ABSTRACT
The properties of the miniature inhibitory junction potentials (M.I.J.P.) and the inhibitory junction potentials (I.J.P.) elicited by nerve stimulation were investigated in longitudinal muscle fibres of the earthworm.
Histograms of the amplitudes (mean, 0·71 mV.) and the intervals (mean, 101 msec.) of the M.I.J.P. showed skew curves.
The polarity of the M.I.J.P. was reversed at about —60 mV. When the external chloride was substituted by glutamate the M.I.J.P. disappeared as an external chloride concentration of 15-20 mM, and further reduction reversed their polarity.
Picrotoxin blocked generation of the M.I.J.P. and the I.J.P.
The cross-over point of the current-voltage relation curves, with and without presence of GABA, occurred at a membrane potential of − 54 mV. in potassium-free solution, and at − 56 mV. in potassium-excess solution.
Iontophoretic application of GABA produced slow hyperpolarization. The equilibrium potential of the GABA-potential was about − 60 mV. During the time course of the GABA-potentials an increase in the membrane conductance was observed.
Miniature excitatory junction potentials (M.E.J.P.) and excitatory junction potentials (E.J.P.) could be recorded from the longitudinal muscle, but the M.E.J.P. were rare.
D-tubocurarine, but not atropine, completely blocked the M.E.J.P. and E.J.P. Prostigmine enhanced their amplitude and duration.
The reversal potential level for the E.J.P. was about o mV. Sodium-free solution lowered the reversal potential level for the M.EJ.P. to −20 mV.
INTRODUCTION
Neuromuscular transmission in invertebrates has been studied extensively (see Bullock & Horridge, 1965). However, the neuromuscular junctions in the earthworm have only been studied by Horridge & Robertes (1960).
The present experiments were carried out to investigate the transmission in the neuromuscular junctions which could be recorded from the postsynaptic membranes of the longitudinal muscle of the earthworm. Excitatory junction potentials (E.J.P.’S) and miniature excitatory junction potentials (M.E.J.P.’S) could be recorded but they appeared only seldom. In contrast, inhibitory junction potentials (I.J.P.’S) and miniature inhibitory junction potentials (M.I.J.P.’S) were recorded from nearly all the cells in which microelectrodes were inserted. Therefore this paper describes mainly the properties of the M.I.J.P.’S and the I.J.P.’S generated from the longitudinal muscle; the properties of the E.J.P.’S and the M.E.J.P.’S are partly described.
EXPERIMENTAL METHODS
The preparation, experimental procedure and the solutions used in these experiments were the same as described in a previous paper (Hidaka, Ito & Kuriyama, 1969).
The concentrations of the chemical agents used will be described elsewhere. The microelectrodes used were filled with 3 M-KCI or 2 M-K-citrate: in particular, mainly 2 M-K-citrate electrodes were used for determination of the reversal potentials. The experiments were carried out during the winter season (October-March), thus input resistance of the membrane showed higher values than those observed during the summer season (Hidaka, Ito & Kuriyama, 1969).
RESULTS
The miniature inhibitory potentials and the inhibitory potentials
During the resting state of the membrane, as well as in the active state of the membrane, small hyperpolarizing potentials, generated spontaneously, could be recorded. The amplitudes of these potential changes varied from the noise level (below 0·1 mV.) to 11 mV. Figure 1 shows the histograms of the amplitudes and intervals of the spontaneously generated small hyperpolarizations recorded from a single cell membrane in the resting state. The mean amplitude in this particular cell was 0·71 mV. (n = 855) and the mean interval was 101 msec, (n = 800). The histograms of both the amplitudes and the intervals were skewed, indicating random generation of the small hyperpolarizing potentials from the focus. The number of generations occurring in 500 msec, was measured many times, and the number of times that each number of generations occurred was plotted. The result was a skew curve which fits well with the theoretical Poisson distribution.
The maximum rates of rise of the small hyperpolarizing potentials varied from 0·3 to 1·1 V./sec., having been measured from the potential changes which exceeded 5 mV. (mean 0·52 V./sec. S.D. = ± 0·006, n = 32).
