Coordination of trophic interactions by separate developmental programs in sensory neurons and their target fields

Transcription

1 Coordination of trophic interactions by separate developmental programs in sensory neurons and their target fields ALUN M. DAVIES, YVES LÄRMET, EDWINA WRIGHT and KRISTINE S. VOGEL Department o f Anatomy, St George s Hospital Medical School, Cranmer Terrace, Tooting, London SW17 ORE, UK Summary In the developing vertebrate nervous system the survival of sensory neurons becomes dependent on neurotrophic factors when their axons reach their target fields, and the synthesis of nerve growth factor (NGF) by target field cells commences with the arrival of the earliest axons. The timing of NGF synthesis and the onset of neurotrophic factor dependence are not, however, reliant on innervation. NGF synthesis occurs on time in developing target fields in which innervation is prevented, and sensory neurons cultured before innervating their targets become dependent on neurotrophic factors for survival after a certain length of time in culture. The length of time neurons survive in culture before becoming neurotrophic factor-dependent is related to the time they would normally contact their targets in vivo: populations of neurons that have nearby targets which are innervated early respond to neurotrophic factors before neurons that have more distant targets which are innervated later. The timing of target field innervation is governed not only by the distance axons have to grow but by the rate at which they grow. Axonal growth rate is also regulated in accordance with target distance: neurons with distant targets extend axons faster than neurons with nearby targets. In addition to reviewing evidence for separate developmental programs that control the timing of neurotrophic factor synthesis in the target field and the onset of neurotrophic factor dependence in early sensory neurons, we will consider the mechanisms that might play a role in regulating the survival of neurons during the phase of neurotrophic factor independence. Key words: neurotrophic factors, nerve growth factor, target fields, axonal growth rate, sensory neurons. Introduction A substantial number of neurons die in the developing vertebrate nervous system shortly after they start innervating their target fields. This matches the number of neurons to the requirements of their target fields and eliminates inappropriately connected neurons (Oppenheim, 1981; Cowen et al. 1984). Target fields play an important role in regulating the number of neurons that survive by producing limited quantities of neurotrophic factors which the neurons require for survival. Neurons that procure an adequate supply of neurotrophic factor Journal of Cell Science, Supplement 15, (1991) Printed in Great Britain The Company of Biologists Limited 1991 survive whereas those that are unsuccessful in this competition die. Evidence for this scheme has mainly come from work on nerve growth factor (NGF), the first neurotrophic factor to be identified and characterised (Levi-Montalcini and Angeletti, 1968; Thoenen and Barde, 1980; Davies, 1988a, 19886; Barde, 1989). The most compelling evidence is that developing neurons whose survival is promoted by NGF in vitro, namely sympathetic and certain kinds of sensory neurons, are also dependent on NGF in vivo. Anti-NGF antibodies eliminate these neurons whereas exogenous NGF rescues neurons that would otherwise die. The target fields of NGF-dependent neurons contain trace quantities of NGF in proportion to their innervation density and specific cell-surface receptors mediate NGF uptake and transport to the perikaryon. Likewise, in vivo manipulation of the availability of the homologous neurotrophic factor, brain-derived neurotrophic factor (BDNF) (Leibrock et al. 1989), also influences the number of BDNF-dependent neurons that survive (Hoffer and Barde, 1988; Kalcheim et al. 1987). In vitro studies of neurotrophin-3 (NT-3), the most recently identified member of the NGF/BDNF family, have shown that like NGF and BDNF it too promotes the survival of certain kinds of neurons during the stage they are innervating their targets (Hohn et al. 1990; Maisonpierre et al. 1990). Work on several developing neuronal systems has shown that neurotrophic factor synthesis and neuronal dependence are closely coordinated with target field innervation. In this review, we will consider these temporal aspects of trophic interactions in the developing vertebrate nervous system and discuss the possible mechanisms that coordinate these events. Timing of neurotrophic factor synthesis The most detailed information on the timing of neurotrophic factor synthesis in the target fields of sensory neurons in relation to their innervation has come from studies of NGF and NGF mrna levels in the maxillary territory of the trigeminal ganglion in the mouse embryo (Davies et al. 1987). This clearly defined cutaneous target field, whose time-course of innervation has been precisely documented (Davies and Lumsden, 1984), can easily be dissected from the embryo at stages prior to and throughout the phase of innervation. Neither NGF nor NGF mrna are detectable prior to the onset of innervation but are first detected as the earliest trigeminal axons come into proximity with the target field epithelium at E ll. Although this conclusion has recently been drawn into question by the claim of Stainier and Gilbert (1990) 111

