DI~VF-J.,OPMENTALBIOLOGY 147, 83-95 (1991)

Developmental Changes in Transmitter Sensitivity and Synaptic Transmission in Embryonic Chicken Sympathetic Neurons Innervated in Vitro n . G A R D E T T E , 1 M. D. L I S T E R U D , A . B. B R U S S A A R D , AND L. W . R O L E 2

Department of Anatomy and Cell Biology, Center for Neurobiology and Behavior, Columbia University, P & S, 630 West 168th Street, New York, New York 10032 Accepted May 30, 1991 Dispersed neurons from embryonic chicken sympathetic ganglia were innervated in vitro by explants of spinal cord containing the autonomic preganglionic nucleus or somatic motor nucleus. The maturation of postsynaptic acetylcholine (ACh) sensitivity and synaptic activity was evaluated from ACh and synaptically evoked currents in voltage-clamped neurons at several stages of innervation. All innervated cells are more sensitive to ACh than uninnervated neurons regardless of the source of cholinergic input. Similarly, medium conditioned by either dorsal or ventral explants mimics innervation by enhancing neuronal ACh sensitivity. This increase is due to changes in the rate of appearance of ACh receptors on the cell surface. There are also several changes in the nature of synaptic transmission with development in vitro, including an increased frequency of synaptic events and the appearance of larger amplitude synaptic currents. In addition, the mean amplitude of the unit synaptic current mode increases, as predicted from the observed changes in postsynaptic sensitivity. Although spontaneous synaptic current amplitude histograms with multimodal distributions are seen at all stages of development, histograms from early synapses are typically unimodal. Changes in the synaptic currents and ACh sensitivity between 1 and 4 days of innervation were paralleled by an increase in the number of synaptic events t h a t evoked suprathreshold activity in the postsynaptic neurons. The early pre- and postsynaptic differentiation described here for interneuronal synapses formed in vitro may be responsible for increased efficacy of synaptic transmission during development in vivo. © 1991AcademicPress, Inc.

nerve-muscle synapse are now approaching molecular resolution. Despite differences in structure and subunit composition between neuronal and muscle-type nicotinic AChRs, some aspects of the regulation of these receptor types may be similar (see Berg et al., 1989; Steinbach and Ifune, 1989 for reviews). For example, AChRs on neurons (nAChRs), like muscle, might be regulated by presynaptic input. Examination of receptor distribution with antibody and toxin probes indicates that nAChRs are concentrated at sites of presynaptic contact (Loring et aL, 1985; Jacob and Berg, 1988). Functional studies of the innervation of sympathetic neurons in developing frog and chick indicate that there are significant increases in neuronal ACh sensitivity and nAChR number following the arrival of presynaptic input both in vivo and in vitro (Leah, et al., 1986; Role, 1988; Brussaard and Role, 1990; L. Marshall, personal communication). In addition, there are changes in the level of nAChR subunit expression (Boyd et al., 1988) and in the properties of the ACh-activated channels expressed in neurons at distinct developmental stages relative to innervation, reminiscent of the developmental regulation of muscletype AChRs (see Schuetze and Role, 1987 and Brehm and Henderson, 1988 for reviews; Marshall, 1985, 1986;

INTRODUCTION

Acetylcholine receptors (AChR) on skeletal muscle are diffusely distributed over the cell surface prior to innervation. The number and distribution of AChRs are regulated by motor nerve input, and factors derived from cholinergic neurons have been shown to enhance both receptor synthesis and clustering in muscle (for review see Salpeter and Loring, 1985; Schuetze and Role, 1987; Brehm and Henderson, 1988). Two such factors, that recently have been purified and cloned (called Agrin and A R I A ) , alter receptor distribution and receptor synthesis, respectively (Usdin and Fischbach, 1986; Harris et al., 1988; McMahan and Wallace, 1989; MagillSolc and McMahan, 1990; Harris, et al., 1991). The level of muscle activity has also been demonstrated to regulate the extent of AChR synthesis and the levels of AChR subunit RNA expression (Merlie et al., 1984; Klarsfeld and Changeux, 1985; Goldman et al., 1988). Thus, the mechanisms of regulation of postsynaptic transmitter sensitivity during development of the

1 Present address: Unite de Neuroendocrinologie, INSERM U 159, Centre Paul Broca, 2 ter, rue d'Alesia, 75014, Paris, France. 2 To whom correspondence should be addressed. 83

0012-1606/91 $3.00 Copyright© 1991by AcademicPress, Inc. All rights of reproductionin any form reserved.

