ACTIVE
PHASE OF FROG’S END-PLATE
POTENTIAL
AKIRA
TAKEUCHIl
AND NORIKO
TAKEUCHF
Department of Physiology, School of Medicine, Juntendo
University, Hongo, Tokyo, Japan
for publication
July 30, 1958)
IT HAS BEEN CONSIDERED that the end-plate potential (e.p.p.) is generated
by the brief ionic flux across the end-plate membrane and the later slowly
declining phase of the e.p.p. is due to the dissipation of the charge along and
across the muscle membrane.
This consideration
was supported by some
authors. Kuffler (21) observed with a single nerve-muscle
preparation
that
the later slowly decaying part of the e.p.p. was destroyed by a propagated
muscle impulse and obtained a duration of transmitter
action (3-4 msec. at
20°C.) by observing the size of the e.p.p. that was built after the invasion of
a propagated muscle impulse. Katz (18) demonstrated
that the neuromuscular transmitter
produced a brief phase of impedance loss at the end-plate
region. Recently Fatt and Katz (10) observed by measuring the displacement of the total charge along and across the muscle membrane during the
e.p.p. that the active depolarization
process at the end-plate had ceased
within 2 msec. On the other hand the time course of the actively depolarizing
phase of the e.p.p. was estimated by an analysis of the time course of the
e.p.p., it being assumed that the exponentially
decaying phase was attributable to the passive repolarization
of the muscle membrane (7,19).
The purpose of the present experiment
was to determine directly the
time course of the active phase of the e.p.p. by using the voltage clamp
method which was originally described by Hodgkin et al. (14) and was also
applied to the squid giant synapse by Tasaki and Hagiwara (29). When the
membrane potential
is clamped at the resting membrane
potential
with
negative feed-back during the neuromuscular
transmission,
the electrotonic
spread of the charge along the muscle fibre membrane can be eliminated.
The feed-back current which flows through the muscle membrane to hold
the membrane potential
at the resting value is due to the brief electric
change at the end-plate, i.e., it will show the active phase of the e.p.p. To
simplify the expression, the feed-back current during neuromuscular
transmission will be called provisionally
the “end-plate
current.” A preliminary
report of the present experiment appeared in 1958 (27).
Materials and solutions
M. sartorius with sciatic nerve was dissected from well-fed winter frogs of species
Rana nigromacuZata. The neuromuscular
transmission
was usually blocked by adding d1 Present address: Department
Utah, Salt Lake City 12, Utah.
of Physiology,
College
of Medicine,
University
of
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 17, 2017
(Recbived
AKIRA
TAKEUCHI
AND
NORIKO
TAKEUCHI
FIG. 1. Schematic
diagram of experimental
arrangement.
current recording
(0.5-5 MO). S: switch.
R: resistor
for
feed-back amplifier was a balanced D.C. ampl%er of three stages and its output was fed to
the current electrode through a cathode-follower
stage in such a way that negative feedback was employed. The voltage gain of the feed-back amplifier was about 4,000, including
a preamplXer
and an output cathode-follower
stage. When the electrode of 30 Ms~ was used
the rise time was 200 psec. The amplifier used for current recording was a balanced D.C.
amplifier of two stages, its frequency characteristics
being flat up to 20 kc. The circuit of
preampmer
for current recording is presented in Fig. 1. When the current electrode of 10
MO was used, the internal resistance of the feed-back circuit was calculated as about 2,500
n and this value was sufficient for the present purpose, because the resistance at the endplate membrane was usually larger than 100 KQ.
Procedure
The sartorius muscle was mounted in the Ringer bath and the sciatic nerve was stimulated with silver electrodes in the second compartment
which was Wed with param
oil.
Usually 10-20 Ma KCl-filled
intracellular
microelectrode
(22, 25) was used for recording
while the low-resistance electrode of about 5-10 MO was used as the current electrode. Before touching the current electrode to the Ringer bath the square voltage was fed to both
grids of the current amplifier in phase and it was confirmed that no potential change appeared in the output of the current ampler.
Then the current electrode was touched to
the Ringer bath and the resistance and the stray capacity of the current electrode was
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 17, 2017
tubocurarine
chloride
(3-5 X10-6 g. /ml.) to the Ringer’s solution. The composition
of
Ringer’s solution used in the present investigation
was as follows: Na+, 111.5; K+, 2.0;
Ca++, 2.0; Cl-, 117.5 mM. and phosphate buffer (1.08 mM.-NazHPO,;
0.42 mM.-NaHr
POJ was added. In some cases the calcium concentration
in Ringer’s solution was increased,
replacing the NaCl, in order to increase the e.p.p. size and also to reduce the variation of
the amplitude of the e.p.p. (cf. 5, ll), and in this case phosphate buffer was omitted. Most
experiments were conducted at room temperature
(14’-19°C.)
in winter. In order to change
the temperature
of Ringer bath, the warm or cold water was circulated through the outer
jacket of Ringer bath and the temperature
of the Ringer bath was measured by a thermisterthermometer.
Voltage clamp method
Voltage clamp method was in principle the same as those of other authors (14,30,32).
In the present investigation,
however, the electrode resistance was high, and a small change
in the arrangement
was made. Recently a similar method was used by Hagiwara and&it0
(13). The schematic diagram of the experimental
arrangement
is presented in Fig. 1. The
ACTIVE
PHASE
OF E.P.P.
397
Sources of errors
LocuZizcztion
of etid-plate.
