Ionic behaviors and nerve growth factor dependence in developing embryonic chick ganglia. I. Studies with intact dorsal root ganglia

Developmental Brain Research, 3 (1982) 419428

419

Elsevier Biomedical Press

IONIC BEHAVIORS A N D NERVE G R O W T H F A C T O R D E P E N D E N C E IN D E V E L O P I N G E M B R Y O N I C C H I C K G A N G L I A . I. STUDIES W I T H I N T A C T D O R S A L ROOT G A N G L I A

STEPHEN D. SKAPER, IVAN SELAK* and SILVIO VARON Department of Biology and School of Medicine, University of California at San Diego, La Jolla, CA 92093 (U.S.A.)

(Accepted July 17th, 1981) Key words: sodium pump -- cations - - nerve growth factor -- sensory ganglia -- development

SUMMARY We have recently shown that intact and dissociated 8-day embryonic (E8) chick dorsal root ganglia (DRG) lose the ability to regulate their intracellular Na +, K ÷ levels when deprived of nerve growth factor (NGF) for 6 h; recovery occurs within minutes of N G F presentation. These ganglionic neurons are believed to depend on N G F for survival and neurite production over a defined period of embryonic life - - between about E6 and El5 in the chick. Using intact D R G from E6-E16 chick embryos we determined developmental changes in: (i) 22Na+ accumulation in the presence and absence of NGF, or in the presence of ouabain; and (ii) intra- and extracellular fluid spaces. Sodium accumulation, in the presence of N G F , increases from E6 to El0. It parallels the total fluid space under ouabain but then decreases conspicuously between El0 and El6, despite little change in the latter. N G F thus prevents Na + accumulation during the early period, and becomes increasingly irrelevant for this behavior in later (after El0) development. These data are interpreted as indicating that: (i) N G F is required for ionic control by D R G neurons up to El0; and (ii) indigenous behaviors for the control of ion pump mechanism(s) are progressively acquired by these cells from El0 to El6, in parallel with the decreasing ionic relevance of NGF. These findings are consistent with the view that the ionic responses to N G F correlate closely with the survival and neurite-promoting effects of this factor.

* Present address: Department of Neurology, University of Liege, Belgium. 0165-3806/82/0000-0000/$02.75 © Elsevier Biomedical Press

420 INTRODUCTION Nerve growth factor (NGF) is traditionally known for its ability to support neuronal survival and to stimulate neuritic production in dorsal root ganglia (DRG) and sympathetic ganglion neurons2,4,13,14. After 30 years of investigation little is yet known about the mechanisms by which N G F exerts its neuronotrophic activities. We have recently demonstrated with 8-day embryonic (E8) chick DRG dissociates that N G F deprivation over 6 h results in the loss by these cells of the ability to control intracellular levels of Na ÷ and K ÷ (and the corresponding ionic gradients across the cell membrane) 7-10. Administration of N G F after 6 h results in full restoration of ionic control, within minutes. Specifically, the cells extrude the accumulated Na + and reestablish high intracellular K ÷ levels. N G F may regulate the activity of a Na, K-pump or equivalent mechanism 1°. Studies in other laboratories have clearly shown a developmental dependence, in chick embryo DRG, of certain biological responses to N G F administration in vitro. For example, chick D R G of increasing embryonic age display decreasing neuritic responses to NGF in explant culture 24. N G F surface receptors on D R G cells decrease in number with embryonic age 3. Lastly, survival in culture of dissociated DRG neurons is supported by N G F only over a limited developmental period, after which other neuronotrophic factors replace N G F L It is therefore conceivable that the difference in ionic behaviors between NGFtreated and NGF-deprived D R G of E8 chick will also diminish with increasing embryonic age. Such a change could reflect: (i) a decreasing capability of N G F to control ionic behaviors in the older DRG; or (ii) an increasing capability of untreated D R G to maintain their intracellular ion balance in the absence of NGF. In this report, we have examined Na + accumulation in vitro by intact chick D R G from 6- to 16-dayold embryos in the presence and absence of NGF, or in the presence of ouabain. Data were also obtained on developmental changes in intra- and extracellular ganglionic spaces, which were then used to further interpret the data on Na + behaviors. The results suggest that older D R G progressively acquire an indigenous capability to regulate their internal Na + environment. MATERIALS AND METHODS Sodium-22 (carrier-free 22NaCl, 4 mCi/ml), [u-aaC]sucrose (673 mCi/mmol), and tritiated water (31-120; 1 mCi/g) were purchased from New England Nuclear. Tris, Hepes (N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid), ouabain, and bovine serum albumin were from Sigma. Nerve growth factor, 7S form, was prepared from the submaxillary glands of adult male mice as previously described and its concentrations expressed in Biological Units (BU)/ml, according to the traditional bioassay 2°. All chemicals were reagent grade. Measurement of 22Na+ accumulation

