Fig. 5F). 9 to 12 days after explantation, in agreement with observations on younger cultures, some cells on the epithelial sheet stained brightly, both their cell bodies and long cell processes. These cells varied in morphology; some were bipolar, others had very well spread cell bodies. From 12 days onwards, many m217c immunoreactive cells were present on the epithelial sheet.

We were interested in determining if the m217c and 192-IgG immunoreactive cells observed on the epi thelial sheet were neural crest or neural tube derived. In order to do so, ventral neural tube cultures were examined for m217c and 192-IgG immunoreactive cells. 2 days after explantation an outgrowth of neural tube cells was present and most cultures had many motor axons associated with them. A mesenchymal outgrowth was never observed. When the cultures were stained with m217c all of them had similar staining patterns. Large quantities of stained whispy fibres were present around the explant and small bundles of small axons had speckled cell surface staining (Fig. 7A). There were also a few stained cells in the predominantly unstained cellular outgrowth. The stained cells were alone or in clumps: some of the cells were similar in morphology to those seen on top of the epithelial sheet i.e. polygonal, whilst others were longer and flatter (Fig. 7B). 2-day cultures were also stained with antibody to anti-NGF receptor. Large numbers of immunoreactive fibres were observed on the epithelial sheet (we believe these to be very fine axons, since we saw a similar staining pattern with anti-neurofilament protein; see later). Some stained cells, flattened or polygonal, were associated with the immunoreactive fibres. There were also some stained cells in the cellular outgrowth. Bundles of motor axons stained brightly. S-100 or GFAP immuno reactivity was never observed on any epithelial sheet that remained in 3-day cultures. 7 to 8 days after explantation, a few S-100 immunoreactive cells were observed on the epithelial sheet. 9 to 21 days after explantation, some laminin immunoreactivity, large numbers of GF AP-positive and S-100-positive cells with astrocytic morphology were observed on the epithelial sheet.

Fig. 7.

(A) Ventral neural tube cultured for 2 days and stained with m217c. Axons and a few associated cells are immunoreactive. (B) Ventral neural tube cultured for 2 days and stained with m217c. Patches of cells in the culture are immunoreactive. Bars, 40μ m.

Fig. 7.

(A) Ventral neural tube cultured for 2 days and stained with m217c. Axons and a few associated cells are immunoreactive. (B) Ventral neural tube cultured for 2 days and stained with m217c. Patches of cells in the culture are immunoreactive. Bars, 40μ m.

Neuronal markers (neurofilament protein)

(i) Neural crest

2 days after explantation, neural crest cells did not express neurofilament protein recognized by either antiserum R-39 or Sternberger-Meyer monoclonal anti body no. 32 (mAb 32). 7 to 8 days after explantation in cultures stained with antiserum R-39 many patches of closely associated cells had immunoreactive filaments in their cell bodies. These cells did not have a neuronal morphology (Fig. 8). The same result was observed in all cultures stained. These patches of stained cells were surrounded by unstained cells. The stained cells varied in morphology. Some were very large and well spread; others were smaller. However, nearer the epithelial sheet were patches of immunoreactive cells which had a much more neuronal morphology. The bipolar neur onal-type cells were also immunoreactive. When a culture was stained with mAb 32, an antibody that recognizes nonphosphorylated (early) neurofilament protein, although immunoreactive cells with neuronal cell bodies and processes were observed near the epithelial sheet, the immunoreactive fibres in cells with a non-neuronal morphology were not observed. Mature neurones (identified with antiserum R-39, mAb 32 and mAb 3A10) were observed in most nests of 192-IgG, m217c, and S-100 immunoreactive cells in long-term (16 days or longer) cultures (Fig. 6B).

Fig. 8.

Neural crest cultured for 7 days and stained with antiserum R-39. Patches of closely associated cells have immunoreactive filaments in their cell bodies but do not have a neuronal morphology. Bar, 40 μm.

Fig. 8.

Neural crest cultured for 7 days and stained with antiserum R-39. Patches of closely associated cells have immunoreactive filaments in their cell bodies but do not have a neuronal morphology. Bar, 40 μm.

