Neuronotrophic effects of skeletal muscle fractions on spinal cord differentiation

Transcription

1 /. Embryol. exp. Morph. Vol. 71, pp , Printed in Great Britain Company of Biologists Limited 1982 Neuronotrophic effects of skeletal muscle fractions on spinal cord differentiation By L. HSU, 1 D. NATYZAK 1 AND G. L. TRUPIN 2 From the Department of Anatomy, New Jersey School of Osteopathic Medicine and Department of Anatomy, Rutgers Medical School, Piscataway SUMMARY Soluble fractions of homogenized skeletal muscle were found to promote neuronal migration and neuritic and glial outgrowth from embryonic chick spinal cord explants. Fractions obtained from skeletal muscle immobilized by prolonged treatment with curare were significantly more effective than normal muscle in accelerating neuronal and glial development. Fractions from other tissues such as brain and lung did not enhance neuronal differentiation, but were effective in stimulating outgrowth of glial cells. Separate measurements of glial and neuronal responses indicate that tissue fractions produce independent effects on the glial and neuronal components. INTRODUCTION Numerous studies support the concept that peripheral target tissues exert a regulatory and trophic effect upon central neurons. It has been demonstrated that surgical removal of chick embryo limb buds produces hypoplasia of ventral horn neurons (Hamburger & Keefe, 1944). Conversely, implantation of supernumerary limbs reduces natural cell death and enhances survival of motorneurons (Hollyday & Hamburger, 1976). In vitro studies have shown that limb explants (Pollack, 198) and target tissues such as cardiac and skeletal muscle (Giller et al 1977; Nishi & Berg, 1977; Ebendal, 1981) are capable of eliciting neuritic outgrowth and synapse formation in cocultured ganglionic or spinal cord cultures. Conditioned media from the same tissues are also effective in promoting neuronal differentiation (Collins, 1978; Helfand, Riopelle & Wessels, 1978; Bennett, Lai & Nurcombe, 198; Dribin & Barrett, 198; Obata & Tanaka, 198; Coughlin, Bloom & Black, 1981; Henderson, 1 Authors' address: Department of Anatomy, New Jersey School of Osteopathic Medicine, P.O. Box 55, Piscataway, New Jersey 8854, U.S.A. 2 Author's address: Department of Anatomy, Rutgers Medical School, Piscataway, New Jersey 8854, U.S.A.

2 84 L. HSU, D. NATYZAK AND G. L. TRUPIN Hachet & Changeux, 1981), suggesting that this trophic regulation is mediated by diffusible factors. Potential neuronotrophic 1 factors (Varon & Bunge, 1978) have also been identified in homogenized extracts of various tissues and organs. Trophic factors that promote neuronal differentiation have been found in fractions of cardiac and skeletal muscle (McLennan & Hendry, 1978; Lindsay & Tarbit, 1979; Smith & Appel, 1981) brain and gut (Jacobson, Ebendal, Hedlund & Norrgren, 198; Riopelle & Cameron, 1981). Biochemical characterization of factors from these different sources is incomplete (Ebendal, Belew, Jacobson & Porath, 1979; Bonyhady, Hendry, Hill & McLennan, 198; Coughlin et al. 1981). The present study has focused on the trophic effects of homogenized skeletal muscle fractions on differentiation of spinal cord cultures. We have used a semiquantitative assay system to evaluate the morphological differentiation of both glial and neuronal components of embryonic chick spinal cord. To assess the trophic influence of skeletal muscle with impaired functional activity, neuronal cultures were also grown in soluble fractions from muscle treated with curare. Our results demonstrate that normal muscle fractions enhance neuritic and glial outgrowth. Soluble fractions from curarized muscle are significantly more effective than normal muscle in accelerating neuronal and glial development. Fractions from other tissues, such as brain and lung were effective only in stimulating outgrowth of non-neuronal cells. MATERIALS AND METHODS Tissue culture Spinal cords of 8-day chick embryos were dissected under aseptic conditions and stripped of meninges. The ventral halves of cord segments from thoracic to lumbar levels were cut into pieces approximately -5 mm 3, explanted onto collagen-coated coverslips and fed one to two drops of growth medium. These lying-drop preparations were sealed in Maximow chambers and incubated at 37 C. Cultures were refed every 1-2 days for a period of 4-8 days. Curare treatment 1-day chick embryos were injected with -5 ml of 1 mg/ml sterile d-tubocurarine chloride (Sigma) in Ringer's chick saline (Hall, 1975). The solution was injected through a pinhole made in the shell and shell membrane above the airspace. After 5-6 days, the curarized embryos were sacrificed and used for the preparation of muscle homogenates as described below. Curarized embryos were slightly smaller than normal embryos of the same age and their limbs often appeared fixed and stiff at the joints. Samples of skeletal muscle from normal and curarized embryos were assayed for acetylcholinesterase by 1 Neuronotrophic factors are trophic factors from neural or non-neural tissues which are directed towards neurons and supporting cells.

