Expression of the gene encoding tyrosine hydroxylase in a subpopulation of quail dorsal root ganglion cells cultured in the presence of insulin or chick embryo extract

Developmental Brain Research, 69 (1992) 23-30 © 1992 Elsevier Science Publishers B.V. All rights reserved 0165-3806/92/$05.00

23

BRESD 51499

Expression of the gene encoding tyrosine hydroxylase in a subpopulation of quail dorsal root ganglion cells cultured in the presence of insulin or chick embryo extract Z h i - G a n g X u e , X i a o Jin X u e , M i r e i l l e F a u q u e t *, J u l i a n S m i t h ** a n d N i c o l e Le D o u a r i n Institut d'Embryologie Cellulaire et MoMculaire du CNRS et du Coll~ge de France, Nogent-sur-Marne (France) (Accepted 19 May 1992)

Key words: Tyrosine hydroxylase; In situ hybridization; Polymerase chain reaction; Dorsal root ganglion; Cell culture; Insulin; Chick embryo extract

Avian sensory ganglia contain a population of normally latent autonomic precursors with catecholaminergic potentialities. The present study examines the expression of the tyrosine hydroxylase (TH) gene in quail dorsal root ganglia (DRG) by both in situ hybridization and polymerase chain reaction (PCR) techniques. In situ hybridization using quail TH eDNA as a probe demonstrated the presence in DRG cell cultures of TH mRNA in a subpopulation of cells that never express the adrenergic phenotype in vivo. Expression of the TH gene in autonomic precursor cells of DRG in culture is totally dependent on the presence either of insulin or chick embryo extract. The numbers of catecholaminergic cells expressing TH mRNA and TH immunoreactivity evolve in a closely similar manner during the culture period. Using two primers, specific for highly conserved 5' regions of TH eDNA, it was possible to detect the same band of DNA amplified by PCR in total RNA from DRG cultures grown in the presence of insulin, sympathetic ganglia and adrenal gland. No amplified DNA was detected in uncultured DRG cells. These data further indicate that, under the influence either of insulin or a still unknown factor contained in the CEE, the TH gene is induced in a subpopulation of DRG cells.

INTRQDUCTION The peripheral nervous system (PNS) arises from a transient embryonic structure, the neural crest. Neural crest cells aggregate to form the various types of ganglia of the PNS after a migratory phase which, at the trunk level, leads them to constitute the dorsal root ganglia (DRG), the sympathetic ganglion chains and the adrenal medulla. Clonal cultures of migrating neural crest cells in a permissive medium 2 as well as in vivo labeling of premigratory cells by vital dyes 4 have shown that a number of crest cells are pluripotent. However, during migration, these cells divide actively and become progressively restricted in their potentialities. As a result of this, the cells reaching the sites of gangliogenesis constitute a heterogeneous population

possessing a variety of developmental capacities as well as different requirements for growth and survival. Gangliogenesis therefore involves a stringent selection process as a result of which only some of the developmental possibilities of the progenitor cells will be expressed. Selection of this sort has been demonstrated in many instances. For example, glial cells potentially able to express the Schwann cell marker, SMP 'J, are repressed from doing so in the gut environment, where enteric gila are SMP- throughout life. Enteric glial cells however, if withdrawn from the gut mesenchymal environment, express the SMP surface glycoprotein. Similarly, we have previously demonstrated in numerous experiments carried out in vivo 23'27 and in vitro 32'35 that the adrenergic phenotype, which is not repre-

Correspondence: Z.G. Xue, lnstitut d'Embryologie Ceilulaire et Mol~culaire du CNRS et du Coll~ge de France, 49 bis, Avenue de la Belle Gabrielle, 94736 Nogent-sur-Marne, Cedex, France. Fax: (33) (1) 4873 4377. * Present address: Laboratoire du D~veloppement et Evolution du syst~me nerveux, Ecole normale sup~rieure, 46 rue d'Ulm, 75005 Paris, France. ** Present address: Centre de biologic du d~veloppement, Universit6 Paul Sabatier, 118 route de Narbonne, 31062 Toulouse, France.