In the present
method,
differing
from
the case in which
voltage
clamp
method
was applied
to the giant
axon,
the longitudinal
resistance
of the
muscle
fibre was not neglected;
therefore
it was necessary
to locate
exactly
the end-plate
focus in order
to minimize
the potential
caused
by the longitudinal
current
flowing
through
the resistance
between
the tip of the electrode
and the end-plate.
The recording
electrode
was moved
along
the muscle
fibre,
and the e.p.p.
of the steepest
rising
phase
as well as
the largest
amplitude
was chosen.
The current
electrode
was then inserted
usually
within
50 p of the recording
electrode.
If the rising
phase of the e.p.p.
was rather
slow everywhere
and the focus of the end-plate
could not be determined
exactly,
the end-plate
was discarded.
In order
to check
the relation
between
the end-plate
current
and the position
of the electrode,
with
the recording
electrode
inserted
at an end-plate
focus,
the current
electrode
was moved
along the muscle
fibre.
As the distance
between
both electrodes
was increased,
the current
which
flowed
during
the neuromuscular
transmission
and the potential
recorded
at the end-plate
focus
became
oscillatory.
Although
this point
was not studied
in detail,
this may be due to the time delay
of the electrotonic
spread
of feed-back
current
to the
position
of the recording
electrode.
When
the current
electrode
and the recording
electrode
were moved
together
along the muscle
fibre from
the end-plate
focus,
keeping
the distance
of electrodes
within
50 p, the time course
of the end-plate
current
became
slower
and its
amplitude
became
smaller.
In order
to test whether
the contribution
of the longitudinal
resistance
between
the end-plate
focus
and the electrodes
was negligible,
recording
and
current
electrodes
were located
accurately
at the end-plate
focus;
then a second
recording
electrode
was inserted
at various
distances
and the membrane
potential
change
was recorded
through
another
D.C.
amplifier
during
neuromuscular
transmission.
When
the
negative
feed-back
was applied,
the potential
change
at the second
recording
electrode
I.
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 17, 2017
measured
by use of the square
voltage
(see below).
In some cases the resistor
for current
recording
R in Fig. 1 was shunted
by a variable
capacitor.
With
this procedure
the capacitative escape
of the feed-back
current
through
the stray
capacity
could
be cancelled,
because
if the time constant
of recording
resistor
and its shunting
capacity
is the same with the time
constant
of the current
electrode
and the stray
capacity,
the current
which
flows
through
the recording
resistor
is solely
the current
through
the resistance
of electrode
and the capacitative
component
is corrected.
The intracellular
microelectrode
was inserted
into an end-plate
focus under
a dissecting microscope
(60 x). The
focus
of end-plate
was decided
by the following
criteria:
(i)
the place at which
the fine nerve
twigs
disappeared
(24),
(ii) the place at which
the e.p.p.
of the shortest
rising
phase
was found
and (iii) the largest
amplitude
was obtained
(10).
Then
the current
electrode
was inserted
into the same muscle
fibre close to the recording
electrode,
usually
within
50 p. With
this method,
however,
the normal
end-plate
could
not
be located
because
of the appearance
of the muscle
action
potential.
The normal
end-plate
was located
by recording
the spontaneous
miniature
e.p.p.
of the steepest
time course,
although
with this procedure
the localization
of the end-plate
was not as correct
as the method
described
above,
because
the amplitude
of the miniature
potential
was too small to determine its time course
exactly,
and also the time course
of each miniature
potential
at the
same point
was somewhat
different.
In some cases the muscle~was
curarized
at the beginning of the experiment
and the end-plate
was located
in the manner
described
above
and
then
the muscle
was washed
with
normal
Ringer’s
solution.
The
feed-back
circuit
was
closed
by switch
S and the feed-back
ampl3erwas
balanced
so that
no current
flowed
through
the current
electrode.
Then
the membrane
potential
and the feed-back
current
during
the neuromuscular
transmission
was recorded
by a dual-beam
oscilloscope.
The
resting
membrane
potential
was measured
during
the experiment
with
another
D.C.
amplifier
(not shown
in Fig. 1).
In order
to change
the potential
at which
the membrane
potential
was to be clamped,
the square
voltage
or the constant
voltage
was fed to the input
of the feed-back
amplifier,
and the end-plate
currents
at various
membrane
potentials
were recorded.
After
recording
the end-plate
currents
at various
membrane
potentials,
the feed-back
circuit
was opened
and the polarizing
current
was fed into the current
electrode
through
100 Ma and the
e.p.p.‘s
at various
membrane
potentials
were recorded
from
the same end-plate.
Further,
the square
pulse
of the constant
current
was fed into the current
electrode
and the electrical
characteristics
of the muscle
membrane
were measured.
398
AKIRA
TAKEUCHI
AND
NORIKO
TAKEUCHI
RESULTS
Active phase in curarized preparation
The active phase was investigated
with a completely curarized muscle.
The end-plate focus was located (see Method) and the feed-back circuit was
closed. The current which flows through the current electrode to hold the
muscle membrane potential at the resting value and the membrane potential
of the end-plate during the neuromuscular
transmission were recorded with
the dual-beam oscilloscope. The inward current through the membrane is
shown as an upward deflexion as in Fig. 2. Record A shows the e.p.p. without feed-back and record B shows the end-plate current and the clamped
membrane notential obtained from the same end-nlate. After nerve stimulation the current began to flow inward through the end-plate membrane and
rose rapidly to a peak in 0.77 msec. and then fell approximately
exponentially, the peak to half decline time being 1.08 msec. (the mean values of 32
experiments),
and total duration was about 4-5 msec. at 17”C., while the
membrane potential remained approximately
constant.
