Transmitter Release Sites Research Articles

Variation among acetylcholine receptor clusters induced by ciliary ganglion neurons in vitro

We have examined the variation in receptor density and area among neurite-associated acetylcholine receptor patches (NARPs) induced by chick ciliary ganglion neurons on nearby myotubes in vitro. Quantitative analysis of rhodamine-α-bungarotoxin (RBTX) NARPs revealed that about 15% of the NARPs were “outstanding” in terms of size (>60 μm 2) and fluorescence intensity (>100 units on a 0–255 scale). The total number of receptors at different NARPs ranged over 3 orders of magnitude. It is likely that variation in NARP size and intensity reflects regional variation in the ability of myotubes to respond to the neuronal influence because (1) no gradient in NARP size or intensity with distance from the soma was evident; (2) the intensities and areas of uninnervated receptor clusters (hot spots) were similar to those of NARPs; (3) acetylcholinesterase was present at the same proportion of hot spots and NARPs at all times examined. We found no physiological or morphological evidence that outstanding NARPs were more effective sites of transmitter release. Outstanding NARPs were restricted to the longest neurite of individual neurons, so they may signal trophic interactions of the sort that promote neurite outgrowth and survival.

Changes in binomial parameters of quantal release at crustacean motor axon terminals during presynaptic inhibition.

1. The effects of presynaptic inhibition on quantal release of transmitter were investigated at neuromuscular junctions of the motor axon supplying one of the limb muscles of a crab (Pachygrapsus crassipes). 2. Binomial analysis of transmitter release recorded at selected neuromuscular junctions with an extracellular 'macro-patch' electrode indicated high probability of release (p) from a limited number of available sites (n). During presynaptic inhibition, both n and p were reduced. 3. The binomial model provided a good description of results from non-inhibited junctions. During presynaptic inhibition, results from some junctions could be described by the binomial model, while those from other junctions could not. An interpretation of this finding is that presynaptic inhibition differentially affects the probability of release at various release sites of the neuromuscular junctional complex. 4. A morphological study of the region of transmitter release under the macropatch electrode was made. Release-dependent uptake of horseradish peroxidase (HRP) into presynaptic terminals was restricted to the region under the recording electrode, by perfusing the preparation with calcium-free solution containing HRP. Transmitter release, and HRP uptake, occurred only at the site of the electrode, which was filled with a calcium-containing solution. Subsequently, serial sections were prepared for electron microscopy and the region of transmitter release was reconstructed. 5. Numerous axo-axonal synapses were found in the HRP-labelled region. Thus, the morphological prerequisite for presynaptic inhibition exists at the site of transmitter release, and not exclusively at a more remote region. 6. The number of morphologically identified excitatory neuromuscular synapses exceeded the 'release sites' estimated from the binomial model (n) by a wide margin. Morphological differences among synapses were observed. It is proposed that not all morphologically identified synapses participated in transmitter release under the experimental conditions employed. Thus, morphologically defined synapses are likely to be non-uniform in their response properties, including probability of transmitter release (p).

Analysis of transmission at interneuronal synapses using a convolution of binomial distributions

The convolution of two binomial distributions has been used for the analysis of the discrete distributions of the elementary EPSPs amplitude in the sensorimotor, proprio- and reticulo-motoneuronal synapses of the frog. It is shown that the quantity of the working transmitter release sites is higher than that estimated by the binomial distribution. The probability of the release sites response to the single action potential was unequal. In most sites they did not exceed 0.1-0.3 and were supported on the constant level. The effect on the transmitter release mechanism induced blocking of the release sites. The disturbance of preparation to the transmitter release sharply decreased the probability of the release sites response.

Highly active neuromuscular system in developing lobsters with programmed obsolescence.