24 sented in DRG during normal development, can be induced in a subset of the ganglion cells if they are subjected to different conditions. This phenomenon occurs in vivo after backtransplanting quail DRG, in which all the sensory neurons are already post-mitotic, into the neural crest cell migration pathway of a young chick embryo. Some undifferentiated latent precursors belonging to the non-neuronal cell population of the DRG migrate in the host and become localized in the sympathetic ganglia and adrenal glands where they express the adrenergic phenotype. Subsequently, an in vitro system was used to analyse these differentiative events further: dissociated DRG cells taken at ages ranging from 9 days of embryonic life (E9) up to 5 days post-hatching (P5) were cultured in the presence of horse serum (HS, 10%) and chick embryo extract (CEE, 5 to 10%). After 3 days, a subpopulation of cells began to express adrenergic properties. Most adrenergic precursors divided before expressing this phenotype and in the absence of CEE virtually no adrenergic cells developed in the culture. Moreover, CEE could be replaced by insulin or IGF-I at physiological concentrations (10 -') M). The adrenergic cells developing in culture displayed tyrosine hydroxylase (TH) immunoreactivity, synthesized and stored catecholarnines, and were able to take up extracellular noradrenaline by a specific high affinity transport process 32''~4'3"s. The quail TH eDNA was recently cloned by one of us t~. We decided to use the TH eDNA 0robe to further document the differentiation of DRG progenitor cells induced in culture by CEE or insulin. We demonstrate here, by using both in situ hybridization and polymerase chain reaction (PCR) techniques, that, under the influence either of a still unknown factor contained in the CEE or insulin, TH gene activity is induced in a subpopulation of DRG ceils that in situ never express the adrenergic phenotype. MATERIALS AND METHODS

Tissue preparation Japanese quail (Cotumix cotumix japonica) eggs, incubated at 38°C, were used in all experiments. The trunk of embryos from day 7 (E7) to day 16 (El6) and adrenal glands from one-month-old quails were fixed in I% (w/v)paraformaldehyde in Ca" +/Mg ~+-free phosphate-buffered saline (PBS) at 4°C for 4-16 h. After rinsing in 15% (w/v) sucrose in PBS for 12-18 h, sections were cut iri a ~y,~.:;tat .at 10/.tin and thawed onto slides treated with gelatine. Sections were stored at -70°C rapidly after cutting.

Cell culture Quail embryonic DRG cells were cultured on collagen-coated glass coverslips as previously described -~2.'~-~.The basic culture medium used was Eagle's minimum essential medium (MEM, Gibco) supplemented with glucose, glutamine, choline, KCI and 10% heat-inactivated horse serum as described in Xue et al) "~. In some experi-

ments, 10% chick embryo extract (CEE) was added to the medium; in others, it was replaced by 10 ng/ml of porcine insulin (Sigma). After fixation with 1% paraformaldehyde in PBS for 1 h at 4°C, the cultures were rinsed for 2 h in PBS containing 15% sucrose and then stored at - 70°C.

Preparation of probes Two TH cDNA probes, QTH3 and QTH2, both subcloned into the pSPT 18 vector (Pharmacia), were used in these experiments. QTH3 (2080 bp) encompasses the complete coding sequence of quail TH I j, whereas QTH2 (2350 bp) begins at nucleotide 1090 of QTH3 and ends with the poly-A tail. After digestion with EcoR I endonuclease, both inserts were purified by gel electrophoresis 24, and labeled by nick-translation using a nick translation kit (BRL) and either [32p]dCTP (3000 Ci/mmol; Amersham) or [35S]dATP (800 Ci/mmol; Amersham). Nick-translated probes were purified with a Sephadex G-50 column and had a specific activity of about 2 to 5 × l0 s cpm/~g. The cRNA probe was produced by using T7 polymerase (Promeg~) to transcribe DNA template derived by insertion of the cDNA fragment into pBluescript.