2 that the first trigeminal axons reach the maxillary epithelium earlier than E ll, this claim is not substantiated by the reported findings of these authors. The section of an E10.5 maxillary process purporting to illustrate axons reaching the maxillary epithelium shows the most distal axons still deep within the mesenchyme. The conclusion that NGF synthesis commences with the onset of target field innervation in the peripheral nervous system has also been reached in a developmental study of NGF levels in the heart ventricle and submandibular gland of the mouse embryo (Korsching and Thoenen, 1988). NGF is first detectable in these tissues around the time of initial innervation by sympathetic neurons. Although NGF synthesis begins with the arrival of axons in the target field, the timing of NGF gene expression does not depend on innervation. When the innervation of chicken embryo hindlimb buds is prevented by ablation of the corresponding neural primordia, NGF mrna expression still occurs on schedule (Rohrer et al. 1988). The mechanism that controls the onset of NGF gene expression in developing target fields independently of their innervation is an enigma. The level of neurotrophic factor synthesis in the target field is another important feature that has a bearing on the trophic interactions between neurons and their targets because it influences the number of neurons that the target field can support. Work on the embyonic mouse trigeminal system has revealed that at the onset of neuronal death there are marked differences in the level of NGF mrna in the cutaneous epithelia of the ophthalmic, mandibular and maxillary territories that are correlated with innervation density (Harper and Davies, 1990). Since regional differences in innervation density are evident in the trigeminal system prior to the onset of neuronal death (Davies and Lumsden, 1984), it is possible that the level of NGF gene expression in each territory is controlled by the number of arriving axons. Several observations, however, suggest that this is unlikely. First, the levels of NGF mrna are identical in normally innervated chicken embryo hindlimb buds and in those in which innervation is prevented (Rohrer et al. 1988). Second, when twice the number of sensory axons are forced to innervate a single hind limb in juvenile Xenopus frogs, the total number of surviving neurons supported by this limb is not significantly greater than the number it normally supports (Lamb et al. 1989). Thus, it seems that initial regional differences in innervation density and regional differences in the level of NGF mrna, though correlated, are independently regulated. Studies of NT-3 gene expression in development suggest that the conclusions based on the timing of NGF synthesis may not be universally applicable to other neurotrophic factors. The level of NT-3 mrna is high relative to BDNF mrna and NGF mrna in the spinal cord, cerebellum and hippocampus during the earlier stages of their development (Maisonpierre et al. 1990). The temporal relationship between NT-3 mrna expression and the innervation of defined target regions within these structures has yet to be determined. Preliminary studies of NT-3 mrna expression in the embryonic mouse trigeminal system, however, suggest that NT-3 mrna is present in the peripheral target field at least a day before the arrival of the first nerve fibres (Harper and Davies, unpublished data). Unlike NGF, which is expressed mainly in the epithelium of cutaneous targets (Davies et al. 1987), NT-3 is expressed predominantly in mesenchyme (Harper and Davies, unpublished data). Thus, early sensory axons may be exposed to NT-3 as they are growing to the periphery. The significance of this may be related to the influence of NT-3 on the maturation of sensory neurons before they become dependent on neurotrophic factors for survival (Wright et al. 1991). Timing and regulation of neurotrophic factor dependence The survival of sensory neurons placed in culture before they have innervated their targets is initially independent of NGF, BDNF and NT-3 (Davies and Lumsden, 1984; Davies, 1989; Vogel and Davies, 1991). Likewise, the survival of sensory neurons that differentiate from progenitors in culture (Ernsburger and Rohrer, 1988) is also initially independent of neurotrophic factors. In a developmental study of the influence of NGF on embryonic mouse trigeminal neurons cultured at closely staged intervals throughout development (Davies and Lumsden, 1984), it was concluded that the onset of NGF responsiveness coincides with the arrival of their axons in the target field. In vitro studies of developing avian retinal ganglion cells have likewise shown that these neurons become dependent on BDNF for survival at the stage their axons reach their targets (Rodriguez-Tebar et al. 1989). The mechanisms that time the onset of neurotrophic factor dependence in developing sensory neurons with the arrival of their axons in the target field have recently been addressed in studies of four populations of placode-derived cranial sensory neurons whose axons have different distances to grow to reach their peripheral and central target fields (vestibular<geniculate<petrosal<nodose). These studies have shown that the neurons control three aspects of their development that influence the timing of target field innervation and the timing of neurotrophic factor dependence: the rate at which their axons grow to their targets, the duration of neurotrophic factor-independent survival and the onset of neurotrophic dependence. The rate at which the axons of cranial sensory neurons grow to their targets in vivo is related to the distance they have to grow; neurons with more distant targets grow faster (Davies, 1989). The finding that similar differences in neurite growth rate are observed in single cell cultures of these neurons set up at the same stage of development suggests that growth rate is an intrinsically regulated property of these neurons. Although neurons with more distant targets have faster axonal growth rates, their axons still take longer to reach their targets than neurons with slower axonal growth rates and nearby targets (Vogel and Davies, 1991). It is perhaps not surprising then that the duration of neurotrophic factor-independent survival and the timing of BDNF-responsiveness in early cultures of these neurons are also related to target distance: neurons with more distant targets survive longer before responding to BDNF with enhanced survival (Vogel and Davies, 1991). These differences in the duration of neurotrophic factor independence are also observed in neurons in dissociated cultures of the otic vesicle and epibranchial placode III set up during the stage when cells are migrating from these structures to form the neurons of the vestibular and nodose ganglia, respectively. These findings suggest that the duration of neurotrophic factor independence is matched to the time-course of target field innervation by a developmental program in neurons initiated prior to gangliogenesis. The timing of neurotrophic factor dependence in vivo 112 A. M. Davies et al.

