IGF-I enhances survival of embryonic chick ciliary ganglion neurons in a calcium-dependent way.

We have shown that neurons from embryonic chick ciliary ganglia in primary culture possess receptors for insulin-like growth factor I (IGF-I). When added to serum- and insulin-free culture medium, the factor potently enhanced neuronal survival as observed after 24 and 48 h of culture. The effect saturated at 5 ng/ml. Laminin was not necessary for the trophic effects of IGF-I; in the absence of the factor, it had no effect on neuronal survival. Insulin exerted a trophic effect similar to that observed with IGF-I, but at higher doses. The trophic effect of IGF-I was sharply and specifically reduced when either a membrane-permeable calcium chelating agent or blockers of voltage-dependent calcium channels were added to the medium.

Alpha-latrotoxin channels in neuroblastoma cells.

The changes in ionic permeability induced by the application of alpha-latrotoxin to NG108-15 neuroblastoma x glioma cells were examined using the nystatin perforated-patch technique for whole-cell recording. Complex single channel activity appeared in the plasmalemmas after delays that ranged from 1-20 min in Krebs' solution. The conductance of a channel fluctuated among at least three broad, approximately equispaced bands, the maximum conductance being about 300 pS, and the reversal potential approximately 0 mV. The channels were permeable to Na+, K+, Ca2+ and Mg2+, poorly permeable to glucosamineH+ and Cl-, and were blocked by La3+. The channels stayed fully open in Ca(2+)-free solutions with 4 mM Mg2+, in solutions with no divalent cations and in solutions with 2 mM Ca2+ and 96 mM Mg2+. They opened infrequently if both internal and external Cl- were replaced by glutamate-. If alpha-latrotoxin opened similar channels in nerve terminals, the flux of ions through them could account for the massive release of neurotransmitter induced by the toxin.

Sodium-activated potassium current in sensory neurons: a comparison of cell-attached and cell-free single-channel activities.

Single-channel currents from Na(+)-dependent K+ channels (KNa) were recorded from cell-attached and inside-out membrane patches of cultured avian trigeminal ganglion neurons by means of the patch-clamp technique. Single-channel properties, such as the high elementary conductance and the occurrence of sub-conductance levels, were unchanged after the patches had been excised from the cells, indicating that they are not under the control of soluble cytoplasmic factors. In cell-attached recordings at the cell resting potential the degree of KNa activity, measured as the probability of the channel being open, Po, was low in most cases (around 0.01) and similar to that observed in the inside-out configuration when the bath solution contained concentrations of Na+ around 30 mM and of K+ close to the physiological intracellular levels. However, in some cell-attached patches Po was high (around 0.2) and comparable to the values measured in cell-free recordings with high Na+ concentrations in the bath (100 mM). The excision of a high-activity patch in the presence of 30 mM Na+ resulted in a fall of Po in about 20 s, which is consistent with the wash-out of a soluble cytoplasmic molecule. After the excision, all KNa displayed a similar Na+ sensitivity, irrespective of the degree of activation observed in the cell-attached mode. In inside-out patches the Po values observed in the presence of either low or high concentrations of Na+ in bath solutions were not modified by internal Ca2+ (0.8-8.5 microM).(ABSTRACT TRUNCATED AT 250 WORDS)

Potassium current activated by intracellular sodium in quail trigeminal ganglion neurons.

Whole-cell voltage clamp and single-channel recordings were performed on cultured trigeminal ganglion neurons from quail embryos in order to study a sodium-activated potassium current (KNa). When KNa was activated by a step depolarization in voltage clamp, there was a proportionality between KNa and INa at all voltages between the threshold of INa and ENa. Single-channel recordings indicated that KNa could be activated already by 12 mM intracellular sodium and was almost fully activated at 50 mM sodium. 100 mM lithium, 100 mM choline, or 5 microM calcium did not activate KNa. The relationship between the probability for the channel to be open (Po) vs. the sodium concentration and the relationship of KNa open time-distributions vs. the sodium concentration suggest that two to three sodium ions bind cooperatively before KNa channels open. KNa channels were sensitive to depolarization; at 12 mM sodium, a 42-mV depolarization caused an e-fold increase in Po. Under physiological conditions, the conductance of the KNa channel was 50 pS. This conductance increased to 174 pS when the intra- and extracellular potassium concentrations were 75 and 150 mM, respectively.

Substance P-like immunoreactivity at the frog neuromuscular junction.