84

DEVELOPMENTALBIOLOGY

Moss et al., 1989; Margiotta and Gurantz, 1989; Brussaard and Role, 1990). Finally, studies of the developmerit of synaptic transmission between neurons in vivo have indicated that early synapses fatigue rapidly and synaptic events are small in amplitude (Rubin 1985a,b; Nelson et al., 1989; L. Marshall, personal communication) similar to reports on early nerve-muscle synapses (Cohen 1980; Hume et al., 1983; Role et al., 1987; Evers et al., 1989). The difficulty of reliably identifying and monitoring the function of early synaptic connections in vivo has limited studies of developing synapses between neurons (McLachlan 1973; Rubin, 1985a, b). For this reason in vitro approaches that permit a detailed analysis of synaptogenesis between neurons have proved invaluable. In a previous study from this laboratory (Role, 1988), embryonic chicken sympathetic neurons were removed prior to receiving synaptic input in vivo and innervated in vitro by explants of spinal cord containing the preganglionic neurons. These studies indicated that presynaptic input increased the sensitivity of postsynaptic neurons to their neurotransmitter, ACh. The present study extends this work by examining the timing of the development of postsynaptic transmitter sensitivity relative to synaptogenesis as well as describing changes in the properties of synaptic transmission at these developing connections. In addition, the present study examines whether the development of transmitter sensitivity and synaptic transmission is restricted to sympathetic preganglionic neurons or whether other cholinergic neurons might substitute for the normal presynaptic input. We find that there is a rapid maturation of synaptic function following initial innervation in vitro. Explants containing either somatic motoneurons or preganglionic neurons both innervate and enhance the ACh sensitivity of sympathetic neurons. In addition, we find that explants containing somatic motoneurons (like those containing preganglionic neurons; Role, 1988) can "condition" the culture medium such that the ACh sensitivity of sympathetic neurons is enhanced in the absence of innervation. This enhanced ACh sensitivity in response to conditioned medium is apparently due to an increase in the net rate of AChR synthesis and/or insertion. Finally, the changes in postsynaptic sensitivity relative to the arrival of presynaptic input are accompanied by maturation of synaptic transmission: both the number and amplitude of spontaneous synaptic currents as well as the extent of suprathreshold synaptic activity are increased over this period of development. Aspects of this work have been previously reported in abstract form (Gardette et al., 1988; Gardette and Role, 1989). MATERIALS AND METHODS

Cell Culture Dispersed embryonic sympathetic neurons were prepared and maintained in vitro as described in Role

VOLUME147, 1991

(1984) with modifications as previously noted (Role, 1988; Mawe et al., 1990). Briefly, lumbar sympathetic ganglia from Embryonic Day (ED) 10 chickens were incubated in a Ca z+, Mg2+-free phosphate-buffered medium with 0.01% trypsin (30 min; Sigma, type III). The tissue was mechanically dispersed to single cells and then plated in medium composed of DMEM (prepared with 2.2 g/liter NaHCOs) supplemented with 10% horse serum, 2% E D l l chicken embryo extract (CEE), 2 mM glutamine (GIBCO), penicillin (50 units/ml), streptomycin (50 ~g/ml), and 2.5S nerve growth factor (NGF, 0.1 ~g/ml; gift of G. Johnson and P. Osbourne, Washington University, St Louis, MO). The cells were plated at five to seven ganglia/35-mm dish (~30,000 cells) on a polyornithine substrate and maintained in a 37°C, 5% COz humidified atmosphere. Under these conditions the cultures are essentially devoid of nonneuronal cells and >90% of the neurons are adrenergic and cholinoceptive (Role, 1984, 1988). To innervate sympathetic neurons in vitro, spinal cord from mid to low thoracic and upper lumbar regions was removed from ED 8 chicken embryos. In avians the preganglionic neurons lie in the dorsal region of the spinal cord, adjacent to the central canal (Langley, 1904; Yip, 1986, 1990). This region was separated from the ventral portion containing the cholinergic motoneurons and added to an established culture of sympathetic neurons (plated 4 days before) according to the previously described procedure (Role, 1988). Under these culture conditions innervation can be detected within 24 hr and results entirely from input from the explants; the sympathetic neurons do not innervate one another and are electrically silent in the absence of input from cocultured explants (Role, 1988). We have restricted our analysis of the development of ACh sensitivity following innervation to times after the first 24 hr of coculture since this is the time required for the explants to settle and attach to the substrate. Replacing the growth media with recording medium earlier than 24 hr disrupts synaptic connections so that we cannot reliably evaluate the innervation status of the sympathetic neurons. Our convention for indicating the time of recording from the cultures is relative to the number of days the sympathetic neurons have been maintained in vitro after culture with or without innervating explants (i.e., Day 0 is the day of explant addition to a culture or to its siblings; Day I is 24 hr after explant addition etc.). To examine the effects of medium conditioned by spinal cord explants, we have used the procedures described in Role (1988) with slight modification. In experiments examining the effects of soluble factors from ventral explants, only that portion of spinal cord was used for explant preparation. In addition, in some experiments, large numbers of explants were added to neu-

GARDETTE ET AL.

Developmental Changes of Chick Neurons

ron-alone cultures under conditions t h a t do not permit explant-substrate attachment (i.e., the volume of culture medium was sufficient for the explants to remain floating). This procedure allows the explants to "condition" the medium without explant-cell contact and obviates the necessity of transferring conditioned medium from explant-alone to neuron-alone cultures. Since this transfer, along with the the several media changes required for the BAC experiments (see below), decreases the number of viable sympathetic neurons, the revised protocol was routinely adapted in later experiments.

