- Email: [email protected]
Identification of nerve growth factor receptors in primary cultures of chick neural crest cells
Developmental Brain Research, 7 (1983) 131-136 Elsevier Biomedical Press
131
Identification of Nerve Growth Factor Receptors in Primary Cultures of Chick Neural Crest Cells DAVID END, LEONID PEVZNER, AARON LLOYD and GORDON GUROFF* Laboratory of Developmental Neurobiology, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD 20205 (U.S.A.) (Accepted July 27th, 1982) Key words: nerve growth factor - - neural crest cells - - NGF receptors
Primary cultures of chick neural crest cells obtained from explanted neural tubes have binding sites for radioiodinated nerve growth factor ([125I]NGF)but not for radioiodinated epidermal growth factor ([125I]EGF).The binding of [125I]NGFwas shown to be a specific and saturable process with a high affinity (Ka = 0.3 nM) for the ligand. Despite the expression of these NGF binding sites, incubation of the neural crest cultures with nerve growth factor did not induce neurite outgrowth; no morphological alterations were observed. This was not due to an inability of the cells to express a neuronal phenotype, since the neural crest cells spontaneously differentiated into neurite-bearing cells. However, the nerve growth factor binding sites do appear to be functional receptors, since nerve growth factor could produce a modest induction of ornithine decarboxylase. The quantity of nerve growth factor binding sites seemed to be independent of the phenotype expressed by the neural crest cells, since both pigmented cells and neuron-like neural crest cells exhibited binding. These findings suggest that the differentiation of neural crest cells into mature nerve growth factor-responsiveneurons may involve the coupling of nerve growth factor receptors to cellular responses important in the expression of the neuronal phenotype. INTRODUCTION When the rat pheochromocytoma cell line, PC12, is treated with nerve growth factor (NGF), there is a cessation of cellular proliferation and a concomitant expression of characteristics similar to those of differentiated sympathetic neurons6,17. Recently, we reported that a near-total loss of receptors for the mitogenic peptide epidermal growth factor (EGF) accompanies the NGF-induced differentiation of PC12 cells10,11. This phenomenon in PC12 cells prompted the hypothesis that the embryonic development of normal N G F target neurons may also involve a loss o f receptors for E G F or similar mitogens as the cells mature under the influence of N G F into terminally differentiated sympathetic or sensory neurons. To explore this question, we have studied primary cultures o f chick neural crest cells. These are the presumed normal counterpart of PC 12 tumor cells since these embryonic precursors give * To whom correspondence should be addressed.
rise to N G F target neurons of the peripheral sensory and sympathetic nervous systemslL This report describes the results of our investigation into the peptide growth factor receptors of neural crest cells and the responses of these cells to N G F . We report herein that primary cultures of chick neural crest cells do not possess receptors for EGF, but do have N G F receptors. These N G F receptors are functional, but do not seem to influence, or be influenced by, the expression of the neuronal phenotype. MATERIALS AND METHODS Cell culture Primary cultures of chick neural crest were prepared from 2-day-old White Leghorn chick embryos purchased from Truslow Farms, Chester, MD. A modification of the procedure described by Cohen and Konigsberg 4 was used. Briefly, the chick embryos were trypsinized at room temperature for
132 30-45 min in a solution of 1 ~o trypsin. Neural tubes were dissected from the surrounding epithelia, mesoderm, and notochord and placed, usually 8 tubes per dish, on collagen-coated (acid-soluble calf skin collagen, Calbiochem-Behring, La Jolla, CA) 35 m m tissue culture dishes. The neural tubes were maintained in Dulbecco's Modified Eagle Medium (Vogt Modification) (Gibco, Grand Island, NY) supplemented with 2 o/chick embryo extract, 7.5 /o horse serum, 7.5~o fetal bovine serum, and antibiotics (100 mU/ml penicillin, 0.25 #g/ml fungizone, and 100/~g/ml streptomycin). After 36 h incubation in a humidified atmosphere containing 5 ~o CO2, the neural tubes were removed by aspiration with a micropipette. The halo of neural crest cells which had migrated out of the tubes was maintained in medium without the chick embryo extract supplement. The medium was changed every 2 days thereafter.