Dorsal root ganglia were dissected from White Leghorn chick embryos at

421 embryonic days (E) 6-16. Five to ten dissected D R G were transferred to 12 × 75 mm plastic tubes containing 125 #1 of a Tris-Hepes albumin medium, THAM (40 mM Tris/Hepes pH 7.4, 140 mM NaC1, 5 mM KC1, 1 mM MgCI2, 0.1 mM CaClz, and 1 (w/v) bovine serum albumin), and in some cases also N G F or ouabain. Z2NaC1 was added at 7.5 #Ci/ml, and incubation carried out at 37 °C. After 6 h the D R G were processed for determination of radioactivity as previously describedL The 22Na+ accumulation can be translated into 'Na spaces', on the assumption that the same final concentration of Na + is attained inside the ganglia as in the incubation medium. Na 'space' (nl) -- (accumulated cpm)/(cpm/nl medium). Measurement o f ganglionic water spaces Replicate sets of D R G (5-10 ganglia each) were incubated, as described above, in T H A M medium to which tracers of [U-14C]sucrose and ~H20 (4 #Ci/ml final concentration) were added. Tracer accumulation was also measured after 6 h incubation at 37 °C, and tracer 'spaces' (nl) calculated as for Na 'spaces'. 3H20 measures total fluid space, and [14C]sucrose the extracellular fluid space; their difference is the intracellular fluid space. Protein measurement The method of Lowry et al. 5 was used, with bovine serum albumin as the standard.

RESULTS Na + accumulation by DRG at different ages Dorsal root ganglia from E6-E16 were incubated for 6 h in ~2Na+-containing T H A M , with or without 10 BU/ml of NGF, or with 1 mM ouabain. Accumulation of 22Na+ reached a steady state during this period; higher concentrations (100 BU/ml) of N G F gave similar results. The results of 4 experiments (2 experiments with ouabain) using independent age series of chick embryos are shown in Fig. 1. NGF-deprived DRG, in the absence of ouabain, accumulated increasing amounts of 22Na+ over the E6-E10 period. This trend was conspicuously reversed between El0 and El6. The earlier increase (E6-10) paralleled that observable under treatment with ouabain, as expected if Na+,K+-pump activity in younger D R G neurons is suppressed in the absence of N G F 1°. The small differences over the E6-E 10 span between ouabain-treated and NGF-deprived D R G probably reflect ouabainsensitive but NGF-independent cells (e.g. nonneurons). The later decline (E10-E16) did not result from a progressive reduction of intracellular space, since Z2Na+ accumulation with ouabain remained at a stable level over the same period. It therefore appears that between El0 and El6 the D R G progressively acquired an indigenous ability to sustain their control of ouabain-sensitive intracellular Na +. NGF-treated D R G failed to accumulate 22Na+ dramatically in the E6-E10 period. The 2ZNa+ content, however, increased over that period, despite the presence of N G F . This increase could reflect: (i) an increase in extracellular space (and intracel-

