Asynchronous Transmitter Release Research Articles

Synchronous versus asynchronous transmitter release: a tale of two types of inhibitory neurons

Inhibitory cortical neurons are thought to generate temporally precise signals important for information processing, but a new study shows that CCK-expressing interneurons continue to release GABA for several hundred milliseconds after bursts of action potentials.

Intraterminal Ca2+ concentration and asynchronous transmitter release at single GABAergic boutons in rat collicular cultures

Neurotransmitter release in response to a single action potential has a precise time course. A significant fraction of the releasable vesicles is exocytosed synchronously, within a few milliseconds after the arrival of an action potential. If repeatedly activated, stimulus-locked phasic synchronous release declines, but synaptic transmission can be maintained through tonic asynchronous transmitter release. The desynchronisation of release during repetitive activation is generally attributed to a build-up of intraterminal Ca2+ concentration. However, the precise relationship between presynaptic Ca2+ level and asynchronous release rate at small central synapses has remained unclear. Here we characterise this relationship for single GABAergic terminals in rat collicular cultures. In the presence of tetrodotoxin, inhibitory postsynaptic currents (IPSCs) and presynaptic Ca2+ transients were recorded in response to direct presynaptic depolarisation of individual boutons. Repetitive stimulation indeed resulted in a shift from phasic to asynchronous neurotransmitter release. A clear dominance of the asynchronous release mode was observed after 10 pulses. The steady-state asynchronous release rate showed a third-power dependency on the presynaptic Ca2+ concentration, which is similar to that of evoked release. The Ca2+ sensor for asynchronous release exhibited a high affinity for Ca2+ and was far from saturation. These properties of the Ca2+ sensor should make the asynchronous release very sensitive to any modification of presynaptic Ca2+ concentration, including those resulting from changes in presynaptic activity patterns. Thus, asynchronous release represents a powerful but delicately regulated mechanism that ensures the maintenance of appropriate inhibition when the readily releasable pool of vesicles is depleted.

Inhibition of mitochondrial Ca2+ uptake affects phasic release from motor terminals differently depending on external [Ca2+

We investigated how inhibition of mitochondrial Ca2+ uptake affects stimulation-induced increases in cytosolic [Ca2+] and phasic and asynchronous transmitter release in lizard motor terminals in 2 and 0.5 mM bath [Ca2+]. Lowering bath [Ca2+] reduced the rate of rise, but not the final amplitude, of the increase in mitochondrial [Ca2+] during 50-Hz stimulation. The amplitude of the stimulation-induced increase in cytosolic [Ca2+] was reduced in low-bath [Ca2+] and increased when mitochondrial Ca2+ uptake was inhibited by depolarizing mitochondria. In 2 mM Ca2+, end-plate potentials (epps) depressed by 53% after 10 s of 50-Hz stimulation, and this depression increased to 80% after mitochondrial depolarization. In contrast, in 0.5 mM Ca2+ the same stimulation pattern increased epps by approximately 3.4-fold, and this increase was even greater (transiently) after mitochondrial depolarization. In both 2 and 0.5 mM [Ca2+], mitochondrial depolarization increased asynchronous release during the 50-Hz train and increased the total vesicular release (phasic and asynchronous) measured by destaining of the styryl dye FM2-10. These results suggest that by limiting the stimulation-induced increase in cytosolic [Ca2+], mitochondrial Ca2+ uptake maintains a high ratio of phasic to asynchronous release, thus helping to sustain neuromuscular transmission during repetitive stimulation. Interestingly, the quantal content of the epp reached during 50-Hz stimulation stabilized at a similar level ( approximately 20 quanta) in both 2 and 0.5 mM Ca2+. A similar convergence was measured in oligomycin, which inhibits mitochondrial ATP synthesis without depolarizing mitochondria, but quantal contents fell to <20 when mitochondria were depolarized in 2 mM Ca2+.

Intraterminal Ca2+ concentration and asynchronous transmitter release at single GABAergic boutons in rat collicular cultures.