If the current which is similar to the end-plate current is assumed to
flow through the muscle fibre membrane,
the e.p.p. can be reconstructed
from the observed end-plate current by a simple numerical analysis using
Hodgkin and Rushton’s
(16) equation. The superimposed
tracings of the
end-plate current and the e.p.p. recorded from the same end-plate
are
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 17, 2017
was suppressed
to a similar
degree
as at the first recording
electrode.
Therefore
it is considered
that if the electrodes
are located
correctly
at the end-plate
focus,
the contribution
of the longitudinal
resistance
will be negligible
and the membrane
potential
of the muscle
fibre can be held at about
the same potential
uniformly
along
the muscle
fibre.
2. Cupacitatizw escape of feed-back current. It must be considered
that there is an escape
of feed-back
current
through
the capacity
of the wall of the current
electrode.
This
may
tend to make the time course
of the end-plate
current
steeper
and to make
the peak amplitude larger.
If the capacity
of the electrode
is assumed
to be 3 pF and the resistance
of the
electrode
to be 10 Ms~, the peak value
of recorded
end-plate
current
can be calculated
to
be 6 per cent larger
than
the true end-plate
current.
Before
the experiment
was started,
the capacitative
escape
of the current
was measured
by applying
the
square
voltage
through
the current
electrode;
the tip of the electrode
was in the Ringer
bath,
and the
capacitative
escape
which
produced
the surge
of the current
appeared
at the beginning
and at the end of the square
current.
In order
to minimize
the capacitative
escape,
(i) lowresistance
electrodes
of 5-10
Ms~ were chosen,
(ii) the current
electrode
was covered
with
the cathode
of cathode-follower
stage
(see Fig.
l), (iii) Ringer’s
fluid
above
the muscle
fibre was not more
than about
1 mm.,
and (iv) in some cases the capacitative
escape
was
compensated
as described
above.
With
these procedures
the surges
of the capacity
current
can be made
negligible.
3. Series
resistance.
The membrane
resistance
at the end-plate
is in general
greater
than 100 KQ. Therefore
in comparison
the resistance
which
is in series with the membrane
resistance
and is mainly
composed
of the resistance
of Ringer’s
fluid or connective
tissue
will be negligibly
small.
The influence
of the polarization
and of the resistance
of the
external
electrode
in the Ringer
bath
was avoided
by using
two external
electrodes
(see
Fig. 1).
4. l!htrinsic
potential. As reported
by Fatt
and Katz
(lo),
the e.p.p.‘s
of neighbouring
fibres
had an influence
on intracellularly
recorded
e.p.p.‘s
and this influence
was greater
when
there
was little
Ringer’s
fluid
above
the preparation.
The depth
of Ringer’s
fluid
above
the muscle
surface
was about
1 mm.,
and in this condition
the influence
of extrinsic
potential
was in general
about
1 per cent.
-
ACTIVE
PHASE
OF E.P.P.
presented in Fig. 2C.. The- potential change was calculated from the endplate current, assuming the time constant of the muscle membrane as 25
msec. and the effective resistance 320 KQ (circles). The size of the e.p.p. was
generally small compared with the resting potential and the time course of
the end-plate current was rapid; therefore it may be considered that the
total charge transported
by the e.p.p. has almost the same value as that
transported
by the end-plate current. This is calculated with the area under
the end-plate current and is in the order of 3 X lo-lo coulombs in the e.p.p. of
10 mV. This corresponds to a net transport
of 3 X10-15 mol of univalent
cations inward or anions outward.
The relationship
between the amplitude of the e.p.p. and the current size
was investigated by altering the e.p.p. size with a change in the concentration of d-tubocurarine
(Fig. 3). The relation obtained from five fibres are
presented in circles in Fig. 4. These points were expected to have a different
gradient but, contrary to this expectation,
most of the points were on approximately
the same curve. Although it is not definite whether this relation
is confirmable in general, the result shows that the muscle fibres on the surface of the sartorius muscle have similar membrane characteristics.
Martin
(23) investigated
the fluctuation
of the e.p.p. size and observed that when
the e.p.p. amplitude
was greater than about 5 mV., the fluctuation
was
smaller than predicted by a simple statistical theory and this discrepancy
was removed when allowance was made for the non-linear relation between
conductance and potential changes at the end-plate membrane. According
to this result the curve presented in Fig. 4 is expected to be concave upwards
because the rate of increase in the e.p.p. size would tend to decrease as the
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 17, 2017
FIG.
2. A: e.p.p. recorded
intracellularly
without
feed-back.
B: end-plate
current
recorded
from
same
end-plate
with
feed-back.
Lower
beam
shows
membrane
potential
recorded
simultaneously
with
end-plate
current.
Voltage
scale:
5 mV.
Current
scale:
1 X10-’
A. Temperature
at
17°C.
C: superimposed
tracings
of e.p.p.
and end-plate
current
recorded
from
same
end-plate.
Circles
indicate
potential
change
calculated
from
end-plate
current,
assuming membrane
time constant
to be 25 msec.
and effective
resistance
320 KS& peak
amplitude
of e.p.p.
being
8.9 mV. and that
of
end-plate
current
1.4 XIOeT
A. Time
in
msec.
