Growth and elimination of nerve terminals at synaptic sites during polyneuronal innervation of muscle cells: a trophic hypothesis

This paper examines the possibility that the elimination of synapses from cells arises from a competition between the nerve terminals for trophic molecules made available by the cells. This idea is applied to the elimination of synapses that occurs during the polyneuronal innervation of muscle cells which accompanies both the development and reinnervation of muscles. In the proposed model, each motorneuron makes the same amount of receptor in its soma for a trophic molecule provided in limited quantities by each muscle cell; this receptor is then distributed to the collateral terminals of the motorneuron in concentrations proportional to the amount of receptor made in the soma by the motorneuron; the more collateral terminals initially possessed by a motorneuron the less will be their concentration of receptor. The receptors in the several collateral terminals on a muscle cell then compete for the trophic molecule provided by the muscle, and terminal growth is proportional to the number of receptor-trophic-molecule bonds formed. An autocatalytic effect has been introduced whereby the increase in size of a terminal accelerates the rate by which the trophic molecule is made available to that terminal for bonding with its receptors. In addition, the affinity between nerve terminal receptors and muscle molecules can be varied in the model. Finally, motorneuron cell death has been analysed as the elimination of neurons that have insufficient terminal area to take up a growth factor in amounts that will allow for the survival of the neuron.     

Probabilistic secretion of quanta and the synaptosecretosome hypothesis: evoked release at active zones of varicosities, boutons, and endplates.

A quantum of transmitter may be released upon the arrival of a nerve impulse if the influx of calcium ions through a nearby voltage-dependent calcium channel is sufficient to activate the vesicle-associated calcium sensor protein that triggers exocytosis. A synaptic vesicle, together with its calcium sensor protein, is often found complexed with the calcium channel in active zones to form what will be called a "synaptosecretosome." In the present work, a stochastic analysis is given of the conditions under which a quantum is released from the synaptosecretosome by a nerve impulse. The theoretical treatment considers the rise of calcium at the synaptosecretosome after the stochastic opening of a calcium channel at some time during the impulse, followed by the stochastic binding of calcium to the vesicle-associated protein and the probability of this leading to exocytosis. This allows determination of the probabilities that an impulse will release 0, 1, 2,... quanta from an active zone, whether this is in a varicosity, a bouton, or a motor endplate. A number of experimental observations of the release of transmitter at the active zones of sympathetic varicosities and boutons as well as somatic motor endplates are described by this analysis. These include the likelihood of the secretion of only one quantum at an active zone of endplates and of more than one quantum at an active zone of a sympathetic varicosity. The fourth-power relationship between the probability of transmitter release at the active zones of sympathetic varicosities and motor endplates and the external calcium concentration is also explained by this approach. So, too, is the fact that the time course of the increased rate of quantal secretion from a somatic active zone after an impulse is invariant with changes in the amount of calcium that enters through its calcium channel, whether due to changes consequent on the actions of autoreceptor agents such as adenosine or to facilitation. The increased probability of quantal release that occurs during F1 facilitation at the active zones of motor endplates and sympathetic boutons is predicted by the residual binding of calcium to a high-affinity site on the vesicle-associated protein. The concept of the stochastic operation of a synaptosecretosome can accommodate most phenomena involving the release of transmitter quanta at these synapses.     

Probabilistic secretion of quanta: spontaneous release at active zones of varicosities, boutons, and endplates.

The amplitude-frequency histogram of spontaneous miniature endplate potentials follows a Gaussian distribution at mature endplates. This distribution gives the mean and variance of the quantum of transmitter. According to the vesicle hypothesis, this quantum is due to exocytosis of the contents of a single synaptic vesicle. Multimodal amplitude-frequency histograms are observed in varying degrees at developing endplates and at peripheral and central synapses, each of which has a specific active zone structure. These multimodal histograms may be due to the near synchronous exocytosis of more than one vesicle. In the present work, a theoretical treatment is given of the rise of intraterminal calcium after the stochastic opening of a calcium channel within a particular active zone geometry. The stochastic interaction of this calcium with the vesicle-associated proteins involved in exocytosis is then used to calculate the probability of quantal secretions from one or several vesicles at each active zone type. It is shown that this procedure can account for multiquantal spontaneous release that may occur at varicosities and boutons, compared with that at the active zones of motor nerve terminals.     