422

+ Ouabain [-NGF1 1600

............l ...........I ............I 1200

.~

800

u

400

~L

E

6

8

I

I

I

10 12 14 Embryonic Age, Days

I

16

Fig. 1. Accumulation of 22Na+ by E6-E16 chick DRG treated with NGF, without NGF, or with ouabain. Ganglia of different ages were incubated (5 DRG/125 #1 THAM) with ~eNa+ (7.5 #Ci/ml) for 6 h at 37 °C. Samples contained: {O) 10 BU/ml NGF; (11) no NGF; ((3) 1 mM ouabain without NGF. The DRG were processed for radioactivity measurement as describedTM. Each point represents the mean ~ S.D. (bars) for 4 values from each of 2 experiments (ouabain curve) or 3 values from each of 4 experiments (± NGF curves).

lular spaces not subject to N G F control); or (ii) a decreasing ability of N G F to regulate an ionic pump in D R G neurons. At the older ages (E12-16), Na ÷ accumulation in NGF-treated ganglia leveled off, as it did in ouabain-treated DRG, suggesting no further expansion of Na÷-accessible spaces. Relationships among the various Na + behaviors The developmental changes revealed by ~2Na+ accumulation under ouabain or N G F treatments relative to untreated D R G were examined further. Age-related differences among the curves in Fig, 1 are plotted in Fig. 2. Curve a in Fig. 2 displays the differences in 22Na+ accumulation between NGF-deprived and NGF-treated D R G , i.e. the developmental changes in net effects of N G F treatment. The amounts of N a ÷ accumulation controlled by N G F increased in the E6-E10 period. Since a similar rise occurred in ouabain-treated cells (compare ouabain- and NGF-treated behaviors in Fig. 1), the increase in N G F effect presumably reflected an expansion in the volume of NGF-responsive neurons rather than an increase in their N G F sensitivity. The sharp decrease between El0 and El6 illustrates the consequences of a nearly constant ZZNa+ accumulation with N G F and a declining one in its absence. This dramatic decrease in net effects of N G F did not reflect a decrease in the ability of N G F to suppress N a + accumulation (since N a + accumulation in the presence of N G F did not increase). Rather, it derived from the apparent increase in the ability of older D R G to control their intracellular N a + independent of exogenous N G F . The latter point is more explicitly demonstrated by curve b, which represents the difference between 22Na+ accumulation in ouabain-treated D R G and in untreated D R G . Over the E6-E10 span, the

423

®

800

600 E o. U 400 = E

20O

~I 6

I 8

I I 10 12 E m b r y o n i c Age, Days

I 14

I 16

Fig. 2. Differences in the Na + behaviors displayed in Fig. 1. The values for 2ZNa+ accumulation obtained at different embryonic ages for the several treatments described in Fig. 1 were subtracted from one another to give the curves shown. Curve a: ( 0 ) difference between NGF-deprived and N G F treated DRG. Curve b: ( 1 ) difference between ouabain-treated (no N G F ) and NGF-deprived DRG.

DRG displayed a constant difference under ouabain treatment, suggesting that the contribution by NGF-insensitive (but ouabain-inhibitable) cells did not vary over that period. Beyond El0, the marked and progressive increase in 22Na+ accumulation imposed differentially by ouabain depicts the concurrent rise of an indigenous, ouabain-sensitive ionic control mechanism. 40

20

30

1:! 15

4

E.C.S.

•

10

'd

I 8

I 10

I 12

[ 14

I 16

Embryonic Age, Days

Fig. 3. Developmental changes in fluid spaces and protein content of chick D R G . Ganglia from ages E6-E16 were incubated (5-10 DRG/125 /~1 T H A M ) with [U-t4C]sucrose (4 #Ci/ml) and sH20 (4 #Ci/ml) for 6 h at 37 °C, with 10 BU/ml N G F or without N G F , and no ouabain. The D R G were processed for their radioactivity contents and the different volumes ('spaces') determined as described under Methods. Similar results were obtained with NGF-treated and -deprived D R G , and the 2 sets of data were pooled.O, total fluid space (TFS); l , extraceilular space (ECS); A , intracellular space (ICS); (3, protein content. Each point for fluid spaces represents the mean 4- S.D. (bars) for 3 values from each of 3-5 experiments. Protein values were derived from analyses on 1 or 2 D R G homogenates.