3 may not be governed solely by the developmental program in early neurons, however, as it has been shown that the onset of BDNF-responsiveness in early nodose neurons is accelerated by transiently exposing them to BDNF within 24 h of the time they normally start responding to BDNF with increased survival (Vogel and Davies, 1991). This raises the possibility that the target field, by its provision of the requisite neurotrophic factor, may also play a role in coordinating the onset of trophic factor dependence with target field innervation. In other words, the developmental program in early neurons may get the timing of dependence approximately correct and the target field provide the final fine adjustment. The need for two separate mechanisms for coordinating neurotrophic factor dependence with target field innervation is unclear. It is likely, however, that an autonomous developmental program in neurons that initiates neurotrophic factor dependence independently of target field innervation may have an error correction function. Neurons whose axons fail to grow to their correct targets, and are thereby not exposed to the appropriate neurotrophic factor at the right time, will be eliminated. Although early sensory neurons can become dependent on neurotrophic factors for survival at the appropriate time independently of target contact and exposure to neurotrophic factor, early sympathetic neurons do not acquire neurotrophic factor dependence in culture (Leah and Kidson, 1983; Emsberger et al. 1989). These early neurons do, however, become dependent on NGF for survival after treatment with retinoic acid (Rodriguez- T6bar and Rohrer, 1991). Whether or not retinoic acid influences the timing of neurotrophic factor dependence in early sensory neurons has yet to be investigated in detail. Neurotrophic factor receptor expression and its relationship to dependence Neurotrophic factor receptors Binding studies using iodinated NGF or BDNF have revealed two kinds of receptors for each factor: low-affinity receptors that have a dissociation constant of 10~9m and high-affinity receptors that have a dissociation constant of m (Sutter et al. 1979; Rodriguez-Tebar and Barde, 1988). A cdna encoding a transmembrane glycoprotein of relative molecular mass (gp75ngfr) which binds NGF with low-affinity has been cloned (Johnson et al. 1986; Radeke et al. 1987). Competition binding studies and gene transfection experiments have shown that gp75ngfr is also a low-affinity receptor for BDNF (Rodriguez-Tebar et al. 1990). In contrast, a 1000-fold excess of either NGF or BDNF is required to reduce binding of the heterologous ligand to its high-affinity receptor, indicating that highaffinity receptors are specific (Rodriguez-Tebar et al. 1990). The finding that the survival-promoting effects of NGF and BDNF are observed at concentrations which result in preferential occupancy of their respective high-affinity receptors (Greene, 1977; Rodriguez-Tebar and Barde, 1988) suggests that these receptors mediate the response of neurons to neurotrophic factors. Recent work indicates that high-affinity receptors have a composite structure. The demonstration that both highand low-affinity NGF binding is restored to NGF receptordeficient PC 12 cells by the introduction of a recombinant retrovirus containing the gp75ngfr cdna suggests that the common low-affinity neurotrophic factor receptor is a component of the high-affinity NGF receptor (Hempstead et al. 1989). Membrane fusion and gene transfection experiments have demonstrated that the other component of the high-affinity NGF receptor is the proto-oncogene trk (Hempstead et al. 1991). This is a transmembrane glycoprotein with an intracellular tyrosine kinase catalytic domain (Martin-Zanca et al. 1989). The demonstration that the addition of NGF to neural cell lines and dorsal root ganglia elicits rapid phosphorylation of trk and stimulates its tyrosine kinase activity suggests that trk plays an important role in NGF signal transduction (Kaplan et al. 1991a, 19916; Klein et al. 1991). An homologous tyrosine kinase, trkb which is expressed predominantly in neural tissues (Klein et al. 1989, 1990a, 19906), has also recently been shown to be a receptor for BDNF and NT-3 and to be rapidly phosphorylated on tyrosine residues when exposed to these neurotrophic factors (Soppet et al. 1991). Timing of neurotrophic factor receptor expression The relationship between the onset of NGF dependence and the expression of NGF receptors has been studied using autoradiography to localize specific binding of iodinated NGF to embryonic mouse trigeminal ganglion neurons in vitro (Davies et al. 1987). When