Ultrastructure of identified fast excitatory, slow excitatory and inhibitory neuromuscular junctions in the locust

Summary The specialized jumping muscle of the locust, the metathoracic extensor tibiae (ETi), is innervated by four physiologically different motoneurons, including FETi, a phasic excitor, SETi, a tonic excitor, and CI, a tonic common inhibitor. FETi neuromuscular junctions were examined in three phasic ETi bundles innervated by FETi. FETi terminals were characterized by patchy contacts on to granular sarcoplasm. The ETi accessory extensor, innvervated by both SETi and CI, contains two morphologically different types of axon ending. When this muscle was soaked in horseradish peroxidase, stimulation of SETi led to selective uptake in vesicles in terminals similar to those of FETi axons but containing smaller vesicles, while stimulation by CI caused increased uptake into terminals with more extensive contact directly on to fibrillar sarcoplasm. As has been observed in excitatory and inhibitory synapses in some crustacean and vertebrate nervous systems, the synaptic vesicles in the locust excitatory endings are round and electron-lucent while those in the inhibitory endings are more irregular in shape. The tonic neuromuscular junctions, SETi and CI, are more densely packed with vesicles, larger in cross-sectional area and appear to be of more complex shape than the smaller, vesicle-sparse, phasic FETi terminals. Following long duration stimulation at 10 Hz, the tonic neuromuscular junctions showed little morphological change. FETi endings, which fatigue within minutes at the same stimulation frequency, showed a 20% decrease in synaptic vesicle density and an increase in irregularly shaped membrane inclusions.     

Characteristics of transmitter release at regenerating frog neuromuscular junctions

1. A study was made of the onset of transmission and the characteristics of transmitter release from regenerating nerve terminals in frog muscle fibres.

2. Soon after transmission had been restored, some junctions were found which responded to nerve stimulation with only subthreshold end-plate potentials.

3. The evoked transmitter release had a non-linear dependence on the external calcium concentration, like that seen at normal junctions.

4. The synaptic delay was only slightly longer than normal, and the amplitudes of single quantum potentials evoked by nerve stimulation seemed to have a normal distribution.

5. The mean amplitude of the spontaneous miniature end-plate potentials was often substantially smaller than the mean amplitude of the evoked quantal potentials at a given end-plate. Some of these small spontaneous potentials were due to transmitter release from the axon terminal. Possible explanations for this discrepancy in size of spontaneous and evoked potentials are discussed. Two to three weeks after reinnervation began, the amplitude of the spontaneous miniature end-plate potentials returned to normal.

     

Postsynaptic specializations at excitatory and inhibitory cholinergic synapses

Summary In serial sections of neurons in the paravertebral ganglia of the frog (Limnodynastes dumerili), the postsynaptic structures termed postsynaptic bar (PSB) and junctional subsurface organ (JSO) were never observed in the same ganglion cell. Further, PSBs were found mostly in small ganglion cells (less than 22 m), while JSOs were found mostly in large ganglion cells (up to 45 m). Between 10 and 22 PSBs were located at both spine and non-spinous somatic synapses of the smaller ganglion cells; while 8 to 16 JSOs were located largely in the axon hillock region of the larger ganglion cells.Based on these observations, it is suggested that the two ganglion cell populations represent the B and C cell types defined according to electrophysiological data. Further, since the nerve terminals adjacent to both these postsynaptic structures appear to be cholinergic according to their vesicular content, this provides some basis for suggesting that JSOs are associated with slow excitatory synapses, while PSBs are present at slow inhibitory synapses.     

Enhanced release of inhibitory and excitatory transmitter quanta in the crayfish neuromuscular junction by glycine and GABA

High concentrations of glycine (greater than or equal to 0.1 mol/l) applied to the bathing medium of voltage-clamped muscle fibres elicited high rates of spontaneous inhibitory miniature currents (sIPSCs) which were identified by means of the noise analysis technique. The rate of spontaneous excitatory miniature currents (sEPSCs) was not raised appreciably in presence of these high concentrations of glycine. Contrary to the effect of glycine, high concentrations of GABA (greater than or equal to 0.1 mol/l) desensitized inhibitory postsynaptic receptors and induced high rates of sEPSCs. When, in addition to GABA, glycine was also applied at high concentrations, the discharge of sEPSCs was suppressed. Moreover, glycine reduced significantly the current noise intensity elicited by activation of excitatory postsynaptic membrane channels with 1.10(-4) mol/l glutamate. The experiments suggest an inhibitory action of glycine on the excitatory receptor--channel complex in the postsynaptic membrane.     