in sire hybridization In situ hybridization was essentially performed according to Block et al) Briefly, sections or cultures were rinsed for 1 h at room temperature in 4 x SSC, 0.02% Denhardt solution. The hybridization was performed at 42°C for 16 h in the hybridization buffer (50% deionized formamide, 4 x SSC, 0.02% Denhardt solution, 0.9% sarcosyl solution and 0.1 M phosphate buffer pH 7.2). Each section or culture was covered with 20 p.I hybridization buffer containing 2-5 ng of probe and covered wilh a coverslip sealed wifl~ tubber cement. Following hybridization, the slides were rinsed in 4 × SSC, washed in 1 × SSC for 45 min at room temperature and then for 45 min at 42°C. The slides were finally dehydrated in ethanol, air-dried and processed for autoradiography using llford K-5 emulsion and an exposure time of 4-9 days at 4°C. Sections were stained with Toluidine blue, dehydrated and mounted with Eukitt. For control experiments, some sections and cultures were hybridized with a ['a'~S]dATP-labeled control'probe consisting of a fragment from the pSPT 18 plasmid, or a ['~'~S]UTP-labeled sensestrand cRNA probe.,

bmnunohistochemisto, A rabbit antiserum against bovine TH (Eugene Tech. International Inc., AIlendale, NY) was used to detect the expression of TH on embryo sections and cultures by indirect immunofluorescence as already described 32"~'a.

Isolation of RNA After dissection, sympathetic ganglia, DRG (from El0 embryos) and adrenal gland (from one-month-old quail) were collected immediately in liquid nitrogen. Cell cultures were washed with PBS, scraped off the dish with a Pasteur pipette and put in liquid nitrogen. All tissues were stored at -800C until used for isolation of RNA. Total RNA from different tissues was extracted by the Chomc~nski and Sacchi s method.

Determinaton of TH m R N A by PCR Two oligonucleotide primers were synthesized. Primer A, 5'CAACCCCCAACAi'CTCCACCTCT-3', begins at c D N A position 23, and primer B (from the complementary e D N A strand), 5'GAGTCTTGTCTCTAAGTGGTGGA-3', begins at position 351 of cDNA' ~.For reverse transcription,5 p.g of total R N A in 24 ~I water were heated at 90°C for 5 min, quenched on ice,added to I0 p,l of 5 × R T C buffer(BRL), 20 U of RNasin (Promega), all 4 dNTP at 2.5 m M (Pharmacia), 50 pmol of primer B and I00 U of M - M L V reverse transcriptase(BRL), and incubated for 2 h at 37oc. Target sequences were amplified in a I00 ~I reaction volume containing half of the reverse transcriptionmixture, 50 pmol of each primer, 2.5 U of Taq polymerase (Perkin-Elmer-Cetus) and I x Taq polymerase buffer (Perkin-Elmer-Cetus). The samples were taken

25 through 45 cycles of amplification by using a step program (94°C, 1.5 min; 560C, 1.5 rain; 72°C, 3 rain). The PCR products were checked by agarose gel. The corresponding band was subcloned into the plasmide pBluescript and sequenced as double-stranded DNA (sequenase kit, United States Biochemical). RESULTS

In vivo detection of TH mRNA in quail embryos We first examined the specificity of the hybridization pattern of the TH cDNA probe on the adrenal gland. Serial sections of adrenal gland from an onemonth-old quail were hybridized either with probe QTH3 or QTH2. With the 2 probed, a strong hybridization activity was observed exclusively in the medullary cords of the adrenal gland; it was absent from the cortical regions (Fig. 1). When an adjacent section was

Fig. 2. In situ hybridization to E l l quail transverse section at the level of the trunk using a 32P-labeled TH cDNA probe. The intense accumulation of TH mRNA is visible over the paravertebral sympathetic ganglia (SO). Signal is never observed in the DRG. Bar = 420 ~m.

hybridized with a control pSPT18 probe, no signal was seen. Secondly, the expression pattern of the TH gene was scrutinized on transverse sections of the trunk in E7El6 embryos. In no case did we observe any signal in sensory ganglia, whereas sympathetic ganglia were strongly positive (Fig. 2). This result is consistent with the results of immunocytochemistry using an antiserum against TH; no TH-immunoreactive cell was found in the DRG. In contrast, many cells expressing different intensities of TH immunoreactivity were observed in sympathetic ganglia (Fig. 3). i

Fig. 1. Section of quail adrenal gland examined by in situ hybridization with 32P-labeled TH cDNA. A: microautoradiography under darkfield exposure. TH mRNA was distributed exclusively in the medullary cords of the adrenal gland. B: the same section photographed under bright field after Toluidine blue staining. Bar = 420/~m.