passing the reaction mixture over Sephadex G-10 (Pharmacia Fine Chemicals, Piscataway, NJ) as described previously 18. Primary cultures of chick neural crest cells were incubated with either [1251]N G F (approximately 8 ng, 120,000 dpm) or with [lzSI]EGF (approximately 5 ng, 300,000 dpm) in 1 xnl of growth medium. After 60 min incubation at 37 °C, the radioactive medium was removed by aspiration and the cultures were quickly rinsed 3 times with ice-cold Hepes-buffered salt solution (HBS). After partial digestion in 1 N NaOH, a portion of the cells was counted in a Searle Model I 185 G a m m a Cotmter while the remainder was used for the determination of total culture protein by the method of Bradford 2. Specific binding of E G F or N G F was defined as the amount of labeled material which could be displaced by either 1 #g/ml of E G F or 5 #g/ml of N G F , respectively. Induction o f ornithine decar,boxylase
Specific binding of radioiodinatedgrowthfactors Epidermal growth factor was prepared by the method of Savage and Cohen 15 and iodinated by the chloramine T procedure described by Cuatrecasas 5. Nerve growth factor was prepared by the method of Bocchini and AngelettiI and iodinated by the lactoperoxidase technique described in BioRad Technical Bulletin No. 1071. Iodinated growth factors were reisolated from the radioactive reaction medium by
Ornithine decarboxylase (ODC) activity was measured by a modification of the method described by Pegg and Williams-Ashman 14. Cultures were incubated in the presence or absence of growth factors for 5 h. Cells were harvested by scraping with a rubber policeman into 2 ml of ice-cold HBS and collected by centrifugation (10,000 g for 5 rain). Pellets were resuspended in 200/zl of a homogenization buffer containing 50 m M Tris (pH 7.4), 5 m M
TABLE I Identification of NGF but not EGF binding sites in primary cultures of chick neural crest cells
[12~I]NGF binding: 130,000 dpm (20 ng) of [12~I]NGFwas added to each culture in 1 ml of medium. [125I]EGF binding: 628,800 dpm (5 ng) of [lZ5I]EGFwas added to each culture in 1 ml of medium. Culture
[125I]NGF binding 1 2 3 4 5 [12~I]EGF binding 1 2 3 4 5
Additions
Amount bound
% Specific binding
dpm
dpm/l~g protein
None None None 5 pg/ml NGF 5/~g/ml NGF
1946 1262 1826 592 733
78.8 97.1 63.4 25.6 17.8
72 79 65 ---
None None None 5/tg/ml EGF 5 #g/ml EGF
658 416 527 496 644
13.9 13.3 14.8 15.8 9.7
None detected
133 TABLE II Specificity of [a251]NGF binding in primary cultures of chick neural crest cells Culture
1 2 3 4 5 6 7 8 9
Additions
None None None NGF (5 pg/ml) NGF (5 #g/ml) Insulin (50 #g/ml) Insulin (50/~g/ml) Cytochrome-C (50 Fg/ml) Cytochrome-C (50/~g/ml)
Amount bound dpm
dpm/l~g protein
1597 1401 1294 592 600 1360 1586 1487 1279
13.1 11.2 10.5 4.5 4.6 11.2 14.1 11.6 10.6
dithiothreitol, and 40 /~M pyridoxal-5'-phosphate and the suspension was frozen on dry-ice for storage at --70 °C overnight. Cell suspensions were thawed and disrupted by sonication. After centrifugation (30,000 g for 30 rain) 75/A portions were brought up to 150/zl with the addition of reaction buffer containing 100 mM Tris (pH 7.4), 10 mM dithiothreitol, 80/zM pyridoxal phosphate, and 6 mM EDTA. The reaction buffer also contained 1.0 /~Ci of L-[114C]ornithine (59 Ci/mmol, Amersham Searle, Arlington Heights, IL) at a final concentration of 150/zM. The reaction mixtures were incubated in rubber-stoppered tubes fixed with hanging center wells containing 200 #1 of hyamine hydroxide for 2 h at 37 °C. The incubations were terminated by the addition of 0.5 ml of 2.5 N H9.SO4 and then allowed to equilibrate at 37 °C for an additional 2 h. Center wells containing the enzymatically liberated 1aco2 were counted in 15 ml of Liquifluor (New England Nuclear, Boston, MA).
centration-dependent displacement of [125I]NGF binding to neural crest cells with the displacement curves generated using NGF-responsive PC12 tumor cells revealed similar binding characteristics for both cell populations and apparent Kas of approximately 0.3 nM (Fig. 1). However, PC12 cells yielded a total specific binding of 60.7 dpm/#g of cell protein while neural crest cells exhibited substantially less (7.1 dpm/Fg of cell protein). The morphological differentiation of trunk neural crest cells can be directed, as described by Greenberg et alp for cranial neural crest cells, by the sera used to grow the cells. Fetal bovine serum (15~o) induces the pigmented melanocytic phenotype and NEURAL CREST CELLS
PC 12
CELLS
]00-
o. . . . .
K d -~o n g l m l
I . . . . . .
RESULTS Primary cultures of chick trunk neural crest cells that had been maintained in tissue culture for 5 days possessed specific binding sites for [125I]NGF but not for [z2SI]EGF (Table I). The binding of [1251]N G F was reduced by the addition of a 1000-fold excess of unlabeled N G F revealing specific binding which accounted for more than 60 ~ of the total binding (Table II). Neither cytochrome-C nor insulin in large excess reduced the binding of [1251]N G F to neural crest cells. Comparison o f the con-
I I ~0 ~0o
• 5~, nglml NGF
I 50
I 10G
I ~o0 ng/ml NGF
Fig. 1. Comparison of the specific binding of [x~5I]NGF to primary cultures of chick neural crest cells and to PC12 tumor cells. Cultures of neural crest cells derived from 7-8 neural tube explants per 35 mm well or PC12 tumor cells grown in 25 cm~ tissue culture flasks were incubated with [z~SI]NGF (approximately 8 ng/ml, 130,000 dpm/ml) in the presence of increasing concentrations of unlabeled NGF. Values are the means of duplicate determinations and are expressed as a percentage of the total binding observed in the absence of unlabeled NGF.