3 Neuronotrophic effects of skeletal muscle fractions 85 the colorimetric method of Ellman, Courtney, Andres & Featherstone (1961). As in previous studies (Hall, 1975; Oppenheim, Pittman, Gray & Maderut, 1978), curarized muscle showed reduced levels (about 8% of normal muscle) of acetylcholinesterase activity. Preparation of tissue fractions Breast and/or thigh muscles from 15- to 16-day normal or curarized chick embryos were dissected, cleared of grossly visible tendinous and neuronal components, and minced to a fine pulp. Approximately -5 gm of muscle were homogenized in 5 ml of Basal Medium Eagle's for 2-3 min at 4 C with a teflon-coated tissue homogenizer. The homogenate was centrifuged at 2 g and at 15 g, each time for 6 min. Particulate components floating at the surface were discarded. The resultant supernatant fraction was sterilized through a -2 /im Sterilet (Amicon) filter and stored at -1 C. Supernatant fractions of brain and lung of 15- to 16-day embryos were also prepared as described above. Total protein concentration of the supernatant fractions was determined by the dye-binding assay of Bradford (1976) with bovine plasma albumin (BioRad) as the standard. Composition of growth media Spinal cord explants were maintained in experimental and control media as indicated in Table 1. Two controls were used to provide both a baseline standard (Group A) and a maximal growth standard (Group D). Group D cultures were grown in a medium enriched with large quantities of serum and embryo extracts (GIBCO) that was reported to promote extensive differentiation of spinal cord explants (Fisher & Federoff, 1977). Group A control cultures were grown in a medium containing moderate proportions of serum and embryo extract formulated to ensure healthy but not maximal neuronal differentiation and glial development (Federoff & Hall, 1979). Experimental Groups B and C were grown in the same defined medium as Group A supplemented with supernatant fractions of normal or curarized muscle. In groups B and C, the amount of serum and chick embryo extracts was reduced so that the total protein content approximated that of the baseline standard (see Table 1). This standardization of protein content ensured that observed differences in neuronal development were not due solely to generalized nutrient effects of the protein components. To establish that observed differences in neuronal development were specific responses to skeletal muscle components, cultures grown in muscle fractions were compared with cultures grown in tissue fractions from lung and brain (Group E).

4 oo HSU Culture groups Control group A (baseline control) Group B (normal muscle) Group C (curarized muscle) Control group D (enriched control) Group E H Table 1. Composition of growth media Defined medium* Heatinactivated horse serum Chick embryo extract Basal medium eagle's 1% 2% Basal medium eagle's 5% 1% Basal medium eagle's 5% 1% Medium 199 with 6 mg % glucose Basal medium eagle's 32% 5% * All media contained 5 /*g/ml gentamicin. Homogenized tissue fraction Normal muscle Curarized muscle 1% normal muscle 1% lung 1% brain Total protein content (/*g/ml) 23 ±4 22 ±14 23 ±3 32 ± N I y Z p H a z