Neurotransmitter release in response to a single action potential has a precise time course. A significant fraction of the releasable vesicles is exocytosed synchronously, within a few milliseconds after the arrival of an action potential. If repeatedly activated, stimulus-locked phasic synchronous release declines, but synaptic transmission can be maintained through tonic asynchronous transmitter release. The desynchronisation of release during repetitive activation is generally attributed to a build-up of intraterminal Ca2+ concentration. However, the precise relationship between presynaptic Ca2+ level and asynchronous release rate at small central synapses has remained unclear. Here we characterise this relationship for single GABAergic terminals in rat collicular cultures. In the presence of tetrodotoxin, inhibitory postsynaptic currents (IPSCs) and presynaptic Ca2+ transients were recorded in response to direct presynaptic depolarisation of individual boutons. Repetitive stimulation indeed resulted in a shift from phasic to asynchronous neurotransmitter release. A clear dominance of the asynchronous release mode was observed after 10 pulses. The steady-state asynchronous release rate showed a third-power dependency on the presynaptic Ca2+ concentration, which is similar to that of evoked release. The Ca2+ sensor for asynchronous release exhibited a high affinity for Ca2+ and was far from saturation. These properties of the Ca2+ sensor should make the asynchronous release very sensitive to any modification of presynaptic Ca2+ concentration, including those resulting from changes in presynaptic activity patterns. Thus, asynchronous release represents a powerful but delicately regulated mechanism that ensures the maintenance of appropriate inhibition when the readily releasable pool of vesicles is depleted.

Development of two transmitter release components during the critical period for imprinting in the chick IMHV.

Transmitter release at an excitatory synapse has two components, fast synchronous and slow asynchronous transmitter release. Using the whole cell recording technique, we investigated the developmental properties of neurotransmitter release, which is composed of the two components in the intermediate and medial part of the hyperstriatum ventral (IMHV) of chicks during the critical period for imprinting. Analysis of the paired-pulse responses revealed that the depression of the excitatory postsynaptic currents (EPSCs), driven mainly by fast synchronous release, was frequently observed in P0-1 chicks but not in those at P5-8. The spontaneous excitatory postsynaptic currents (sEPSCs) after the paired-pulse stimulation, which were thought to be driven by asynchronous transmitter releases, were observed more frequently in P0-1 chicks than P5-8 chicks. Furthermore, examination of Ca2+ dependency in the evoked EPSCs showed that the amplitudes in P5-8 chicks were more sensitive to reduction of the extracellular Ca2+ concentration than younger chicks. Considering that the Ca2+ dependency of EPSCs is defined by both Ca2+ sensitivity and the proportion of each type of release machineries at the release site, these results indicate that the ratio of fast synchronous to slow asynchronous transmitter release machinery changed during the critical period. These changes may play critical roles in the capacity of the avian brain to consolidate novel experience in the immediate period after hatching.

Synchronous and asynchronous exocytosis induced by subthreshold high K+ at Cs(+)-loaded terminals of rat hippocampal neurons.

Transmitter release at Cs(+)-loaded autaptic terminals was selectively activated by the subthreshold concentration of external K+, and Ca(2+) channel types and transmitter pools involved in synchronous and asynchronous exocytosis were studied. When a neuron was depolarized to +30 mV by applying a current through a pipette containing Cs(+) for >30 s, a rapid external K+ jump to 3.75-10 mM, otherwise ineffective, produced an outward current (K10 response). K10 responses were initially graded (type-1) and then became a spike and plateau-shape with (type-2) or without a latency (type-3). On repolarization to -60 mV, a high K+ jump induced inward currents (called also K10 response) similar to those at +30 mV, whose shape changed from that of type-3, then type-2 and finally type-1 over 30 min. During a period favorable for inducing a type-3 response, a current similar to this response was generated by a voltage pulse (+ 80 or 90 mV, 20 or 30 ms) to the cell soma. Currents similar to K10 responses were rarely induced by a high K+ jump without a conditioning depolarization except for some cells, but consistently produced when 3 mM Cs(+) and 50 microM 4-aminopyridine were externally applied for tens of minutes. Picrotoxin, 6-cyano-7-nitroquinoxaline-2,3-dione with 3-[(RS)-2-carboxypiperazin-4-yl]-propyl-1-phosphonic acid or Cd(2+) in, or Ca(2+) removal from, a high-K+ solution blocked all the K10 responses, while a plateau remaining after a high K+ jump was not blocked by Ca(2+) removal immediately after the K+ jump. Thus Cs(+) loading and decreased K+ concentration in autaptic terminals by a conditioning depolarizing current selectively sensitize the terminals to a subthreshold high K+ jump for depolarization to activate synchronous or asynchronous transmitter release. Nicardipine (5-10 microM) blocked type-1 and -2 responses but not type-3 responses, while omega-conotoxin (10 microM) blocked all the types of K10 response in the presence of nicardipine. Increasing the interval of high K+ jumps biphasically increased the magnitude of K10 response, preferentially in the postjump fraction reflecting purely the asynchronous activation of exocytotic machinery, and decreased the reduction of miniature postsynaptic current frequency after a K10 response. These results suggest the roles of N(P/Q)-type Ca(2+) channels in synchronous exocytosis at the terminals, L-type Ca(2+) channels in initiating a Ca(2+) action potential at the parent axon and both types in asynchronous exocytosis and also suggest the different releasable pools of transmitter for two modes of exocytosis in cultured hippocampal neurons.