134 horse serum ( 1 5 ~ ) produces a neuronal morphology. After 10 days culture under these conditions, the neuronal (Fig. 2A) and the pigmented cell (Fig. 2B) phenotypes were obtained. The binding of [125I]NGF by these two divergent phenotypes was comparable; the pigmented cells exhibited binding of 15.4 dpm/#g of cell protein and the neuronal cells had 11.8 dpm/#g of cell protein bound.
TABLE III 77~e effects o f N G F o n the levels o f ornithine deearboxyh~se in neural crest cells
Cultures were treated for 6 h. Values are means of duplicate determination. Experiment
Treatment
O DC activity (Jhtol 14C O~/l~gprotein/ 2 h)
1
None NGF (20 ng/ml) None NGF (20 ng/ml) None NGF (20 ng/ml)
82 317 155 22"7 76 174
2 !
3
i/~i!
Despite the presence of specific, high-affinity N G F binding sites and the persistence of these binding sites during the morphological divergence of neural crest cells, treatment of neural crest cells with 50 ng/ml of N G F for up to 20 days did not induce the morphological differentiation which can be obtained in P C I 2 cells. While the cells were able to express the neurite-bearing neuronal phenotype, neither the extent nor the speed of this response was influenced by N G F . However, N G F treatment did elicit an elevation in O D C activity in neural crest cells (Table III). While this 2- to 3-fold induction of O D C is relatively modest in comparison to the response obtained in P C I 2 cells 11, it suggests that the specific binding sites for N G F are, in fact, functional receptors. DISCUSSION
Fig. 2. The effects of different sera on the morphological development of trunk neural crest cells. A: primary cultures of chick neural crest cells were grown in the presence of 15 horse serum, conditions which induced the appearance of a predominantly neural phenotype. Cultures exhibited a total specific binding of [x~5I]NGFof 11.8 dpm//~g of cell protein. B: neural crest cells grown in 15~ fetal bovine serum, conditions which induced the appearance of a predominantly pigmented cell phenotype. Cultures exhibited a total specific binding of [12'~I]NGFof 15.4 dpm//~g of cell protein. Magnification : 125 ×.
Unlike the NGF-responsive PC12 tumor cell line, neural crest cells do not have specific binding sites for the peptide growth regulator EGF. Thus, our previous observation that the NGF-induced differentiation of PCI2 cells is accompanied by a 90~o reduction in functional E G F receptors 11 may not hold significance for the ontological development of N G F target neurons. Rather, the observation may simply reflect the transition of the proliferating PC12 tumor cell to a quiescent neuronal phenotype. In the presence of N G F , the cells may assume a more normal phenotype and lose some of their tumorigenic characteristics, one of which may be the presence of E G F receptors.
135 In the light of the extensive evidence that demonstrates that N G F fails to alter the biochemical or morphological development of neural crest cultures 7,13, the identification of specific [I~51]NGF binding sites on neural crest cell cultures was unexpected. Again, as described in these previous reports, N G F did not alter the gross morphology of neural crest cultures or induce the extension of neurites. If anything, N G F seemed to have a deleterious effect on neural crest cell cultures and suppressed the spontaneous appearance of the neuronal phenotype (data not shown). Such a finding is difficult to reconcile with the widely documented neurotrophic effects of N G F on target neurons derived from neural crest cells. N G F is an absolute requirement for the normal development and survival of sympathetic and sensory neurons s. However, Norr 13 observed a similar phenomenon in crest cell cultures exposed to N G F prior to their interaction with the neutrophic influence of the surrounding mesoderm and notochord tissue. This failure of N G F to elicit a morphological response would not seem to be due to peculiarities in the crest cell N G F receptor, since it displayed an affinity for N G F similar to that of the PC12 cell receptor. Although the relative binding of N G F to crest cell cultures, expressed as a function of total cell protein, was approximately 10 ~ of the binding to PC12 cell cultures, the crest cell cultures were clearly not a homogeneous population of cells, as are the PCI2 tumor cells. Thus, the presence of noncrest derivatives could have diluted out the binding. This dilution with unresponsive cells might also explain the rather modest elevation of ODC activity; the response might be quite a bit larger in some o f the cells, but others might not be responding at all. The fact that the N G F binding sites were coupled to ODC induction, an induction characteristic of REFERENCES 1 Bocchini, V. and Angeletti, P. U., The nerve growth factor: purification as a 30,000 molecular weight protein, Proc. nat. Acad. Sci. U.S.A., 64 (1969) 787-794. 2 Bradford, M. M., A rapid and sensitive method for the quantitation of microgram quantities of protein using the principle of protein-dye binding, Analyt. Biochem., 72 (1976) 248-254. 3 Bronner-Fraser, M., Sieber-Blum, M. and Cohen, A. M.,