cultured before innervating their targets, trigeminal neurons were unlabelled by iodinated NGF. Labelled neurons were first observed in cultures set up at the time when the earliest neurons start innervating their targets, and in later cultures the proportion of labelled neurons increased to reach a maximum when the last axons reach their targets. Although this finding suggests that appreciable levels of NGF receptors are first expressed on developing sensory neurons when their axons reach their targets, autoradiography fails to detect very low levels of NGF receptors and cannot adequately distinguish between low- and highaffinity receptors. In a study of developmental changes in the level of gp75ngfr mrna in the trigeminal ganglion (Wyatt et al. 1990), a low level of expression was detected from the earliest stages of axonal outgrowth, prior to the onset of target field innervation. From a knowledge of the change in the number of neurons in the trigeminal ganglion during development and the demonstration that gp75 FR mrna expression is restricted to neurons in the ganglion, it has been possible to calculate the mean level of gp75ngfr mrna per neuron throughout development. When trigeminal axons are growing to their targets there is a low constant level of gp75ng mrna of about 30 molecules per neuron. Shortly after the first axons reach the target the mean level of gp75ngfr mrna per neuron increases five-fold to reach a second plateau shortly after the last axons have reached their targets. These findings are consistent with the labelling of trigeminal neurons by iodinated NGF in relation to target field innervation and additionally suggest that low levels of at least the common low-affinity neurotrophic factor receptor are expressed on developing sensory neurons before their axons have reached their targets. This is further supported by the detection of gp75 FR mrna by in situ hybridization in embryonic chick DRG during the earliest stages of axonal outgrowth (Heuer et al. 1990). Indeed, the detection of gp75ngfr mrna in premigratory neural crest by in situ hybridization suggests that at least some neural crest cells may also express the common low-affinity neurotrophic factor receptor (Heuer et al. 1990). The functional significance of neurotrophic factor receptor expression prior to the onset of target field innervation Regulation of sensory neurons in target fields 113

4 has recently been demonstrated by blocking the synthesis of the common low-affinity receptor in early, neurotrophic factor independent DRG neurons using antisense oligonucleotides that hybridize in the region of the translation initiation codon of gp75ngfr mrna (Wright et al. 1991). Neurons in cultures treated with these antisense oligonucleotides fail to undergo an early maturational change from having small, spindle-shaped, phase-dark cell bodies to having large, spherical, phase-bright cell bodies. This maturational change is accelerated by treating the early neurons with either BDNF or NT-3 but not NGF. Whether or not trkb is also expressed in immature, neurotrophic factor independent sensory neurons has yet to be ascertained. The low level of neurotrophic factor receptors on early sensory neurons may also play a role in coordinating the onset of neurotrophic factor dependence with the arrival of axons in the target field. As mentioned above, transient exposure of early nodose ganglion neurons to BDNF in vitro accelerates the acquisition of BDNF-dependence. Thus, it is possible that binding of neurotrophic factor to receptors may lead to an increase in receptor expression and the onset of neurotrophic factor dependence. Although this has not yet been tested experimentally in early sensory neurons, it has been shown that NGF treatment increases the level of gp75ngfr mrna in PC12 cells and adult DRG neurons (Landreth and Shooter, 1980; Doherty et al. 1988; Verge et al. 1989; Lindsay et al. 1989). In early, neurotrophic factor independent sympathetic neurons, retinoic acid has been shown to induce the expression of high-affinity NGF receptors and concomitantly bring about the dependence of these neurons on NGF (Rodriguez-Tebar and Rohrer, 1991). Whether or not retinoic acid has any effect on neurotrophic factor receptor expression in early sensory neurons has yet to be investigated. During and beyond the stage of target field innervation there are marked developmental changes in the levels of gp75ngfr mrna and NGF receptors in sensory ganglia. The interpretation of such changes is complicated by the expression of low-affinity NGF receptors on Schwann cells at these later stages (Yan and Johnson, 1987; Scarpini et al. 