Developmental increase in asynchronous GABA release in cultured hippocampal neurons

Developmental changes in GABAergic synaptic transmission were examined in cultured hippocampal neurons using patch-clamp recordings and Ca 2+ imaging. In paired recordings, tetanization of the presynaptic GABAergic neuron with 80 pulses at either 40 or 80 Hz was accompanied by tetanic depression of inhibitory postsynaptic responses. In neurons that had been cultured for more than two weeks, asynchronous inhibitory postsynaptic currents often appeared during the tetanus and continued for several seconds following stimulation. There was little asynchronous activity in neurons that had been cultured for shorter times. However, no age-related changes were observed in the amplitude of single synchronous inhibitory postsynaptic currents, paired-pulse depression or post-tetanic potentiation of inhibitory postsynaptic currents. Following equimolar replacement of extracellular Ca 2+ with strontium ions (Sr 2+), single autaptic inhibitory postsynaptic currents were depressed in amplitude and asynchronous inhibitory postsynaptic currents were present on the decaying phase. Sr 2+-induced asynchronous inhibitory postsynaptic currents showed no dependence on age in culture. Imaging of Ca 2+ in single GABAergic boutons was performed by including Fluo-3 in the patch pipette. During action potential firing induced by stimulating at 80 Hz for 1 s, intracellular calcium [Ca 2+] i increased rapidly in individual boutons. Following the stimulus, [Ca 2+] i decayed back to baseline within 10–15 s. The half-time of decay increased from 1.7±0.2 s at 15 days in vitro to 4.0±0.2 s at 30 days in vitro ( P<0.05), with a developmental profile that closely matched the increase in asynchronous inhibitory postsynaptic currents. We propose that the increase in tetanus-induced asynchronous GABA-release during the first month of synapse maturation in vitro is caused by a slowing of the Ca 2+-clearing mechanisms in the GABAergic boutons. This results in larger and more prolonged elevations of [Ca 2+] i during tetanic stimulation, which leads to enhanced asynchronous transmitter release. We propose that the results of this study demonstrate a potentially important aspect of synapse maturation during development, and also imply that GABA release is up-regulated in conditions of decreased Ca 2+ buffering and clearing.

Inhibitory Transmission Mediated by Asynchronous Transmitter Release

At fast CNS synapses, the role of asynchronous release following initial synchronous release is poorly understood. We examined the contribution of asynchronous release to GABAergic transmission in the cochlear nucleus across a 40-fold range of electrical stimulus frequencies. Whereas quantal release was highly synchronized at low frequencies, it was largely continuous and desynchronized at high frequencies. Despite the change in release mode, intense and steady inhibitory transmission was virtually maintained. Experimental analyses and modeling studies indicated that this "desynchronization" process was dependent on presynaptic Ca2+ accumulation, facilitation of vesicle release, and short-term depletion of available vesicles. Asynchronous release at high frequencies may help generate a smooth inhibitory "tone" by minimizing the consequences of random timing of presynaptic action potentials.

Serotonin, via 5-HT 2A receptors, increases EPSCs in layer V pyramidal cells of prefrontal cortex by an asynchronous mode of glutamate release