Physiological Recordings Electrophysiological recordings from sympathetic neurons in vitro employed the whole-cell tight seal recording configuration of the patch-clamp technique (Hamill et al., 1981). Patch-clamp electrodes (6-13 M~) were obtained by a two-stage pull on a vertical electrode puller (Kopf 700D or Narashighe PP-83) and filled with 150 mM KC1, 2 mM MgClz, 1 mM CaC12, 10 mM Hepes, 5 mM Mg-ATP, 100 ttM GTP, and 11 mM EGTA, at pH 7.2. The external patch recording medium contained 140 mM NaC1, 6 mM KC1, 0.8 mM MgC12, 10 mM CaC12, 12.5 mM Hepes, and 5 mM glucose, at pH 7.4. Synaptic and ACh-induced currents were recorded with a List EPC7 patch-clamp amplifier. Neurons were voltage-clamped to -60 mV, unless otherwise noted. Data were stored on videotape with a PCM digitizer (Neurodata DR384) and currents measured directly with a digital oscilloscope (Nicolet 3091) or off line with the DEC 11/73 system described below. A m p l i t u d e h i s t o g r a m s of s p o n t a n e o u s s y n a p t i c currents were generated from recordings of synaptic currents in the presence of TTX (3 #M). The amplitude of the individual events was measured by an interactive algorithm developed by the late S. Schuetze and revised by L. Simmons (Eli Lily Research, Indianapolis, IN). The synaptic currents were first evaluated with respect to rise time and included only if the time to peak was

10 8

z

,

•

,

,

,

,

,

,

,

,

,

,

,

,

,

,O~OTO'IfOTOlrOTOTOTO'rOTOI'O'rOTOTO'IrOYOTOFOI

6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 pA 16

Day 4

14

12 lO 8 =~

6 4 2 0

6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96

pA FIG. 5. A m p l i t u d e h i s t o g r a m s of s p o n t a n e o u s s y n a p t i c c u r r e n t s recorded f r o m two different cells a f t e r 1 a n d 4 days of culture in t h e presence of dorsal spinal cord explants. S p o n t a n e o u s s y n a p t i c c u r r e n t s were recorded f r o m sibling c u l t u r e s i n n e r v a t e d for 1 a n d 4 days in vitro. T h e r e c o r d i n g m e d i u m included TTX (3 ttM) to exclude evoked s y n a p t i c activity. T h e d a t a h e r e were f r o m 9.4 a n d 4.0 m i n of recording, respectively. The h i s t o g r a m s a r e fit by t h e s u m of t w o t h r e e G a u s s i a n d i s t r i b u t i o n s as described u n d e r M a t e r i a l s a n d M e t h ods. Note t h a t in t h i s e x p e r i m e n t in t h e D a y 4 h i s t o g r a m t h e r e w a s n o t a sufficient n u m b e r of e v e n t s >75 pA (clear b a r s ) to p e r m i t a f o u r t h G a u s s i a n to be fit to t h e distribution.

GARDETTE ET AL.

Developmental Changes of Chick Neurons

100 >-

v.>

0.76 sp/min

-,- 7-- 80 I.-- 0

m "~ 60

~ Z

4o I1:

~ 2o 0,09~sp/min a.

o ]

4

DAYS OF CO-CULTURE WITH DORSAL EXPLANTS

FIG. 6. Development of suprathreshold activity with time of coculture with dorsal explants. The number of innervated cells with synaptically driven action potentials increases between 1 and 4 days of coculture. The firing frequency of action potentials for cells displaying synaptically driven suprathreshold activity is also enhanced, from 0.09 to 0.76 spikes/min during the same developmental period.

following the formation of synapses on sympathetic ganglion neurons. In addition, over a more protracted time course, synaptic transmission matures reflecting further pre- and postsynaptic maturation that can be readily studied in this simplified in vitro setting. ACh sensitivity is significantly enhanced as soon as innervation can be detected. This rapid increase in postsynaptic neuronal ACh sensitivity occurred with both appropriate and inappropriate cholinergic presynaptic partners. In addition to this initial increase in ACh sensitivity following innervation there is a slower development of postsynaptic ACh sensitivity, with continued innervation in vitro. Many studies of the regulation of AChRs by innervation of muscle have indicated that the innervating nerve plays a critical role in determining both the number and the distribution of AChRs on the muscle cell surface. These studies also indicate that AChR subunit mRNA and AChR protein are increased at the region of the synapse (see Salpeter and Loring, 1985; Steinbach and Bloch, 1986; Schuetze and Role, 1987; Brehm and Henderson, 1988 for reviews). The search for a neurally derived factor or factors that regulate AChR expression in muscle has yielded at least one activity that increases the expression of both new AChR protein on the muscle surface and enhanced levels of c~-subunit mRNA (Usdin and Fischbach, 1986; Harris et aL, 1988, 1991). Using a strictly functional assay of neuronal AChRs, we find that innervation by cholinergic neurons--either somatic or autonomic--increases the ACh sensitivity of sympathetic neurons in vitro. Presynaptic regulation of postsynaptic transmitter sensitivity has been previously suggested for ACh (Role, 1988), glutamate (O'Brien and Fischbach, 1986), and adrenergic responses (Pun et al., 1985) in neurons. O'Brien and Fisch-