1988). Studies on dissociated DRG cell suspensions (containing both neurons and non-neuronal cells) have revealed that between E14 and E16 in chick embryos there is a four- to six-fold decrease in capacity of the cells to bind iodinated NGF specifically (Herrup and Shooter, 1975). Likewise, the intensity with which chick DRG and neural crest-derived cranial sensory ganglia are labelled with iodinated NGF in tissue sections also decreases over this same period of development (Raivich et al. 1985,1987). The level of gp75ngfr mrna is also markedly lower in 3 month rat DRG than at birth (Buch et al. 1987) and immunoprecipitation studies have shown that the level of gp75ng per milligram protein is also appreciably lower in adult rat DRG than in late fetal DRG (Yan and Johnson, 1987). The finding that the proportion of embryonic chick DRG neurons specifically labelled by iodinated NGF is lower in E16 cultures compared with E10 cultures and that the proportion of labelled neurons in E10 cultures decreases after several days (Rohrer and Barde, 1982) suggests that at least a proportion of sensory neurons lose NGF receptors during development. This apparent decrease in the level of NGF receptors on developing DRG neurons correlates with the reduction in the magnitude of neurite outgrowth elicited by NGF from DRG explants (Herrup and Shooter, 1975). Although several in vivo and in vitro studies suggest that the majority of adult DRG neurons are not dependent on NGF for survival (Levi-Montalcini and Angeletti, 1968; Johnson et al. 1980; Johnson and Yip, 1985; Lindsay, 1988), a proportion of DRG neurons continue to express high-affinity NGF receptors (Verge et al. 1989). In mature DRG neurons these receptors probably mediate the effects of NGF on neuropeptide synthesis (Lindsay and Harmer, 1989). Mechanisms regulating neuronal survival prior to neurotrophic factor dependence The expression of gp75ngfr in sensory neurons before their axons have reached their targets (Wyatt et al. 1990) and the demonstration that these receptors are involved in the maturation of neurons during this early stage of their development (Wright et al. 1991) raise the possibility that neuronal survival during this stage is also dependent on neurotrophic factors, produced perhaps by neurons themselves or by other cells in their vicinity. It is unlikely that members of the NGF family are involved, however, because antisense oligonucleotides that inhibit the synthesis of gp75ngfe and prevent the maturation of early sensory neurons do not accelerate the rate at which these neurons die compared with neurons in control cultures treated with sense oligonucleotides (Wright et al. 1991). The demonstration that moderate elevation in the cytosolic free-calcium ion concentration promotes the survival of neurotrophic factor dependent neurons (Koike et al. 1989; Collins and Lile, 1989; Koike and Tanaka, 1991) and that neurotrophic factors seem to promote neuronal survival without affecting intracellular calcium (Koike et al. 1989; Tolkovsky et al. 1990) has raised the possibility that processes that elevate the cytosolic free- Ca2+ level may promote the survival of neurons before they become dependent on neurotrophic factors for survival and that the acquisition of dependence may might be associated with a fall in the level of cytosolic calcium. This proposal is consistent with the demonstration that the intracellular Ca2+ chelator BAPTA/AM kills neurotrophic factor-independent early nodose neurons at concentrations that have little affect on the survival of older nodose neurons growing in the presence of BDNF. Furthermore, chronic application of either caffeine or thapsigargin (which deplete intracellular Ca2+ stores) kills neurotrophic factor-independent early nodose neurons (Larmet, Y. and Davies, A. M., unpublished data). This is not due to a non-specific toxic effect of these reagents because they do not affect the survival of older nodose neurons grown in the presence of either BDNF or medium containing 30 nm KC1. This work was supported by grants from The Wellcome Trust, Medical Research Council, Science and Engineering Research Council, Cancer Research Campaign and Action Research. Yves Larmet was supported by an EMBO longterm fellowship. References B a r d e, Y.-A. ( ). What, if anything, is a neurotrophic factor? Trends Neurosci. 1 1, B a r d e, Y.-A. ( ). Trophic factors and neuronal survival. Neuron 2, B u c h, C. R., M a r t i n e z, H. J., B l a c k, I. B. a n d C h a o, M. V. ( ). Developmentally regulated expression of nerve growth factor receptor gene in the periphery and brain. Proc. natn. Acad. Sci. U.S.A. 8 4, A. M. Davies et al.