The effect of calcium ions on the secretion of quanta evoked by an impulse at nerve terminal release sites

A study has been made of the effects of calcium ions on the number of quanta secreted from all the release sites at an amphibian motor nerve terminal recorded with an intracellular microelectrode (m) compared with the number secreted simultaneously from a small number of release sites recorded with an extracellular microelectrode (me). If the endplate potential was made subthreshold by lowering the external calcium concentration ([Ca]o less than or equal to 0.4 mM), it was possible to find small groups of release sites for which me was comparable to m, indicating considerable nonuniformity in the probability of release of a quantum at different groups of release sites (Pe) in a given [Ca]o. Increasing [Ca]o in the range from 0.25 to 0.4 mM increased the probability of release of a quantum at groups of release sites (Pe), independent of the initial value of Pe, and the dependence of Pe on [Ca]o followed a fourth power relationship. A conditioning impulse enhanced the probability of release of a quantum by a subsequent test impulse at release sites, if Pe was less than 1.0 during the conditioning impulse. It is shown that the present observations regarding the dependence of Pe on [Ca]o and on conditioning impulses can be quantitatively predicted from previous observations regarding the dependence of the binomial parameters m, p, and n on [Ca]o and on conditioning impulses determined with intracellular electrodes, if the probability of secretion of a quantum at a release site (Pj) is different for different release sites and Pj is distributed as a beta random variable.     

Probabilistic secretion of quanta in the central nervous system: granule cell synaptic control of pattern separation and activity regulation.

The implications of probabilistic secretion of quanta for the functioning of neural networks in the central nervous system have been explored. A model of stochastic secretion at synapses in simple networks, consisting of large numbers of granule cells and a relatively small number of inhibitory interneurons, has been analysed. Such networks occur in the input to the cerebellum Purkinje cells as well as to hippocampal CA3 pyramidal cells and to pyramidal cells in the visual cortex. In this model the input axons terminate on granule cells as well as on an inhibitory interneuron that projects to the granule cells. Stochastic secretion at these synapses involves both temporal variability in secretion at single synapses in the network as well as spatial variability in the secretion at different synapses. The role of this stochastic variability in controlling the size of the granule cell output to a level independent of the size of the input and in separating overlapping inputs has been determined analytically as well as by simulation. The regulation of granule-cell output activity to a reasonably constant value for different size inputs does not occur in the absence of an inhibitory interneuron when both spatial and temporal stochastic variability occurs at the remaining synapses; it is still very poor in the presence of such an interneuron but in the absence of stochastic variability. However, quite good regulation is achieved when the inhibitory interneuron is present with spatial and temporal stochastic variability of secretion at synapses in the network. Excellent regulation is achieved if, in addition, allowance is made for the nonlinear behaviour of the input-output characteristics of inhibitory interneurons. The capacity of granule-cell networks to separate overlapping patterns of activity on their inputs is adequate, with spatial variability in the secretion at synapses, but is improved if there is also temporal variability in the stochastic secretion at individual synapses, although this is at the expense of reliability in the network. Other factors which improve pattern separation are control of the output to very low activity levels, and a restriction on the cumulative size of the excitatory input terminals of each granule cell. Application of the theory to the input neural networks of the cerebellum and the hippocampus shows the role of stochastic variability in quantal transmission in determining the capacity of these networks for pattern separation and activity regulation.     