The termination of transmitter action at the crustacean excitatory neuromuscular junction.

1. Excitatory junctional currents (e.j.c.s) and miniature excitatory junctional currents (mine.e.j.c.s) have been followed by recording the focal extracellular potential at excitatory neuromuscular junctions of Maia squinado, the Spider Crab. 2. If L-aspartate (concentrations less than or equal to 1 mM) is present in the saline, the average e.j.c. is prolonged by an increase in the duration of its falling phase. 3. No change occurs in the time course of the probablility of release of quanta in the e.j.c. as determined from the histogram of first quantal latencies. 4. The min.e.j.c. is also prolonged by L-aspartate (concentrations less than or equal to 0-5 mM). The rise time increases slightly, the pre-exponential period of the decay phase almost doubles, and the time constant of the final exponential decay increases to a value larger than the average lifetime of the elementary conductance event produced by L-glutamate. 5. Changes in e.j.c.s and min.e.j.c.s produced by aspartate show a striking similarity to the action of neostigmine on the time course of the vertebrate end-plate current. It is proposed that aspartate exerts its action by blocking a process normally reponsible for clearing the synaptic cleft of transmitter.     

L-glutamate as an excitatory transmitter at the Drosophila larval neuromuscular junction.

The possibility that L-glutamate is the excitatory transmitter at the Drosophila larval neuromuscular junction and the ionic basis of its action on the muscle membrane are examined. 2. Iontophoretically applied L-glutamate causes muscle depolarization (L-glutamate potential) if and only if the L-glutamate pipette is within a few mum of the nerve ending. D-glutamate, substance P, ACh and GABA are ineffective. 3. Bath-applied L-glutamate produces similar changes in the time course and amplitude of miniature excitatory junctional potential (m.e.j.p.), excitatory junctional potential (e.j.p.) and the L-glutamate potential. 4. Neuromuscular transmission and excitation-contraction coupling are operative in a haemolymph-like solution containing 1 mM L-glutamate. 5. The reversal potentials of the e.j.p. and the L-glutamate potential are identical to each other, changing similarly with changes in the ionic compositions of the external medium (twelve solutions). 6. The ionic dependence of the reversal potentials is predicted from an extended constant-field equation using a ratio of sodium:potassium permeabilities of PNa/PK=1-3, and a ratio of magnesium:potassium permeabilities of PMg/PK=4-7. 7. It is concluded that L-glutamate is, or is an agonist of, the excitatory transmitter at certain Drosophila larval neuromuscular junctions.     

An excitatory action of adrenaline, noradrenaline, and 5-hydroxytryptamine on the spinal cord

1. Fasciculation was produced in the trapezius and sternomastoid muscles of anaesthetized cats by adrenaline, noradrenaline and 5-hydroxytryptamine (5-HT) put into the cisterna magna. Isoprenaline, similarly applied, was ineffective.

2. Each of the three active amines was without effect when applied for a second time after the fasciculation in response to the first application had passed off, but 5-HT was effective after adrenaline and, similarly, adrenaline after 5-HT.

3. Fasciculation produced by adrenaline, but not by 5-HT, was inhibited by ergotamine or phenoxybenzamine, given intravenously.

4. In producing fasciculation, the adrenaline appeared to be acting on or through the lateral aspect of the upper cervical cord, about the line of emergence of the roots of the spinal accessory nerve, mainly in C1 and C2.

5. The electromyograms of the fasciculation due to adrenaline and 5-HT showed intermittent bursts of activity. After adrenaline, the bursts consisted of fewer spikes than after 5-HT. The intervals between consecutive spikes were 3·75-5 msec after either amine, and the bursts occurred irregularly, at frequencies between 7 and 12/sec.

6. It is suggested that adrenaline and 5-HT have excitatory actions on the dendrites or somata of the spinal accessory motor neurones, and the possible role of these amines as synaptic transmitters in the spinal nucleus of the accessory nerve is discussed.

     

The synergistic action of L-glutamate and L-aspartate at crustacean excitatory neuromuscular junctions.

1. When L-glutamate and L-aspartate are simultaneously applied to the excitatory neuromuscular junctions of Maia squinado, they produce an increase in the conductance of the post-synaptic membrane much larger than the sum of conductance effects produced by the individual amino acids alone. 2. An examination of the synaptic noise occurring during this synergistic action reveals that the elementary conductnace events produced by aspartate are suppressed and that normal elementary conductance events produced by glutamate are occurring at an enormously increased rate. 3. It is suggested that aspartate causes this potentiation by inhibiting a system for transmitter inactivation in the region of the post-synaptic receptors and that this system, under normal conditions, prevents the access of externally applied glutamate to the synaptic receptors.     