5 Neuronotrophic effects of skeletal muscle fractions 87 Table 2. Rating system for quantifying growth of spinal cord explants Growth parameters Score Neuronal migration - number of neurons beyond original explant Neuritic outgrowth - number of fibres, outgrowth beyond original explant Glial outgrowth - outgrowth beyond original explant None neurons neurons 3 21 or more neurons No fibres Fibres sparse, outgrowth = explant diameter Fibres moderately dense, outgrowth = explant diameter Fibres abundant, outgrowth = 2 or more times explant diameter No glial outgrowth Glial outgrowth = explant diameter Glial outgrowth = explant diameter Glial outgrowth = 2 or more times explant diameter Rating of growth in spinal cord explants After 4 or 8 days, cultures were fixed in 1% neutral buffered formalin and stained with silver (Holmes, 1943). Explants were examined with the light microscope and rated for three growth parameters to provide a measure of neuronal and glial development (Table 2). For each growth parameter, cultures were assigned a numerical score of -3. The frequency distribution of these scores served as an index of the degree of growth and differentiation achieved by different culture groups. Frequency distribution of scores was subjected to the Chi-square test. The overall significance of difference in the distribution of scores between groups for each growth parameter was tested. Each group was checked against other control or experimental groups (e.g. Group A against B, C, D, respectively) in order to compare and evaluate the growth profiles and developmental trends produced by the different media. RESULTS Group A cultures - baseline control medium (Table 3) The morphological differentiation of ventral horn cultures grown in baseline control medium becomes evident by 4 days in vitro (DIV), and shows further progression by 8 DIV. Soon after attachment to the collagen substrate, nonneuronal cells begin to spread out from the core of the explant, followed by an outgrowth of short neuritic fibres and multipolar neurons (Fig. 1). Outward neuronal migration in Group A cultures is relatively slow; at 4 DIV, most cultures (54%) remain dense with no visible neurons beyond the explant (score of ), while at 8 DIV, 31 % of the cultures still remain unspread. In contrast, neuritic outgrowth displays a sharp increase between 4 and 8 days,

6 L. HSU, D. NATYZAK AND G. L. TRUPIN

7 Neuronotrophic effects of skeletal muscle fractions 89 with a much higher frequency of cultures having scores of 3 (1 % at 4 DIV versus 24 % at 8 DIV). Glial outgrowth shows a similar shift towards higher scores between 4 and 8 DIV. Group B cultures - normal muscle fractions {Table 3) As compared with control Group A, Group B cultures show a marked acceleration of early glial outgrowth (Fig. 2), a moderate enhancement of neuritic outgrowth, but no appreciable increase in neuronal migration. Twentytwo percent of the 4-day cultures in Group B display a maximal glial outgrowth (score of 3), while only 1 % of 4-day cultures in Group A show such development (P < 1). This early spurt of glial outgrowth in Group B does not continue, however, and there is relatively little change by 8 DIV. Comparison of neuritic outgrowth reveals almost identical scores for Group A and B at 4 DIV. By 8 DIV however, 73 % of Group B cultures have long fibre outgrowths (scores of 2, 3) while only 61 % of Group A baseline control cultures have comparable growth (P < 5). This enhancement of neuritic outgrowth produced by normal muscle fractions in Group B media was in fact equal to the dense outgrowth produced in Group D cultures grown in enriched media (29 % of scores of 3 in Group B versus 32 % in Group D). Group B displays a moderate increase in neuronal migration between 4-8 days, but this is not significantly different from trie parallel increase in neuronal migration occurring in Group A control cultures during the same period. Group C cultures - curarized muscle fractions (Table 3) When compared with Groups A and B, Group C cultures showed significant advancement in both neuronal and glial development by 4 DIV. Within Group C, only 28 % of the 4-day cultures showed an absence of neuronal migration (score of ). This is indicative of a rapid initiation of neuronal Fig. 1. Spinal cord explant grown in baseline control medium (Group A) after 4 DIV. A few multipolar neurons (arrows) and short neuritic fibres appear at the edge of the explant (E). Neuronal elements overlie a background of small glial cells, x 5. Scale bar = 2 /mi. Fig. 2. Spinal cord culture grown in normal muscle fraction (Group B) after 4 DIV. The culture shows a marked acceleration of glial outgrowth. A few neurons have migrated out from the explant which was located at lower left, x 12. Scale bar = 83 /im. Fig. 3. Spinal cord explant grown in curarized muscle fraction (Group C) after 8 DIV. A large number of neurons and an extensive neuritic network can be seen. x25o. Scale bar = 18/mi. Fig. 4. Spinal cord explant grown in 1% normal muscle fraction (Group E) after 5 DIV. Two neurons are clearly visible (arrows) and other darkly stained neurons can be seen emerging from the explant (at top). The culture shows extensive neuritic outgrowth, x 55. Scale bar = 18/tm.