400
AKIRA
TAKEUCHI
AND
NORIKO
TAKEUCHI
shunting conductance at the end-plate increases. Although the lower curve
in Fig. 4 tends to be slightly concave, this is much less than that expected
from Martin’s results and an almost linear relation was obtained until the
amplitude
of the e.p.p. reached 10 mV. This discrepancy may be explained
as follows: the time course of the active phase is much more rapid than the
time course of the e.p.p. and at the moment when the active phase reached
the peak value, the e.p.p. had not yet reached its peak. Therefore the charge
displaced during the neuromuscular
transmission
may be little influenced
by the membrane potential change. In order to investigate this point further, the e.p.p. size-current size relation was measured with the eserinized
end-plate in sodium-deficient
Ringer’s solution. Since in this case the e.p.p.
FIG. 4. Lower curve (circles): relationship between amplitude
of e.p.p. and endplate current obtained from five curarized
end-plates.
Upper curve (triangles):
relationship obtained from eserinized preparation in sodium-deficient
Ringer’s solution.
Ordinate:
amplitude
of end-plate
current,
Abscissa: amplitude
of e.p.p.
u
.z :!
+
E
Z
2
’ -
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 17, 2017
FIG. 3. Change in amplitude
of e.p.p.
and end-plate current recorded from same
end-plate
in various concentrations
of dtubocurarine
at 18°C. Left: e.p.p.‘s. Right:
end-plate
currents. Voltage scale: 2 mV.
Current
scale: 5 X10-8 A. Time scale: 2
msec.
ACTIVE
PHASE
OF E.P.P.
401
size could not be altered by changing d-tubocurarine
concentration,
the endplates which had different e.p.p. sizes were chosen. In this preparation
the
time course of the active phase was prolonged and the difference in the time
course of the active phase and that of the e.p.p. was small. Then it is expected that the membrane potential change influences the displacement
of
the charge, and the current size-e.p.p. size relation will show a non-linearity.
The upper curve in Fig. 4 shows an example obtained from the eserinized
preparation
and this agrees with the above hypothesis.
It is well known that when double stimuli are applied to the nerve, two
e.p.p.‘s summate and in the frog the e.p.p. produced by the second nerve
volley is greater than the first, although the time course is unchanged (2,
7). The e.p.p.‘s and the end-plate currents which are produced by two nerve
volleys at 20°C. and 9°C. are presented in Fig. 5. The end-plate current
which was produced by the second nerve volley was consistently
greater
than the conditioning
one and its time course was unchanged. Although in
one case out of 14 end-plates examined, the second end-plate current was
somewhat smaller than the conditioning
one, but in this case when the
stimulus strength was changed, the end-plate current showed two different
amplitudes;
this end-plate might be supplied by two motor fibres. It is
worthy of note that when the interval separating two volleys was made
shorter, the end-plate current appeared to be delayed and the main part of
the end-plate currents did not summate. This is clearly shown in Fig. 5D
and E. It is known that if the shock interval is short the conduction velocity
of the second nerve impulse decreases (12, 28). Then the fact that the end-
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 17, 2017
FIG. 5. Superimposed
e.p.p.‘s
(A, C) and end-plate
currents
(B, D, E) set up by double
nerve
volleys,
recorded
from
same end-plate
at 20°C.
(A, B) and at 9°C.
(C-E).
Voltage
scale:
2 msec. In D five test stimuli
are applied
scale: 5 mV. Current
scale: 5 x 1Oe8 A. Time
and four responses
appear
and in E two responses
are observed.
Note
delay
in appearance
of response
when
shock
interval
is short.
402
AKIRA
TAKEUCHI
AND
NORIKO
TAKEUCHI
plate current does not summate may be due to the decrease in the conduction velocity of the second nerve impulse. Although this result is indirect,
this will suggest the intimate relation between the duration of the end-plate
current and of the nerve impulse (see p. 409).
Some influences on time course of active phase
of temperature on active phase. It has been shown by some
7, 26) that lowering the temperature
lengthened
the time
e,p.p. and in curarized preparations
decreased its amplitude.
currents recorded from a curarized end-plate at different
are presented in Fig. 6. It is clear from this that lowering
FIG.
6. End-plate
currents
recorded
from curarized end-plate at different temperatures (from above downwards
at 20”,
17.5” and 15°C.). Current scale: 1 X10e7 A.
Time scale: 2 msec.
the temperature
decreased the amplitude
of the end-plate
current and
lengthened its time course. Q10 for the rising phase was 1.95 and that for
the falling phase was 2.05 in the temperature
range between 10°C. and
20°C. (mean values of four experiments).
The lengthening of the end-plate
current may be partly due to the lengthening of the nerve action potential,
but Q10 for the nerve action potential is different from that for the end-plate
current in the point that Q10 for the rising phase of the nerve impulse is
relatively small and that for the falling phase is large (15). Other factors
which mainly influence the time course of the end-plate current may be the
velocity of the reaction between the transmitter
and the receptor at the
end-plate and the activity of the cholinesterase. At the lower temperature
the change in the reaction velocity will tend to lengthen the rising phase of
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 17, 2017
1. Efect
authors (1,
course of the
The end-plate
temperatures
ACTIVE
PHASE
OF E.P.P.
403
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 17, 2017
the end-plate current and the change in the activity of the choline&erase
will prolong the falling phase.
2. Effect of eserine on active phase. The action of anticholinesterase
was
investigated by some authors (8,9,10). They found that there was a marked
lengthening
of the active phase during which the e.p.p. was built up. In
Fig. 7A the change in the time course of the end-plate current with addition
of eserine (concentration
1 X1O-5 g./ml.) to curarine Ringer’s solution is
presented. The duration of the rising phase was 1.28 msec. and peak to half
declining time was 2.01 msec. at 18OC. (mean values of 10 experiments).