91

bach (1986) examined changes in sensitivity to glutamate following coculture of motoneurons with interneurons. These culture conditions resulted in a high degree of synaptic input to the motoneurons, enhanced glutamate sensitivity, and an apparent redistribution of glutamate sensitivity to "hotspots" at sites of neuritecell contact. Similar areas of high agonist sensitivity on spinal cord neurons were reported by MacDermott and collaborators, (Arancio and MacDermott, 1991) who examined glutamate-sensitive regions on neurites for the relative contribution of NMDA vs non-NMDA-type receptors. Cocultures of specific populations of brainstem neurons and prospective target neurons have yielded qualitatively similar results. For example, studies by Nelson and colleagues (Pun et al., 1985) suggest that locus ceruleus neurons might induce a-adrenergic receptor expression in innervated spinal cord neurons. Several studies have attempted to determine the role of presynaptic input in regulating transmitter sensitivity by examining the effects of denervation. Whereas denervation results in a dramatic, activity-dependent increase in AChR number in muscle, the effects of denervation on the ACh sensitivity of autonomic neurons are more controversial. Early results indicate a denervation-dependent increase in ACh sensitivity similar to that in muscle. More recent studies suggest that this result was due to decreased acetylcholinesterase activity and demonstrate that denervation causes (at most) small decreases in neuronal AChRs (Schuetze and Role, 1987 for review; Kuffier et al., 1971; Roper, 1976; Dennis and Sargent, 1979; Dunn and Marshall, 1985; Jacob and Berg, 1988; McEachern et aL, 1989). Molecular analyses of changes in nAChR subunit expression also demonstrate a small decrease in the levels of expression of ~3-subunit mRNA following denervation (Boyd et al., 1988). The decreases in neuronal AChRs following denervation that are detected by physiological or biochemical assays are so small that investigators have concluded that presynaptic input is not important in controlling neuronal AChRs (Jacob and Berg, 1988). However, in view of findings presented here and previous work on changes in neuronal transmitter sensitivity following innervation, we would suggest that initial neural input might play an important role in regulating postsynaptic transmitter sensitivity but that maintenance of presynaptic input is not required to retain the high transmitter sensitivity once established. Previous studies have also examined the regulation of ACh sensitivity in neurons relative to contact with target. In particular, postganglionic axotomy of ciliary ganglia apparently results in a decrease in ACh sensitivity (Brenner and Martin, 1976; but see Engisch and Fischbach, 1990), a decrease in AChR antibody binding sites (Jacob and Berg, 1988), and a decrease in AChR subunit

92

DEVELOPMENTALBIOLOGY

mRNA (Boyd et al., 1988). Thus, target may also play a critical regulatory role in determining transmitter receptor expression in neurons. The data presented here indicate t h a t not only cholinergic neurons but also soluble factors from these cholinergic neurons can regulate neuronal AChRs. Previous studies demonstrated t h a t neither ACh itself nor functional synaptic transmission was required for the observed increase in ACh sensitivity following innervation by dorsal spinal cord explants (Role, 1988). We extend these findings to demonstrate t h a t ventral spinal cord explants can also rapidly induce a significant increase in the ACh sensitivity of sympathetic neurons in the presence or absence of functional synaptic connections. The regulation of neuronal ACh sensitivity by either dorsal or ventral spinal cord explants (but not by explants of noncholinergic neurons or by nonneuronal cells from spihal cord; Role, 1988) is consistent with previous findings on the regulation of muscle AChRs. Increases in AChR number and clustering in muscle are induced by cholinergic neurons other than spinal motoneurons and extracts of such tissues (e.g., ciliary ganglion neurons) but not by noncholinergic neurons or their extracts (see Schuetze and Role, 1987 for review). Additional insight into the mechanism of regulation of neuronal AChRs by neurally derived factors is provided by our observation t h a t conditioned medium from spinal cord explants specifically increases the rate of appearance of new nAChRs on the neuronal surface. This suggests t h a t soluble factors increase nAChR synthesis and/or insertion or decrease the rate of degradation of newly synthesized nAChRs. The effects of conditioned medium on neuronal ACh sensitivity are consistent with the notion t h a t neuronal AChRs might be regulated by mechanisms similar to those of muscle AChRs. Although these experiments cannot determine the relative contribution of innervation-independent vs innervation-dependent factors in the development of ACh sensitivity in neurons, it is interesting to note that in experiments where the explant density was high the conditioned medium was quantitatively as effective as innervation. The relative efficiency of dorsal explants containing preganglionic neurons versus ventral spinal cord explants in forming synapses with sympathetic neurons may be related to adhesion and/or recognition. Fewer ventral explants successfully attached to the substrate and, although the initial extension of processes from dorsal and ventral explants was similar, neurites from ventral spinal cord explants cocultured with sympathetic neurons were more fasciculated and did not extend as far as those from dorsal spinal cord explants under the same conditions. As a result the number of sympathetic neurons contacted by ventral explants was

VOLUME147, 1991

somewhat less than the number contacted by dorsal spihal cord explants. Recent studies by Moorman and Hume (1990) describe distinct patterns of interaction of preganglionic axons with sympathetic vs sensory neurons. These studies show t h a t preganglionic neurons prefer (i.e., will contact and extend across) their appropriate target but overtly retract upon contact with sensory neurons. In this view, neurites from preganglionic neurons in the dorsal explants might have a higher probability of maintaining a stable contact with sympathetic neurons than neurites from motoneurons in the ventral explants. Even once a contact is formed, however, the ventral explants are less effective then the dorsal explants in developing a functional synapse. We think it is unlikely that the relative success of dorsal versus ventral explants in innervating sympathetic neurons represents cell death since media conditioned by relatively few ventral explants increased ACh sensitivity two- to fourfold compared with untreated controls.