424 T. F S + ouabain

40

Na space + ouabain T.F.S,

30

i".. ~

h=no NGF

t

=~/

.aspa¢o

E 20 ,,"

o

E C

,o ~ / t I

6

I

8

Na space

/

,o

B*.GF

1;

,

,,

1'

Ernbtyonic Age. Days

Fig. 4. Developmental changes in DRG spaces traced with ~2Na: effect of NGF and ouabain treatments. A: (upper curves), the fluid spaces traced with 2ZNa+, in the absence of NGF under ouabain treatment ( I ) or no ouabain (A), were obtained by converting the data from Fig. 1 (cpm/ DRG) to nl/DRG (see Methods). The arrow for the Na + space curve in the absence of ouabain indicates the direction for this parameter after El0 (see Fig. 1). Total fluid space in the absence of ouabain (---) was taken from Fig. 3. Total fluid space was also determined in the presence of 1 mM ouabain (©). The shaded areas indicate the difference between total fluid and Na+ spaces in the presence or absence of ouabain. B: (lower curves), the fluid space traced with ~Na + for NGF-treated DRG~O) was taken from the data in Fig. 1, and converted to nl/DRG. Extracellular space (---) was taken from Fig. 3. Error bars were omitted for the sake of clarity (see Figs. 1 and 3 for the respective standard deviations).

Developmental changes in ganglionicfluid spaces Several of the interpretations drawn for the 22Na+ experiments might be better defined if the developmental changes in total, intra- and extracellular fluids which the D R G undergo in vivo were measured. Using radiolabeled, membrane impermeant molecules (e.g. inulin or sucrose) the extracellular space (ECS) can be determined after tissue equilibration (see Methods). Total fluid space (TFS) can be similarly measured after equilibration with 3H20. The intracellular space (ICS) is the difference of the 2 former values. Here, D R G sets from each age were incubated for 6 h in THAM-containing [14C]sucrose and 3H20, with or without NGF. No significant differences were found in the final measurements for NGF-treated and NGF-deprived DRG. The average results, from 3-5 independent experiments, are shown in Fig. 3. Total fluid space increased by only 20 ~ between E6 and E8, but by 150 ~ from E9 to El2; no further increase was noticeable after El2. This pattern correlates well with the developmental increase in protein content (broken line in Fig. 3) and overall size (not shown). Both ICS and ECS contributed to the total fluid increase: ICS was the larger component, being nearly the same as ECS at E6-E9 but becoming 1.6-fold greater than ECS by El2. These results confirmed that the increases in 22Na+ accumulation observed over approximately the same age span (cf. Fig. 1) reflected mainly developmental expansions of fluid compartments accessible to exogenous N a +.