In vitro analysis of TH gene activity in DRG cultures In previous experiments, we demonstrated that TH immunoreactivity could be detected in cultures of dissociated DRG cells grown in a medium containing either 10% CEE or insulin at physiological concentration (10 -9 M) 35. Therefore, we wished to determine whether TH mRNA was synthesized by DRG cells in culture. With the basic medium composed of MEM and 10% HS, no TH mRNA could be detected at any time point from day 1 to 7 (data not shown). In

Fig. 3, lmmunocytochemicalstaining with the TH antibody on sections of sympathetic ganglia (A) and DRG (B) of El5 quail. Intensely TH-immunoreactivecells are obsirvcd in the sympatheticganglion, whereas no immunoreactivecells are found in the DRG. Bar - 20 &m.

contrast, in the presence of IO% of CEE or 10 ng/ml

of insulin, cells which expressed TH mRNA were present after 3 days of culture. These cells were small and morphologically different from sensory neurons (Fig. 4). Their number increased rapidly during the next few days of culture. The number of TH mRNA containing cells counted in the cultures over a period of 3-7 days

was very similar to that of TH-immunoreactive cells (Fig. 5). The silver grains of the hybridizing mRNAs were always located in the cell bodies; they were never seen in the fibers, where TH immunoreactivity could be found. The morphology of the cells expressing TH mRNA was very like to that of TH immunoreactive cells (Fig. 4). In no case was the signal detected when a

Fig. 4. A,B: culturesexaminedby in situ hybridizationwith %-labeled TH cDNA. Autoradiogramswere photographedunder bright field after Toluidine blue staining.A: the first cells with TH mRNA appear after 3 days in culture. Note that the cell expressingTH mRNA is smaller than the sensory neuron (solid arrow).B: a labeled cell and a negative cell in a 5-day culture. C: no specific signal was found when a control probe (3sS-labeledplasmid)was used. D: the morphologyof TH-immunoreactivecells revealed by fluorescence photomicrographin a similar culture. Bar = 9 Frn.

.~

I

~~

.~

,~_

~-_~

,~:,

, : ~ i ~ ~..... ~

i!~¸~

,?

~ il i.~ ¸¸¸¸¸

e

o

I

~o~,i

7 ~ ~

o

g

o

o

o

28 control vector probe or a sense-strand cRNA probe were used, (data of cRNA experiment not shown). Detection of TH mRNA in rico and in citro by PCR Because of the small amount of TH mRNA produced in sympathetic ganglia and DRG cultures, detection of TH transcripts by Northern blot was difficult. Therefore, the amplification of a target sequence by polymerase chain reaction (PCR) seemed to be appropriate to further document the differentiation of adrenergic cells in DRG cultures and to confirm the specificity of in situ hybridization. Using 2 primers, specific for the 5' region of TH cDNA, only a single band of amplified DNA (about 330 bp) was detected in El0 sympathetic ganglia as well as in the adult adrenal gland. The size and nucleotid sequence of this band corresponded exactly to the region expected from the quail TH eDNA sequence ~. Whereas no amplified DNA was detected from DRG RNA prior to culture, a TH DNA band appeared after amplification of mRNA from DRG cells cultured for 6 days in the presence of 10 ng/mi insulin (Fig. 6).'This experiment therefore

1

2

3

4

v-.,Wm

-600

4WP

Fig. 6. Amplification of TH eDNA from total RNA of El0 sympathetic ganglia (lane 1), 6-day cultures of El0 DRG in the presence of 10 ng/ml insulin (lane 2) and adrenal gland (lane 4). No amplified DNA was detected from RNA of uncultured DRG (lane 3). The size markers are shown in base pairs (bp). The picture is a negative image of an ethidium bromide-stained gel.

confirms the results obtained from in situ hybridization with the TH ~DNA probe and immunocytochemistry with the anti-TH antiserum. DISCUSSION

4= W tn tm

•

1000

m m

a) o

t~

0

6

z

500

1

2

3

4

5

6

?