It has been recently shown that frog motoneurons and their nerve terminals contain calcitonin gene-related peptide-like immunoreactivity in large dense-core vesicles (Matteoli et al., Proc. natn. Acad. Sci. U.S.A. 85, 7366-7370, 1988). We report here by an immunofluorescence approach that the same neurons and nerve terminals also contain substance P-like immunoreactivity. The demonstration of substance P-like immunoreactivity in the frog motor nerve endings supports previous data suggesting a physiological role for this peptide in the modulation of cholinergic transmission.

Dual effect of potassium on transmitter exocytosis.

The time course of exocytosis of quanta of acetylcholine induced by 20 mM K+ was studied at the frog neuromuscular junction. Images of vesicle fusion on freeze-fracture replicas were mostly localized at the active zones in resting preparations fixed in 20 mM K+. Fusions appeared also outside the active zones in preparations fixed after 1 min exposure to 20 mM K+ and were evenly distributed over the presynaptic membrane after 5 min in 20 mM K+ (even though secretion was prevented by withdrawing Ca2+ until 30 s before fixation). The mean densities of vesicle fusions were comparable in all conditions, as were the total number of quanta released during the fixation period. This indicates that fusions outside active zones represent ectopic exocytosis, slowly activated by potassium. Partial inactivation of K(+)-induced quantal release (time and concentration-dependent) was observed electrophysiologically; this may be related to the observed decrease in density of vesicle fusions along the active zones, with time. Consistently, after 5 min in 15 mM K+ fusion density at the active zones remained high. It is concluded that active zone-associated and ectopic fusions are two exocytotic processes activated with differential time courses and concentration-dependence by K+.

Differential effect of alpha-latrotoxin on exocytosis from small synaptic vesicles and from large dense-core vesicles containing calcitonin gene-related peptide at the frog neuromuscular junction.

The regulatory peptide called calcitonin gene-related peptide (CGRP) was detected by immunofluorescence in frog motor neurons and motor nerve terminals. In motor nerve terminals, CGRP-like immunoreactivity was found to be segregated within large dense-core vesicles. To determine whether exocytosis from acetylcholine-containing small synaptic vesicles and from CGRP-containing large dense-core vesicles can be independently stimulated, nerve-muscle preparations were exposed to alpha-latrotoxin. This toxin induced complete depletion of acetylcholine-containing small synaptic vesicles but did not induce a parallel depletion of CGRP-like immunoreactivity and of large dense-core vesicles. These effects were independent of the presence of extracellular Ca2+ and occurred both at room temperature and at low temperature (1-3 degrees C). These findings suggest that exocytosis from the two vesicle populations is mediated by distinct biochemical mechanisms, which might be differentially regulated by physiological stimuli.

The effect of potassium on exocytosis of transmitter at the frog neuromuscular junction.

1. Electrophysiology and morphology have been combined to investigate the time course of the exocytosis of quanta of neurotransmitter induced by elevated concentrations of K+ at the frog neuromuscular junction. 2. Replicas of freeze-fractured resting nerve terminals fixed in the presence of 20 mM-K+ showed images of fusion of synaptic vesicles with the presynaptic axolemma which were closely associated with the active zones. After 1 min in 20 nM-K+ fusions appeared also outside the active zones, and by 5 min they became uniformly distributed over the presynaptic membrane. 3. The average total density of fusions was not significantly different at the various times examined since it decreased at the active zones while it increased over the rest of the membrane. 4. Resting terminals fixed in 20 mM-K+ released 33,000-45,000 quanta after the addition of fixative; terminals stimulated by 20 mM-K+ for 1-5 min released 50,000-100,000 quanta during fixation. The fixative potentiated K+-induced transmitter release. 5. Fusions were uniformly distributed in terminals pre-incubated for 5 min in 20 mM-K+ without added Ca2+, stimulated by adding Ca2+ for 30 s, and then fixed. Conversely, after 5 min stimulation in hypertonic Ringer solution fusions remained predominantly located near the active zones. A similar distribution was observed after 15 min stimulation by a lower concentration of K+ (15 mM). 6. At all concentrations of K+ tested (10, 15, 20, 25 mM) miniature end-plate potential (MEPP) rate attained a steady-state value within 10-15 min. Values from a single junction were generally lower at higher concentrations of K+, which indicates partial inactivation of the secretion-recycling process. 7. The data indicate that K+ initially activates exocytosis at the active zones. Subsequently, ectopic exocytosis is activated while sites at the active zones appear to undergo partial inactivation. These phenomena are not related to the intensity or to the amount of previous secretion.