The primary locomotory apparatus in the three larval stages of the lobster, Homarus americanus, are paddlelike structures on the thoracic appendages called exopodites, which beat almost continuously. Consequently their power and return-stroke muscles are examples of highly active but short-lived neuromuscular systems. The muscles, which are well vascularized, are of the fast type with 2-3-micron sarcomere lengths and 6 thin filaments surrounding a thick one. The most striking feature, however, is the large volume of mitochondria making up 40-50% of the fiber. They appear as simple cylinders packed several layers deep along the periphery of the fiber and as large, multibranched forms distributed throughout the fiber and subdividing it into smaller units. The motor innervation to the return-stroke muscle is via 3 excitatory axons, which generate large junctional potentials and twitch contractions. The muscle is densely populated with large neuromuscular synapses, most of which have a well-defined active site or dense bar denoting the site of transmitter release. Altogether this motor system is specialized for prolonged activity. Atrophy of the neuromuscular system occurs by the late larval third stage. The muscle fibers lose their identity, fuse, and become vacuolated. The myofibrils condense and erode and the mitochondria are lost. Atrophy of motor innervation is gradual with individual axons dropping out. The largest axon providing most of the innervation is the first to degenerate. Early degenerative changes affect the axon and neuromuscular terminals but not the synaptic contacts, dense bars, and vesicles, which appear intact. Continued atrophy in the postlarval fourth stage reduces the exopodites to vestiges. Thus the return-stroke muscle of the larval exopodites in which muscle fiber and motoneurons are identifiable permits study of the interaction between a neuron and its target muscle undergoing programmed obsolescence.

Disruption of active zones in frog neuromuscular junctions following treatment with proteolytic enzymes.

Frog neuromuscular junction treated with proteolytic enzymes to remove the basal lamina were studied with freeze-fracture techniques in order to examine the influence of the basal lamina in the maintenance of active zone ultrastructure. The active zone is believed to be the site of transmitter release and has a unique membrane organization and location in the neuromuscular junction. After removal of the basal lamina by successive treatment of 0.01% collagenase and 0.1% protease for 1 h each, active zone disruption was observed. Some active zones became segmented, and some were also randomly located and oriented, but they still had normal double-row particle organization. Others contained only clusters of large intramembrane particles. These disorganized active zones were still functional as indicated by the presence of vesicle openings. Some enzyme-treated junctions were also exposed to the membrane cholesterol probe, filipin, to examine the expression of membrane lipid heterogeneity in disrupted active zones. As in normal active zones, filipin-sterol complexes were absent. The densities of background particles in the presynaptic membranes and of large particles thought to be acetylcholine receptors were not significantly altered by the enzyme treatment. Although a direct effect of the enzymes on active zone ultrastructure can not be totally excluded, the present work is consistent with a maintenance role of the basal lamina in active zone organization and location.

Morphology, ultrastructure and synapse distribution of putative serotonergic salivary neurons in the locust

Neurons innervating the locust salivary glands have been investigated using the electron microscope. We have examined the projections in the suboesophageal ganglion of two identified salivary neurons by differential double labelling with cobalt ions, followed by silver intensification. Numerous synaptic inputs occur on the arborizations of the salivary neurons, particularly on fine branches and on small spines arising from larger branches. Although a few instances occur of common input to both salivary neurons from a single presynaptic element, many of the appositions between branches seen in the light microscope do not represent functional connections. A few structures resembling presynaptic dense bodies have been observed in salivary neuron profiles, but with few synaptic vesicles. Large dense granules are present in some labelled profiles, but not in the vicinity of synapse-like membrane specializations. We have also examined unidentified neuron profiles within the acini of the salivary glands, which contain large dense granules and small vesicles. There is good evidence that these unidentified terminals correspond to the suboesophageal salivary neurons. The central arborizations of the salivary neurons appear to serve largely as sites of synaptic input, whereas the peripheral terminals are likely sites of transmitter release. The results are considered in the context of the known immunoreactivity of the salivary neurons with antibodies to 5-hydroxytryptamine, and compared with analagous systems in other insect groups.

Non-uniform changes in nerve-terminal calcium homeostasis during aging

Currently unpublished studies indicate reduced rates of calcium clearance from the transmitter release sites in presynaptic terminals of some, but not all, motor nerves in aged rats. Further evidence indicates that deficits in calcium regulation may not necessarily underlie observed age-related changes in evoked transmitter release.