(b) Epithelial sheet

2 days after explantation large numbers of immuno-reactive fibres were observed in cultures stained with mAb 32 and antiserum R-39. 7 to 8 days after explan tation, when cultures were stained with antiserum R-39 a few axons were stained. Associated with the centrally remaining neural tube cells were a few immunoreactive cells. At this time, in a culture stained with mAb 32, large numbers of immunoreactive fibres were observed on the epithelial sheet.

Controls

No immunoreactivity was observed in cultures either when the primary antibody was omitted from the staining procedure or when it was replaced with rabbit serum. Other monoclonal antibodies from the same immunoglobulin class as m217c and 192-IgG gave different staining patterns to these antibodies. These control antibodies were 151-IgG, a mAb that recognizes the rat EGF receptor (which was produced by the same fusion that produces 192-IgG) (Parsons Chandler et al. 1985) and anti-GFAP (Boehringer). The third control class matched mAb was F16.4.4, an antibody to Class 1 antigen of the rat major histocompatibility complex (Hart & Fabre, 1981; Jeff Butcher, personal communi cation). This antibody recognized many cells in sections of adult rat liver but failed to recognize any cells in rat neural crest cultured for 3 days. The binding pattern of antibodies to S-100 protein and laminin antisera were also unique to each of these antisera. In addition, antiserum to PO protein stained sections of adult rat sciatic nerve but failed to stain 3- or 4-week neural crest cultures.

The aim of these experiments was to establish when neural crest cells first start to express a Schwann cell phenotype. A number of antibodies are now available which make it possible to identify Schwann cells at various stages of differentiation. Many of these anti bodies are specific to rat cells. We have therefore developed a technique for culturing rat neural crest cells which can be cultured from 18-to 24-somite (E12) rat embryo neural tube. There are four lines of evidence that indicate that our cultures contain neural crest cells. First, the cells are morphologically and behaviourly very similar to those cultured from chick or quail neural tube. Second, these cells are not observed in cultures derived from ventral neural tube (i.e. neural tube from which the dorsal half which contains the neural crest has been removed). Third, as do avian neural crest cells, the cells in our cultures grew very well on a fibronectin substrate and, fourth, subpopulations of them ex pressed Schwann cell, neuronal and premelanocyte phenotypes, all of which cell types are known to be neural crest derivatives. We do not, however, presently have an immunological marker for rat neural crest cells similar to the HNK-1 antigen, which is found on avian neural crest cells.

A subpopulation of our cultured rat premigratory neural crest cells express the antigen recognized by m217c and a similar pattern of cells express NGF receptor recognized bymAB 192-IgG. NGF receptor has been demonstrated previously on avian neural crest cells using a variety of indirect techniques (End et al. 1983; Bernd, 1985; Greiner et al. 1986).

Immunostaining of cultured neural tube and dorsal root ganglion shows that both m217c and 192-IgG bind to neurofilament-staining sensory neurones and motor axons as well as Schwann cells. Neural crest cells immunoreactive to these two antibodies may therefore either be bipotential precursors which could form either Schwann cells or sensory neurones depending on the environmental cues encountered during development or they may already be committed to forming one of these cell types. These antigens are observed on neural crest cells as early as 1 and 2 days in culture. Thus, even before neural crest cells migrate from the neural tube or shortly after the commencement of migration some neural crest cells may already be partially or totally committed to forming Schwann cells. These cells could then proceed to form Schwann cells and express other glial markers characteristic of Schwann cells (e.g. S-100) providing that they encounter the correct devel opmental cues in the embryo. We have also observed the antigens recognized by m217c and 192-IgG on migrating neural crest cells in sections of gelatin embedded E12 rats, demonstrating that these antigens are expressed on neural crest cells in vivo.