Acetylcholine receptors in regenerating muscle accumulate at original synaptic sites in the absence of the nerve

We examined the role of nerve terminals in organizing acetylcholine receptors on regenerating skeletal-muscle fibers. When muscle fibers are damaged, they degenerate and are phagocytized, but their basal lamina sheaths survive. New myofibers form within the original basal lamina sheaths, and they become innervated precisely at the original synaptic sites on the sheaths. After denervating and damaging muscle, we allowed myofibers to regenerate but deliberately prevented reinnervation. The distribution of acetylcholine receptors on regenerating myofibers was determined by histological methods, using [125I] alpha-bungarotoxin or horseradish peroxidase-alpha-bungarotoxin; original synaptic sites on the basal lamina sheaths were marked by cholinesterase stain. By one month after damage to the muscle, the new myofibers have accumulations of acetylcholine receptors that are selectively localized to the original synaptic sites. The density of the receptors at these sites is the same as at normal neuromuscular junctions. Folds in the myofiber surface resembling junctional folds at normal neuromuscular junctions also occur at original synaptic sites in the absence of nerve terminals. Our results demonstrate that the biochemical and structural organization of the subsynaptic membrane in regenerating muscle is directed by structures that remain at synaptic sites after removal of the nerve.     

Calcium homeostasis and nerve growth factor secretion from vascular and bladder smooth muscle cells

Bladder and vascular smooth muscle cells cultured from four rat strains (WKY, SHR, WKHA, WKHT) differing in rates of nerve growth factor (NGF) production were used to determine whether a relationship exists between intracellular calcium and NGF secretion. Basal cytosolic calcium was related to basal NGF secretion rates in bladder and vascular smooth muscle cells from all four strains with the exception of WKHT bladder muscle cells. Thrombin is a calcium-mobilizing agent and increases NGF production from vascular but not bladder smooth muscle cells. Strain differences were found in the magnitude of the calcium peak induced by thrombin in vascular smooth muscle cells, but these differences did not correlate with NGF secretion. Thrombin caused a calcium response in bladder smooth muscle cells without influencing NGF production. Quenching the calcium transient with a calcium chelator had no effect on thrombin-inducted NGF secretion rates in vascular smooth muscle cells. Thus, basal intracellular calcium may establish a set point for NGF secretion from smooth muscle. In addition, transient elevations in cytosolic calcium were unrelated to the induction of NGF output.     

The neuromuscular junction: Selective remodeling of synaptic regulators at the nerve/muscle interface

The peripheral synapses between motoneurons and skeletal muscle fibers, the neuromuscular junctions, are ideal to investigate the general principles of synaptogenesis that depend on the interaction of activity-dependent and activity-independent signals. Much has been learned from gene “knock out” mouse models that helped to identify major synaptic regulators. The “knock out” approach, however, may not distinguish between changes arising from the disruption of molecular signaling pathways and changes caused by the absence of synaptic transmission. To circumvent these problems, postsynaptic activity was modulated in mouse models by specifically targeting endplate receptors or the activity of synaptic regulators such as MuSK. Both regulators have multiple functions and acetylcholine receptors are not just signal transducers but regulate the localization and architecture of endplates. The results show that detailed analysis of mouse models will help to understand the complexity in mechanisms that regulate synaptic remodeling.     

Efficacy of thalamocortical and intracortical synaptic connections: quanta, innervation, and reliability.

Thalamocortical (TC) synapses carry information into the neocortex, but they are far outnumbered by excitatory intracortical (IC) synapses. We measured the synaptic properties that determine the efficacy of TC and IC axons converging onto spiny neurons of layer 4 in the mouse somatosensory cortex. Quantal events from TC and IC synapses were indistinguishable. However, TC axons had, on average, about 3 times more release sites than IC axons, and the mean release probability at TC synapses was about 1.5 times higher than that at IC synapses. Differences of innervation ratio and release probability make the average TC connection several times more effective than the average IC connection, and may allow small numbers of TC axons to dominate the activity of cortical layer 4 cells during sensory inflow.     