Noradrenaline and β-adrenoceptor agonists increase activity of voltage-dependent calcium channels in hippocampal neurons

The predominance of unconventional transmitter release sites at noradrenaline-containing synapses and the diffuse projections of noradrenaline-containing fibres originating in locus coeruleus have led to speculation that noradrenaline may act as a neuromodulator in the central nervous system. Evidence suggests that it has a modulatory function in the plasticity of the developing nervous system, in controlling behavioural states of an organism, and in learning and memory. Recently, Hopkins and Johnston demonstrated that noradrenaline enhances the magnitude, duration and probability of induction of long-term potentiation (LTP) at mossy fibre synapses in the hippocampal formation, and LTP is widely believed to be a cellular substrate for aspects of memory. To investigate the membrane effects of noradrenaline on central neurons, we used a newly developed preparation in which patch-clamp techniques can be applied to exposed adult cortical neurons. We report here that noradrenaline produces an enhancement in the activity of voltage-dependent calcium channels in granule cells of the hippocampal dentate gyrus. This action appears to be mediated by beta-adrenoceptors and can be mimicked by cyclic AMP.

Transmitter localization and vesicle turnover at a serotoninergic synapse between identified leech neurons in culture

An electron microscopic study has been made of chemical synapses that develop between identified nerve cells isolated from the CNS of the leech and maintained in culture. Structures resembling synapses were observed in pairs of Retzius cells and P sensory cells at which chemical transmission had been demonstrated by recording with microelectrodes. Vesicle recycling was shown by following the uptake of extracellular markers after stimulation. The membrane separation between the presynaptic Retzius cell (which is known to liberate serotonin) and the postsynaptic P cell was wider in synaptic than in extrasynaptic regions. The Retzius cell contained clusters of clear vesicles apposed to thickenings of the presynaptic membrane. These clear vesicle clusters were capped by a layer of dense core vesicles that did not contact the presynaptic membrane thickenings. Subsynaptic cisternae were found in the postsynaptic cell opposite the presynaptic membrane thickenings. Occasional slight postsynaptic membrane thickenings were seen. Extracellular material was observed within the synaptic cleft. Similar synaptic structures developed between pairs of Retzius cells in culture; even a single Retzius cell was able to form autapses upon itself. Structures resembling transmitter release sites were found in Retzius cells at a distance from any postsynaptic membranes. These are presumed to be locations for the diffuse release of transmitter. Presynaptic structures resembling release sites were never observed in P cells apposed to Retzius cells. Antibody to serotonin (5-HT) labelled with colloidal gold showed serotonin to be localized in the dense core vesicles in Retzius cells. Stimulation of pairs of Retzius and P cells by raised concentrations of K+ resulted in uptake of extracellular markers. Only Retzius cells became labelled. Ferritin was found in cisternae, in dense core vesicles, and in clear vesicles. HRP was found in cisternae and in clear vesicles. Colloidal gold was taken up by coated vesicles and was occasionally found in both clear and dense core vesicles. The uptake of extracellular markers following stimulation was blocked by high Mg++. These results show that structures develop between pairs of cells at which chemical transmission develops and that transmitter release leads to turnover of dense core and clear vesicles.

The distribution of acetylcholine receptor clusters and sites of transmitter release along chick ciliary ganglion neurite-myotube contacts in culture.