Previously, serotonin (5-HT) was found to induce a marked increase in glutamatergic spontaneous excitatory postsynaptic currents (EPSCs) in apical dendrites of layer V pyramidal cells of prefrontal cortex; this effect was mediated by 5-HT 2A receptors, a proposed site of action of hallucinogenic and atypical antipsychotic drugs. Unexpectedly, although the effect of 5-HT was Ca 2+-dependent and tetrodotoxin-sensitive, it did not appear to involve the activation of excitatory afferent impulse flow. This paradox prompted us to investigate (in rat brain slices) whether 5-HT was acting through an atypical mode of excitatory transmitter release. We found that the frequency of 5-HT-induced spontaneous EPSCs was fully supported by Sr 2+ in the absence of added Ca 2+, implicating the mechanism of asynchronous transmitter release which has been linked to the high-affinity Ca 2+-sensor synaptotagmin III. Although the early, synchronous component of electrically evoked EPSCs was reduced while 5-HT was being applied, late, nonsynchronous components were enhanced during 5-HT washout and also by the 5-HT 2 partial agonist 1-(2,5-dimethoxy-4-iodophenyl-2-aminopropane (DOI); the effect of DOI was blocked by a selective 5-HT 2A antagonist (MDL 100,907). This late, nonsynchronous component was distinct from conventional polysynaptic EPSCs evoked in the presence of the GABA A antagonist bicuculline, but resembled asynchronous glutamatergic excitatory postsynaptic potentials (EPSPs) evoked in the presence of Sr 2+. An enhancement of asynchronous EPSCs by a specific neurotransmitter receptor has not been reported previously. The possible role of excessive asynchronous transmission in the cerebral cortex in mediating the hallucinogenic effects of 5-HT 2A agonists such as DOI is discussed.

Simultaneous measurement of intracellular Ca2+ and asynchronous transmitter release from the same crayfish bouton.

1. A technique has been developed to monitor neurotransmitter release simultaneously with intracellular Ca2+ concentration ([Ca2+]i) in single release boutons whose diameters range from 3 to 5 microns. 2. Using this technique, we have found a highly non-linear relationship between the rate of asynchronous release and [Ca2+]i. The Hill coefficient lies between 3 and 4. 3. The affinity (Kd) of the putative release-related Ca2+ receptor for asynchronous release was calculated to be in the range of 2-4 microM. 4. The same range of values of Hill coefficient and Kd were obtained when [Ca2+]i was elevated both by bath application of ionomycin and by repetitive stimulation at high frequency. 5. Our results show that the Ca2+ receptor(s) associated with asynchronous release exhibits high affinity for Ca2+.

Rapid kinetics and inward rectification of miniature EPSCs in layer I neurons of rat neocortex.

With the use of the whole cell patch-clamp technique combined with visualization of neurons in brain slices, we studied the properties of miniature excitatory postsynaptic currents (mEPSCs) in rat neocortical layer I neurons. At holding potentials (-50 to -70 mV) near the resting membrane potential (RMP), mEPSCs had amplitudes of 5-100 pA and were mediated mostly by alpha-amino-3-hydroxy-5-methyl-4-isoxazoleproprionate (AMPA) receptors. Amplitude histograms were skewed toward large events. An N-methyl-D-aspartate (NMDA) component was revealed by depolarization to -30 mV or by the use of a Mg2+-free bathing solution. At RMP, averaged AMPA mEPSCs had a 10-90% rise time of approximately 0.3 ms (uncorrected for instrument filtering). The decay of averaged mEPSCs was best fit by double-exponential functions in most cases. The fast, dominating component had a decay time constant of approximately 1.2 ms and comprised approximately 80% of the total amplitude. A small slow component had a decay time constant of approximately 4 ms. Positive correlations were found between rise and decay times of both individual and averaged mEPSCs, indicative of dendritic filtering. Some large-amplitude mEPSCs and spontaneous EPSCs (recorded in the absence of tetrodotoxin) had slower kinetics, suggesting a role of asynchronous transmitter release in shaping EPSCs. The amplitudes of mEPSCs were much smaller at +60 mV than at -60 mV, indicating that synaptic AMPA-receptor-mediated currents were inwardly rectifying. These results suggest that neocortical layer I neurons receive both NMDA- and AMPA-receptor-mediated synaptic inputs. The rapid decay of EPSCs appears to be largely determined by AMPA receptor deactivation. The observed rectification of synaptic responses suggests that synaptic AMPA receptors in layer I neurons may lack GluR-2 subunits and may be Ca2+ permeable.

Asynchronous transmitter release: control of exocytosis and endocytosis at the salamander rod synapse.