NGF-responsive cells 11, indicates that the N G F binding sites represent functional receptors. They do not, however, influence the choice of phenotype. Furthermore, when cells were induced to differentiate into the divergent pigmented, melanocytic phenotype and the neuronal phenotype, the levels of N G F binding were unchanged. Thus, N G F receptors may not be unique to neural crest neurons, but, rather, may be expressed by several crest derivatives. The presence of N G F receptors in the pigmented cell phenotype is consistent with reports of N G F receptors in melanomas TM. Further studies on the presence of N G F receptors in normal melanocytes may prove enlightening. In conclusion, trunk neural crest cells express N G F receptors independently of their phenotype, but do not have their initial development regulated by this peptide. Thus, N G F may not be the primary signal for the differentiation of sympathetic and sensory neurons, but, rather, may participate later in the development after the expression of the neuronal phenotype and after the biochemical features of this phenotype are linked to the N G F receptor system. Alternatively, the procedures used to prepare neural crest cells for tissue culture may be so traumatic that N G F receptors become uncoupled from their normal morphological and biochemical responses. Thus, normal differentiation in response to N G F is prevented. This latter possibility can probably be discounted, however, on the grounds that when cultured neural crest cells are reinjected into embryos, the cells undergo normal differentiation 3. Therefore, the normal ontogenesis of neural crest cells would seem to require the coupling of N G F receptors to appropriate biological responses. ACKNOWLEDGEMENT D.E. is recipient of a National Research Service Award 1-F32-HD05905-1. Clonal analysis of the avian neural crest: migration and maturation of mixed neural crest clones injected into host chicken embryos, J. comp. Neurol., 193 (1980) 423434. 4 Cohen, A. M. and Konigsberg, I. R., A clonal approach to the problem of neural crest determination, Develop. BioL, 46 (1975) 262-280. 5 Cuatrecasas, P., Insulin-receptor interactions in adipose cells: direct measurement and properties, Proc. nat. Acad. Sci. U.S.A., 68 (1971) 1264-1268. 6 Dichter, M. A., Tischler, A. S. and Greene, L. A., Nerve
136
7
8
9
10
I1
growth factor-induced increase in electrical excitability and acetylcholine sensitivity of a rat pheochromocytoma cell line, Nature (Lond.), 268 (1977) 501-504. Fauquet, M., Smith, J., Ziller, C. and LeDouarin, N. M., Differentiation of autonomic neuron precursors in vitro: chotinergic and adrenergic traits in cultured neural crest cells, J. Neurosei., 1 (1981) 478492. Gorin, P. D. and Johnson, E. M., Experimental autoimmune model of nerve growth factor deprivation: effects on developing sympathetic and sensory neurons, Proc. nat. Acad. Sci. U.S.A., 76 (1979) 5382-5386. Greenberg, J. H., Foidart, J. M. and Greene, R. M., Collagen synthesis in cultures of differentiating neural crest cells, Cell. Diff, 9 (1980) 153-163. Huff, K. R. and Guroff, G., Nerve growth factor-induced reduction in epidermal growth factor receptors in PC12 cells: an aspect of cell differentiation, Bioehem. Biophys. Res. Commun., 89 (1979) 175-180. Huff, K., End, D. and Guroff, G., Nerve growth factorinduced alteration in the response of PC12 pheochromocytoma cells to epidermal growth factor, J. Cell Biol., 88 (1981) 189-198.
12 LeDouarin, N. M., The ontogeny of the neural crest in avian embryo chimaeras, Nature (Lond.), 286 (1980) 663-669. 13 Norr, S, In vitro analysis of sympathetic neuron differentiation from chick neural crest cells, Develop. Biol., 34 (1973) 16-38. 14 Pegg, A. E. and Williams-Ashman, H. G., Biosynthesis of putrescine in the prostate gland of the rat, Biochem. J., 108 (1968) 533-539. 15 Savage, C. R., Jr. and Cohen, S., Epidermal growth factor and a new derivative, J. biol. Chem., 247 (1972) 7609-7611. 16 SheJwin, S. A., Sliski, A. H. and Todaro G. J., Human melanoma cells have both nerve growth factor and nerve growth factor-specific receptors on their cell surfaces, Proc. nat. Acad. Sci. U.S.A., 76 (1979) 1288-1292. 17 Tischler, A. S. and Greene, L. A., Nerve growth factorinduced process formation by cultured rat pheochromocytoma cells, Nature (Lond.), 258 (1975) 341-342. 18 Tuszynski, G. P., Knight, L., Piperno, J. R. and Walsh, P. N., A rapid method for removal of [l~sI]iodide following iodination of protein solutions, Analyt. Biochem., 106 (1980) 118-122.