Examination of neural crest cultured for 16 to 24 days and then stained with m217c or 192-IgG revealed that many nests of immunoreactive cells, many of which had a bipolar morphology, were present throughout the culture and these cells also stained for S-100 protein and laminin. Because these cells contain S-100 protein, are associated with laminin, often have a bipolar mor phology and stain with m217c and 192-IgG (neither of which stains CNS glia) we conclude that these cells are Schwann cells. The observation that these Schwann cells differentiate in the areas previously occupied by neural crest cells, many of which were m217c and 192-IgG positive, provides strong evidence that at least some m217c-, 192-IgG-positive neural crest cells are Schwann cell precursors. PO protein and GalC were not observed associated with the Schwann cells, indicating they had not been induced to myelinate. We observed neurones associated with most of these nests of cells. Many of these neurones appeared to have no connec tion with, and to be at some distance from, the epithelial sheet, so it is possible that they may have differentiated from neural crest cells. The fact that most S-100-positive cells colocalized with neurones strongly suggests that neurones are involved in the Schwann cell differentiation although the fact that some nests of Schwann cells did not contain mature neurones suggests that other factors present in the culture may be involved or that immature neurones may be sufficient. These results do, however, clearly demonstrate that neural crest cells do not require an interaction with tissues on the migratory pathway such as somites or notochord in order to differentiate into Schwann cells. We suggest that the undifferentiated m217c-, 192-IgG-positive neural crest cells did not receive appropriate develop mental cues (e.g. contact with neurones) to enable them to differentiate.

The significance of the seemingly identical staining patterns obtained with m217c and 192-IgG is unknown although these antigens appear to be present or absent on similar cell types and cell lines. For example, in addition to Schwann cells, both m217c and 192-IgG bind to cultured rat dorsal root ganglia neurones and CNS axons, neither m217c (our observations) nor 192-IgG (DiStefano & Johnson, 1988) binds to cultured rat astrocytes and both m217c and 192-IgG bind to PC12 cells (Fields & Dammerman, 1985; DiStefano & John son, 1988). m217c has also been reported to bind to some tumour cell lines (a rat oligodendroglioma and a human glioma cell line) but not to other cell lines (a nontransformed glial cell line and a nontransformed rat fibroblast line). The binding of 192-IgG to these cells has not been tested to our knowledge. The identical staining patterns seen with m217c and 192-IgG and double-labelling experiments with both of these anti bodies suggest that neural crest cells that express the antigen recognized by m217c also express the NGF receptor. However, it has not been possible to demon strate unequivocally that there is no crossover of the secondary antibodies when cells are stained with the two monoclonals and therefore this question remains unresolved.

The significance of the presence of NGF receptor on Schwann cells and neural crest cells is not known. Studies of NGF binding to cultured Schwann cells have previously shown that the NGF-binding sites are of the low affinity type (DiStefano & Johnson, 1988) for which a clear function remains to be established. Addition of NGF to cultured Schwann cells has no effect on survival or morphological differentiation (Zimmerman & Sut ter, 1983; Taneuchi et al. 1986) and NGF does not have any effect on the growth or survival of neural crest cells (Greiner et al. 1986; End et al. 1983). It has been postulated that the presence of NGF receptors on Schwann cells is involved in interactions between axons and Schwann cells during development and regener ation (Taneuchi et al. 1986 and DiStefano & Johnson, 1988).

Some cells that stained with m217c and 192-IgG migrated out from ventral neural tube, and are there fore probably not neural crest derived. It is possible that these cells are satellite cells for ventral root axons. Small stellate HNK-1 immunoreactive cells have been identified in avian ventral neural tube cultures (Loring & Erickson, 1987) and, in chick-quail chimaeras, quail cells emigrate from a grafted crest ablated neural tube along with motor axons (Lunn et al. 1987). The fact that cells in ventral neural tube cultures stain with m217c and 192-IgG, both of which antigens are not present on CNS glia, suggests that these cells resemble PNS rather than CNS glia. This result adds to the accumulating evidence that not all PNS glia are neural crest derived. It has also recently been demonstrated that 5-HT-containing cells migrate from ventral neural tube in vitro, which suggests that the neural tube may contribute to neurones as well as glia of the PNS (Loring et al. 1988).

We have developed a technique for culturing rat neural crest cells and have demonstrated that a sub population of these cells express Schwann cell antigens. Under these culture conditions, some cells eventually differentiate into morphologically and immunocyto chemically recognizable Schwann cells. We now aim to define the developmental cues that are responsible for the differentiation of Schwann cells from neural crest.

This work is supported by the Medical Research Council (UK). We thank Peter Starling for help with the photography and Ors Roger Keynes and Ron Meyer for their helpful comments on the manuscript.