Figure 2 shows the effect of picrotoxin (5 × 10−6 g./ml.) on the generation of the small hyperpolarizing potentials. Generation of the small hyperpolarizing potentials ceased completely after 20 min. in the perfusing solution. This observation, and others described later, suggests that these small hyperpolarizations might be M.I.J.P.’S generated by a chemical transmitter released from the inhibitory nerve terminals.
In fact, peripheral nerve stimulation generated an I.J.P. and transiently blocked the spontaneous discharges. Figure 3 shows the effects of peripheral nerve stimulation on the spontaneous spike discharges. In the present experiment dissection of the small peripheral nerve was difficult. Therefore, one stimulating electrode (an Ag-AgCl electrode of 0 × ·1 mm. diameter) was placed on the peripheral nerve and another indifferent electrode was placed about 3 cm. away from the tissue. Short pulses (0·5 msec.) blocked the spontaneous discharges (a, b), and even when the membrane was quiescent a hyperpolarization could be recorded (c). The hyperpolarization lasted 1·5−3 sec. and the maximum hyperpolarization was 14 mV. at a resting potential of — 37 mV. This I.J.P. sometimes blocked only the spontaneous discharges without changing the membrane potential. The time course of the I.J.P. was not always smooth and the hyperpolarization often had an irregular time course. Figure 4 shows the two different types of I.J.P. : (a) and (b) show smooth transitions of the I.J.P. in response to peripheral nerve stimulation, and (c) and (d) show irregular transitions of the I.J.P.’S.
The latter resembled the end-plate potential in the frog skeletal muscle in low-calcium and high-magnesium solution, but the polarity of the potential was reversed. These irregular hyperpolarizations might indicate the sum of the quantal release of the chemical transmitter from the nerve terminals. Furthermore, an off response of the membrane could be observed (e, f) Figure 5 shows the effects of the I.J.P. elicited by external stimulation during the application of the depolarizing current to the muscle membrane through the recording electrode. Depolarizing current elicited the spikes (a), and the external stimulation produced the I.J.P. (b); when the external stimulation was applied during the depolarizing current preceding the onset of the spike the inhibitory potential prevented the generation of the spike (c). The delays in generation of the I.J.P. after nerve stimulation varied from 20 to 90 msec. These variations in the latency might be mainly due to the slow conduction velocity of the nerve, and also to the complicated peripheral arrangement of the nerve terminals since the stimulating electrode and the recording electrode were placed approximately 2 mm. apart in all experiments.
Reversal potential
When the membrane was hyperpolarized in stages by intracellular polarization using the Wheatstone bridge method, the M.I.J.P.’S were proportionally reduced in amplitude. Stronger hyperpolarization applied to the membrane reversed the polarity of the M.I.J.P.’S. Figure 6 shows an example of the reversal of the M.I.J.P.’S. In this particular cell, in which the membrane potential was − 36 mV. (b), when the membrane was hyperpolarized to − 58 mV. the M.I.J.P.’S nearly ceased and often small depolarizations appeared (c). At a membrane potential of −92 mV., reversed polarity of the M.I.J.P.’S with high amplitude could be observed (d). In contrast, when the membrane was depolarized to + 20 mV. by a depolarizing current increased gradually to avoid spike generation, the amplitudes of the M.I.J.P.’S were enhanced. This result suggests that the equilibrium potential of this M.I.J.P. was located at nearly —60 mV.
Reversal of polarity of the I.J.P.’S evoked by the field stimulation could also be observed. When the muscle membrane was hyperpolarized stepwise by the hyperpolarizing currents the amplitude of the I.J.P. was gradually lowered, and when the membrane was hyperpolarized to — 55 mV. (n = 8) the I.J.P.’S reversed their polarity. Figure 7 shows the relationship between the amplitude of the I.J.P. and the membrane potential. The vertical line shows the membrane potential and the horizontal line shows the amplitude of the I.J.P.’S. When the membrane was polarized from —35 to — 53 mV., the polarity of the I.J.P.’S were reversed (two experiments). The right-hand side of Fig. 7 shows the actual records of the changes of the amplitude of I.J.P. and membrane potential by intracellular hyperpolarizing currents.