Coated vesicles and pits during enhanced quantal release of acetylcholine at the neuromuscular junction.

Frog neuromuscular junctions were stimulated by different methods to secrete quanta of ACh, and the attendant changes in the ultrastructure of the nerve terminal were assessed by morphometric analysis of electron micrographs. Secretion was stimulated by electrical stimulation at 2 Hz or by application of the secretagogues, lanthanum, ouabain or black widow spider venom, either in the presence or in the absence of extracellular Ca2+. The numbers of synaptic vesicles, coated vesicles and coated pits, and the length of axolemma and area of axoplasm were measured on the micrographs. There was a significant increase (about threefold) in the total number of coated structures (vesicles plus pits) per micron2 of axoplasm, but the fractional increase in the number of coated pits exceeded the fractional increase in the number of coated vesicles. These increases were positively correlated with the increase in the length of axolemma per unit area and negatively correlated with the changes in concentration of synaptic vesicles, suggesting that they were due to the increases in the surface area of the terminal that accompany a loss of vesicles. However, the increase in the concentration of coated structures was not related to the number of quanta secreted or to the estimated number of vesicles recycled. The lack of correspondence between the fractional increases in the coated pits and coated vesicles and the poor correlation between the numbers of these structures and the overall parameters of the secretory process suggest that, in contrast to the situation in other secretory systems, coated pits and coated vesicles may not play a crucial role in maintaining the functional population of synaptic vesicles at rapidly secreting neuromuscular junctions.

The phorbol ester, 12-O-tetradecanoyl-phorbol-13-acetate, enhances the evoked quanta release of acetylcholine at the frog neuromuscular junction.

The possible role of protein kinase C in the regulation of quantal transmitter release was studied at the frog neuromuscular junction by using the phorbol ester 12-O-tetradecanoyl-phorbol-13-acetate (TPA), a compound known to mimic the effects of the physiological activator of the enzyme, endogenous diacylglycerol. The main effect of the phorbol ester was to increase the quantal content, m, of the endplate potential. The initial values of m were adjusted over a wide range by changing the Ca2+ concentration of the extracellular medium, and the TPA-induced fractional increase in m was significantly greater at junctions with a lower initial quantal content. On the other hand, the absolute increases in m induced by the phorbol ester were positively correlated with the square root of the initial quantal content. The possible physiological significance of this correlation is discussed in view of the well known relationship between extracellular Ca2+ concentration and the quantal content of the end plate potential.

Measurement of quantal secretion induced by ouabain and its correlation with depletion of synaptic vesicles.

Ouabain (0.1 and 0.05 mM) was applied to frog cutaneous pectoris nerve-muscle preparations bathed in modified Ringer's solution containing either 1.8 mM Ca2+ (and 4 mM Mg2+) or no added Ca2+ (4 mM Mg2+ and 1 mM EGTA). During the intense quantal release of acetylcholine (ACh) induced by ouabain, the parameters of the miniature endplate potentials (mepps) were deduced from the variance, skew, and power spectra of the endplate recordings by applying a recently described modification of classical fluctuation analysis. Often the high frequency of mepps is not stationary; therefore, the signal was high-pass filtered (time constant of the resistance-capacitance filter of 2 ms) to remove the errors introduced by nonstationarity. When ouabain was applied in the presence of Ca2+, mepp frequency started to rise exponentially after a lag of 1.5-2 h, reached an average peak frequency of 1,300/s in approximately 30 min, and then suddenly subsided to low level (10/s). In Ca2+-free solution, after a shorter lag (1-1.5 h), mepp frequency rose to peak rate of 700/s in approximately 20 min and then gradually subsided. In spite of the different time course of secretion in the two experimental conditions, the cumulative quantal release was not significantly different (7.4 +/- 1.3 X 10(5) in Ca2+-containing and 8.8 +/- 2.7 X 10(5) in Ca2+-free solutions). 60 min after the peak secretion, the muscles were fixed for observation in the electron microscope. Morphometric analysis on micrographs of neuromuscular junctions revealed in both cases a profound depletion of synaptic vesicles and deep infoldings of presynaptic membrane. This rapid depletion and the lack of uptake of horseradish peroxidase suggest that ouabain impairs the recycling process that tends to conserve the vesicle population during intense secretion of neurotransmitter. The good correlation observed between the reduction in the store of synaptic vesicles and the total number of quanta of ACh secreted in the absence of a vigorous membrane recycling strongly supports the view that the secretion of a quantum of ACh requires the fusion of a synaptic vesicle with the axolemma.