Bee
,
J.
&
Thorogood
,
P. V.
(
1980
).
The role of tissue interactions in the skeletogenic differentiation of avian neural crest cells
.
Devi Biol
78
,
47
62
.
Bernd
,
P.
(
1985
).
Appearance of nerve growth factor receptor on cultured neural crest cells
.
Devi Biol
112
,
145
156
.
Brockes
,
J.P.
,
Fields
,
K. L.
&
Raff
,
M. C.
(
1977
).
A surface antigenic marker for rat schwann cells
.
Nature, Land
.
266
,
364
366
.
Brockes
,
J.P.
,
Fields
,
K. L.
&
Raff
,
M. C.
(
1979
).
Studies on cultured rat Schwann cells. 1. Establishment of purified populations from cultures of rat peripheral nerve
.
Brain Res
165
,
105
118
.
Bunge
,
M. B.
,
Bunge
,
R. P.
,
Carey
,
D. J.
,
Cornbrooks
,
C. J.
,
Eldridge
,
C. F.
,
Williams
,
A. K.
&
Wood
,
P. M.
(
1983
).
Developing and regenerating nervous systems
.
In Tarbox Parkinsons Disease Symposium
(ed.
P. W.
Coates
,
R. R.
Markwald
&
A. D.
Kenny
), pp.
71
105
.
New York
:
Alan R. Liss
.
Bunge
,
M. B.
,
Williams
,
A. K.
&
Wood
,
P. M.
(
1982
).
Neuron-Schwann cell interaction in basal lamina formation
.
Devi Biol
92
,
449
460
.
Chandler
,
C. E.
,
Parsons
,
L. M.
,
Hosang
,
M.
&
Shooter
,
E. M.
(
1984
).
A monoclonal antibody modulates the interaction of nerve growth factor with PC12 cells
.
J. biol. Chem
.
259
,
6882
6889
.
Cohen
,
A. M.
&
Konisberg
,
I. R.
(
1975
).
A clonal approach to the problem of neural crest determination
.
Devi Biol
46
,
262
280
.
Cornbrooks
,
C. J.
,
Cary
,
D. J.
,
Mcdonald
,
J. A.
,
Timpl
,
R.
&
Bunge
,
R. P.
(
1983
).
In vivo and in vitro observations on laminin production by schwann cells
.
Proc. natn. Acad. Sci. U.S.A
.
80
,
3850
3854
.
Derby
,
M. A.
&
Newgreen
,
D. F.
(
1982
).
Differentiation of avian neural crest cells in vitro: Absence of a developmental bias toward melanogenesis
.
Cell Tissue Res
225
,
365
378
.
Distefano
,
P. S.
&
Johnson Jr
,
E.
M
. (
1988
).
Nerve growth factor receptors on cultured rat Schwann cells
.
J. Neurosci
8
,
231
241
.
End
,
D.
,
Pevzner
,
L.
,
Lloyd
,
A.
&
Guroff
,
G.
(
1983
).
Identification of nerve growth factor receptors in primary cultures of chick neural crest cells
.
Devi Brain Res
7
,
131
136
.
Fields
,
K. L.
&
Dammerman
,
M.
(
1985
).
A monoclonal antibody equivalent to anti-rat neural antigen-! as a marker for Schwann cells
.
Neuroscience
15
,
877
885
.
Greiner
,
C. A. M.
,
Lloyd
,
A. T.
&
Guroff
,
G.
(
1986
).
Ontogeny of the nerve growth factor receptor in primary cultures of neural crest cells
.
Devl Brain Res
26
,
145
150
.
Hart
,
D. N. J.
&
Fabre
,
J. W.
(
1981
).
Major Histocompatibility Complex Antigens in rat kidney, ureter and bladder
.
Transplantation
31
,
318
325
.
Holton
,
B.
&
Weston
,
J. A.
(
1982a
).
Analysis of glial cell differentiation in peripheral nervous tissue. 1. S-100 accumulation in quail embryo spinal ganglion cultures
.
Devl Biol
89
,
64
71
.
Holton
,
B.
&
Weston
,
J. A.
(
1982b
).
Analysis of glial cell differentiation in peripheral nervous tissue. 2. Neurons promote SlOO synthesis by purified glial precursor cell populations
.
Devi Biol
89
,
72
81
.
Howard
,
M. J.
&
Bronner-Fraser
,
M.
(
1985
).
The influence of neural tube derived factors on the differentiation of neural crest cells in vitro
.
J. Neurosci