1. We have studied exocytosis and endocytosis in the synaptic terminal of salamander rods using a combination of Ca2+ imaging, capacitance measurement and the photolysis of Ca2+ buffers. 2. The average cytoplasmic Ca2+ concentration at the dark resting potential was 2-4 microM. 3. An average cytoplasmic Ca2+ concentration of 2-4 microM maintained a high rate of continuous exocytosis and endocytosis. 4. Changes in the rate of exocytosis were followed in less than 0.7 s by compensatory changes in the rate of endocytosis. 5. Vesicle cycling in the rod synapse is specialized for graded transmission and differs from that previously described for synapses that release synchronized bursts of transmitter.

Counting quanta: Direct measurements of transmitter release at a central synapse

Contradictory hypotheases regarding the nature of synaptic transmission in the CNS have arisen from indirect methods of quantal analysis. In this study, we directly count the quanta released following nerve stimulation to examine synaptic transmission at a fast glutamatergic synapse in the mammalian auditory brainstem. Our results demonstrate the relationship between spontaneous and nerve-evoked synaptic events, indicate that asynchronous transmitter release governs the time course of evoked transmission, and show that the stochastic quantal release process, as originally proposed at the neuromuscular junction, is highly conserved at this central synapse.

A study of synchronization of quantal transmitter release from mammalian motor endings by the use of botulinal toxins type A and D.

1. The effects of botulinum toxin (BoTx) types A and D on spontaneous and evoked phasic transmitter release were studied in the isolated extensor digitorum longus muscle of the rat or the levator auris longus muscle of mice. 2. The toxins were injected subcutaneously into the hindleg of adult rats or the dorsal aspect of the neck of mice. At various times after the injection the muscles were removed from the anaesthetized animal and neuromuscular transmission examined in vitro by conventional intracellular techniques. 3. Both toxins reduced spontaneous transmitter release recorded as the frequency of miniature end-plate potentials but BoTx type D was less effective in that respect than the type A toxin. 4. With both toxins the block of evoked phasic transmitter release, recorded as end-plate potentials, was almost complete. As previously reviewed by Simpson (1986) the block produced by BoTx type A was partially reversed by procedures which elevate the intraterminal level of calcium ions. However, in BoTx type D-paralysed muscles such procedures failed to restore phasic transmitter release but caused a period of high-frequency asynchronous transmitter release following each nerve impulse. 5. To investigate if the lack of synchronization of evoked transmitter release observed in BoTx type D-paralysed muscles was due to alterations in presynaptic currents we examined, by perineural recordings, the Na+, fast K+, slow K+, K+-Ca2+-dependent and the Ca2+ currents in BoTx type D-paralysed muscles. These presynaptic currents were not altered as compared to unpoisoned controls. 6. We suggest that there exists a presynaptic process, which in addition to Ca2+ influx participates in transmitter synchronization and which is a main target for BoTx type D action.

Presynaptic membrane potential and transmitter release at the crayfish neuromuscular junction.

Release of transmitter was evoked at neuromuscular junctions of the crayfish opener muscle by passage of current through an intracellular electrode impaling a branch of the motor axon close to a muscle fiber. Membrane-potential changes in the presynaptic axon branch were monitored, together with postsynaptic potentials. Depolarization of impaled secondary axonal branches by more than 10 mV led to an increase in asynchronous transmitter release. The release was facilitated by prolonged (50-500 ms) depolarizations and it decayed rapidly when depolarization was terminated. Ca2+ was essential for facilitated release; however, no indication of a Ca spike was found at the recording site. Input-output curves for the synapse were obtained by applying depolarizing pulses of varying amplitude to the axon branch. Transmitter output was strongly influenced by both amplitude and duration of the applied depolarization. During normal synaptic transmission, propagated Na+-dependent action potentials were recorded in the secondary axonal branches but there was no evidence for a calcium-dependent component for these action potentials. Evoked release was dependent on Ca2+ and was steeply dependent on the amplitude of the action potential, which could be made variable in size by application of tetrodotoxin (TTX). Prolonged depolarization of axonal branches resulted in enhancement of transmitter release evoked by an action potential. The enhancement occurred in spite of a simultaneous reduction of the amplitude of the action potential. Morphological features of the terminals were investigated after injection of lucifer yellow into the axon. An electrical model incorporating the morphological features suggests that membrane-potential changes set up in the main axon reach the nearest terminals with 30-40% attenuation, while events originating in the terminals would be severely attenuated in the main axon. Comparison of the crayfish synapse with other frequently studied synapses shows both similarities and differences, suggesting that it is not possible to apply findings made in one synapse to all others.