8 \J1 \jvv 111 parameter Neuronal migration Neuritic outgrowth Glial outgrowth Table 3. Effect of soluble muscle fractions on differentiation of spinal cord cultures % Distribution by s scores of cultures grouped after 4 DIV* % Distribution of cultures grouped by scores after 8DIVt Culture groups A (baseline control) B (normal muscle) C (curarized muscle) D (enriched control) A B C D A B C D * Number t Number of cultures/group: of cultures/group: A = A = ; B = 185; 186; B = 218; C = 11; D C = 88; D = = 115. = o X d p z H CM p p d

9 Neuronotrophic effects of skeletal muscle fractions 91 Table 4. Effect /1% soluble tissue fractions on differentiation of spinal cord cultures after 5 DIV % Distribution grouped by of culture scores parameter Group E* cultures Neuronal migration Neuritic outgrowth Glial outgrowth 1% normal muscle 1% brain 1% lung 1% muscle 1% brain 1% lung 1% muscle 1% brain 1% lung * Number of cultures/group: normal muscle = 84; brain = 58>; lung = migration, since all other 4-day control and experimental groups have a higher percentage of cultures showing no sign of migration (score of ). This rapid neuronal migration is not significantly different from the extensive neuronal migration in 4-day Group D cultures maintained in enriched medium. By 8 DIV, neuronal migration in Group C (Fig. 3) actually surpasses that of Group D cultures (P < 1). With respect to neuritic outgrowth, Group C cultures are not significantly different from Group D cultures at 4 DIV, but again surpass Group D cultures by 8 DIV (68 % versus 32% of cultures with scores of 3; P < -1). Glial development in both experimental groups that contain muscle fractions (Groups B and C) is more advanced than that of baseline and maximal growth controls (Groups A and D). Between 4 and 8 days, Groups B and C undergo relatively little change, but continue to show more extensive glial outgrowth than Groups A and D. Curarized muscle fractions are more effective than normal muscle fractions in eliciting glial outgrowth at 4 DIV (35% versus 22% with scores of 3; P < 5). In control Groups A and D, only 1 and 4%, respectively, achieved the same scores for glial outgrowth. Group D cultures - enriched control medium {Table 3) Group D medium, enriched with large amounts of serum proteins and embryo extracts, stimulated extensive neuronal differentiation of ventral horn cultures. Cultures differentiate rapidly and show marked neuronal migration and neuritic outgrowths by 4 DIV. Twenty-one percent of the cultures have over 2+ neurons (score of 3), and the majority of cultures show long, dense neuritic fibres. Glial development is more limited, with only 4% of the cultures