The formation and function of single transmitter release sites at mature amphibian motor—nerve terminals

The extent of quantal transmitter release from single sites of synaptic vesicle accumulations along the length of motor-nerve terminal branches at the amphibian neuromuscular junction has been investigated. Such a determination involves development of a model for the generation of quantal potential fields at single styryl-dye stained sites along the length of a branch. Successful testing and application of this model indicates that the extent of quantal release at a dye-stained site is proportional to the total length of active zone at the site. The stability of these sites and of their ensheathing terminal Schwann cell processes was also investigated. Following simultaneous injection of the terminal Schwann cell and nerve terminal with different fluorescent dyes, terminal branches were observed to show dynamic changes in their length, with these occurring in relatively short periods of hours or less. Redistribution of styryl dye stained sites at the ends of branches also occurred in such short periods of time. These were accompanied by changes in the configuration of terminal Schwann cells, which generally occurred prior to changes in the length of nerve terminal branches.     

The influence of basal lamina on the accumulation of acetylcholine receptors at synaptic sites in regenerating muscle

If skeletal muscles are damaged in ways that spare the basal lamina sheaths of the muscle fibers, new myofibers develop within the sheaths and neuromuscular junctions form at the original synaptic sites on them. At the regenerated neuromuscular junctions, as at the original ones, the muscle fiber plasma membrane is characterized by infoldings and a high concentration of acetylcholine receptors (AChRs). The aim of this study was to determine whether or not the synaptic portion of the myofiber basal lamina sheath plays a direct role in the formation of the subsynaptic apparatus on regenerating myofibers, a question raised by the results of earlier experiments. The junctional region of the frog cutaneous pectoris muscle was crushed or frozen, which resulted in disintegration and phagocytosis of all cells at the synapse but left intact much of the myofiber basal lamina. Reinnervation was prevented. When new myofibers developed within the basal lamina sheaths, patches of AChRs and infoldings formed preferentially at sites where the myofiber membrane was apposed to the synaptic region of the sheaths. Processes from unidentified cells gradually came to lie on the presynaptic side of the basal lamina at a small fraction of the synaptic sites, but there was no discernible correlation between their presence and the effectiveness of synaptic sites in accumulating AChRs. We therefore conclude that molecules stably attached to the myofiber basal lamina at synaptic sites direct the formation of subsynaptic apparatus in regenerating myofibers. An analysis of the distribution of AChR clusters at synaptic sites indicated that they formed as a result of myofiber-basal lamina interactions that occurred at numerous places along the synaptic basal lamina, that their presence was not dependent on the formation of plasma membrane infoldings, and that the concentration of receptors within clusters could be as great as the AChR concentration at normal neuromuscular junctions.     

Reciprocal interactions between perisynaptic Schwann cells and regenerating nerve terminals at the frog neuromuscular junction

The perisynaptic Schwann cell (PSC) has gained recent attention with respect to its roles in synaptic function, remodeling, and regeneration at the vertebrate neuromuscular junction (NMJ). Here we test the hypothesis that, following nerve injury, processes extended by PSCs guide regenerating nerve terminals (NTs) in vivo, and that the extension of sprouts by PSCs is triggered by the arrival of regenerating NTs. Frog NMJs were double-stained with a fluorescent dye, FM4-64, for NTs, and fluorescein isothiocyanate (FITC)-tagged peanut agglutinin (PNA) for PSCs. Identified NMJs were imaged in vivo repeatedly for several months after nerve injury. PSCs sprouted profusely beginning 3-4 weeks after nerve transection and, as reinnervation progressed, regenerating NTs closely followed the preceding PSC sprouts, which could extend tens to hundreds of microns beyond the original synaptic site. The pattern of reinnervation was dictated by PSC sprouts, which could form novel routes joining neighboring junctions or develop into new myelinated axonal pathways. In contrast to mammals, profuse PSC sprouting in frog muscles was not seen in response to axotomy alone, and did not occur at chronically denervated NMJs. Instead, sprouting coincided with the arrival of regenerating NTs. Immunofluorescent staining revealed that in muscle undergoing reinnervation 4 weeks after axotomy, 91% of NMJs bore PSC sprouts, compared to only 6% of NMJs in muscle that was chronically denervated for 4 weeks. These results suggest that reciprocal interactions between regenerating NTs and PSCs govern the process of reinnervation at frog NMJs: regenerating NTs induce PSCs to sprout, and PSC sprouts, in turn, lead and guide the elaboration of NTs.     