Acetylcholine receptors accumulate along the length of cholinergic neuron-skeletal muscle contacts in vitro. The main purpose of this study was to describe, in a quantitative way, the distribution of acetylcholine receptor clusters induced by ciliary ganglion neurons over a period of time extending from hours to weeks after contacts are established. Neurites were filled with Lucifer Yellow and receptor clusters were identified with rhodamine-bungarotoxin. A cluster located within 5 micron of a nerve process or 10 micron of the base of a growth cone was considered to be a neurite-associated receptor patch (NARP). The first synaptic potentials were evoked 20 min after growth cone-myotube contact, and, after 24 h of co-culture, greater than 60% of the nerve-muscle pairs tested were functionally connected. NARPs appear rapidly; the first clusters were detected approximately 6 h after the neurons were plated. They were composed of several small subclusters or speckles of rhodamine-bungarotoxin fluorescence. The initial accumulation of receptors may occur at the advancing tips of nerve processes because NARPs were found at greater than 80% of the growth cone-muscle contacts examined between 12 and 24 h of co-culture. Over the 3-wk period examined, the mean incidence of NARPs ranged between 1.0 and 2.6 per 100 micron of neurite-myotube contact, with the peak observed on the second day of co-culture. During the first 3 d in culture, when the neurons were multipolar, nearly all of the primary processes induced one or more clusters. With time, as the neurons become unipolar (Role and Fischbach, 1987) NARPs persisted along the remaining dominant process. Measurements made during the third day of co-culture suggest that NARPs disappear along shorter neurites before they retract. Synaptic currents were detected by focal extracellular recording at 55% of the NARPs. The fact that spontaneous or evoked responses were not recorded at 45% suggests that contacts with clusters exhibit two functional states. Two types of presynaptic specialization at identified NARPs observed by scanning electron microscopy appear to be correlated with the functional state.

Distribution of opiate receptor subtypes and enkephalin and dynorphin immunoreactivity in the hippocampus of squirrel, guinea pig, rat, and hamster.

The distribution of enkephalin and dynorphin immunoreactivity in the hippocampus of four rodent species (gray squirrel, guinea pig, rat, and hamster) is compared with the pattern of opiate receptor subtypes (mu, delta, and kappa). The distribution of opioid peptides is fairly consistent in the anterior hippocampus of these four species. Intense immunoreactivity for dynorphin and enkephalin is found in the hilus of the dentate gyrus and in the mossy fiber system. Occasional immunoreactive processes are seen in the dentate molecular layer and scattered throughout the CA1 and CA3 fields. In the rat and hamster, an additional plexus of enkephalinergic fibers straddles both sides of the hippocampal fissure. Cells immunoreactive for both opioid peptides are located in and just superficial to the dentate granule cell layer. Opiate receptors are variably distributed in these rodent species. In the squirrel, guinea pig, and hamster, mu and kappa binding is dense in the stratum lucidum of CA3 and the molecular layer of the dentate gyrus. In the rat, dense mu and kappa binding is localized within and adjacent to the pyramidal and granule cell layers. Delta receptor patterns show additional species differences. In the rat, the delta distribution is similar to the mu and kappa patterns. In the other species, the delta binding pattern is generally the inverse of the mu/kappa pattern: most areas of the hippocampus are enriched in delta sites, whereas the stratum lucidum and the pyramidal cell layer are receptor-sparse. Thus, the stratum lucidum--site of dense terminations of mossy fibers containing opioid peptides--is characterized by selectively sparse delta receptors in four species and by selectively dense kappa receptors in three species. The three receptor subtypes, taken either individually or together and compared to the peptides, are more variably and more widely distributed throughout the hippocampus and fail to show a correspondence with opioid-peptide-containing terminals. The mismatches suggest that receptor locations and densities are organized without relation to the sites of relevant transmitter release.

Formation of Specific Connections in Culture by Identified Leech Neurones Containing Serotonin, Acetylcholine and Peptide Transmitters

ABSTRACT Connections made in culture by identified leech neurones have been studied using pairs of cells that contain various transmitters. These cells were: motoneurones innervating the heart; the anterior pagoda neurones (which contain both acetyl-choline and an FMRFamide-like peptide); the Retzius cells (which contain serotonin) ; and the pressure sensory neurones, which contain an unidentified transmitter. Heart motoneurones (HE) and anterior pagoda (AP) cells in culture reacted with antiserum against the peptide FMRFamide. The immunoreactive peptide was found in dense-core vesicles at presumptive sites of transmitter release. In culture the HE, AP and Retzius cells formed non-rectifying electrical synapses with one another within 2 days. Some pairs of cells made connections resembling those seen in normal leech ganglia; others formed novel connections seen only in culture. Electrical connections made by sensory P cells with HE and AP cells showed rectification. Confirming earlier results, P cells never established electrical connections with Retzius cells. In certain pairs of neurones, chemically mediated synaptic interactions developed. Thus, stimulation of AP cells evoked hyperpolarizing synaptic potentials that arose after a delay in Retzius cells and in P cells. Similarly, stimulation of HE cells gave rise to delayed, slow synaptic potentials in AP cells and in isolated heart muscle fibres in culture. P cells, which in culture are never presynaptic to Retzius cells, made chemical connections with AP cells. These results support the conclusion that identified leech neurones in culture make synaptic connections with certain specific target cells while ignoring others.