5
,
3302
3309
.
Ito
,
K.
&
Taneuchi
,
T.
(
1984
).
The differentiation in vitro of the nerual crest cells of the mouse embryo
.
J. Embryol. exp. Morph
.
84
,
49
62
.
Jessen
,
K. R.
,
Mirsky
,
R.
&
Morgan
,
L.
(
1987
).
Axonal signals regulate the differentiation of non-myelin forming Schwann cells: An immunohistochemical study of galactocerebroside in transected and regenerating nerves
.
J. Neurosci
.
7
,
3362
3369
.
Jesson
,
K. R.
&
Mirsky
,
R.
(
1984
).
Nonmyelin-forming Schwann cells coexpress surface proteins and intermediate filaments not found in myelin-forming cells: a study of Ran-2, A5E3 antigen and glial fibrillary acidic protein
.
J. Neurocytology
13
,
923
984
.
Le Douarin
,
N. M.
(
1982
).
The Neural Crest
.
Cambridge University Press
.
Loring
,
J. F.
,
Barker
,
D. L.
&
Erickson
,
C. A.
(
1988
).
Migration and differentiation of neural crest and ventral neural tube cells in vitro: Implications for in vitro and in vivo studies of the neural crest
.
J. Neurosci
.
8
,
1001
1015
.
Loring
,
J. F.
&
Erickson
,
C. A.
(
1987
).
Neural crest cell migratory pathways in the trunk of the chick embryo
.
Devi Biol
121
,
220
236
.
Lumsden
,
A.
(
1984
).
In Tooth Morphogenesis and Differentiation. Colloque lnserm
25
(ed
A. B.
Belcourt
&
J. V.
Ruch
), pp.
29
40
.
Paris
:
Inserm
.
Lunn
,
E. R.
,
Scourfield
,
J.
,
Keynes
,
R. J.
&
Stern
,
C. D.
(
1987
).
The neural tube origin of ventral root sheath cell in the chick embryo
.
Development
101
,
247
254
.
Mirsky
,
R.
,
Winter
,
J.
,
Abeny
,
E. R.
,
Pruss
,
R. M.
,
Gavrilovic
,
J.
&
Raff
,
M. C.
(
1980
).
Myelin specific proteins and glycolipids in rat schwann cells and oligodendrocytes in culture
.
J. Cell Biol
.
84
,
483
494
.
Parsons Chandler
,
L.
,
Chandler
,
C. E.
,
Hosang
,
M.
&
Shooter
,
E. M.
(
1985
).
A monoclonal antibody which inhibits epidermal growth factor binding has opposite effects on the biological action of epidermal growth factor in different cells
.
J. biol Chem
.
260
,
3360
3367
.
Peng
,
W. W.
,
Bressler
,
J.P.
,
Tiffany-Castigolioni
,
E.
&
De Vellis
,
X.
(
1982
).
Development of a monoclonal antibody against a tumour-associated antigen
.
Science
215
,
1102
1104
.
Smith
,
L. C.
&
Thorogood
,
P. V.
(
1983
).
Transfilter studies on the mechanism of epithelio-mesenchymal interaction leading to chondrogenic differentiation of neural crest cells
.
J. Embryol. exp. Morph
.
75
,
165
188
.
Smith-Thomas
,
L. C.
,
Davis
,
J.P.
&
Epstein
,
M. L.
(
1986
).
The gut supports neurogenic differentiation of periocular mesenchyme, a chondrogenic neural crest-derived cell population
.
Devi Biol
ll5
,
293
300
.
Sobue
,
G.
,
Shuman
,
S.
&
Pleasure
,
D.
(
1986
).
Schwann cell responses to cyclic AMP. Proliferation, change in shape and appearance of surface galactocerebroside
.
Brain Res
.
362
,
23
32
.
Taneuchi
,
M.
,
Brent Clark
,
H.
&
Johnson
Jr,
E. M.
(
1986
).
Induction of nerve growth factor receptor in schwann cells after axotomy
.
Proc. nam. Acad. Sci. U.S.A
.
83
,
4094
4098
.
Zimmerman
,
A.
&
Sutter
,
A.
(
1983
).
Beta Nerve growth factor (NGF) receptors on glial cells. Cell-cell interaction between neurons and schwann cells in cultures of chick sensory ganglia
.
EMB0 J
.
2
,
879
885
.