131
Identification of Nerve Growth Factor Receptors in Primary Cultures of Chick Neural Crest Cells DAVID END, LEONID PEVZNER, AARON LLOYD and GORDON GUROFF* Laboratory of Developmental Neurobiology, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD 20205 (U.S.A.) (Accepted July 27th, 1982) Key words: nerve growth factor - - neural crest cells - - NGF receptors
Primary cultures of chick neural crest cells obtained from explanted neural tubes have binding sites for radioiodinated nerve growth factor ([125I]NGF)but not for radioiodinated epidermal growth factor ([125I]EGF).The binding of [125I]NGFwas shown to be a specific and saturable process with a high affinity (Ka = 0.3 nM) for the ligand. Despite the expression of these NGF binding sites, incubation of the neural crest cultures with nerve growth factor did not induce neurite outgrowth; no morphological alterations were observed. This was not due to an inability of the cells to express a neuronal phenotype, since the neural crest cells spontaneously differentiated into neurite-bearing cells. However, the nerve growth factor binding sites do appear to be functional receptors, since nerve growth factor could produce a modest induction of ornithine decarboxylase. The quantity of nerve growth factor binding sites seemed to be independent of the phenotype expressed by the neural crest cells, since both pigmented cells and neuron-like neural crest cells exhibited binding. These findings suggest that the differentiation of neural crest cells into mature nerve growth factor-responsiveneurons may involve the coupling of nerve growth factor receptors to cellular responses important in the expression of the neuronal phenotype. INTRODUCTION When the rat pheochromocytoma cell line, PC12, is treated with nerve growth factor (NGF), there is a cessation of cellular proliferation and a concomitant expression of characteristics similar to those of differentiated sympathetic neurons6,17. Recently, we reported that a near-total loss of receptors for the mitogenic peptide epidermal growth factor (EGF) accompanies the NGF-induced differentiation of PC12 cells10,11. This phenomenon in PC12 cells prompted the hypothesis that the embryonic development of normal N G F target neurons may also involve a loss o f receptors for E G F or similar mitogens as the cells mature under the influence of N G F into terminally differentiated sympathetic or sensory neurons. To explore this question, we have studied primary cultures o f chick neural crest cells. These are the presumed normal counterpart of PC 12 tumor cells since these embryonic precursors give * To whom correspondence should be addressed.
rise to N G F target neurons of the peripheral sensory and sympathetic nervous systemslL This report describes the results of our investigation into the peptide growth factor receptors of neural crest cells and the responses of these cells to N G F . We report herein that primary cultures of chick neural crest cells do not possess receptors for EGF, but do have N G F receptors. These N G F receptors are functional, but do not seem to influence, or be influenced by, the expression of the neuronal phenotype. MATERIALS AND METHODS Cell culture Primary cultures of chick neural crest were prepared from 2-day-old White Leghorn chick embryos purchased from Truslow Farms, Chester, MD. A modification of the procedure described by Cohen and Konigsberg 4 was used. Briefly, the chick embryos were trypsinized at room temperature for
132 30-45 min in a solution of 1 ~o trypsin. Neural tubes were dissected from the surrounding epithelia, mesoderm, and notochord and placed, usually 8 tubes per dish, on collagen-coated (acid-soluble calf skin collagen, Calbiochem-Behring, La Jolla, CA) 35 m m tissue culture dishes. The neural tubes were maintained in Dulbecco's Modified Eagle Medium (Vogt Modification) (Gibco, Grand Island, NY) supplemented with 2 o/chick embryo extract, 7.5 /o horse serum, 7.5~o fetal bovine serum, and antibiotics (100 mU/ml penicillin, 0.25 #g/ml fungizone, and 100/~g/ml streptomycin). After 36 h incubation in a humidified atmosphere containing 5 ~o CO2, the neural tubes were removed by aspiration with a micropipette. The halo of neural crest cells which had migrated out of the tubes was maintained in medium without the chick embryo extract supplement. The medium was changed every 2 days thereafter.