On the elementary conductance event produced by L-glutamate and quanta of the natural transmitter at the neuromuscular junctions of Maia squinado.

1. The membrane potential of giant muscle fibres of Maia squinado was measured with an intracellular wire electrode. On applying L-glutamate to the fibre the cell deplorized and fluctuations of the membrane potential around its mean level--glutamate noise--were seen. 2. The variance of the glutamate voltage noise is proportional to the mean level of depolarization. The noise can be regarded as being caused by numerous exponentially decaying elementary voltage events about 5 X 10(-10) V in amplitude. The miniature excitatory junctional potential (min.e.j.p.) is approximately 6000 times the amplitude of the elementary voltage event produced by L-glutamate. 3. The power spectrum of glutamate voltage noise is a Lorentzian with a half-power frequency of approximately 20 Hz. 4. Min. e.j.p.s. decay exponentially with a time constant that coincides with the average lifetime of the elementary glutamate voltage event. 5. When glutamate is applied locally to a spot where extracellular min. e.j.p.s. can be recorded with a focal glass pipette, extracellular glutamate noise is seen. Glutamate noise could not be detected from elsewhere on the fibre. 6. The variance of the extracellular noise is proportional to the mean extracellular potential, and its power spectrum is a Lorentzian with a half-power frequency of about 110 Hz. 7. The extracellular min. e.j.p.s decay exponentially with a time constant that coincides with average lifetime of the elementary glutamate current event. 8. It is suggested that the decay of the quantal currents flowing at the excitatory junction is limited by the closure of the conductance channels in the post-synaptic membrane and not by the relaxation of the transmitter concentration in the synaptic cleft.     

Competition between sodium and calcium ions in transmitter release at mammalian neuromuscular junctions

1. Frequencies of miniature end-plate potentials (m.e.p.p.s) were recorded at neuromuscular junctions in rat diaphragm-phrenic nerve preparations in vitro.2. In the presence of raised [K] (15-20 mM) lowering [Na] caused a rapid increase in m.e.p.p. frequency whether [Ca] was low or normal. Raising [Na] towards the normal concentration (162 mM) caused a slow fall in frequency and raising [Ca] in the range 0.32-2 mM caused a slow increase in frequency. These effects were less in the normal [K] (5 mM).3. Mean m.e.p.p. frequencies were determined for solutions containing 15 mM-K and combinations of [Ca] and [Na]. M.e.p.p. frequency varied inversely with [Na] when [Ca] was constant. In each of the three Na concentrations used (162, 113 and 65 mM) raising [Ca] in the range 0.32-2 mM increased m.e.p.p. frequency but when raised above 2-3 mM, Ca depressed frequency.4. A model was proposed in which Ca affected transmitter release by changing the concentration in the presynaptic membrane of a complex CaX to which the rate of transmitter release was directly proportional. Higher concentrations of Ca depressed transmitter release by inactivating CaX. Sodium ions competitively depressed release either by competing with calcium ions for association with X or by reducing the affinity of X for Ca.5. When [Na] was lowered in solutions containing raised [Mg] and [Ca], the increase of mean m.e.p.p. frequency was greater than that observed in raised [Ca] and normal [Mg] and was of the same order as the increases seen in low [Ca]. The result was interpreted to indicate either that Na and Mg do not compete with Ca at the same site or that Mg affects the affinity of X for Ca and Na.6. The effect of lowering [Na] on m.e.p.p. frequency was a specific effect of Na ions. When LiCl was substituted for NaCl, the increase of m.e.p.p. frequency persisted. Changes in [Cl] had no effect on m.e.p.p. frequency.7. There was a linear relation between the mean logarithm of m.e.p.p. frequencies and [K], the slope of the relation increasing as [Na] was lowered. Conversely, lowering [Na] caused a greater increase in m.e.p.p. frequency as [K] was raised.8. The variation of m.e.p.p. frequencies in a diaphragm was roughly proportional to a second or higher power of [Na] and inversely proportion to [Ca]. It was thought that this could be due to differences in chelation of Ca which were more apparent at low Ca concentrations.9. The similarities between the effects of Na, Ca and K on m.e.p.p. frequency and the effects of these ions on Ca-influx in heart muscle led to the suggestion that transmitter release is proportional to the concentration of a negatively charged complex of a carrier X with one calcium ion (CaX) at the internal surface of the membrane and that changes in membrane potential affect transmitter release by changing the distribution or location of CaX in the membrane.     