10 92 L. HSU, D. NATYZAK AND G. L. TRUPIN having maximal outgrowth (score of 3). After 8 DIV, moderate increases in both fibre and glial development are noted, but neuronal migration remains unchanged. The primary differences between control Groups A and D are the more extensive neuronal migration and neuritic outgrowth produced by the enriched Group D medium (P < 1). There is no significant difference in glial outgrowth in these two control groups. Group E cultures - lung, brain and normal muscle fractions {Table 4) Cultures grown in 1% normal muscle fractions produce significantly higher levels of neuronal migration and neuritic outgrowth (P < 1; Fig. 4) than cultures maintained in 1% brain or lung fractions. After 5 DIV, 95% of cultures grown in lung fractions remain dense with no neuronal migration beyond the core of the explant. In contrast, only 45 % of the cultures grown in 1% normal muscle fractions show an absence of neuronal migration (score of ), and a combined 17% of these latter cultures display at least neurons (scores of 2, 3). Neuritic outgrowth from cultures grown in 1% muscle fractions is moderate, but superior to neuritic outgrowths in cultures grown in brain or lung fractions (P < 1). By 5 DIV, brain, lung and muscle fractions all stimulate a moderate glial extension, with no significant difference in glial response to the three tissue fractions. These cultures showed no further neuronal and glial differentiation after 5 DIV. DISCUSSION The results of this study indicate that soluble fractions from skeletal muscle tissue promote growth and differentiation of embryonic chick spinal cord explants. The migration of multipolar neurons from the core of the explant and the outgrowth of both neuritic fibres and underlying glial cells are enhanced by fractions of curarized muscle and to a lesser extent by normal muscle fractions. This neuronotrophic effect cannot be attributed to a general nutritional response to soluble tissue proteins since fractions of lung or brain did little to promote neuronal migration or neuritic extension. These observations are in agreement with reports that conditioned medium from skeletal muscle cultures promotes neuritic outgrowth (Dribin & Barrett, 198; Obata & Tanaka, 198; Henderson et al. 1981) and enhanced motorneuron survival (Bennett et al. 198). Similar studies on effects of brain extracts have yielded somewhat conflicting results. Tissue extracts from brain and heart have been reported to promote neuritic outgrowth from sensory (Lindsay & Tarbit, 1979) and ciliary ganglia (Jacobson et al. 198). Our study indicates, however, that soluble brain fractions do not significantly affect neuronal migration or fibre outgrowth in spinal cord explants. These latter results are in agreement with a recent survey which showed that soluble brain extracts had a minimal effect on neurite outgrowth in dissociated spinal cord cultures (Riopelle & Cameron, 1981).

11 Neuronotrophic effects of skeletal muscle fractions 93 Studies on target tissue regulation of spinal cord development have usually focused on neuronal differentiation, with little emphasis on accompanying glial development. In our study, we separately evaluated both glial and neuronal responses to potential trophic factors. Our results suggest that muscle fractions produce independent effects on the glial and neuronal components. This view is supported by two observations: 1) both eurarized and normal muscle fractions stimulate early glial development, but only curarized muscle fractions promote extensive neuronal migration; 2) tissue fractions from brain, lung and muscle produce equivalent glial development, but only muscle fractions elicit a significant neuronal response. These observations are consistent with reports that enhancement of neuritic outgrowth by muscle conditioned medium was not blocked when glial proliferation was inhibited by 5-fluoro-2'-deoxyuridine, (Dribin & Barrett, 198). Furthermore, Obata & Tanaka (198) observed reduced glial outgrowth in spinal cord explants at a time when neuritic outgrowth was enhanced by muscle conditioned medium. Neuronotrophic factors may affect differentiation by promoting substrate interactions during neuronal migration and axonal elongation (Varon & Bunge, 1978; Adler & Varon, 1981). Of the many cell types surveyed, only nerve, glial and skeletal muscle cells have been shown to produce significant amounts of substrate attachment material (Schubert, 1977). Similar attachment substances are also found within medium conditioned by cardiac muscle, and have been shown to be critical in contact guidance and control of directionality of neuritic growth (Collins, 1978; Collins & Garrett, 198). Homogenized muscle fractions may also contain substances that facilitate neuronal and glial attachment or act to stimulate proliferation of glial cells (Hanson & Partlow, 1978). We are presently investigating whether skeletal muscle fractions may also have mitogenic effects. Muscle fractions from embryos treated with curare are significantly more effective than normal muscle fractions in promoting neuronal and glial development. Curarized muscle fractions were also more effective in inducing differentiation than an enriched medium specifically designed to promote maximal neuronal development (Fisher & Federoff, 1977). These observations are consistent with recent reports that embryos immobilized with botulinum toxin or curare showed a distinct increase in motorneuron survival during periods of normal cell death (Pittman & Oppenheim, 1978). Blocking physiological activity at the neuromuscular synapse with these drugs produced a delay in motorneuron degeneration, although the mechanism by which peripheral muscle prevented death of central neurons was not identified. Our results, utilizing fractions of homogenized muscle, suggest that an actual component within curarized muscle acts to promote and sustain neuronal differentiation. It remains to be determined whether curarized muscle fractions contain factors specific to curarized tissues or have an altered level of a neuronotrophic component found in normal muscle. Preliminary biochemical analysis of both normal and curarized muscle fractions is now in progress. 4 EMB 71