passing the reaction mixture over Sephadex G-10 (Pharmacia Fine Chemicals, Piscataway, NJ) as described previously 18. Primary cultures of chick neural crest cells were incubated with either [1251]N G F (approximately 8 ng, 120,000 dpm) or with [lzSI]EGF (approximately 5 ng, 300,000 dpm) in 1 xnl of growth medium. After 60 min incubation at 37 °C, the radioactive medium was removed by aspiration and the cultures were quickly rinsed 3 times with ice-cold Hepes-buffered salt solution (HBS). After partial digestion in 1 N NaOH, a portion of the cells was counted in a Searle Model I 185 G a m m a Cotmter while the remainder was used for the determination of total culture protein by the method of Bradford 2. Specific binding of E G F or N G F was defined as the amount of labeled material which could be displaced by either 1 #g/ml of E G F or 5 #g/ml of N G F , respectively. Induction o f ornithine decar,boxylase
Specific binding of radioiodinatedgrowthfactors Epidermal growth factor was prepared by the method of Savage and Cohen 15 and iodinated by the chloramine T procedure described by Cuatrecasas 5. Nerve growth factor was prepared by the method of Bocchini and AngelettiI and iodinated by the lactoperoxidase technique described in BioRad Technical Bulletin No. 1071. Iodinated growth factors were reisolated from the radioactive reaction medium by
Ornithine decarboxylase (ODC) activity was measured by a modification of the method described by Pegg and Williams-Ashman 14. Cultures were incubated in the presence or absence of growth factors for 5 h. Cells were harvested by scraping with a rubber policeman into 2 ml of ice-cold HBS and collected by centrifugation (10,000 g for 5 rain). Pellets were resuspended in 200/zl of a homogenization buffer containing 50 m M Tris (pH 7.4), 5 m M
TABLE I Identification of NGF but not EGF binding sites in primary cultures of chick neural crest cells
[12~I]NGF binding: 130,000 dpm (20 ng) of [12~I]NGFwas added to each culture in 1 ml of medium. [125I]EGF binding: 628,800 dpm (5 ng) of [lZ5I]EGFwas added to each culture in 1 ml of medium. Culture
[125I]NGF binding 1 2 3 4 5 [12~I]EGF binding 1 2 3 4 5
Additions
Amount bound
% Specific binding
dpm
dpm/l~g protein
None None None 5 pg/ml NGF 5/~g/ml NGF
1946 1262 1826 592 733
78.8 97.1 63.4 25.6 17.8
72 79 65 ---
None None None 5/tg/ml EGF 5 #g/ml EGF
658 416 527 496 644
13.9 13.3 14.8 15.8 9.7
None detected
133 TABLE II Specificity of [a251]NGF binding in primary cultures of chick neural crest cells Culture
1 2 3 4 5 6 7 8 9
Additions
None None None NGF (5 pg/ml) NGF (5 #g/ml) Insulin (50 #g/ml) Insulin (50/~g/ml) Cytochrome-C (50 Fg/ml) Cytochrome-C (50/~g/ml)
Amount bound dpm
dpm/l~g protein
1597 1401 1294 592 600 1360 1586 1487 1279
13.1 11.2 10.5 4.5 4.6 11.2 14.1 11.6 10.6
dithiothreitol, and 40 /~M pyridoxal-5'-phosphate and the suspension was frozen on dry-ice for storage at --70 °C overnight. Cell suspensions were thawed and disrupted by sonication. After centrifugation (30,000 g for 30 rain) 75/A portions were brought up to 150/zl with the addition of reaction buffer containing 100 mM Tris (pH 7.4), 10 mM dithiothreitol, 80/zM pyridoxal phosphate, and 6 mM EDTA. The reaction buffer also contained 1.0 /~Ci of L-[114C]ornithine (59 Ci/mmol, Amersham Searle, Arlington Heights, IL) at a final concentration of 150/zM. The reaction mixtures were incubated in rubber-stoppered tubes fixed with hanging center wells containing 200 #1 of hyamine hydroxide for 2 h at 37 °C. The incubations were terminated by the addition of 0.5 ml of 2.5 N H9.SO4 and then allowed to equilibrate at 37 °C for an additional 2 h. Center wells containing the enzymatically liberated 1aco2 were counted in 15 ml of Liquifluor (New England Nuclear, Boston, MA).
centration-dependent displacement of [125I]NGF binding to neural crest cells with the displacement curves generated using NGF-responsive PC12 tumor cells revealed similar binding characteristics for both cell populations and apparent Kas of approximately 0.3 nM (Fig. 1). However, PC12 cells yielded a total specific binding of 60.7 dpm/#g of cell protein while neural crest cells exhibited substantially less (7.1 dpm/Fg of cell protein). The morphological differentiation of trunk neural crest cells can be directed, as described by Greenberg et alp for cranial neural crest cells, by the sera used to grow the cells. Fetal bovine serum (15~o) induces the pigmented melanocytic phenotype and NEURAL CREST CELLS
PC 12
CELLS
]00-
o. . . . .
K d -~o n g l m l
I . . . . . .
RESULTS Primary cultures of chick trunk neural crest cells that had been maintained in tissue culture for 5 days possessed specific binding sites for [125I]NGF but not for [z2SI]EGF (Table I). The binding of [1251]N G F was reduced by the addition of a 1000-fold excess of unlabeled N G F revealing specific binding which accounted for more than 60 ~ of the total binding (Table II). Neither cytochrome-C nor insulin in large excess reduced the binding of [1251]N G F to neural crest cells. Comparison o f the con-
I I ~0 ~0o
• 5~, nglml NGF
I 50
I 10G
I ~o0 ng/ml NGF
Fig. 1. Comparison of the specific binding of [x~5I]NGF to primary cultures of chick neural crest cells and to PC12 tumor cells. Cultures of neural crest cells derived from 7-8 neural tube explants per 35 mm well or PC12 tumor cells grown in 25 cm~ tissue culture flasks were incubated with [z~SI]NGF (approximately 8 ng/ml, 130,000 dpm/ml) in the presence of increasing concentrations of unlabeled NGF. Values are the means of duplicate determinations and are expressed as a percentage of the total binding observed in the absence of unlabeled NGF.