To investigate the nature of the M.I.J.P. the external chloride concentration, [Cl]0, was replaced with D-glutamate, an anion to which the membrane is impermeable. When the [Cl]0, normally 148·3 mM, was reduced to 5·6 mM, the polarity of the M.I.J.P.’S was reversed. Figure 8 shows the effect of chloride-deficient solution (5·6 mM) on the M.I.J.P.’S. After 15 min. exposure to chloride-deficient solution the M.I.J.P.’S reversed their polarities completely; after washing with the normal Ringer’s solution their polarities changed back again to their direction before treatment.
As observed in the previous paper (Hidaka, Ito, Kuriyama & Tashiro, 1969), changes in [Cl]0 changed the membrane potential only slightly; but this experiment showed that reversal of the M.I.J.P.’S occurred. Therefore, the chloride equilibrium potential (EC1) differed from the membrane potential level.
It has also been observed that the excitatory chemical transmitter in crustacean muscle was L-glutamate (Robbins, 1959; van Harreveld, 1959; van Harreveld & Mendelson, 1959; Takeuchi & Takeuchi 1964); and therefore, to avoid the doubt raised from the use of L-glutamate in substitution for chloride, the effect of various concentrations of D-glutamate on the M.I.J.P.’S was observed. When [Cl]0 was reduced to 41·4 mM (106·9 mM D-glutamate) the M.I.J.P.’S showed a polarity similar to that of the control. In [Cl]0 of 17·7 mM (130·6 mM D-glutamate) the M.I.J.P.’S nearly disappeared, and in 5·6 mM (142·5 mM D-glutamate), as described previously, the polarity of the M.I.J.P.’S was completely reversed. This fact might indicate that the reversal of M.I.J.P.’S in the glutamate substituted solution was due not to the D-glutamate itself but to a change in the EC1. Furthermore, the generation of the reversed M.I.J.P.’S in the D-glutamate substituted solution (5·6 mM-Cl still remaining) was blocked by treatment with 10−5 g./ml. picrotoxin. Furthermore, substitution of L-asparate and by sulphate had similar effects on the M.I.J.P.’S to those observed in the chloride-deficient D-glutamate solution.
The effects of GABA and picrotoxin on the postsynaptic membrane
It has already been established that in Crustacea and some kinds of insects the inhibitory neuromuscular junction potential is due to the release of γ-aminobutyric acid (GABA) from the inhibitory nerve terminals (Hoyle, 1955; Hoyle & Wiersma, 1958; Cerf et al. 1959; Boistel & Fatt, 1958;,Kuffler & Edwards, 1958; Krawitz, Kuffler & Potter, 1963; Usherwood & Grundfest, 1965; Takeuchi & Takeuchi, 1965, 1966). The present experiment also investigated the effect of GABA and picrotoxin on the membrane conductance of longitudinal muscle in the same way as Usherwood & Grundfest (1965) studied the inhibitory innervation in insect muscle. When the longitudinal muscle was bathed in a potassium-free solution, the membrane was hyperpolarized from − 34·1 mV. (s.D. = ± 1·9, n = 25) to-42·8 mV. (S.D. = +2·5,71 = 25) and the effective resistance of the membrane was increased from 99 MΩ (n = 4) to 186 MΩ (n = 8). When treated with 10−5 g./ml. GABA in the potassium-free solution, the membrane was further hyperpolarized to −46·5 mV. (S.D. = ± 1·8, n = 25), and the input resistance was reduced to 138 MΩ (n = 8). The current-voltage curves in potassium-free solution and in the potassium-free GABA solution crossed at about − 54 mV. When the membrane was depolarized by five times the normal [K]o (13·5 mM) the membrane potential decreased from —36·0 mV. (S.D. = ±2·1, n = 25) to —28·6 mV. (S.D. = ±0·95, n = 25). On application of GABA (10−5 g./ml.) in the above solution the membrane hyperpolarized to —33·4 mV. (S.D. = ±2·8, n = 25). In the solutions containing excess potassium the effective resistance of the membrane was decreased from 99 MΩ (n = 4) to 83 MΩ (M = 5) and when 10−6 g./ml. GABA was added to the above solution the membrane input resistance was reduced to 54·1 Mil (n = 7). The current-voltage curves made before and during the application of GABA crossed at about − 53 mV. The discrepancies observed in the crossing potentials between the potassium-free and potassium-excess solutions might be expected from the fact that an increase in the external KC1 will lead to an increased level of KC1 inside the cell. Therefore the slight difference in the potential of the crossing seems insignificant. Figure 9 shows the current-voltage relations of the postsynaptic membrane in potassium-free and a high-potassium solution (13 · 5 mM) with or without presence of GABA (10−5 g./ml.). The current-voltage relations are drawn from the mean values of several different cells, since it was difficult to insert the microelectrode in the same cell before and during treatment with GABA. In this figure the cross-over point of the current-voltage relation in the 5 times normal (K)o and 5 times normal (K)o with GABA (10−8 g./ml.) was − 56 mV., and in potassium-free and in potassium-free with GABA was − 55 mV.
Picrotoxin (10−5 g./ml.) itself increased the input resistance of the membrane and restored the reduced membrane resistance produced by pre-treatment with GABA (10−5 g./ml.) to the value observed before treatment with GABA. However, picro-toxin (10−5 g./ml.) did not affect the spontaneous spike discharges of the membrane or the spikes elicited by intracellular depolarizing currents and only blocked the generation of the M.I.J.P.’S and the I.J.P.’S elicited by stimulation of peripheral nerves.
Iontophoretic application of GABA
To investigate the nature of the chemical transmitter at the neuromuscular junction GABA was applied by iontophoresis manner. A glass microcapillary of tip diameter 2 − 3 μ was filled with GABA at an approximate concentration of 1 M. The recording electrode was inserted into the fibre and the GABA-filled capillary was placed on the surface of a muscle fibre about 25 μ distance from the recording electrode. The duration of stimulation was always 500 msec, and only the intensity of the stimulus was varied.
Figure 10 shows the hyperpolarization of the membrane produced by the ionto-phoretic application of GABA. The latencies of the hyperpolarization after the applied currents varied from 5 to 15 sec., and the amplitude of the hyperpolarization also varied from 2 to 10 mV. at a current intensity of 5 × 10−7 A. The latency preceding hyperpolarization was very long compared with that observed in crustacean muscle (Takeuchi & Takeuchi, 1965). Histological study shows that the longitudinal muscle layer is covered by relatively thick amorphous tissue, and hence GABA may be expected to take some time to reach the cell surface. As shown in Fig. 10 b the amplitude and frequency of the M.I.J.P.’S were often increased before and during the course of hyperpolarization. However, these phenomena were not observed when an electrical current of the opposite polarity was passed.
When repetitive short polarizations were applied intracellularly, to the cell membrane, for a duration of 100 msec, each and with a frequency of 5 c./sec., changes in the membrane conductance during the course of GABA-potential could be observed (Fig. 10c). The most significant reduction in membrane resistance was observed during the peak of the hyperpolarization and the gradual increase of the membrane resistance to the initial value was accompanied by the repolarization phase of the hyperpolarization.
The reversal of the GABA-potential could be recorded by the application of conditioning hyperpolarization. When the conditioning hyperpolarization reached — 62 mV., no potential change was observed. Further strong hyperpolarization of the membrane reversed the polarity of the GABA-potential. Furthermore, the M.I.J.P.’S generated spontaneously during the generation of the GABA-potential reversed their polarity at the same membrane potential for the GABA-potential.