12 94 L. HSU, D. NATYZAK AND G. L. TRUPIN This investigation was supported by Grant No. RRO9O85-O3 from the NIH, No from the National Osteopathic Foundation to L.H. and Grant No. HD from the NIH to G.L.T. REFERENCES ADLER, R. & VARON, S. (1981). Neuritic guidance by polyornithine-attached materials of ganglionic origin. Devi Biol. 81, BENNETT, M. R., LAI, K. & NURCOMBE, V. (198). Identification of embryonic motoneurons in vitro: Their survival is dependent on skeletal muscle. Brain Res. 19, BONYHADY, R. E., HENDRY, I. A., HILL, C. E. & MCLENNAN, I. S. (198). Characterization of a cardiac muscle factor required for the survival of cultured parasympathetic neurones. Neurosci. Lett. 18, BRADFORD, M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein dye-binding. Analyt. Biochem. 72, COLLINS, F. (1978). Induction of neurite outgrowth by a conditioned-medium factor bound to the culture substratum. Proc. natn. Acad. Sci., U.S.A. 75, COLLINS, F. & GARRETT, J. E. (198). Elongating nerve fibers are guided by a pathway of material released from embryonic non-neuronal cells. Proc. natn. Acad. Sci., U.S.A. 77, COUGHLIN, M. D., BLOOM, E. M. & BLACK, I. B. (1981). Characterization of neuronal growth factor from mouse heart cell conditioned medium. Devi Biol. 82, DRIBIN, L. B. & BARRETT, J. N. (198). Conditioned medium enhances neuritic outgrowth from rat spinal cord explants. Devi Biol. 74, EBENDAL, T. (1981). Control of neurite extension by embryonic heart explants. /. Embryol. exp.morph. 61, EBENDAL, T., BELEW, M., JACOBSON, C.-O. & PORATH, J. (1979). Neurite outgrowth elicited by embryonic chick heart: partial purification of the active factor. Neurosci. Lett. 14, ELLMAN, G. L., COURTNEY, K. D., ANDRES, JR., V. & FEATHERSTONE, R. M. (1961). A new and rapid colorimetric determination of acetylcholinesterase activity. Biochem. Pharmacol. 7, FEDEROFF, S. & HALL, C. (1979). Effect of horse serum on neural cell differentiation in tissue culture. In Vitro 15, FISHER, K. R. S. & FEDEROFF, S. (1977). The development of chick spinal cord in tissue culture. I. Fragment cultures from embryos of various developmental stages. In Vitro 13, GILLER, E. L., NEALE, J. H., BULLOCK, P. N., SCHRIER, B. K. & NELSON, P. G. (1977). Choline acetyltransferase activity of spinal cord cell cultures increased by co-culture with muscle and by muscle-conditioned medium. J. Cell Biol. 74, HALL, B. K. (1975). A simple, single-injection method for inducing long-term paralysis in embryonic chicks, and preliminary observations on growth of the tibia. Anat. Rec. 181, HAMBURGER, V. & KEEFE, E. L. (1944). The effects of peripheral factors on the proliferation and differentiation in the spinal cord of chick embryos. /. exp. Zool. 96, HANSON, G. R. & PARTLOW, L. M. (1978). Stimulation of non-neuronal cell proliferation in vitro by mitogenic factors present in highly purified sympathetic neurons. Brain Res. 159, HELFAND, S. L., RIOPELLE, R. J. & WESSELS, N. K. (1978). Non-equivalence of conditioned medium and nerve growth factors for sympathetic, parasympathetic and sensory neurons. Expl Cell. Res. 113, HENDERSON, C. E., HACHET, M. & CHANGEUX, J.-P. (1981). Neurite outgrowth from embryonic chicken spinal neurons is promoted by media conditioned by muscle cells. Proc. natn Acad. Sci., U.S.A. 78, HOLLYDAY, M. & HAMBURGER, V. (1976). Reduction of the naturally occurring motor neuron loss by enlargement of the periphery. /. comp. Neur. 17,