425

Comparison of Na spaces and fluid spaces 22Na+ may also be used as a 'tracer' for Na+-accessible spaces. As with [14C]sucrose and 3H~O, one must make the assumption that the 22Na+ accumulated at equilibrium reaches the same concentration in the fluids it occupies as it has in the incubation medium. It is possible to calculate the Na ÷ spaces corresponding to the data of Fig. 1, and to compare them with the fluid spaces reported in Fig. 3, as well as with an additional TFS curve obtained from ouabain-treated DRG. Fig. 4 presents 2 sets of comparisons. The first comparison (Fig. 4B) is between the extracellular space and the Na + space derived from NGF-treated DRG. To the extent to which N G F ensures full activity of the Na +, K+-pump in its target neurons, 22Na+ should only have access to the extracellular fluid (plus small amounts in the intracellular space due to normal Na + exchange processes, and in cells irreversibly damaged by the experimental procedure) and, therefore, occupy essentially the same space as does [14C]sucrose. This, in fact, was the case for the entire developmental period examined. The second comparison (Fig. 4A) is drawn between Na ÷ spaces and total fluid spaces in the absence of N G F : (i) in the presence of ouabain, Na ÷ should be free to acquire in both the intra- and extracellular compartments the same concentration as in the external medium. Sodium spaces should therefore match the TFS measured with 3H20. This was indeed the case at all ages, with the exception of E6; (ii) in the absence of ouabain, N G F deprivation should result in, over the E6-E10 age span, an inactivation of existing Na ÷, K+-pump mechanisms similar to that imposed by ouabain. Agerelated changes in Na ÷ space under such conditions should then, firstly, parallel those observed in the TFS of ouabain-treated D R G (measured by either 22Na or 3H20) and, secondly, coincide with the TFS measured by 3H20 in the absence of ouabain. The experimental data confirm the first prediction, but not the second one (see Discussion). Note that this comparison is of necessity confined to the E6-E10 age span. Beyond El0, Na ÷ spaces can no longer be assumed to measure intracellular as well as extracellular fluid (Fig. 4A, arrow), in view of the apparent development of indigenous control of Na ÷ accumulation (cf. Figs. 1 and 2); and (iii) comparison of these various curves also reveals that both ~2Na+ and 3H20 total fluid spaces were generally higher in the presence than in the absence of ouabain, over the developmental period (E6-E10), during which they undergo a substantial increase. DISCUSSION The discovery that N G F regulates Na + and K + control in its target ganglionic neurons s-l° provides 2 directions for future study TM. One concerns the molecular steps through which N G F achieves the regulation of Na +, K+-pump mechanisms11, lz. The second direction, addressed in part here, involves possible relationships that might exist between the ionic effects of N G F and the trophic ones for which N G F is traditionally known, in particular the support of neuronal survival. The present study provides the first evidence that the ionic response to N G F changes with the developmental age of the chick DRG, and does so along the very

426 temporal patterns that have been reported for other D R G responses to NGF1,3, 24. This finding is explicitly depicted in Fig. 2 (curve a). The net effect induced by N G F on Z2Na+ accumulation increases between E6 and El0, then decreases dramatically between El0 and El6. It would appear, then, that the NGF-related Na ÷ behavior fits precisely the traditional views that N G F becomes increasingly: (i) important for D R G neurons over their early development; and (ii) irrelevant in later development (cf. refs. 1, 2). The disappearance of an overt response to N G F need not reflect exclusively a loss of cellular responsiveness to N G F (e.g. loss of surface receptorsZ), since it could equally result from the loss of a dependence on N G F (with receptors still present). In dissociated cell cultures of older embryonic chick DRG, neuronal survival is less and less dependent on N G F but also fails in its absence 1. This suggests a decreased competence of N G F with a continuing need for some trophic support. Indeed, it has been shown that survival of more mature chick embryo D R G neurons can be promoted by other neuronotrophic factorsa,5,1z,zL The situation appears to differ in intact DRG, however. Sodium control at older embryonic ages is progressively less affected by NGF, but also increasingly occurs in the absence of N G F (cf. Fig. 1). The development of an NGF-independent competence for ionic control could be due to: (i) an acquisition of complete self-sufficiency (i.e. independence from any extrinsic influence); or (ii) an increased intraganglionic availability of the factor(s) responsible for ionic regulation, whether this be N G F itself or any one of the new ganglionic neuronotrophic factors recently discovered 15. In either case, the indigenous source could be the nonneuronal ganglionic cells which are known to supply trophic agents to their homologous neurons z1-~3. By disrupting the intimate relationship that normally occurs within the ganglion between the neurons and nonneurons, it should be possible to determine whether the indigenous ionic control mechanism occurs in isolated neurons from the older D R G and, if not, whether the addition of nonneur onal cells or isolated neuronotrophic factors can restore such ionic control. Experiments are now under way to examine this possible rote for nonneurons or soluble neuronotrophic supplements, using D R G dissociates. Examination of fluid spaces and their comparison to 22Na+-measured spaces provided additional information. There was an increase of both ECS and ICS in D R G of increasing age, as could be expected from the conspicuous developmental increase in ganglionic size. Complete identity was found between extracellular space ([14C]sucrose) and Na + space in NGF-maintained DRG. This confirms that the net effect of N G F on Na + accumulation (Fig. 2, curve a) is exclusively directed to intracellular spaces. Full identity was also found (with the single exception of E6 DRG) between Na + space and total fluid space (3HeO) in ganglia treated with ouabain, demonstrating complete equilibration by the 22Na+ and the absence of Na+-sequestering mechanisms under such conditions. The total fluid spaces measured under ouabain, however, were generally larger than those in the absence of the drug over the E6-EI0 period. Additional work will be needed to evaluate the possibility of a 'swelling' response to ouabain by the younger but not the older DRG. Lastly, Na + and H20 spaces failed to match each other in E6-E10 D R G (though they did so at older ages) in the absence of