Excitatory neuromuscular transmission
The M.E.J.P.’S could be recorded from the longitudinal muscle, but very rarely. Figure 11 shows the appearances of the M.E.J.P.’S. The duration of the M.E.J.P. was shorter than that of the M.I.J.P., and often single M.E.J.P. could trigger the spike.
On very rare occasions both of the M.I.J.P.’S and M.E.J.P.’S could be recorded from the same cell. The small transient hyperpolarization and depolarization appeared without any synchronization. When the membrane was conditionally depolarized by the intracellularly applied outward current, the amplitudes of the M.I.J.P.’S were enhanced but the amplitudes of the M.E.J.P.’S were reduced. When the membrane was depolarized to o mV. or more no transient depolarization of the membrane could be observed. All the junction potentials showed hyperpolarization, i.e. reversal potential of the M.E.J.P.’S showed a distinct difference from that of the M.I.J.P.’S. The measured reversal potential for the M.E.J.P.’S ranged between —10 mV. and o mV. The M.E.J.P.’S were completely blocked by treatment with D-tubocurarine (10−6 g./ml.).
The E.J.P.’S could also be recorded by external stimulation of the peripheral nerve. To prevent the.generation of the I.J.P. the preparation was pretreated with picrotoxin (5 × 10−4 g./ml.). Two different types of the E.J.P. could be observed, i.e. irregular transition and the smooth transition of the E.J.P. The former resembled that observed in the neuromuscular transmission of frog muscle under treatment with calcium-deficient and excess magnesium solutions, and was presumably due to irregular quantal release of the excitatory chemical transmitter.
Figure 12 shows the response of the membrane elicited by weak external stimulation. The E.J.P. with irregular transitions could be elicited by external stimulation. When the E.J.P.’S reached the critical membrane potential, the spike appeared on the E.J.P. The E.J.P. showed like the sum of the M.EJ.P.’S as described previously.
Not only the M.E.J.P. but also the E.J.P. was abolished by treatment with D-tubocurarine but not with atropine.
In contrast with the action of D-tubocurarine, prostigmine (10−8 g./ml.) enhanced the amplitude and prolonged the duration of the E.J.P. elicited by external stimulation. When the amplitude of the E.J.P. exceeded the critical firing level of the membrane, the spike was elicited.
The reversal potential of the E.J.P. was determined. It was difficult to determine the reversal potential by the application of the depolarizing current to the membrane because the E.J.P. elicited the spike even with weak depolarization of the membrane. The reversal potential was therefore estimated from the extrapolated line of the amplitude changes of the E.J.P. caused by the conditioning hyperpolarization of the membrane. Figure 13 shows the changes in amplitudes of the E.J.P.’S under the various membrane potential levels fixed by the intracellularly applied inward currents.
The reversal potential was about 0 ≅ −2 mV. (S.D. = ± 3-·1, n = 5). This value was in agreement with that measured from the M.E.J.P.
Whether or not an excitatory nerve fibre diffusely innervates a single muscle fibre has not yet been investigated. If the innervation is multiple, the reversal potential might be more negative than o mV., as determined in the tonic muscle fibre of the frog by Burke & Ginsborg (1956). When the reversal potential was measured in sodium-free solution, it was shifted to − 20 mV. (S.D. = ± 4·0, n = 11) from o mV. in the normal solution as described previously. This indicates that acetylcholine as a chemical transmitter presumably increased GNa and GK but not GC1 because the reversal potential was not influenced by treatment with picrotoxin which was thought to inhibit the chloride permeability.
DISCUSSION
Inhibitory neuromuscular transmission
The present results might indicate that the M.I.J.P.’S and the I.J.P.’S recorded from the longitudinal muscle of the earthworm are due to release of a chemical transmitter from the peripheral nerve terminals, and that the chemical transmitter is γ-amino-butyric acid. The criteria for recognizing chemical synapses and transmitters can be applied to the present results obtained by electrophysiological method, as follows.
(i) The M.I.J.P.’S reversed their polarities under a conditioning hyperpolarization of − 60 mV.