8

days in culture Fig. 5. Evolution of the number of TH-immunoreactive cells (e) and TH mRNA-containing cells (o) as a function of time, Each culture dish initially contained 7× l0 4 cells from El0 quail DRG. Insulin was added in the culture medium at 10 ng/ml. The symbols represent the means+SEM of counts performed on at least 3 separate cultures at each time point.

In this study we demonstrate a specific hybridization of TH eDNA in a subpopulation of cells in cultures of quail DRG. The specificity of the hybridization was shown by the absence of signal in sections or cultures treated with a control plasmid probe that does not contain the TH sequence. This analysis was completed by the identification of TH transcripts using the PCR technique. Adrenergic differentiation is not seen in vivo in DRG cells at any developmental stage or in the adult by immunocytochemistry (unpublished result). We show here that this is also the case at the mRNA level. It therefore appears that in birds, apart from the brain, the catecholaminergic phenotype is exclusively expressed by autonomic sympathetic neurons, adrenal medulla and adrenergic paraganglia. Such is not the case in mammals, where transient adrenergic cells have been found in the gut of mouse embryos t'6a2'3°'31 and where, in certain species n9,2°,25, DRG contain neurons exhibiting TH immunoreactivity. In the experiments described here, the subset of DRG cells able to express catecholaminergic phenotype was found to require certain growth factors in

29

order to differentiate along this pathway. These include insulin and IGF-135. CEE also contains powerful, but uncharacterised inducer(s) of catecholamine synthesis in these precursors. Such an effect of CEE on adrenergic cell differentiation in neural crest cell cultures has already been reported in several instances I°'m4,es'36. It has to be noted that, even in the presence of CEE or insulin, the expression of TH mRNA was never detected before 3 days of cu!tt, re. The same result was found when TH immunoreactivity was used to identify the cells 3:''35. Incorporation of [3H]TdR in the cultures showed that most TH expressing cells had divided, in contrast to the post-mitotic sensory neurons, which never incorporated the isotope. Moreover, the number of cells expressing TH immunoreactivity and TH mRNA (see Fig. 5) increased in a strikingly similar manner during the culture period. Although we were not able to show TH immunoreactivity and mRNA in the same cells, we consider that both techniques identify the same cell type: the cells that express TH have the same morphology as those containing TH mRNA, they are multipolar and they are smaller than the sensory neurons. According to the model presented by one of us 2m'22, a selection among various developmental potencies takes place at the sites of gangliogenesis. A specific array of growth or survival factors triggers, in each type of ganglion, the expression of one or another of these potencies. Factors deriving from the neural tube have been shown to be necessary for sensory neuron survival and differentiation in DRG a.~,~s,m6,ma.This effect of neural tube seems to be mediated by BDNF and laminin, as far as the neurons are concerned ~'~H, and by bFGF, in the case of the non neuronal cells of the developing DRG ts. The early steps of development of the sympathoadrenal lineage have been shown to be dependent on the survival-promoting effect of bFGF in the rat 29 and on insulin and IGF-I in avian species 35. The production of bFGF by the embryonic neural tube at the critical stage of sympathetic ganglion development has been demonstrated t5. Similarly IGF-I is widely distributed in various tissues of the young embryo s'26. This is also true for insulin, although it is present at very low concentration 7. The involvement of these factors in sympathetic ganglion differentiation in vivo, although not demonstrated, is a likely possibility. Their selective effect on certain cell types may be due to the specific expression of the appropriate receptor by certain subsets of precursors at the appropriate developmental stage. Acknowledgements. We gratefully thank Dr. B. Bloch for introducing us to in situ hybridization techniques. We also thank Y. Rantier, B.