427 both ouabain and N G F : N a + spaces were consistently greater than those measured by aHzO. Experiments in which the 2~Na+ and 3H20 accumulations were measured concurrently in the very same ganglia also displayed such a difference, ruling out sampling variabilities. We have, at present, no explanation for this discrepancy and the possible implication of a ouabain-preventable, intracellular N a + sequestration mechanism in these younger ganglia. In conclusion, the present study strengthens the view that the ionic response to N G F closely correlates with the more traditional response to N G F , i.e. neuronal survival and/or neurite elongation. Some important correlations have already been established in previous studies by: (i) the failure, thus far, to sustain ionic control in D R G cells with any other extrinsic agent (serum constituents, hormones) except N G F s; and (ii) the consistent expression of the ionic response in all ganglionic systems displaying survival or neurite responses to N G F , but not in any lacking such N G F responsivenessL In addition, the new data demonstrate a developmental acquisition by intact D R G of indigenous ways to regulate their Na, K - p u m p mechanisms, thereby providing additional support to the concept 15,17-19,2s that ionic control - - whether imposed by N G F or by other means - - may play a critical role for maintenance and differentiation of ganglionic neurons. ACKNOWLEDGEMENTS This work was supported by USPHS Grant NS-07606 from the National Institute of Neurological and Communicative Disorders and Stroke. I.S. was supported by the Neurology Department of the School of Medicine, University of Liege (Belgium). REFERENCES 1 Barde, Y.-A., Edgar, D. and Thoenen, H., Sensory neurons in culture: changing requirements for survival factors during embryonic development, Proc. nat. Acad. Sci. (U.S.A.), 77 (1980) 11991203. 2 Greene, L. A. and Shooter, E. M., The Nerve Growth Factor: biochemistry, synthesis, and mechanism of action, Ann. Rev. Neurosci., 3 (1980) 353-402. 3 Herrup, K. and Shooter, E. M., Properties of the fl-nerve growth factor receptor in development, J. Cell BioL, 67 (1975) 118-125. 4 Levi-Montalcini, R. and Angeletti, P. U., Nerve Growth Factor, PhysioL Rev., 48 (1968) 534-569. 5 Lowry, O. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J., Protein measurement with the Folin phenol reagent, J. biol. Chem., 193 (1951) 265-275. 6 Manthorpe, M., Skaper, S. D., Barbin, G. and Varon, S., Ciliary neuronotrophic factors (CNTFs). Concurrent activities on certain Nerve Growth Factor-responsive neurons, J. Neurochem., in press. 7 Skaper, S. D. and Varon, S., Nerve Growth Factor influences sodium ion extrusion across chick embryonic dorsal root ganglionic neurons, Biochem. Biophys. Res. Commun., 88 (1979) 563-568. 8 Skaper, S. D. and Varon, S., Properties of the Na + exclusion mechanism controlled by Nerve Growth Factor in chick embryo dorsal root ganglionic neurons, J. Neurochem., 196 (1980) 1654-1660. 9 Skaper, S. D. and Varon, S., Maintenance by Nerve Growth Factor of the intracellular sodium environment in spinal sensory and sympathetic cells, Brain Res., 197 (1980) 379-389. 10 Skaper, S. D. and Varon, S., Nerve Growth Factor influences potassium movements in chick embryo dorsal root ganglionic cells, Exp. Cell Res., 131 (1981) 353-361.