Development of Synaptic Transmission Studies of synaptic function in developing rat and frog sympathetic ganglia in vivo (Rubin, 1985a,b; L. Marshall; personal communication) have indicated that early synapses are weak, fatigue rapidly, and display little spontaneous activity. However, because of the difficulty of identifying newly innervated neurons in developing animals, there is little information on the nature of transmission of these early contacts or on the development of transmitter sensitivity relative to the time of onset of synaptic function. The advantages of studying neuronal innervation in vitro lie in both the ability to control the timing of innervation and the relative ease in identifying neurons before and shortly after initial innervation. These in vitro studies reveal t h a t increased postsynaptic transmitter sensitivity of sympathetic neurons was accompanied by maturation of synaptic transmission. This includes an increased frequency of spontaneous synaptic epsc's and a small but significant increase in the average amplitude of the unit synaptic current. It is, therefore, likely that both pre- and postsynaptic changes contribute to the observed increase in synaptic efficacy, reflected in increases in both the extent and the frequency of suprathreshold activity with development in vitro. Similar to previous findings in vivo (Rubin, 1985a,b; L. Marshall, personal communication) the earliest contacts between explants of preganglionic neurons and primary sympathetic neurons are not at all robust. The frequency of spontaneous events is very low and examination of amplitude histograms of the spontaneous

GARDETTEETAL.

Developmental Changes of Chick Neurons

epsc's indicates that most events fall into the first amplitude mode. In addition, the amplitude of the unit synaptic current mode is small--only 13 pA on average. This finding is particularly striking in view of previous work on the properties of nAChR channels expressed by these neurons (Moss et al., 1989; Brussaard and Role, 1990) which suggests that these synaptic currents would be the result of activation of, at most, ~13 nAChR channels. This is in contrast to results at developing neuromuscular synapses (e.g., Frank and Fischbach, 1979; Role et al., 1987; but see Cohen, 1980) but consistent with previous data on interneuron-spinal motoneuron synapses (O'Brien and Fischbach, 1986). This result underscores the potential importance of even small changes in the properties and/or number of AChRs expressed at developing synapses which might dramatically alter the impact of a given contact. After 4 days of innervation several features of the synaptic transmission change in a manner that may result from pre- and postsynaptic development. Increases in the minifrequency and unit miniamplitude are accompanied by a significant decrease in the percentage of synaptic currents within the unit mode. The small increase in the average amplitude of the unit mode is most likely due to the increase in postsynaptic ACh sensitivity, although it is theoretically possible t h a t this too is presynaptic in origin (i.e., due to increased ACh/vesicle). Whether pre- or postsynaptic in origin, concurrent with these observed changes in the synaptic currents there is an increase in the frequency of synaptically driven suprathreshold responses in the sympathetic neurons. These findings are consistent with previous reports documenting the maturation of synaptic transmission in developing sympathetic ganglia in vivo reported by Rubin (1985a,b). The analysis of the histograms of synaptic currents indicates t h a t they are best fit, early on, by one and later, by the sum of at least two or three Gaussian distributions. Furthermore, statistical analysis of the mean amplitudes of the individual peaks in the multimodal histograms indicates that the mean of the second mode is about twice that of the first. Since the peaks of the synaptic current amplitude histogram are integral multiples of the unit mode, one explanation is that the spontaneous release at these early synapses could be quantal. Quantal release at newly formed synapses has been previously suggested from work on the developing neuromuscular junction, although the low frequency and skewed nature of the spontaneous and evoked synaptic current histograms make it difficult to establish this point conclusively (Kidokoro et al., 1980; Kidokoro and Yeh, 1982; Kidokoro, 1984, Role et al., 1987). It should be noted that these multimodal histograms were obtained in the presence of a concentration of TTX

93

t h a t blocks all evoked action potential activity in both the preganglionic and the postganglionic sympathetic neurons in chick (Moorman and Hume, personal communication; W. Bug and L. Role, unpublished observations; but not so in frogs: see Bowers, 1985) and hence are comprised entirely of spontaneous synaptic currents. The observation t h a t the mean of the second peak of the amplitude histograms is twice that of the first could be due to (a) concurrent release of two quanta, (b) synaptic vesicles loaded with twice the amount of neurotransmitter, or (c) activation of postsynaptic sites of twice the conductance (i.e., twice as many nAChRs or nAChRs of twice the conductance inserted at the synaptic site). The latter possibility, although not typically invoked, is supported by previous nAChR single channel studies on these cells, indicating both clustering and segregation of channels of a particular conductance class on innervated neurons (Moss et al., 1989; Moss and Role, 1991). In addition, these studies have revealed that innervation results in the expression of a class of channels of twice the conductance of one that is expressed prior to innervation (Moss et al., 1989; Moss and Role, 1991; Brussaard and Role, unpublished observations). Studies of nerve-muscle synaptic development indicate that increases in minifrequency and changes in postsynaptic receptor distribution can occur rapidly following initial nerve-muscle contact (Anderson and Cohen, 1977; Kullberg et al., 1977; Frank and Fischbach, 1979; Fischbach et al., 1979; Olek et al., 1983; Xie and Poo, 1986; Evers et al., 1989). The studies presented here show t h a t innervation of sympathetic neurons is also followed by a rapid increase in postsynaptic ACh sensitivity as well as more protracted changes in transmitter release and postsynaptic responses. Our studies suggest t h a t the maturation of postsynaptic function is due, at least in part, to presynaptic-derived factor(s). Previous work also suggests t h a t interactions of pre- and postsynaptic cells enhance the frequency of spontaneous release from cholinergic neurons in a manner similar to t h a t reported here (Xie and Poo, 1986; Evers et al., 1989). If the changes revealed by these in vitro studies also occur in vivo, inductive interactions between the pre- and the postsynaptic neurons are likely to have an important role in synaptic development. This study was supported by grants from the NIH (NS22061), McKnight Foundation, and The Council for Tobacco Research (to L.R.), an individual predoctoral award from the National Science Foundation (to M.L.),a NATOgrant and MuscularDystrophyFellowship Award (ABB)and grants from the CNRS,NATO,Fondationpour la Recherche Medicale,and Philippe Foundation(to R.G.). We thank Drs. A. MacDermottand T. Jessell for criticalcommentsand adviceon a previous version of the manuscript. Note added in proof. A recent study of AChR expression during synaptogenesis in chicken ciliary ganglia in vivo also concludes that