5 C o l l i n s, F. a n d L i l e, J. D. (1989). Brain Res. 502, C o w a n, W. M., F a w c e t t, J. W., O L e a r y, D. D. M. a n d S t a n f i e l d, B. B. (1984). Regressive events in neurogenesis. Science 225, D a v i e s, A. M. (1988a). Role of neurotrophic factors in development. Trends Genet. 4, D a v i e s, A. M. (19886). The emerging generality of the neurotrophic hypothesis. Trends Neurosci. 11, D a v i e s, A. M. (1989). Intrinsic differences in the growth rate of early nerve fibres related to target distance. Nature 337, D a v i e s, A. M., B a n d t l o w, C., H e u m a n n, R., K o r s c h i n g, S., R o h r e r, H. a n d T h o e n e n, H. (1987). Timing and site of nerve growth factor synthesis in developing skin in relation to innervation and expression of the receptor. Nature 326, D a v i e s, A. M. a n d L u m s d e n, A. G. S. (1984). Relation of target encounter and neuronal death to nerve growth factor responsiveness in the developing mouse trigeminal ganglion. J. comp. Neurol. 253, D o h e r t y, P., S e a t o n, P., F l a n i g a n, T. P. a n d W a l s h, F. S. (1988). Factors controlling the expression of the NGF receptor in PC12 cells. Neurosci. Lett. 92, E r n s b e r g e r, U. a n d R o h r e r, H. (1988) Neuronal precursor cells in chick dorsal root ganglia: Differentiation and survival in vitro. Devi Biol. 126, E r n s b e r g e r, U., E d g a r, D. a n d R o h r e r, H. (1989). The survival of early chick sympathetic neurons in vitro is dependent on a suitable substratum but independent of NGF. Devi Biol. 135, G r e e n e, L. A. (1977). Quantitative in vitro studies on the nerve growth factor (NGF) requirements of neurons. I. Sensory neurons. Devi Biol. 58, H a r p e r, S. J. a n d D a v i e s, A. M. (1990). NGF mrna expression in developing cutaneous epithelium related to innervation density. Development 110, H e m p s t e a d, B. L., M a r t i n - Z a n c a, D., K a p l a n, D. R., P a r a d a, L. F. a n d C h a o, M. V. (1991). High-affinity NGF binding requires coexpression of the trk proto-oncogene and the low-affinity NGF receptor. Nature 350, H e m p s t e a d, B. L., S c h l e i f e r, L. S. a n d C h a o, M. V. (1989). Expression of functional nerve growth factor receptors after gene transfer. Science 243, H e r r u p, K. a n d S h o o t e r, E. M. (1975). Properties of the 5 nerve growth factor receptor in development. J. Cell Biol. 67, H e u e r, J. G., F a t e m i e -N a i n i e, S., W h e e l e r, E. F. a n d B o t h w e l l, M. (1990). Structure and expression of the chicken NGF receptor. Devi Biol. 137, H o f f e r, M. M. a n d B a r d e, Y.-A. (1988). Brain-derived neurotrophic factor prevents neuronal death in vivo. Nature 331, H o h n, A., L e i b r o c k, J., B a i l e y, K. a n d B a r d e, Y.-A. (1990). Identification and characterization of a novel member of the nerve growth factor/brain-derived neurotrophic factor family. Nature 344, J o h n s o n, D., L a n a h a n, A., B u c k, C. R., S e h g a l, A., M o r g a n, C., M e r c e r, E., B o t h w e l l, M. a n d C h a o, M. ( ). Expression and structure of the human NGF receptor. Cell 47, J o h n s o n, E. M., G o r i n, P., B r a n d e i s, L. D. a n d P e a r s o n, J. (1980). Dorsal root ganglion neurons are destroyed by exposure in utero to maternal antibody to nerve growth factor. Science 210, J o h n s o n, E. M. a n d Y i p, H. K. (1985). Central nervous system and peripheral nerve growth factor provide trophic support critical to mature sensory neuronal survival. Nature 314, K a l c h e i m, C., B a r d e, Y.