428 11 Skaper, S. D. and Varon, S., Na +, K+-ATPase and ouabain binding activities of Nerve Growth Factor-supported and -deprived chick embryo dorsal root ganglia, J. Neurosci. Res., 6 (1981) 133-141. 12 Skaper, S. D. and Varon, S., Mutually independent cyclic AMP and sodium responses to Nerve Growth Factor in embryonic chick dorsal root ganglia, J. Neurochem., 37 (1981) 222-228. I3 Thoenen, H. and Barde, Y.-A., Physiology of nerve growth factor, Physiol. Rev., 60 (1980) 12841335. 14 Varon, S., Nerve Growth Factor and its mode of action, Exp. Neurol., 48 (1975) 75-92. 15 Varon, S. and Adler, R., Trophic and specifying factors directed to neuronal cells, Advanc. Cell. Neurobiol., 2 (1981) 115-163. 16 Varon, S. and Skaper, S. D., Short-latency effects of Nerve Growth Factor: an ionic view. In E. Giacobini, A. Vernadakis and A. Shahar (Eds.), Tissue Culture in Neurobiology, Raven, New York, 1980, pp. 333-347. 17 Varon, S. and Skaper, S. D., Short-latency effects of Nerve Growth Factor deprivation and readministration on ganglionic cells, J. Supramolec. Structure, 13 (1980) 329-337. 18 Varon, S. and Skaper, S. D., Ionic features in the mode of action of Nerve Growth Factor. In R. Rodnight, H. Bachelard and W. Stahl (Eds.), Chemisms of the Brain, Churchill-Livingstone, Edinburgh, 1981, in press. 19 Varon, S. and Skaper, S. D., In vitro responses of sympathetic neurons to Nerve Growth Factor and other macromolecular agents. In R. D. G. Milner and G. Burnstock (Eds.), Development of the Autonomic Nervous System, Ciba Foundation Symposium 83, Pitman Medical, London, 1981, pp. 151-176. 20 Varon, S., Nomura, J., Perez-Polo, J. R. and Shooter, E. M., The isolation and assay of the nerve growth factor proteins. In R. Fried (Ed.), Methods and Techniques of Neuroscience, M. Dekker, New York, 1972, pp. 203-229. 21 Varon, S., Raiborn, C. and Norr, S., Association of antibody to Nerve Growth Factor with ganglionic non-neurons (glia) and consequent interference with their neuron-supportive action, Exp. Cell Res., 88 (1974) 247-256. 22 Varon, S., Skaper, S. D. and Manthorpe, M., Trophic activities for dorsal root and sympathetic ganglionic neurons in media conditioned by Schwann and other peripheral cells, Develop. Brain Res., 1 (1981) 73-87. 23 Varon, S., Manthorpe, M., Skaper, S. D. and Adler, R., Neuronotrophic factors; problems and perspectives. In B. Haber, R. Perez-Polo and J. D. Coulter (Eds.), Proteins of the Nervous System - Structure and Function, Progr. Clin. Biol. Res., Liss, New York, 1981, in press. 24 Winick, M. and Greenberg, R. E., Chemical control of sensory ganglia during a critical period of development, Nature (Lond.), 205 (1965) 180-181.