These values were markedly longer than those obtained from a curarized
end-plate (p. 398). It was shown by Fatt and Katz (10) that when the nervemuscle transmission was blocked by replacing the external NaCl with isotonic sucrose, prostigmine
produced a dramatic lengthening
of the e.p.p.
The end-plate current in sodium-deficient
Ringer’s solution and that after
addition of eserine (concentration
1 X10-? g. /ml.) are presented in Fig. 7B.
The end-plate current in sodium-deficient
Ringer’s solution has a slower
time course than curarized muscle, rise time being 1.02 msec. and peak to
half decline time 1.59 msec. at 18°C. (mean values of 10 experiments).
After
addition of eserine the end-plate current had a rounded peak in 1.58 msec.
and then fell to one half in another 2.72 msec. at 18OC. (mean values of five
experiments).
3. Effect of sinomenine on active phase. It has been reported that sinomenine prolonged the descending phase of nerve action current (31). In
Fig. 7C an example of the effect of sinomenine hydrochloride
(Shionogi &
Co.) on the e.p,p. and the end-plate current is shown. After an addition of
1~10-~ g./ml. sinomenine to the curarine Ringer’s solution, the later part
of the falling phase of the e.p.p. was strikingly
lengthened
and the total
duration of the e.p.p. became more than 200 msec., although the rise time
of the e.p.p. was little influenced. Since the time constant and the effective
resistance of the muscle membrane were more than doubled by adding
sinomenine, the lengthening
of the time course of the e.p.p. with addition
of sinomenine may be partly due to the increase in the time constant of
the muscle membrane, -but the main effect must be due to the prolongation of the time course of the active phase. The end-plate currents recorded
from the same end-plate before and after an addition of 1 x10-4 g./ml.
sinomenine are shown in lower records of Fig. 7C. Sinomenine reduced the
amplitude of the e.p.p. and of the end-plate current. In this case d-tubocurarine concentration
in Ringer’s solution was decreased when sinomenine
was added. The effect of sinomenine is similar to that of anticholinesterase,
except for the finding that sinomenine has little influence on the time course
of the early part of the end-plate current or even shortens it, the rise time
being 0.64 msec. and the peak to half decline time being 0.84 msec. (mean
values of 8 experiments).
At the same time the later part of the end-plate
current is remarkably prolonged.
404
AKIRA
TAKEUCHI
AND
NORIKO
TAKEUCHI
Relation between end-plate current and resting membrane potential
The relation between the e.p.p. size and the resting membrane potential
was found by Fatt and Katz (10) to be approximately
proportional.
In
the present experiment the relation between the amplitude of the end-plate
current and the membrane potential was investigated.
Figure 8 illustrates
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 17, 2017
FIG. 7. Effects of eserine and sinomenine on e.p.p. and on end-plate current. A: (left),
e.p.p. (upper) and end-plate current (lower) recorded from curarized end-plate.
(Right),
recorded in curarine +eserine (1 X10m6 g./mh) Ringer’s solution. B: (left), e.p.p. (upper)
and end-plate current (lower) recorded from end-plate blocked in sodium-deficient
Ringer’s
solution, replacing
NaCl by isotonic sucrose. (Right),
recorded from sodium-deficient
+ eserine (1 X1O-6 g-/ml.) treated muscle. C: (left), e.p.p. (upper) and end-plate current
(lower) recorded from curarized end-plate
(3 X lo-” g. /ml. d-tubocurarine).
(Right), after
addition
of sinomenine
(1 X10-6 g. /ml. d-tubocurarine
-t-I X lo-’ g. /ml. sinomenine).
Voltage scale: 2 mV. Current scale: 1 X lo-’ A. Time scale: 2 msec. Temperature
at 18°C.
ACTIVE
PHASE
OF E.P.P.
405
the end-plate currents and the e.p.p.‘s of the curarized preparation recorded
from the same end-plate at various membrane potentials. The amplitude of
the e.p.p. and the end-plate current are plotted against the membrane potential in Fig. 9. There are approximately
linear relations between the
amplitude
of the end-plate current and the membrane potential,
and also
between the e.p.p. size and the membrane potential. If both lines are extrapolated, it seems that they cross the abscissa at a point about lo-20 mV.
negative to the Ringer bath. With the curarized preparation
the membrane
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 17, 2017
FIG. 8. E.p.p.‘s
and end-plate
currents
recorded
from
same
end-plate
at various
membrane
potentials.
Left:
e.p.p.‘s
recorded
from above
downwards
at 60,75,90,100
and
112 mV. respectively.
Right:
end-plate
currents
recorded
from
above
downwards
at 55,
65, 85, 95 and 110 mV. respectively.
Voltage
scale 10 mV. Current
scale:
1 X10-7
A. Time
scale:
3 msec. Temperature:
17°C.
406
AKIRA
TAKEUCHI
AND
NORIKO
TAKEUCHI
potential could not be reduced less than 50 mV., because of the appearance
of the action potential and contraction,
and the point at which the line
crossed the abscissa could not be determined accurately. With the preparation treated with sodium-deficient
Ringer’s solution the membrane potential
could be reduced below 50 mV. and the end .-plate current was then reversed.
mV
30
0
O0
a
0
a@
20
0
0
08
0.
0
0
IO
0
-50
membrona
Q)
N
‘J;
2
i
between
membrane
FIG. 9. Relation
potential and amplitude
of e.p.p. and endplate current obtained fkom same end-plate.
Abscissa:
membrane
potential
in mV.