Henry and S. Tissot for the illustrations. This work was supported by the Centre National de la Recherche Scientifique and the Institut National de la Recherche M6dicale, the Fondation pour la Recherche M6dicale Fran~aise, the Association pour la Recherche contre le Cancer and the Ligue Nationale Fran~aise contre le Cancer. REFERENCES 1 Baetge, G., Pintar, J.E. and Gershon, M.D., Transiently catecholaminergic (TC) cells in the bowel of the fetal rat: precursors of noncatecholaminergic enteric neurons, Dec. Biol., 141 (1990) 353-380. 2 Baroffio, A., Dupin, E. and Le Douarin, N.M., Clone-forming ability and differentiation potential of migratory neural crest cells, Proc. Natl. Acad. Sci. USA, 85 (1988) 5325-5329. 3 Bloch, b., Popovici, T., Le Gueilec, D., Normand, E., Chouham, S., Guitteny, A.F. and Bohlen, P., In situ hybridization histochemistry for the analysis of gene expression in the endocrine and central nervous system tissues: a 3-year experience, J. Neurosci. Res., 16 (1986) 183-200. 4 Bronner-Fraser, M. and Fraser, S.E., Cell lineage analysis reveals multipotency of some avian neural crest cells, Nature, 335 (1988) 161-164. 5 Chomczynski, P. and Sacchi, N., Single-step method of RNA isolation by acid guanidium thiocyanate-phenol-chloroform extraction, Anal. Biochem., 162 (1987) 156-159. 6 Cochard, P., Goldstein, M. and Black, I.B., Ontogenetic appearance and disappearance of tyrosine hydroxylase and catecholamines in the rat embryo, Proc. NatL Acad. ScL USA, 75 (1978) 2986-2990. 7 De Pablo, F., Roth, J., Hernandez, E. and Pruss, R., Insulin is present in chicken eggs and early chick embryos, Endocrinology, 111 (1982) 1909-1916. 8 De Pablo, F., Scott, L.A. and Roth, J., Insulin and insulin-like growth factor I in early development: peptides, receptors and biological events, Endocrinology, 11 (1990) 558-577. 9 Dulac C. and Le Douarin N.M., Phenotypic plasticity of Schwann cells and enteric glial cells in response to the microenvironment, Proc. Natl. Acad. Sci. USA, 88 (1991) 6358-6362. 10 Fauquet, M., Smith, J., Ziller, C. and Le Douarin, N.M. Differentiation of autonomic neuron precursors in vitro: cholinergic and adrenergic traits in cultured neural crest, J. Neurosci., 1 (1981) 478-492. I 1 Fauquet, M., Grima, B., Lamouroux, A. and Mallet, J., Cloning of quail tyrosine hydro~lase: amino acid homology with other hydroxylase discloses functional domains, J. Neurochena., 50 (1988) 142-148. 12 Howard, M.J. and Bronner-Fraser, M., The influence of neural tube-derived factols on differentidtio~l of neural crest cells in vitro, l. !-li~tochemistry study on the appearance of adrenergic cells, J. Neurosci., 5 (1985), 3302-3309. 13 Gershon, M.D., Rothman, T.P. and Teitelman, G.N., Transient and differential expression of aspects of the catecholaminergic phenotype during development of the fetal bowel of rat and mice, J. Neurosci., 4 (1984) 2269-2280. 14 Hofer, M.M. and Barde, Y.A., Brain-derived neurotrophic factor prevents neuronal death in vivo, Nature, 331 (1988) 261-262. 15 Kalcheim, C., Basic fibroblast growth factor stimulates survival of non-neurnnal cells developing from trunk neural crest, Dec. BioL, 134 (1989) 1-10. 16 Kalcheim, C. and Le Douarin, N.M., Requirement of a neural tube signal for the differentiation of neural crest cells into dorsal root ganglia, Dec. BioL, 116 (1986) 451-466. 17 Kalcheim, C. and Gendreau, M., Brain-derived neurotrophic factor stimulates survival and neuronal differentiation in cultured avian neural crest, Dev. Brain Res., 41 (1988) 79-86. 18 Kalcheim, C., Barde, Y.A., Thoenen, H. and Le Douarin, N.M., In vivo effect of brain-derived neurotrophic factor on the survival of developing dorsal root ganglion cells, EMBO J., 6 (1987) 2871- 2873. 19 Katz, D.M., Markey, K.A., Gt,'dstein, M. and Black, I.B., Expres-