(ii) In potassium-free solution the current-voltage curves made before and during application of GABA crossed at about − 54 mV. and in excess-potassium solution at about − 56 mV.
(iii) The I.J.P.’S evoked by external stimulation of the peripheral nerve reversed their polarities when the membrane was hyperpolarized to −55 mV.
(iv) The hyperpolarizations of the membrane produced by iontophoretically applied GABA reversed their polarities at a conditioning hyperpolarization of about − 60 mV.
The above four observations show the existence of the equilibrium potential, and the range might be from − 54 to − 60 mV. However, the difficulty found in comparing the reversal potential for the GABA-potential and the I.J.P.’S arises from the fact that the longitudinal muscle might have a widely distributed innervation. Also, when the polarizing current is applied at a point on the muscle fibre, the measured reversal potential of the I.J.P. and M.I.J.P. may therefore slightly shift from the exact level of the equilibrium potential (Fatt & Katz, 1953; Burke & Ginsborg, 1956). The important finding was that the M.I.J.P. and the GABA-potential reversed their polarities at the same membrane potential, i.e. the ionic mechanism involved in the above phenomena might be the same. Furthermore:
(v) Picrotoxin did not block the spontaneous spike discharges of the membrane but blocked the M.I.J.P.’S and I.J.P.’S elicited by stimulation of the peripheral nerve.
(vi) Picrotoxin restored to its initial value the membrane resistance which was reduced by treatment with GABA
From the above results it might be possible to conclude that the neuromuscular inhibitory transmission is due to release of chemical transmitter and that this transmitter is GABA.
The inhibitory effect upon the longitudinal muscle produced by GABA might be explained by the increased chloride conductance, as follows.
(viii) The M.I.J.P.’S and I.J.P.’S elicited by stimulation of the inhibitory nerve reversed their polarities in a chloride-deficient solution.
Reversal potential for the I.J.P. remained the same in potassium-free and excess potassium solutions as described in (ii).
(ix) The membrane conductance was increased during the course of the hyperpolarization produced by iontophoretically applied GABA.
(x) In chloride-deficient solutions the I.J.P. reversed its polarity and the I.J.P. could elicit the spike. The reversed I.J.P. was blocked by treatment with picrotoxin, i. e. the polarity of the I.J.P. is determined by the level of the chloride equilibrium potential(EC1)
The histogram of the intervals of the M.I.J.P.’S indicates a random release of chemical transmitter from the nerve terminals, and the histogram of the amplitude of the M.I.J.P.’S shows a skew curve as observed in diffusely innervated muscle (cf. Martin, 1966). The duration of the M.I.J.P.’S was much longer than the value expected from the passive decay of electrotonic potential. This tendency was also observed in the I.J.P. recorded from crustacean muscle (Fatt & Katz, 1953; Takeuchi & Takeuchi, 1965). They reported that the time course of the I.J.P.’S was much slower than that of the E.J.P.’S and also that the time course of the GABA-potential was much slower than that of the glutamate-induced potential. Our observations also confirmed the above observations. However, it is unlikely that glutamate is an excitatory chemical transmitter in the earthworm.
Excitatory neuromuscular transmission
From the present results it might be concluded that the M.I.J.P.’S and EJ.P.’S elicited by stimulation of the peripheral nerve are due to release of a chemical transmitter, presumably acetylcholine, from the nerve terminals. The reasons are as follows: (i) The M.E.J.P.’S and E.J.P.’S reversed their polarities when the membrane was depolarized to about o mV. (ii) The M.EJ.P.’S and E.J.P.’S were blocked by treatments with D-tubocurarine. (iii) Atropine had no effect on either of these potential changes, (iv) Prostigmine enhanced the duration and the amplitude of these potential changes.
From the existence of equilibrium potentials relating to the M.EJ.P.’S and E.J.P.’S it might be possible to assume that the excitatory chemical transmitter increased the permeability of the membrane to sodium and to potassium, since the reduction of [Na]0 shifted the equilibrium potential to a more negative value than that recorded in normal solution.