94

DEVELOPMENTALBIOLOGY

signals from presynaptic input are important for the induction of AChR expression in neurons (M. H. Jacob. Acetylcholine receptor expression in developing chick ciliary ganglion neurons. J. Neurosci. 11, 1701-1712.). REFERENCES ANDERSON,M. J., and COHEN,M. W. (1977). Nerve-induced and spontaneous redistribution of acetylcholine receptors on cultured muscle cells. J. PhysioL (London) 268, 757-773. ARANCIO,0., and MACDERMOTT,A. B. (1991). Differential distribution of excitatory amino acid receptors on embryonic rat spinal cord neurons in culture. J. NeurophysioL 65, 899-913. BERG, D. K., BOYD, R. T., HALVORSEN, S. W., HIGGINS, L. S., JACOB, M. H., and MARGIOTTA,J. F. (1989). Regulating the number and function of neuronal acetylcholine receptors. Trends Neurosc£ 12, 16-21. BOWERS, C. (1985). A cadmium sensitive, tetrodotoxin-resistant sodium channel in bullfrog autonomic axons. Br. Res. 340, 143-147. BOYD, R. T., JACOB, M. H., COUTURIER,S., BALLIVET, M., and BERG, D. K. (1988). Expression and regulation of acetylcholine receptor mRNA in chick ciliary ganglia. Neuron 1,495-502. BREHM, e., and HENDERSON, L. (1988). Regulation of acetylcholine receptor channel function during development of skeletal muscle. Dev. Biol. 129, 1-11. BRENNER, H. R., and MARTIN,A. R. (1976). Reduction in acetylcholine sensitivity of axotomized ciliary ganglion cells. J. Physiol. (London) 260, 159-175. BROWN, D. A., and KWIATKOWSKI,D. (1976). A note on the effect of dithiothreitol (DTT) on the depolarization of isolated sympathetic ganglia by carbachol and bromo-acetylcholine. Br. J. PharmacoL 56, 128-130. BRUSSAARD,A. B., and ROLE, L. W. (1990). Regulation of four distinct nicotinic ion channel types in sympathetic neurons following innervation in vitro. Soc. Neurosci. Abstr. 16, 93.11. CHOI, D. W., and FISCHBACH,G. D. (1981). GABA-mediated synaptic potentials in chick spinal cord and sensory neurons. J. NeurophysioL 45, 605-620. COHEN, S. i . (1980). Early nerve-muscle synapses in vitro release transmitter over postsynaptic membrane having low acetylcholine sensitivity. Proc. Natl. Acad. Sci. USA 77, 644-648. DENNIS, M. J., and SARGENT,P. B. (1979) Loss of extrasynaptic acetylcholine sensitivity upon reinnervation of parasympathetic ganglion cells. J. Physiol. (London) 289, 263-275. DUNN, P. M., and MARSHALL,L. M. (1985). Lack of nicotinic supersensitivity in frog sympathetic neurones following denervation. J. Physiol. (London) 363, 211-225. ENGISCH, K. L., and FISCHBACH, G. D. (1990). The development of ACH- and GABA-activated currents in normal and target-deprived embryonic chick ciliary ganglia. Dev. Biol. 139, 417-426. EVERS, J., LASER, M., SUN, Y. A., XIE, Z. P., and Poo, M. M. (1989). Studies of nerve-muscle interactions in Xenopus cell culture: Analysis of early synaptic currents. J. Neurosci. 9, 1523-1539. FISCHBACH, G. D., FRANK, E., JESSELL, T. M., RUBIN, L. L., and SCHUETZE, S. M. (1979). Accumulation of acetylcholine receptors and acetylcholinesterase at newly formed nerve-muscle synapses. Pharmacol. Rev. 30, 411-428. FRANK, E., and FISCHBACH,G. D. (1979). Early events in neuromuscular junction formation in vitro. Induction of acetylcholine receptor clusters in the postsynaptic membrane and morphology of newly formed synapses. J. Cell Biol. 83, 143-158. GARDETTE, R., MAWE, G. M., D'AGOSTARO,L., and ROLE, L. W. (1988). Early innervation and transmitter sensitivity of sympathetic neurons in vitro: Evaluation of synaptic transmission by electrophysiol-