-A., T h o e n e n, H. a n d L e D o u a r i n, N. M. (1987). In vivo effect of brain-derived neurotrophic factor on the survival of developing dorsal root ganglion cells. EMBO J. 6, K a p l a n, D. R., H e m p s t e a d, B. L., M a r t i n -Z a n c a, D,, C h a o, M. V. a n d P a r a d a, L. F. (1991a). The trk proto-oncogene product: A signal transducing receptor for nerve growth factor. Science 252, K a p l a n, D. R., M a r t i n -Z a n c a, D. a n d P a r a d a, L. F. (19916). Tyrosine phosphorylation and tyrosine kinase activity of the trk proto-oncogene produce induced by NGF. Nature 350, K l e i n, R., C o n w a y, D., P a r a d a, L. F. a n d B a r b a c i d, M. (1990a). The trkb tyrosine protein kinase gene codes for a second neurogenic receptor that lacks the catalytic kinase domain. Cell 61, K l e i n, R., J i n g, S., N a n d u r i, V., O R o u r k e, E. a n d B a r b a c i d, M. (1991). The trk proto-oncogene encodes a receptor for nerve growth factor. Cell 65, K l e i n, R., M a r t i n -Z a n c a, D., B a r b a c i d, M. a n d P a r a d a, L. F. (19906). Expression of the tyrosine kinase receptor gene trkb is confined to the murine embryonic and adult nervous system. Development 109, K l e i n, R., P a r a d a, L. F., C o u l i e r, F. a n d B a r b a c i d, M. (1989). trkb, a novel tyrosine protein kinase receptor expressed during mouse neural development. EMBO J. 8, K o i k e, T., M a r t i n, D. P. a n d J o h n s o n, E. M. (1989). Role of Ca2+ channels in the ability of membrane depolarization to prevent neuronal death induced by trophic-factor deprivation: evidence that levels of internal Ca2+ determine nerve growth factor dependence of sympathetic ganglion cells. Proc. natn. Acad. Sci. U.S.A. 86, * K o i k e, T. a n d T a n a k a, S. (1991). Evidence that nerve growth factor dependence of sympathetic neurons for survival in vitro may be determined by levels of cytoplasmic free Ca2+. Proc. natn. Acad. Sci. U.S.A. 88, K o r s c h i n g, S. a n d T h o e n e n, H. (1988). Developmental changes of nerve growth factor levels in sympathetic ganglia and their target organs. Devi Biol. 126, L a m b, A. H., F e r n s, M. J. a n d K l o s e, K. (1989). Peripheral competition in the control of sensory neuron numbers in Xenopus frogs reared with a single bilaterally innervated hindlimb. Devi. Brain Res. 45, L a n d r e t h, G. E. a n d S h o o t e r, E. M. (1980). Nerve growth factor receptors on PC12 cells: ligand-induced conversion from low- to highaffinity states. Proc. natn. Acad. Sci. U.S.A. 77, L e a h, J. a n d K i d s o n, C. (1983). Survival of chick embryo sympathetic neurons in cell culture. Int. J. Devi. Neuroscience 1, L e i b r o c k, J., L o t t s p e i c h, F., H o h n, A., H o f e r, M., H e n g e r e r, B., M a s i a k o w s k i, P., T h o e n e n, H. a n d B a r d e, Y.-A. (1989). Molecular cloning and expression of brain-derived neurotrophic factor. Nature 341, L e v i -M o n t a l c i n i, R. a n d A n g e l e t t i, P. U. (1968). Nerve growth factor. Physiol. Rev. 48, L i n d s a y, R. M. (1988). Nerve growth factors (NGF, BDNF) enhance axonal regeneration but are not required for survival of adult sensory neurons. J. Neurosci. 8, L i n d s a y, R. M. a n d H a r m e r, A. J. (1989). Nerve growth factor regulates expression of neuropeptide genes in adult sensory neurons. Nature 337, L i n d s a y, R. M., S h o o t e r, E. M., R a d e k e, M. J., D e c h a n t, G., T h o e n e n, H. a n d L i n d h o l m, D. (1989). Nerve growth factor regulates expression of the NGF-receptor gene in adult sensory neurons. Eur. J. Neurosci. 2, M a i s o n p i e r r e, P., B e l u s c i o, L., S q u i n t o, S., I p, N., F u r t h, M., L i n d s a y, R. a n d Y a n c o p o u l o s, G. (19906). Neurotrophin-3: a neurotrophic factor related to NGF and BDNF. Science 247, M a r t i n -Z a n c a, D., B a r b a c i d, M. a n d P a r a d a, L. F. (1990). Expression of the tnrk proto-oncogene is restricted to the sensory cranial and spinal ganglia of neural crest origin in mouse development. Genes Dev. 6, M a r t i n - Z a n c a, D., O s k a m, R., M i t r a, G., C o p e l a n d, T. a n d B a r b a c i d, M. (1989). Molecular and biochemical characterization of the human trk proto-oncogene. Molec. cell Biol. 9, O p p e n h e i m, R. (1981). Neuronal death and some related phenomena during neurogenesis. In Studies in Developmental Biology (ed. W. M. Cowan), pp Oxford University Press. R a d e k e, M. J., M i s k o, T. P., Hsu, C., H e r z e n b e r g, L. A. a n d S h o o t e r, E. M. (1987). Gene transfer and molecular cloning of the rat nerve growth factor receptor. Nature 325, R a i v i c h, G., Z i m m e r m a n, A. a n d S u t t e r, A. (1985). The spatial and temporal pattern of /3NGF receptor expression in the developing chick embryo. EMBO J. 4, R a i v i c h, G., Z i m m e r m a n, A. a n d S u t t e r, A. (1987). Nerve growth factor (NGF) receptor expression in chicken cranial development. J. comp. Neurol. 