The distribution of acetylcholine receptors in the normal and denervated neuromuscular junction of the frog.

A combined light and electron microscopic investigation was performed to study the distribution and fate of clusters of horseradish peroxidase (HRP)-labelled or rhodamine-labelled alpha-bungarotoxin binding sites in normal and denervated cutaneous pectoris muscles of the frog. After staining for axon and cholinesterase (ChE) it appears that, in normal muscles, binding sites for rhodamine alpha-bungarotoxin are strictly confined to those parts of the synaptic gutter occupied by the nerve terminal. Binding sites are missing in axon-abandoned gutters or gutters occupied only by the Schwann cell. Similarly, in partially occupied gutters, HRP-toxin binding sites are confined to the parts of the muscle fibre membrane apposed to the nerve; they are missing at axon-free lateral parts at which secondary clefts and ChE are present. These observations suggest that junctional acetylcholine (ACh) receptor clusters are strictly controlled by the nerve. Thirty-five days after denervation, receptor density was apparently reduced in some parts of the gutters while other parts of the same gutters showed high receptor density. In addition, the length of gutter totally devoid of receptor clusters increased from an average fraction of 9% in control muscles to 14% in denervated muscles. Loss of receptors occurred both at Schwann cell-free and at Schwann cell-occupied gutters. In muscles denervated for 500-750 days, no toxin binding sites could be detected in the junctional membrane, whereas ChE was still present. Schwann cells had apparently abandoned some gutters but were present at others. Upon arrival of spontaneously reinnervating axons, muscle fibres accumulate ACh receptor clusters at the junctional membrane. Patches intensely labelled with alpha-bungarotoxin and associated with ChE reaction product were found near former junctions in long-term denervated muscles. It is concluded that after long-term denervation the muscle fibre cannot maintain junctional ACh receptor clusters by itself. In normally innervated muscles, receptor clusters are actively maintained by nerve-borne factors near the transmitter release sites.

The ultrastructural basis for synaptic transmission between primary muscle afferents and neurons in Clarke's column of the cat

The synaptic connection between primary muscle afferents and dorsal spinocerebellar tract (DSCT) neurons has been studied in an attempt to reveal some of the mechanisms underlying excitatory transmission in the mammalian central nervous system. Previous electrophysiological experiments have shown that the excitatory postsynaptic potentials (EPSPs) evoked DSCT neurons by impulses in a single muscle afferent fluctuate in amplitude. These fluctuations occur between discrete amplitudes which are separated by quantal increments. Two alternative hypotheses relate such a quantal increment to all-or-nothing transmitter release from either (1) an entire synaptic bouton or (2) an individual transmitter release site, given that a bouton may contain multiple release sites. The present study was undertaken primarily to gain ultrastructural evidence on these proposals. Electrodes filled with horseradish peroxidase (HRP) were used to label single identified group Ia afferent fibers and DSCT neurons in the lumbar spinal cord of anesthetized cats. HRP-labeled Ia synaptic boutons, and the contacts formed between HRP-labeled Ia boutons and the dendrites of a DSCT neuron labeled intracellularly with HRP, were examined in serial sections under the electron microscope. Group Ia boutons were found to contain multiple synaptic specializations, as evidenced by pre- and postsynaptic thickenings and presynaptic clusters of vesicles. Careful examination of a bouton in serial sections revealed each specialization as a separate structure. These observations support the proposal that synaptic transmission between group I muscle afferents and DSCT neurons occurs with discrete all-or-nothing EPSPs associated with transmitter release sites, rather than boutons per se.