134 horse serum ( 1 5 ~ ) produces a neuronal morphology. After 10 days culture under these conditions, the neuronal (Fig. 2A) and the pigmented cell (Fig. 2B) phenotypes were obtained. The binding of [125I]NGF by these two divergent phenotypes was comparable; the pigmented cells exhibited binding of 15.4 dpm/#g of cell protein and the neuronal cells had 11.8 dpm/#g of cell protein bound.
TABLE III 77~e effects o f N G F o n the levels o f ornithine deearboxyh~se in neural crest cells
Cultures were treated for 6 h. Values are means of duplicate determination. Experiment
Treatment
O DC activity (Jhtol 14C O~/l~gprotein/ 2 h)
1
None NGF (20 ng/ml) None NGF (20 ng/ml) None NGF (20 ng/ml)
82 317 155 22"7 76 174
2 !
3
i/~i!
Despite the presence of specific, high-affinity N G F binding sites and the persistence of these binding sites during the morphological divergence of neural crest cells, treatment of neural crest cells with 50 ng/ml of N G F for up to 20 days did not induce the morphological differentiation which can be obtained in P C I 2 cells. While the cells were able to express the neurite-bearing neuronal phenotype, neither the extent nor the speed of this response was influenced by N G F . However, N G F treatment did elicit an elevation in O D C activity in neural crest cells (Table III). While this 2- to 3-fold induction of O D C is relatively modest in comparison to the response obtained in P C I 2 cells 11, it suggests that the specific binding sites for N G F are, in fact, functional receptors. DISCUSSION
Fig. 2. The effects of different sera on the morphological development of trunk neural crest cells. A: primary cultures of chick neural crest cells were grown in the presence of 15 horse serum, conditions which induced the appearance of a predominantly neural phenotype. Cultures exhibited a total specific binding of [x~5I]NGFof 11.8 dpm//~g of cell protein. B: neural crest cells grown in 15~ fetal bovine serum, conditions which induced the appearance of a predominantly pigmented cell phenotype. Cultures exhibited a total specific binding of [12'~I]NGFof 15.4 dpm//~g of cell protein. Magnification : 125 ×.
Unlike the NGF-responsive PC12 tumor cell line, neural crest cells do not have specific binding sites for the peptide growth regulator EGF. Thus, our previous observation that the NGF-induced differentiation of PCI2 cells is accompanied by a 90~o reduction in functional E G F receptors 11 may not hold significance for the ontological development of N G F target neurons. Rather, the observation may simply reflect the transition of the proliferating PC12 tumor cell to a quiescent neuronal phenotype. In the presence of N G F , the cells may assume a more normal phenotype and lose some of their tumorigenic characteristics, one of which may be the presence of E G F receptors.
135 In the light of the extensive evidence that demonstrates that N G F fails to alter the biochemical or morphological development of neural crest cultures 7,13, the identification of specific [I~51]NGF binding sites on neural crest cell cultures was unexpected. Again, as described in these previous reports, N G F did not alter the gross morphology of neural crest cultures or induce the extension of neurites. If anything, N G F seemed to have a deleterious effect on neural crest cell cultures and suppressed the spontaneous appearance of the neuronal phenotype (data not shown). Such a finding is difficult to reconcile with the widely documented neurotrophic effects of N G F on target neurons derived from neural crest cells. N G F is an absolute requirement for the normal development and survival of sympathetic and sensory neurons s. However, Norr 13 observed a similar phenomenon in crest cell cultures exposed to N G F prior to their interaction with the neutrophic influence of the surrounding mesoderm and notochord tissue. This failure of N G F to elicit a morphological response would not seem to be due to peculiarities in the crest cell N G F receptor, since it displayed an affinity for N G F similar to that of the PC12 cell receptor. Although the relative binding of N G F to crest cell cultures, expressed as a function of total cell protein, was approximately 10 ~ of the binding to PC12 cell cultures, the crest cell cultures were clearly not a homogeneous population of cells, as are the PCI2 tumor cells. Thus, the presence of noncrest derivatives could have diluted out the binding. This dilution with unresponsive cells might also explain the rather modest elevation of ODC activity; the response might be quite a bit larger in some o f the cells, but others might not be responding at all. The fact that the N G F binding sites were coupled to ODC induction, an induction characteristic of REFERENCES 1 Bocchini, V. and Angeletti, P. U., The nerve growth factor: purification as a 30,000 molecular weight protein, Proc. nat. Acad. Sci. U.S.A., 64 (1969) 787-794. 2 Bradford, M. M., A rapid and sensitive method for the quantitation of microgram quantities of protein using the principle of protein-dye binding, Analyt. Biochem., 72 (1976) 248-254. 3 Bronner-Fraser, M., Sieber-Blum, M. and Cohen, A. M.,