Differential effects of alpha-latrotoxin on mouse nerve endings and fibers.

The action of alpha-latrotoxin (alpha-LTx), the major toxic component from black widow spider venom, was studied on mammalian motor nerve endings and fibers. Electrophysiologic recordings of presynaptic and postsynaptic responses at early stages after the application of different doses of alpha-LTx showed massive transient increases in transmitter release and changes in the configuration of presynaptic currents. Later, a loss of invasion of terminals by nerve impulses occurred, whereas the impulse conduction in nerve fibers was unmodified. These results support the idea of a selective effect of alpha-LTx on presynaptic membrane excitability.

Response of ventrobasal thalamic neurons activated by dorsal and dorsolateral funiculi: an intracellular study.

Intracellular responses to stimulation of the dorsal column (DC) and dorsolateral funiculus (DLF) were recorded in cells of the thalamic ventrobasal complex (VB) in anesthetized cats, with the dorsal funiculi either intact or isolated. The responsiveness of VB neurons was tested using graded stimulation, paired-shock, and interaction techniques. Of the 60 VB neurons thoroughly studied, 50 responded to stimulation of the DC with excitatory postsynaptic potentials (EPSPs) followed by inhibitory postsynaptic potentials (IPSPs); half of these 50 neurons responded to stimulation of the DLF with the same pattern, whereas no IPSPs could be elicited in the remaining neurons. The majority of EPSPs could be fractionated into unitary components during graded electrical stimulation. The number of such components observed was greater after DLF than after DC stimulation. In most neurons, the DLF-evoked EPSPs were smaller in amplitude than the DC-evoked EPSPs. Paired-shock stimulation facilitated the DLF excitatory responses. The amplitude of IPSPs induced by DLF stimulation was significantly smaller than that evoked by DC stimulation, and DC stimulation reduced the excitatory response to subsequent DLF stimulation. The data support the known dominance of the DC pathway in the cat.

Patterns of motor innervation in the pectoral muscle of adult Xenopus laevis: evidence for possible synaptic remodelling.

1. An anatomical and electrophysiological study was performed on the pectoral muscle of Xenopus laevis. 2. Silver-impregnated preparations revealed immature endings, collateral and terminal sprouting and signs of synaptic regression. 3. Twenty percent of the fibres received a dual innervation from two different nerves. The synapses of 25% of these fibres are formed in close vicinity. 4. Some of the singly innervated and most of the dually innervated end-plates generated only subthreshold electrical activity. Synaptic efficacy in dually innervated muscle fibres with closely spaced or distant endings was, on the average, one third and two thirds, respectively, of that obtained in singly innervated fibres. 5. Fibres with subthreshold electrical activity displayed normal ACh sensitivity. 6. The existence of non-transmitting synapses, of dually innervated end-plate sites and of morphological signs of the sprouting of new endings and the degeneration of old ones suggests that synaptic remodelling may occur in normal adult muscles.

Interaction between motor axons from two different nerves reinnervating the pectoral muscle of Xenopus laevis.

1. An electrophysiological and morphological study of sprouting and regeneration of motor nerves has been performed in the dually innervated pectoral muscle of Xenopus laevis. 2. Section of one of the nerves induced axon sprouting in the intact nerve. Synapse formation by the sprouting axons was slow since the intact nerve took more than 3 months to increase its field of innervation by 70%. The rate of axon regeneration was faster than that of axon sprouting since the cut nerve reinnervated its former territory in less than 1 month. 3. At early stages of synapse formation the sprouted or regenerated endings were poorly branched but any terminal branches were, as a rule, longer than normal. Signs of degeneration and replacement of endings have been observed. 4. Low levels of transmitter release persisted for several months in newly formed endings. This depression was more pronounced at the endings formed outside the normal field of innervation of the nerve. 5. Poly-innervated muscle fibres have been observed during reinnervation by regenerated or sprouted axons. Their number decreases gradually in the months that follow the beginning of reinnervation. Synaptic efficacy was lower at poly than at mono-innervated muscle fibres. At doubly innervated spots and at separated spots on the same fibre average end-plate potential (e.p.p.) amplitude was 1/3 and 2/3 respectively of that recorded at singly innervated fibres. 6. Electrophysiological and morphological data have been compared at individual doubly innervated end-plate sites. End-plate potential amplitude was positively correlated with the degree of ending development. 7. Sprouted endings remain functional after periods of reinnervation of 30 months, although signs of regression have been observed. These are probably mediated by spontaneous degeneration of the terminals and replacement by endings from the regenerating nerve.