Ultrastructural correlates of naturally occurring differences in transmitter release efficacy in frog motor nerve terminals.

Motor nerve terminals in cutaneous pectoris muscles of the frog Rana pipiens release more transmitter and form synapses with higher levels of effectiveness than do those in sartorius muscles. Neuromuscular junctions from these two muscles were compared in the electron microscope to search for ultrastructural correlates of differences in transmitter release and synaptic effectiveness. The following measurements were made from cross-sections of junctions with known levels of effectiveness: (a) the presence of active zones, the presumed sites of transmitter release, (b) active zone size, (c) the perimeter, cross-sectional area, height and width of nerve terminals, (d) number of mitochondria, (e) vesicle density, and (f) the extent to which Schwann cells wrap terminals. Nerve terminals in the two muscles did not differ in size, shape or vesicle density. The more strongly releasing cutaneous pectoris terminals did, however, have significantly larger active zones due to deeper invagination of the terminal into the postsynaptic gutter and lesser interposition of Schwann cell processes between presynaptic and postsynaptic membranes. Cutaneous pectoris terminals also contained more mitochondria, presumably to supply the greater energy demand imposed by high release levels.

Localization of α-bungarotoxin binding sites in the ciliary ganglion of the embryonic chick: An autoradiographic study at the light and electron microscopic level

Binding sites for [ 125I]α-bungarotoxin were localized in the chick ciliary ganglion by light and electron microscopic autoradiography. Groups of four ganglia were incubated for 4 h with 20 nM [ 125I]α-bungarotoxin alone or with radioactive α-bungarotoxin plus either 1 μM unlabeled α-bungarotoxin or 100 μ M d-tubocurarine. Specific binding to various morphological regions was determined by subtracting the densities of autoradiographic grains in the presence of competing nonradioactive ligands from the densities in the absence of those ligands. Most of the specific binding in the ganglion (91%) was associated with neurons. Seventy-four per cent was found within 1.2 μm of neuronal plasma membranes and 17% was found overlying neuronal cell bodies. Analysis of the specific binding associated with neurons, but not with the neuronal plasma membranes, revealed that lysosomes and multivesicular bodies were 9- and 32-fold more heavily labeled than other cellular organelles. The grain density over choroid cell bodies was signicantly higher than that over ciliary neurons. Most(> 75%) of the autoradiographic grains within 0.25 μ m of neuronal plasma membranes were found in “complex” contact regions of the membranes, which are characterized by extensive membrane evaginations. However, after correcting for the amount of plasma membrane present in the various regions of the membrane studied, α-bungarotoxin binding was found to be uniform. Few (<10%) of the specialized membranes between pre- and postsynaptic neurons were found in these “complex” contact regions suggesting that the bulk of α-bungarotoxin binding to neuronal membranes is located at some distance from the sites of transmitter release. The density of α-bungarotoxin binding sites in the neuronal plasma membrane was low (~ 100–200 sites/μm 2 of neuronal membrane) compared to published values of the density of binding sites at the neuromuscular junction. Since α-bungarotoxin does not bind preferentially to specialized synaptic membranes, it seems unlikely that the binding sites for this toxin are the neuronal nicotinic receptors.

Location and functioning of transmitter release sites in frog neuromuscular junction

End-plate potentials (EPP) and miniature EPP (MEPP) were recorded in a single neuromuscular synapse of the frog sartorius muscle by means of two microelectrodes with a resistance of 0.5–2.0 Mθ. Groups of signals (fields), reflecting transmitter secretion in spatially distinct release sites were identified by extracellular recording on MEPP amplitude scatter diagrams. Release sites in the nerve ending were found to be unevenly distributed, to be grouped in certain areas, and to differ in their probability of secretion of a quantum of transmitter. Comparison of fields on MEPP and uniquantal EPP amplitude scatter diagrams in solution with low Ca++ concentration (0.2–0.4 mM) showed that ability to induce evoked and spontaneous transmitter release at the release site differs, and that sometimes a release site does not participate in evoked secretion. The results of simultaneous recording of synaptic potentials using extra- and intracellular electrodes indicate that transmitter secretion in spatially distinct groups of release sites leads to the appearance of polymodality in the distribution of amplitudes of intracellularly recorded MEPP and uniquantal EPP.