3O

20

21 22

23

24

25

sion of catechalaminergic characteristics by primary sensory neurons in the normal adult rat in vivo, Proc. Natl. Acad. Sci. USA, 80 (1983) 3526-3530. Jonakait, G.M., Markey, K.A., Goldstein, M. and Black, I.B., Transient expression of selected catecholaminergic traits in cranial sensory and dorsal root ganglia of the embryonic rat, Do'. Biol., 101 (1984) 51-60. Le Douarin, N.M., Cell line segregation during peripheral nervous system ontogeny, Science, 231 (1986) 1515-1522. Le Douarin, N.M., Dupin, E., Baroffio, A. and Dulac, C., New insights into the development of neural crest derivatives, b~t. Ret'. Cytol., in press. Le Lievre, C.S., Schweizer, G.G., Ziller, C.M. and Le Douarin, N.M., Restrictions of developmental capabilities in neural crest derivatives as tested by in vivo transplantation experiments, Det'. Biol., 77 {1980} 362-378. Maniatis, T., Fritsch, E.F. and Sambrook, J., Mok'cular Cloning, a Laborato~' Mmmal, 2nd edn, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1989. Price, J. and Mudge, A.W., A subpopulation of rat dorsal root ganglion neurons is catecholaminergic, Nature. 301 (1983) 241.4.'L

26 Ralphs, J.R., Wylie, L. and Hill, D.J., Distribution of insulin-like growth peptides in the developing chick embryo, Det'elopment, l0t) (1990) 51-58. 27 Schweizer, G., Ayer-Lc Lievre, C. and Le Douarin, N.M., Restrictions of developmental capacities in the dorsal root ganglia during the course of development, (.'ell Diffi, r., 13 (1983) 191-200. 28 Sieber-Blum, M. and Cohen, A.M., Clonal analysis of quail neural crest cells: they are pluripotent and differentiate in vitro in the absence of non crest cells, Det'. Biol., 80 (1980)96-1(16.

29 Stemple, D.L., Mahanthappa, N.K. and Anderson, D.J., Basic FGF induces neuronal differentiation, cell division and NGF dependence in chromaffin cells: a sequence of events in sympathetic development, Neuron, 1 (1988)517-525. 30 Teitelman, G., Joh, T.H. and Reis, D.J., Transient expression of a noradrenergic phenotype in cells of the rat embryonic gut, Brain Res., 158 (1978) 229-234. 31 Teitelman, G., Gershon, M.D., Rothman, T.P., Joh, T.H. and Reis, D.J., Proliferation and distribution of cells that transiently express a catecholaminergic phenotype during development in mice and rats, Det'. Biol., 86 (1981) 348-355. 32 Xue, Z.G., Smith, J. and Le Douarin, N.M., Differentiation of catecholaminergic cells in cultures of embryonic avian sensory ganglia, Proc. NatL Acad. Sci. USA, 82 (1985) 8800-8804. 33 Xue, Z.G., Smith, J. and Le Douarin, N.M., Developmental capacities of avian embryonic dorsal root ganglion cells: neuropeptides and tyrosine hydroxylase in dissociated cell cultures, Det'. Brain Res., 34 (1987) 99-109. 34 Xue, Z.G. and Smith, J., High-affinity uptake of noradrenaline in quail dorsal root ganglion cells that express tyrosine hydroxylase immunoreactivity in vitro, J. NeuroscL, 8 (1988) 806-813. 35 Xue, Z.G., Le Douarin, N.M. and Smith, J., Insulin and insulinlike growth factor-I can trigger the differentiation of catecholaminergic precursors in cultures of dorsal root ganglia, Cell Differ. Det'., 25 (1988) 1-10. 36 Ziller, C., Fauquet, M,, Kalcheim, C., Smith, J. and Le Douarin N.M., Cell lineages in peripheral nervous system ontogeny: medium-induced modulation of phenotypic expression in neural crest cultures, Det'. Biol., ! 20 (1987) 101-111.