The physiology of a coelenterate neuromuscular synapse

1. Postsynaptic potentials can be recorded intracellularly from epitheliomuscular cells overlying the inner nerve-ring in the medusaPolyorchispenicillatus (Cnidaria, Hydrozoa) (Fig. 1). These postsynaptic potentials lead to the generation of muscle action potentials which propagate through the swimming-muscle sheets. It is the swimming motor neuron network that innervates this epithelium. An alternative motor pathway is present that involves a network of small, multipolar neurons that is interpolated between swimming motor neurons and overlying epithelial cells (Fig. 2). 2. That the postsynaptic potentials recorded are due to release of a chemical transmitter is supported by the following evidence: (a) PSPs have a constant delay\((\bar x = 3.2{\text{ms)}}\) following the presynaptic spike (Figs. 3, 4); (b) high Mg++ concentrations reduce the amplitude of PSPs and eventually block transmission (Fig. 10); (c) there is no electrical coupling between presynaptic neurons and postsynaptic epithelial cells. 3. There is an inverse relationship between the duration of presynaptic action potentials and the amplitude of PSPs (Figs. 5, 6). The duration of presynaptic action potentials is a reflection of the degree of synchrony of spiking in the motor network so that short duration motor spikes are associated with synchronous firing. In such cases, the simultaneous release of transmitter substance at a number of neighbouring synapses will cause rapid temporal summation of PSPs in postsynaptic cells. Similarly, long duration presynaptic spikes are associated with asynchronous transmitter release and consequently with small PSPs (Figs. 8, 9). 4. Changes in PSP amplitude are seen in all postsynaptic cells of a localised region (Fig. 7). 5. The muscle action potential can be separated into two components, the velar and subumbrellar action potentials (Fig. 11). This biphasic nature of muscle action potentials recorded in the synaptic region results from all-or-none action potentials that are generated at the velar and subumbrellar borders of this region conducting back electrotonically. These action potentials made to conduct antidromically towards the synapses by electrical stimulation of the muscle sheets decrement as they travel through the synaptic region (Fig. 12). The nature of electrical coupling between epithelial cells in the synaptic non-muscular region and the muscle sheets proper must be different. 6. Larger amplitude PSPs are associated with muscle action potentials that follow with a shorter latency, and that have the two components (velar and subumbrellar) following each other more rapidly (Figs. 5, 9). 7. Action potentials in the motor network are brought into phase as they conduct around the margin. This leads to more synchronous activation of synapses and hence larger PSPs at regions distant from the initiation site of the motor spike. The resulting decrease in the latency of muscle APs at these distant sites will automatically compensate for the conduction delay of motor spikes.

Alterations of synaptic action in chromatolysed motoneurones of the cat.

1. Monosynaptic EPSPs in lumbosacral motoneurones undergoing chromatolysis were studied by intracellular recording from 7 to 20 days after section of the appropriate ventral roots of the cat.2. The maximum monosynaptic EPSPs evoked in chromatolysed motoneurones by afferent volleys from the biceps-semitendinosus or the triceps surae muscles ranged from 1.0 to 9.5 mV in amplitude. The time-to-peak of these EPSPs was 1.7 msec on the average. These values were significantly smaller and longer, respectively, than the amplitude and the time-to-peak of monosynaptic EPSPs observed in normal motoneurones. The long time-to-peak of EPSPs in chromatolysed motoneurones could not be accounted for by asynchronous transmitter release.3. The mean number of unit EPSPs responding to a single afferent impulse (m) in chromatolysed motoneurones was comparable to that found in normal motoneurones.4. The amplitude of unit EPSPs estimated from the mean EPSP amplitude and the m value following stimulation of a single afferent fibre was significantly smaller in chromatolysed motoneurones than in normal motoneurones. This difference was attributed to a difference in synaptic location.5. The shape of monosynaptic EPSPs evoked in chromatolysed motoneurones by stimulation of single afferent fibres was analysed on the basis of Rall's compartment model. The analysis suggested that there is a lack of the excitatory synaptic input to the cell body in chromatolysed motoneurones.6. Similar alterations were also found in IPSPs. The degree of change in synaptic responses evoked by stimulation of various pathways appears to depend on the synaptic location.7. Following the study of the interaction of several inputs on the motoneurone and of their dependence on membrane potential, a tentative model of the synaptic distribution of different pathways is proposed.