Ordinates:
amplitude
of e.p.p. and endplate current, in 10-T A. Hollow circles:
e.p.p.‘s. Full circles: end-plate
currents.
0
Omv
-100
potential
When the membrane potential is clamped at constant voltage during
the neuromuscular
transmission,
the electrotonic
spread of the charge
along the muscle fibre membrane can be neglected and the electrical change
at the end-plate membrane can be measured. Therefore as a fist approximation the electrical change at the end-plate membrane is assumed to be the
change in the resistance and the electromotive
force which are inserted in
series at the end-plate membrane and are the function of the time. If this
condition is assumed the following equation is obtained:
I(E,
t) =
E - E(t)
R(t)
I(E,
t>: the end-plate current at time t from the onset and at the membrane potential E.
E: the potential at which the membrane potential is clamped.
E(t) : the electromotive
force inserted in the end-plate membrane at
time t from the onset of the end-plate current.
R(t) : the resistance inserted in the end-plate membrane
at time t
from the onset of the end-plate current.
At a fixed time ti from the onset of the end-plate current, the end-plate
current depends only on the membrane potential, Then if I(E, tJ is plotted
against E, a linear relation will be observed and its gradient shows l/R(&)
and the membrane potential at which the line crosses the abscissa (I(E, t)
=0) shows E(tJ. An example obtained from an end-plate blocked with
sodium-deficient
Ringer’s solution, replacing the NaCl with isotonic sucrose,
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 17, 2017
0
ACTIVE
PHASE
407
OF E.P.P.
is shown in Fig. 10. Although the lines obtained at various times. from the
onset of the end-plate current have different gradients, they cross the abscissa at about the same point. This result shows that in practice E(t) is not
the faction
of the time but a constant voltage inserted at the end-plate
membrane and the time course of the end-plate current depends on the
change in the resistance R(t). In Fig. 9 the gradient of the line which shows
the relation between the amplitude of the end-plate current and the membrane potential will represent the peak value of the shunting resistance and
this is about 380 Ka.
FIG. 10. Size of end-plate
current
at
various
times
firom
start
of current
are
plotted
against
displacement
of membrane
potential
from
its resting
value.
End-plate
currents
are measured
at A, 0.56 msec.;
B,
1.12 msec.;
C, 1.67 msec.;
and
D, 2.22
msec.
hm
onset
of end-plate
current.
Recorded
from
end-plate
blocked
in sodiumdeficient
Ringer’s
solution,
replacing
NaCl
by isotonic
sucrose.
-50
0
+3olnv
In some cases the time course of the end-plate current, especially its
falling phase, was somewhat lengthened at the higher membrane potential
and in this case the relation as shown in Fig. 10 did not hold exactly and the
lines did not converge on a point, although the discrepancy was not very
great. No satisfactory explanation is developed for this lengthening
at present, but it suggests that the hyperpolarizing
current interferes with some
processes at the end-plate membrane, e.g., the process whereby transmitter
substance is removed from the proximity of the end-plate membrane.
Active phase in normal end-plate
In the normal muscle fibre the e.p.p. leads to a propagating
spike and
contraction.
When the normal end-plate is located exactly, the membrane
potential can be clamped at the resting value during the neuromuscular
transmission and the spike does not appear from this end-plate. An example
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 17, 2017
1.5
408
AKIRA
TAKEUCHI
AND
NORIKO
TAKEUCHI
lower than that they used and also the present experiment was performed
at relatively lower temperatures
(l6”-18”C.).
Although in the later part of
the end-plate current the current size could not be measured exactly because
of the appearance of the rapid inward current and the movement artefact,
the charge displaced by the end-plate current was approximately
l-l.5 X 10-g
coulombs. The mean of the rise time was 0.70 msec. and the peak to half
decline time was 1.56 msec. at 18OC. (mean values of nine experiments).
The falling phase of the normal end-plate current was somewhat slower than
the curarized end-plate current, and this suggests that curarine shortens the
time course of the end-plate current perhaps by competing the receptor of
the end-plate with the transmitter
(4).
In the falling phase of the end-plate current, an inward current of rapid
time course was observed (Fig. 11). This rapid inward current was observed
in most muscle fibres (11 of 15) and might be due to the local current of the
propagated muscle spike which originated from the other end-plate than the
one clamped in the present experiment.
This result coincides with the report that most muscle fibres have double innervation
(17, 20).
DISCUSSION
The application of the voltage clamp method to the end-plate has some
weak points that are mainly due to the cable-like property of the muscle
fibre and to the high resistance of the electrode, but has an advantage that
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 17, 2017
of the end-plate current in normal Ringer’s solution is presented in Fig. 11.
The end-plate current at the normal end-plate was much greater than in
curarized muscle and was about 1 X10+ A. (0.7-1.2). In this experiment the
relation between the membrane potential and the size of end-plate current
was not measured; therefore the shunting resistance at the end-plate membrane was not measured directly. But if the constant voltage inserted at the
end-plate membrane (see previous section) was assumed to be about 15 mV.
(Fig. 9), the driving voltage for this end-plate current will be about 70 mV.
(the resting potential being about 85 mV.), and the shunting resistance is
about 60-100 KQ. Although this value is somewhat greater than that reported by Fatt and Katz (10) and de1 Castillo and Katz (3), this may be
due to the difference in the experimental
conditions, i.e., in the present experiment
the calcium concentration
in Ringer’s solution was in general
ACTIVE
PHASE
OF E.P.P.