Effects of physiologic alterations on binomial transmitter release at magnesium-depressed neuromuscular junctions.

1. Transmitter release from Mg2+-treated frog neuromuscular junctions can be described by binomial statistics. Good agreement between the observed amplitude-frequency distribution of e.p.p.s. and that predicted by binomial statistics is observed with relatively low concentrations of Mg2+. Conversely, good agreement is found with Poisson predictions when higher concentrations of Mg2+ are used to depress transmission. 2. Binomial analysis at these junctions shows that Mg2+ reduces quantal content (m), the probability of release (p) and to a lesser extent the available stores of transmitter (n). Raising Ca2+ causes an increase in n and p and a small but significant increase in n. K+ increases m and p but not n. 3. During 'frequency-facilitation' (1-6 Hz), e.p.p.s., m and n are increased but p is unaffected. 4. It is concluded that binomial statistics can be used to estimate the quantal parameters of transmitter release and that these parameters can be identified as discrete entities.     

Binomial analysis of quantal transmitter release at glycerol treated frog neuromuscular junctions.

1. The actions of ether and methoxyflurane on the evoked potentials of in vitro preparations of the guinea-pig olfactory cortex were studied. Following stimulation of the lateral olfactory tract (l.o.t.) evoked potentials could be recorded from the cortical surface; these potentials consisted of an initial wave (the compound action potential of the l.o.t.) followed by a negative field potential which was associated with the synchronous excitation of many superficial excitatory synapses (population e.p.s.p.). Superimposed on the population e.p.s.p. was a number of positive peaks. These positive peaks reflect the synchronous discharge of many neurones and so have been called population spikes. 2. When ether or methoxyflurane was added to the gas stream that superfused the surface of the preparations, the population e.p.s.p.s. and population spikes were depressed at lower concentrations than those required to depress the compound action potential of the afferent fibres. 3. The evoked activity of individual cells in the cortex was depressed by ether and methoxyflurane. However, five of the twelve cells tested in ether showed an increase in their evoked activity at concentrations below 4-5%, but at higher concentrations these cells also became depressed. 4. Both ether and methoxyflurane depressed the sensitivity of cortical neurones to iontophoretically applied L-glutamate and may similarly depress the sensitivity of the post-synaptic membrane to the released transmitter substance. 5. Neither anaesthetic appeared to increase the threshold depolarization required for nerve impulse generation. Thus, the decrease of the discharge of the post-synaptic cells was primarily caused by a depression of chemical transmission. 6. Ether caused some cells in the cortex to alter their normal pattern of synaptically evoked discharge and both anaesthetics induced similar changes during excitation by glutamate.     

Does 5-hydroxytryptamine influence "purinergic" inhibitory neurons in the intestine?

Intrinsic inhibitory neurons to guinea pig taenia coli and small bowel circular muscle were activated by transmural electrical stimulation, and the postinhibitory contractile response of the muscle was utilized to evaluate whether or not the neuronal action of 5-hydroxytryptamine (5HT) was associated with the inhibitory neurons. The postinhibitory contractile responses of the small intestinal circular muscle were unaffected by 5HT. The 5HT antagonist methysergide also did not affect the poststimulus contractile response of the circular muscle. The amplitude and area under the contractile curve of the poststimulus contractile response of the taenia coli were reduced and the amplitude of the relaxation response to electrical stimulation was increased in one-half of the preparations after application of 5HT. Methysergide did not alter the poststimulus contractile response of the taenia coli. 5HT is implicated as a neurotransmitter substance for slow synaptic excitation within the enteric nervous system of the guinea pig small intestine; however, the 5HT synapses do not appear to be present on the "purinergic" inhibitory neurons nor on neurons that synaptically influence the inhibitory neurons.     

Interaction between inhibitory and excitatory synaptic potentials at a peripheral neurone.

1. The interaction between inhibitory and excitatory synaptic potentials in neurones lying in the submucous plexus of guinea-pig ileum has been examined. 2. It was found that during an inhibitory conductance change, electrotonic potentials were more depressed in amplitude than were excitatory synaptic potentials. 3. It is suggested that inhibitory conductance changes may have only a slight effect on the impedance seen by excitatory synaptic currents as much of the excitatory synaptic current flow is likely to be capacitive. 4. A part of the depression of excitatory synaptic potential amplitude was not associated with changes in electrical properties of neurones and it is suggested that inhibitory transmitter may reduce the release of excitatory transmitter.