NGF-responsive cells 11, indicates that the N G F binding sites represent functional receptors. They do not, however, influence the choice of phenotype. Furthermore, when cells were induced to differentiate into the divergent pigmented, melanocytic phenotype and the neuronal phenotype, the levels of N G F binding were unchanged. Thus, N G F receptors may not be unique to neural crest neurons, but, rather, may be expressed by several crest derivatives. The presence of N G F receptors in the pigmented cell phenotype is consistent with reports of N G F receptors in melanomas TM. Further studies on the presence of N G F receptors in normal melanocytes may prove enlightening. In conclusion, trunk neural crest cells express N G F receptors independently of their phenotype, but do not have their initial development regulated by this peptide. Thus, N G F may not be the primary signal for the differentiation of sympathetic and sensory neurons, but, rather, may participate later in the development after the expression of the neuronal phenotype and after the biochemical features of this phenotype are linked to the N G F receptor system. Alternatively, the procedures used to prepare neural crest cells for tissue culture may be so traumatic that N G F receptors become uncoupled from their normal morphological and biochemical responses. Thus, normal differentiation in response to N G F is prevented. This latter possibility can probably be discounted, however, on the grounds that when cultured neural crest cells are reinjected into embryos, the cells undergo normal differentiation 3. Therefore, the normal ontogenesis of neural crest cells would seem to require the coupling of N G F receptors to appropriate biological responses. ACKNOWLEDGEMENT D.E. is recipient of a National Research Service Award 1-F32-HD05905-1. Clonal analysis of the avian neural crest: migration and maturation of mixed neural crest clones injected into host chicken embryos, J. comp. Neurol., 193 (1980) 423434. 4 Cohen, A. M. and Konigsberg, I. R., A clonal approach to the problem of neural crest determination, Develop. BioL, 46 (1975) 262-280. 5 Cuatrecasas, P., Insulin-receptor interactions in adipose cells: direct measurement and properties, Proc. nat. Acad. Sci. U.S.A., 68 (1971) 1264-1268. 6 Dichter, M. A., Tischler, A. S. and Greene, L. A., Nerve
136
7
8
9
10
I1
growth factor-induced increase in electrical excitability and acetylcholine sensitivity of a rat pheochromocytoma cell line, Nature (Lond.), 268 (1977) 501-504. Fauquet, M., Smith, J., Ziller, C. and LeDouarin, N. M., Differentiation of autonomic neuron precursors in vitro: chotinergic and adrenergic traits in cultured neural crest cells, J. Neurosei., 1 (1981) 478492. Gorin, P. D. and Johnson, E. M., Experimental autoimmune model of nerve growth factor deprivation: effects on developing sympathetic and sensory neurons, Proc. nat. Acad. Sci. U.S.A., 76 (1979) 5382-5386. Greenberg, J. H., Foidart, J. M. and Greene, R. M., Collagen synthesis in cultures of differentiating neural crest cells, Cell. Diff, 9 (1980) 153-163. Huff, K. R. and Guroff, G., Nerve growth factor-induced reduction in epidermal growth factor receptors in PC12 cells: an aspect of cell differentiation, Bioehem. Biophys. Res. Commun., 89 (1979) 175-180. Huff, K., End, D. and Guroff, G., Nerve growth factorinduced alteration in the response of PC12 pheochromocytoma cells to epidermal growth factor, J. Cell Biol., 88 (1981) 189-198.
12 LeDouarin, N. M., The ontogeny of the neural crest in avian embryo chimaeras, Nature (Lond.), 286 (1980) 663-669. 13 Norr, S, In vitro analysis of sympathetic neuron differentiation from chick neural crest cells, Develop. Biol., 34 (1973) 16-38. 14 Pegg, A. E. and Williams-Ashman, H. G., Biosynthesis of putrescine in the prostate gland of the rat, Biochem. J., 108 (1968) 533-539. 15 Savage, C. R., Jr. and Cohen, S., Epidermal growth factor and a new derivative, J. biol. Chem., 247 (1972) 7609-7611. 16 SheJwin, S. A., Sliski, A. H. and Todaro G. J., Human melanoma cells have both nerve growth factor and nerve growth factor-specific receptors on their cell surfaces, Proc. nat. Acad. Sci. U.S.A., 76 (1979) 1288-1292. 17 Tischler, A. S. and Greene, L. A., Nerve growth factorinduced process formation by cultured rat pheochromocytoma cells, Nature (Lond.), 258 (1975) 341-342. 18 Tuszynski, G. P., Knight, L., Piperno, J. R. and Walsh, P. N., A rapid method for removal of [l~sI]iodide following iodination of protein solutions, Analyt. Biochem., 106 (1980) 118-122.