409
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 17, 2017
with this method the experimental
condition is made simpler. If the voltage
clamp method is applied to the end-plate, the current which flows through
the end-plate membrane
during the neuromuscular
transmission
depends
only on the membrane potential
and on the electrical change at the endplate membrane and it will not be necessary to consider the electrical characteristics of the muscle membrane around the end-plate.
The time course of the active phase obtained in the present experiments
was in general agreement with the results reported by Eccles et al. (7) and
Katz (19) which were obtained by analyzing the e.p.p. The total duration of
the active phase was difficult to measure exactly, however; it was about
4-5 msec. at 17OC.and this value coincided with the value obtained by
Kuffler (21) who observed the size of the e.p.p. that was rebuilt after the invasion of a propagating
muscle impulse. Fatt and Katz (10) reported by
measuring the total charge displaced by e.p.p. that the active depolarization
process at the end-plate ceased within 2 msec. Although this value is shorter
than that obtained in the present experiment,
the end-plate current had a
rapid rising phase and a relatively slow declining phase and most of the current terminated within 2 msec. at higher temperature
(cf. Fig. 6). When the
displacement
of the charge along the muscle fibre was calculated with
Hodgkin and Rushton’s (16) equation 5.2, a curve similar to that reported
by Fatt and Katz (10) was obtained.
After the transmitter
was liberated from the nerve terminal, if dsusion
were to occur transversely across the cleft 500 A. wide to the end-plate membrane, the liberated transmitter
substance would be fairly uniformly
distributed in 10 /~sec. (6). Then the transmitter
substance would combine with
the receptor of the end-plate membrane and cause an electrical change of the
end-plate membrane. The transmitter
would then be removed from the endplate membrane by hydrolysis or diffursion. In the present condition it will
not be necessary to consider the desensitization
produced by acetylcholine
at the end-plate receptor because of the short time application of the transmitter to the end-plate membrane. Therefore the factors which determine
the time course of the active phase will be (i) the process of the liberation of
the transmitter
from the nerve terminals,
(ii) the reaction velocity of the
transmitter with the receptors at the motor end-plate, (iii) the process of the
electrical change at the end-plate membrane,
and (iv) the rapidity
with
which the transmitter
is removed from the end-plate membrane.
These
factors are speculative and which of them play the main part to determine
the time course cannot be decided at present but the fact that the main part
of the end-plate current did not summate and that sinomenine prolonged the
later part of the end-plate current will show the intimate relationship
between the time course of the end-plate current and that of the nerve impulse.
Of course these results are indirect, but suggest that the main factor which
determines the time course of the end-plate current is the process of the liberation of the transmitter
from the nerve terminals and the other factors,
such as (ii) and (iv), may have some influence on the time course of the end-
410
AKIRA
TAKEUCHI
AND
NORIKO
TAKEUCHI
plate current especially at lower temperature
(see p. 403). The contribution
of factor (iii) cannot be decided at present and further investigation
on this
problem will be necessary.
SUMMARY
ACKNOWLEDGMENT
The authors wish to express their thanks to Professor S. Sakamoto for his constant
advice and encouragement.
Thanks are also due to Professor Y. Katsuki and Dr. S. Hagiwara for reading the manuscript
and for making valuable suggestions. The authors are
also indebted to Dr. S. Ebashi for kindly supplying d-tubocurarine
chloride for this study.
1. BOYD, I. A. AND MARTIN, A. R. The end-plate
potential
in mammalian
muscle.
J. Physiol., 1956, 132: 74-91.
2. DEL CASTILLO, J. AND KATZ, B. Statistical factors involved in neuromuscular
facilitation and depression. J. Physiol., 1954, 124: 574-585.
3. DEL CASTILLO, J. AND KATZ, B. The membrane
change produced by the neuromuscular transkitter.
J. Physiol., 1954, 125: 546-565.
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 17, 2017
The active phase of the end-plate potential of frog’s sartorius muscle was
investigated
with the voltage clamp method. The feed-back current which
flowed through the muscle membrane during neuromuscular
transmission
was called the “end-plate current.”
1. The time from onset to peak of the end-plate current was 0.77 msec.,
peak to half decline time was 1.08 msec., and total duration was about. 4-5
msec. at 17OC.
2. In the curarized preparation
there was an almost linear relation between the amplitudes
of the e.p.p. and end-plate current until the e.p.p.
reached 10 mV. With the eserinized preparation in sodium-deficient
Ringer’s
solution, an upward concavity of the curve was observed.
3. When two nerve volleys were applied, the end-plate current produced
by the second nerve volley was consistently
larger than the first and the
main part of the end-plate currents did not summate.
4. Lowering the temperature
lengthened
the time course of the endplate current and reduced its amplitude.
Q10 for rising phase was 1.95 and
for falling phase was 2.05 at temperatures
between 10’ and 20°C.
5. Eserine lengthened the time course of the end-plate current, especially
in sodium-deficient
Ringer’s solution. Sinomenine lengthened the later part
of the end-plate current.
6. There was an approximately
linear relationship
between the membrane potential and the end-plate current. If the line was extrapolated
it
crossed the zero-current line at the membrane potential lo-20 mV. negative
to the outside Ringer’s solution.
7. In the normal end-plate the peak value of the end-plate current was
about l-l.5 X1O-6 A. The time course of the normal end-plate current was
somewhat longer than that of a curarized preparation.
On the falling phase
of the normal end-plate current a rapid inward current was observed which
might be due to the propagating
muscle action potential from another endplate of the same muscle fibre.
ACTIVE
PHASE
OF E.P.P.
411
126: 293-303.