VOLUME147, 1991

ogy and cytochrome oxidase histochemistry. Soc. Neurosc~ Abstr. 18, 330.4 GARDETTE,R., and ROLE,L. W. (1989). Development of synaptic transmission and transmitter sensitivity of sympathetic neurons innervated in vitro by appropriate vs inappropriate presynaptic input. Soc, Neurosci, Abstr. 19, 70.10. GOLDMAN,D., BRENNER, H. R., and HEINEMANN,S. (1988). Acetylcholine receptor a-, fl-, and ~-subunit mRNA levels are regulated by muscle activity. Neuron 1,329-333. HAMILL, O. P., MARTY, A., NEHER, E., SAKMANN, B., and SIGWORTH, F. J. (1981). Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pflugers Arch. 391, 85-100. HARRIS, D. A., FALLS, D. L., DILL-DEVOR,R. M., and FISCHBACH,G. D. (1988). Acetylcholine receptor-inducing factor from chicken brain increases the level of mRNA encoding the receptor a-alpha subunit. Proc. NatL Acad. Sci. USA 85, 1983-1987. HARRIS, D. A., FALLS, D. L., JOHNSON, F., and FISCHBACH,G. D. (1991). A prion-like protein from chicken brain co-purifies with acetylcholine receptor-inducing activity. Proc. Natl AcacL Sci. USA, in press. HARTZELL,H. C., and FAMBROUGH,D. M. (1972). Acetylcholine receptors: Distribution and extrajunctional density in rat diaphragm after denervation correlated with acetylcholine sensitivity. J. Gen. Physiol. 60, 248-262. HIGGINS, L. S., and BERG, D. K. (1988). Metabolic stability and antigenic modulation of nicotinic acetylcholine receptors on bovine adrenal chromaffin cells. J. Cell Biol. 107, 1147-1156. HUME, R. I., ROLE, L. W., and FISCHBACH, G. D. (1983). ACh release from growth cones detected with patches of ACh receptor-rich membranes. Nature (London) 305, 632-635. JACOB, M. H., and BERG, D. K. (1988). The distribution of acetylcholine receptors in chick ciliary ganglion neurons following disruption of ganglionic connections. J. Neurosci 8, 3838-3849. KARLIN, A., and BARTELS, E. (1966). Effects of blocking sulfhydryl groups and of reducing disulfide bonds on the acetylcholine-activated permeability system of the electroplax. Biochim. Biophys. Acta 126, 525-535. KIDOKORO, Y. (1984). Two types of miniature endplate potentials in Xenopus nerve-muscle cultures. NeuroscL Res. 1, 157-170. KIDOKORO, Y., ANDERSON,M. J., and GRUENER,R. (1980). Changes in synaptic potential properties during acetylcholine receptor accumulation and neurospecific interactions in Xenopus nerve-muscle cell culture. Dev. BioL 78, 464-483. KIDOKORO,Y., and YEH, E. (1982) Initial synaptic transmission at the growth cone in Xenopus nerve-muscle cultures. Proc. NatL Acad. Sci. USA 79, 6727-6731. KLARSFELD, A., and CHANGEUX,J. P. (1985). Activity regulates the levels of acetylcholine receptor a-subunit mRNA in cultured chicken myotubes. Proc. Natl. Acad. Sc£ USA 82, 4558-4562. KUFFLER, S. W., DENNIS, M. J., and HARRIS, A. J. (1971). The development of chemosensitivity in extrasynaptic areas of the neuronal surface after denervation of parasympathetic ganglion cells in the heart of the frog. Proc. R. Soc. London B 177, 555-563. KULLBERG,R. W., LENTZ, T. L., and COHEN, M. W. (1977). Development of the myotomal neuromuscular junction in Xenovas laevis: An elec= trophysiological and fine structural study. Dev. Biol. 60, 101-129. LANGLEY,J. N. (1904). On the sympathetic system of birds and on the muscles which move the feathers. J. PhysioL (London) 30, 221-252. LEAH, J. D., GIBBS, E., and KIDSON, C. (1986). Synapse formation and induction of acetylcholine receptors by spinal neurones in cocultures with sympathetic ganglion and muscle cells. Dev Neurosc£ 8, 76-88. LISTERUD, M., BRUSSAARD,A. B., DEVAY, P., COLMAN, D., and ROLE, L. W. (1991). Functional contribution of individual subunits to nico-

GARDETTE ET AL.