Electrical activity of mouse motor nerve terminals.

Focal extracellular recording was made from presynaptic motor nerve terminals made visible directly in in vitro preparations of mouse intercostalis intimi muscles. The demonstration of a net inward current distally in these terminals at sites of transmitter release allow the conclusion that the mammalian terminal is electrically excitable, contrary to a recent claim that it is not.

Regeneration of the active zone at the frog neuromuscular junction.

The active zone is a unique specialization of the presynaptic membrane and is believed to be the site of transmitter release. The formation of the active zone and the relationship of this process to transmitter release were studied at reinnervated neuromuscular junctions in the frog. At different times after a nerve crush, the cutaneous pectoris muscles were examined with intracellular recording recording and freeze-fracture electron microscopy. The P face of a normal active zone typically consists of two double rows of particles lined up in a continuous segment located opposite a junctional fold. In the initial stage of reinnervation, clusters of large intramembrane particles surrounding membrane elevations appeared on the P face of nerve terminals. Like normal active zones, these clusters were aligned with junctional folds. Vesicle openings, which indicate transmitter release, were seen at these primitive active zones, even though intramembrane particles were not yet organized into the normal pattern of two double rows. The length of active zones at this stage was only approximately 15% of normal. During the secondary stage, every junction was reinnervated and most active zones had begun to organize into the normal pattern with normal orientation. Unlike normal, there were often two or more discontinuous short segments of active zone aligned with the same junctional fold. The total length of active zone per junctional fold increased to one-third of normal, mainly because of the greater number of segments. In the third stage, the number of active zone segments per junctional fold showed almost no change when compared with the secondary stage. However, individual segments elongated and increased the total length of all active zone segments per junctional fold to about two-thirds of the normal length. The dynamic process culminated in the final stage, during which elongating active zones appeared to join together and the number of active zone segments per junctional fold decreased to normal. Thus, in most regions, regeneration of the active zones was complete. These results suggest that the normal organization of two double rows is not necessary for the active zone to be functional. Furthermore, localization of regenerating active zones is related to junctional folds and/or their associated structures.

Synaptic input from identified muscle afferents to neurones of the dorsal spinocerebellar tract in the cat.

Single identified group I a and I b muscle afferent fibres were injected with horseradish peroxidase in the lumbar dorsal columns of anaesthetized cats. The morphological details of the axon collaterals and terminal boutons of these muscle afferents within Clarke's column were subsequently reconstructed. The rostro-caudal extent of synaptic terminals from a single afferent fibre within Clarke's column was found to be restricted to less than 1 mm. In the same experiments, dorsal spinocerebellar tract (d.s.c.t.) neurones were retrogradely labelled by injection of horseradish peroxidase into the cerebellum. Synaptic contacts between labelled group Ia and Ib afferent fibres and the soma and proximal dendrites of d.s.c.t. neurones were found. The synaptic contacts from both Ia and Ib fibres varied greatly in size, from 1 X 1 micron up to 'giant' synapses of 20 X 3 micron. Excitatory post-synaptic potentials (e.p.s.p.s) were evoked in d.s.c.t. neurones by impulses in single group I muscle afferent fibres. The fluctuations in peak amplitude of each e.p.s.p. were determined from e.p.s.p. and noise recordings, using a numerical deconvolution procedure. In general, these single-fibre e.p.s.p.s fluctuated between discrete amplitudes separated by an incremental amplitude which was approximately constant. This incremental amplitude did not depend on the average peak amplitude of the particular e.p.s.p. examined. Our anatomical observations of 'giant' boutons arising from Ia and Ib afferent fibres contacting d.s.c.t. neurones raises the possibility of multiple transmitter release sites within an individual synaptic bouton. It is proposed that synaptic transmission between group I muscle afferents and d.s.c.t. neurones occurs with discrete all-or-nothing e.p.s.p.s associated with transmitter release sites.