18.
KATZ,
plate’
19,
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
B.
Impedance
changes
in frog’s
muscle
associated
potentials.
J. Neumphysiol.,
1942, 5: 169-184.
B.
The electrical
properties
of the muscle
fibre
with
electrotonic
and
‘end-
membrane.
Pmt. Roy. Sot.,
1948, B135: 506-534.
KATZ,
B. AND KUFFLER,
S. W.
Multiple
motor
innervation
of the frog’s
sartorius
muscle.
J. Neurophysiol.,
1941, 4: 209-223.
KUFFLER,
S. W.
Further
study
on transmission
in an isolated
nerve
muscle
fibre
preparation.
J. Neurophysiol.,
1942, 5: 309-322.
LING,
G. AND GERARD,
R. W.
The
normal
membrane
potential
of frog
sartorius
fibres. J. cell. camp. Physiol., 1949, 34: 383-396.
MARTIN,
A. R.
A further
study
of the statistical
composition
of the end-plate
potential.
J. Physiol., 1955,130:
114-122.
NASTUK, W. L. The electrical
activity
of the muscle
cell membrane
at the neuromuscular
junction.
J. cell. camp. Physiol., 1953, 42: 249-272.
NASTUK, W. L. AND HODGKIN,
A. L.
The electrical
activity
of single
muscle
fibres.
J. cell. camp. Physiol., 1950, 35: 39-73.
TAKEUCHI,
N. The effect
of temperature
on the neuromuscular
junction
of the frog.
Jap. J. Physiol., 1958,8: 391404.
Direct
determination
of the active
phase
of endTAKEUCHI,
N. AND TAKEUCHI,
A.
plate potential.
Nature, 1958, 181: 779.
TASAKI, I. Nervous transmission. Springfield,
Illinois,
C. C Thomas,
1953, pp. 86-87.
TASAKI, I. AND HAGIWARA, S. Analysis
of impulse
transmission
in the squid
giant
synapse.
Fed, Proc., 1957,16:
127-128.
TASAKI, I. AND HAGIWARA, S. Demonstration
of two stable
potential
states
in the
squid
giant
axon
under
tetraethylammonium
chloride.
J. gen. PhysioZ., 1957, 40:
KATZ,
859485.
TASAKI, I. AND MIZUGUCHI,
K.
The changes
in the electric
impedance
during
activity
and the effect
of alkaloids
and polarization
upon the bioelectric
processes
in the myelinated nerve
fibre.
B&him,
Biophys. Acta, 1949, 3: 484-493.
WEIDMANN, S. The effect of the cardiac
membrane
potential
on the rapid
availability
of the sodium-carrying
system.
J. Physiol., 1955, 127: 213-224.
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 17, 2017
4. DEL CASTI=,
J. AND KATZ, B.
Interaction
at end-plate
receptors
between
different
choline
derivatives.
Pmt. Roy. SOC., 1957, B146: 369-381.
5. DEL CASTI=,
J. AND STARK, L. The effect of calcium
ions on the motor
end-plate
potentials.
J. PhysioZ., 1952, 116: 507-515.
6. ECCLES, J. C. AND JAEGER, J. C.
The relationship
between
the mode
of operation
and the dimensions
of the junctional
regions
at synapses
and motor
end-organs.
Proc.
Roy. Sot., 1958, B148: 38-56.
7. ECCLES,
J.C.,
KATZ, B., AND KUFFLER, S. W. Nature
of the ‘endplate
potential’
in
curarized
muscle.
J. Neurophysiol.,
1941, 4: 362-387.
8. ECCLES,
J. C., KATZ,
B., AND KUFFLER,
S. W.
Effect
of eserine
on neuromuscular
transmis&m.
J. Neumphysiol.,
1942, 5: 211-230.
9. Ecc~~s,
J. C. AND MACFARLANE,
W. V.
Actions
of anticholinesterases
on end-plate
potential
of frog muscle.
J. Neumphysiol.,
1949, 12: 59-80.
An analysis
of the end-plate
potential
recorded
with an intra10. FATC, P. AND KATZ, B.
cellular
electrode.
J. Physiol., 1951, 115: 320-370.
11. FAIT,
P. AND KATZ,
B.
Spontaneous
subthreshold
activity
at motor
nerve
endings.
J. Physiol., 1952,117:
109-128.
The
nature
of conduction
of an impulse
in the
12. GAFFER,
H. S. AND ERLANGER,
J.
relatively
refractory
period.
Amer. J. PhysioZ., 1925, 73: 613-635.
Mechanism
of action
potential
production
in the nerve
13. HAGIWARA,
S. AND SAITO, N.
cell of a puffer.
Proc. Jap. Acad., 1957,33:
682-685.
14. HODGKIN,
A. L., HUXLEY,
A. F., AND KATZ,
B.
Measurement
of current-voltage
relations
in the membrane
of the giant
axon of LoZigo. J. Physiol., 1952,116:
424-448.
15. HODGKIN,
A. L. AND KATZ,
B,
The effect
of temperature
on the electrical
activity
of the giant
axon of the squid.
J. Physiol., 1949,109:
240-249.
16. HODGKIN,
A. L. AND RUSHTON,
W. A. H.
The electrical
constants
of a crustacean
nerve
fibre.
PIVC. ROY. SOC., 1946, B133: 444-479.
17. HUNT,
C. C. AND KUFFLER,
S. W.
Motor
innervation
of skeletal
muscle:
multiple
innervation
of individual
muscle
fibres
and motor
unit
function.
J. Physiol., 1954,