Olesoxime,a cholesterol-like neuroprotectant restrains synaptic vesicle exocytosis in the mice motor nerve terminals: Possible role of VDACs

Olesoxime is a cholesterol-like neuroprotective compound that targets to mitochondrial voltage dependent anion channels (VDACs). VDACs were also found in the plasma membrane and highly expressed in the presynaptic compartment. Here, we studied the effects of olesoxime and VDAC inhibitors on neurotransmission in the mouse neuromuscular junction. Electrophysiological analysis revealed that olesoxime suppressed selectively evoked neurotransmitter release in response to a single stimulus and 20 Hz activity. Also olesoxime decreased the rate of FM1–43 dye loss (an indicator of synaptic vesicle exocytosis) at low frequency stimulation and 20 Hz. Furthermore, an increase in extracellular Cl⁻ enhanced the action of olesoxime on the exocytosis and olesoxime increased intracellular Cl⁻ levels. The effects of olesoxime on the evoked synaptic vesicle exocytosis and [Cl⁻]_i were blocked by membrane-permeable and impermeable VDAC inhibitors. Immunofluorescent labeling pointed on the presence of VDACs on the synaptic membranes. Rotenone-induced mitochondrial dysfunction perturbed the exocytotic release of FM1–43 and cell-permeable VDAC inhibitor (but not olesoxime or impermeable VDAC inhibitor) partially mitigated the rotenone-driven alterations in the FM1–43 unloading and mitochondrial superoxide production. Thus, olesoxime restrains neurotransmission by acting on plasmalemmal VDACs whose activation can limit synaptic vesicle exocytosis probably via increasing anion flux into the nerve terminals.     

Single GABAergic synaptic terminals from rat midbrain exhibit functional P2X and dinucleotide receptors, able to induce GABA secretion

GABAergic terminals from rat midbrain characterized by immunolocalization of glutamic acid decarboxylase and/or the vesicular inhibitory amino acid transporter respond to ATP or P(1),P(5)-di(adenosine-5') pentaphosphate (Ap(5)A) with an increase in the intrasynaptosomal calcium concentration measured by a microfluorimetric technique in single synaptic terminals. The ATP response is mediated through the activation of P2X receptors with an abundant presence of P2X(3) subunits. Ap(5)A, however, exerts its effects by acting through a different receptor termed the dinucleotide receptor. Both receptors, once activated in the presence of extrasynaptosomal calcium, induce a concentration-dependent GABA release from synaptosomal populations with EC(50) values of 16 and 20 microM for ATP and Ap(5)A, respectively. Specific inhibition of GABA release is obtained with pyridoxal phosphate-6-azophenyl-2',4'-disulphonic acid (80 microM) on the ATP effect and with P(1),P(5)-di(inosine-5') pentaphosphate (100 nM) on the dinucleotide receptor.     

The dual phosphatase activity of synaptojanin1 is required for both efficient synaptic vesicle endocytosis and reavailability at nerve terminals

Phosphoinositides have been implicated in synaptic vesicle recycling largely based on studies of enzymes that regulate phosphoinositide synthesis and hydrolysis. One such enzyme is synaptojanin1, a multifunctional protein conserved from yeast to humans, which contains two phosphoinositol phosphatase domains and a proline-rich domain. Genetic ablation of synaptojanin1 leads to pleiotropic defects in presynaptic function, including accumulation of free clathrin-coated vesicles and delayed vesicle reavailability, implicating this enzyme in postendocytic uncoating of vesicles. To further elucidate the role of synaptojanin1 at nerve terminals, we performed quantitative synaptic vesicle recycling assays in synj1(-/-) neurons. Our studies show that synaptojanin1 is also required for normal vesicle endocytosis. Defects in both endocytosis and postendocytic vesicle reavailability can be fully restored upon reintroduction of synaptojanin1. However, expression of synaptojanin1 with mutations abolishing catalytic activity of each phosphatase domain reveals that the dual action of both domains is required for normal synaptic vesicle internalization and reavailability.