256, R o d r i g u e z -T b a r, A. a n d B a r d e, Y.-A. (1988). Binding characteristics of brain-derived neurotrophic factor to its receptors on neurons from the chick embryo. J. Neurosci. 8, R o d r i g u e z -T b a r, A., D e c h a n t, G. a n d B a r d e, Y.-A. (1990). Binding of brain-derived neurotrophic factor to the nerve growth factor receptor. Neuron 4, R o d r i g u e z -T jsb a r, A., J e f f r e y, P.L., T h o e n e n, H. a n d B a r d e, Y.-A. (1989). The survival of chick retinal ganglion cells in response to brain-derived neurotrophic factor depends on their embryonic age. Devi Biol. 136, R o d r i g u e z -T b a r, A. a n d R o h r e r, H. (1991). Retinoic acid induces NGF-dependent survival response and high-affinity NGF receptors in immature chick sympathetic neurons. Development 112, R o h r e r, H. a n d B a r d e, Y.-A. (1982). Presence and disappearance of nerve growth factor receptors on sensory neurons in culture. Devi Biol. 89, R o h r e r, H., H e u m a n n, R. a n d T h o e n e n, H. (1988). The synthesis of nerve growth factor (NGF) in developing skin is independent of innervation. Devi Biol. 128, S c a r p i n i, E., R o s s, A. H., R o s e n, J. L., B r o w n, M. J., R o s t a m i, A., K o p r o w s k i, H. a n d L i s a k, R. P. (1988). Expression of nerve growth factor receptor during human peripheral nerve development. Devi Biol. 125, Regulation of sensory neurons in target fields 115

6 S o p p e t, D., E s c a n d o n, E., M a r a g o s, J., M id d l e m a s, D. S., R e id, S. W., B l a ir, J., B u r t o n, L. E., St a n t o n, B. R., K a p l a n, D. R., H u n t e r, T., N ik o l ic s, K. a n d P a r a d a, L. F. (1991). The neurotrophic factors brain-derived neurotrophic factor and neurotrophin-3 are ligands for the trkb tyrosine kinase receptor. Cell 65 (in press). S t a in ie r, D. Y. a n d G il b e r t, W. (1990). Pioneer neurons in the mouse trigeminal sensory system. Proc. natn. Acad. Sci. U.S.A. 87, S u t t e r, A., R io p e l l e, R. J., H a r r is-w a r r ic k, R. M. a n d S h o o t e r, E. M. (1979). Nerve growth factor receptors. J. biol. Chem. 254, T h o e n e n, H. a n d B a r d e, Y.-A. (1980). Physiology of nerve growth factor. A. Rev. Physiol. 60, T o l k o v sk y, A. M., W a l k e r, A. E., M u r r e l l, R. D. a n d Su id a n, H. S. (1990). Ca2+ transients are not required as signals for long-term neurite outgrowth from cultured sympathetic neurons. J. Cell Biol. 110, V e r g e, K. M. K., R io p e l l e, R. J. a n d R ic h a r d s o n, P. M. (1989). Nerve growth factor receptors on normal and injured sensory neurons. J. Neurosci. 9, V o g e l, K. S. a n d D a v ie s, A. M. (1991). The duration of neurotrophic factor independence in early sensory neurons is matched to the timecourse of target field innervation by a developmental program initiated prior to gangliogenesis. Neuron (in press). W r ig h t, E., V o g e l, K. S. a n d D a v ie s, A. M. (1991). Neurotrophic factors promote neuronal maturation before the onset of survival dependence (submitted). W y a t t, S., S h o o t e r, E. M. a n d D a v ie s, A. M. (1990). Expression of the NGF receptor gene in sensory neurons and their cutaneous targets prior to and during innervation. Neuron 4, Y a n, Q. a n d J o h n s o n, E. G. Jr (1987). A quantitative study of the developmental expression of nerve growth factor (NGF) receptor in rats. Devi Biol. 121, A. M. Davies et al.