13 Neuronotrophic effects of skeletal muscle fractions 95 HOLMES, W. (1943). Silver staining of nerve axons in paraffin sections. Anat. Rec. 86, JACOBSON, C.-O., EBENDAL, T., HEDLUND, K.-O. & NORRGREN, G. (198). Factors stimulating neurite outgrowth in chick embryo ganglion. In Control Mechanisms In Animal Cells (ed. L. Jimenez de Asua et al), pp New York: Raven Press. LINDSAY, R. M. & TARBIT, J. (1979). Developmental^ regulated induction of neurite outgrowth from immature chick sensory neurons (DRG) by homogenates of avian or mammalian heart, liver and brain. Neurosci. Lett. 12, MCLENNAN, I. S. & HENDRY, I. A. (1978). Parasympathetic neuronal survival induced by factors from muscle. Neurosci. Lett. 1, NISHI, R. & BERG, D. K. (1977). Dissociated ciliary ganglion neurons in vitro: survival and synapse formation. Proc. natn. Acad. Sci., U.S.A. 74, OBATA, K. & TANAKA, H. (198). Conditioned medium promotes neurite growth from both central and peripheral neurons. Neurosci. Lett. 16, OPPENHEIM, R. W., PITTMAN, R., GRAY, M. & MADERUT, J. L. (1978). Embryonic behavior, hatching and neuromuscular development in the chick following a transient reduction of spontaneous motility and sensory input by neuromuscular blocking agents. /. comp. Neurol. 179, PITTMAN, R. H. & OPPENHEIM, R. W. (1978). Neuromuscular blockade increases motoneurone survival during normal cell death in the chick embryo. Nature, Lond. 271, POLLACK, E. (198). Target-dependent survival of tadpole spinal cord neurites in tissue culture. Neurosci. Lett. 16, RIOPELLE, R. J. & CAMERON, D. A. (1981). Neurite growth promoting factors of embryonic chick-ontogeny, regional distribution and characteristics. /. Neurobiol. 12, SCHUBERT, D. (1977). The substrate attached material synthesized by clonal cell lines of nerve, glia and muscle. Brain Res. 132, SMITH, R. G. & APPEL, S. H. (1981). Evidence for a skeletal muscle protein that enhances neuron survival, neuritic extension and acetylcholine (Ach) synthesis. Abstr. Soc. Neurosci. 7, 144. VARON, S. & BUNGE, R. P. (1978). Trophic mechanisms in the peripheral neurons system. Ann. Rev. Neurosci. 1, {Received 12 October 1981, revised 1 March 1982) 4-3

14