Developmental Changes o f Chick Neurons

tinic AChR channels in sympathetic neurons revealed by antisense oligonucleotides, submitted for publication. LORING, R. H., DAHM, L. M., and ZIGMOND, R. E. (1985). Localization of a-bungarotoxin binding sites in the ciliary ganglion of the embryonic chick: An autoradiographic study at the light and electron microscopic level. Neuroscience 14, 645-660. MAGILL-SOLC, C., and MCMAHAN, U. J. (1990). Agrin-like molecules in motor neurons. J. Physiol. (Paris) 84, 78-81. MARGIOTTA, J. F., and GURANTZ, D. (1989). Changes in the number, function, and regulation of nicotinic acetylcholine receptors during neuronal development. Dev. Biol. 135, 326-339. MARSHALL, L. M. (1985). Presynaptic control of synaptic channel kinetics in sympathetic neurones. Nature 317, 621-623. MARSHALL, L. M. (1986). Different synaptic channel kinetics in sympathetic B and C neurons of the bullfrog. J. Neurosci. 6, 590-593. MAWE, G. M., GARDETTE, R., D'AGOSTARO, L., and ROLE, L. W. (1990). Synaptic transmission at autonomic synapses in vitro revealed by cytochrome oxidase histochemistry. J. Neurobiol. 21, 578-591. MCEACHERN, A. E., JACOB, M. H., and BERG, D. K. (1989). Differential effects of nerve transection on the ACh and GABA receptors of chick ciliary ganglion neurons. J. Neurosci. 9, 3899-3907. MCMAHAN, U. J., and WALLACE, R. G. (1989). Molecules in basal lamina t h a t direct the formation of synaptic specializations at neuromuscular junctions. Dev. Neurosci. l l , 227-247. MCLACHLAN, E. M. (1973). The formation of synapses in mammalian sympathetic ganglia reinnervated with preganglionic or somatic nerves. J. Physiol. London 237, 217-242. MERLIE, J. P., ISENBERG, K. E., RUSSELL, S. D., and SANES, J. R. (1984). Denervation supersensitivity in skeletal muscle: Analysis with a cloned cDNA probe. J. Cell Biol. 99, 332-335. MOORMAN, S. J., and HUME, R. I. (1990). Growth cones of chick sympathetic preganglionic neurons in vitro interact with other neurons in a cell-specific manner. J. Neurosci. 10, 3158-3163. MOSS, B. L., SCHUETZE, S. M., and ROLE, L. W. (1989). Functional properties and developmental of regulation of nicotinic acetylcholine receptors on embryonic chicken sympathetic neurons. Neuron 3, 597-607. MOSS, B. L., and ROLE, L. W. (1991). Regulation of acetylcholine receptor channel properties and segregation of receptor subtypes on embryonic chicken sympathetic neurons during innervation in vivo, submitted for publication. NELSON, P. G., Yu, C., FIELDS, R. D., and NEALE, E. A. (1989). Synaptic connections in vitro: Modulation of number and efficacy by electrical activity. Science 244, 585-587. 0'BRIEN, R. J., and FISCHBACH, G. D. (1986). Excitatory synaptic transmission between interneurons and motoneurons in chick spinal cord cell cultures. J. Neurosci. 6, 3284-3289. O'LAGUE, P. H., POTTER, D. D., and FURSHPAN, E. J. (1978). Studies on r a t sympathetic neurons developing in cell culture. Dev. BioL 67, 384-443.

95

OLEK, A. J., PUDIMAT, P. A., and DANIELS, M. P. (1983). Direct observation of the rapid aggregation of acetylcholine receptors on identified cultured myotubes after exposure to embryonic brain extract. Cell 34, 255-264. PUN, R. Y. K., MARSHALL, K. C., HENDELMAN, W. J., GUTHRIE, P. B., and NELSON, P. G. (1985). Noradrenergic responses of spinal neurons in locus coeruleus-spinal cord co-cultures. J. Neurosci. 5, 181191. ROLE, L. W. (1984). Substance P modulation of acetylcholine-induced currents in embryonic chicken sympathetic and ciliary ganglion neurons in vitro. Proc. Natl. Acad. Sci. USA 81, 2924-2928. ROLE, L. W. (1988). Neural regulation of acetylcholine sensitivity in embryonic sympathetic neurons. Proc. Natl. Acad. Sci. USA 85, 2825-2829. ROLE, L. W., ROUFA, D. G., and FISCHBACH, G. D. (1987). The distribution of acetylcholine receptor clusters and sites of t r a n s m i t t e r release along chick ciliary ganglion neurite-myotube contacts in culture. J. Cell Biol. 104, 371-379. ROPER, S. (1976). The acetylcholine sensitivity of the surface membrane of multiply-innervated parasympathetic ganglion cells in the mudpuppy before and after partial denervation. J. PhysioL (London) 254, 455-473. RUBIN, E. (1985a). Development of the r a t superior cervical ganglion: Ingrowth of preganglionic axons. J. Neurosci. 5, 685-696. RUBIN, E. (1985b). Development of the r a t superior cervical ganglion: Initial stages of synapse formation. J. Neurosci. 5, 697-704. SALPETER, M. M., and LORING, R. H. (1985). Nicotinic acetylcholine receptors in vertebrate muscle: Properties, distribution and neural control. Prog. Neurobiol. 25, 297-325. SCHUETZE, S. M., and ROLE, L. W. (1987). Developmental regulation of nicotinic acetylcholine receptors. Annu. Rev. Neurosci. 10, 403-457. SNEDECOR, G. W., and COCHRAN, W. G. (1989). "Statistical Methods" Eighth ed. Iowa State Univ. Press, Ames, IA. STEINBACH, J. H., and BLOCH, R. J. (1986). The distribution of acetylcholine receptors on vertebrate skeletal muscle cells. In "Receptors in Cellular Recognition and Developmental Processes," pp. 183-213. Academic Press, New York. STEINBACH, J. H., and IFUNE, C. (1989). How many kinds of nicotinic acetylcholine receptors are there? Trends Neurosci. 12, 3-6. USDIN, T. B., and FISCHBACH, G. D. (1986). Purification and characterization of a polypeptide from chick brain t h a t promotes the accumulation of acetylcholine receptors in chick myotubes. J. Cell Biol. 103, 493-507. XIE, Z. P., and POO, M. M. (1986). Initial events in the formation of neuromuscular synapse: Rapid induction of acetylcholine release from embryonic neuron. Proc. NatL Acad. Sci. USA 83, 7069-7073. YIP, J. W. (1986) Specific innervation of neurons in the paravertebral sympathetic ganglia of the chick. J. Neurosci 6, 3459-3464. YIP, J. W. (1990). Identification of location and timing of guidance cues in sympathetic preganglionic axons of the chick. J. Neurosci 10, 2476-2484.