jf. exp. Biol. 128, 1-17 (1987)
Printed in Great Britain © The Company of Biologists Limited 1987
\
STUDIES OF THE CARDIAC-LIKE ACTION POTENTIAL
IN CRAYFISH GIANT AXONS INDUCED BY PLATINIZED
TUNGSTEN METAL ELECTRODES
BY LISTON A. ORR AND EDWARD M. LIEBERMAN
Department of Physiology, School of Medicine, East Carolina University,
Greenville, NC 27834, USA
Accepted 22 October 1986
SUMMARY
A lightly platinized tungsten (Pt-W) wire electrode, axially inserted into a crayfish
giant axon, causes the development of cardiac-like action potentials with durations of
up to 4 s. The plateau in membrane potential typically occurs within lOmin of the
start of action potential elongation. The effect occurs without passing current
through the Pt-W electrode and is temporally related to a dramatic decrease in
intracellular pH (pH,). Such an effect cannot be induced by a decrease in pH,
produced by equilibrating the axon with HCO3~-CO2 solution (pH6), an NH4C1
rebound or direct intracellular injection of PO43~ buffer (pH45). Action potential
elongation is accompanied by a block of delayed rectification and the possibility that
inward rectification also develops cannot be ruled out. Plateau generation requires
Na + and Ca z+ inward currents as demonstrated by abolition of the plateau by
[Na + ] o or [Ca 2+ ] o depletion or treatment with tetrodotoxin (TTX) or verapamil.
The block of outward rectification by Pt-W requires external Na + or Ca 2+ . Action
potential elongation produced by 3,4-diaminopyridine is not sensitive to verapamil
and the waveform is different from that produced by Pt-W. The data support the
possibility that different classes of excitable membranes have similar channel
populations and that the functional differences between them reside in the inhibitory
or masking influences that are present in the microenvironments of the various
membrane channels.
INTRODUCTION
This study arose from a chance observation made while using tungsten (W) wire as
a substitute for platinum (Pt) as an intracellular current-passing electrode. Tungsten
was expected to be a good alternative to Pt because of its superior mechanical
properties and equivalent electrical properties.
When crayfish giant axons were cannulated with a platinized tungsten (Pt-W)
wire, prepared similarly to platinized platinum electrodes, modification of the action
potential (AP) occurred within 15min. Whereas the normal action potential has a
duration of approximately 1 ms, action potentials with durations of up to 4 s were
Key words: axons, crayfish, cardiac-like action potentials, tungsten electrodes, K + , Ca2+,
channels.
2
L. A. ORR AND E. M. LIEBERMAN
observed in axons exposed to the Pt-W wire. The rising phase of the AP was largely
unaffected, while the falling phase developed a plateau similar to that seen in the
cardiac action potential. It was not necessary for current to be passed through the
wire for the effect to develop. Some component of the Pt-W wire or product of a
chemical reaction between the Pt-W wire and axoplasm was evidently altering the
membrane ionic permeability control mechanisms responsible for the production of
the action potential.
Long-duration action potentials of cardiac muscle are due to a prolonged increase
of Na + and Ca2+ conductances combined with a low K + conductance (Gettes &
Reuter, 1974; Reuter, 1967; Weidmann, 1951). Elongated action potentials have
been produced in nerve preparations only by the use of pharmacological or toxic
agents which delay Na + inactivation and/or K + activation (Shrager, Macey &
Stockholm, 1969; Shrager, Strickholm & Macey, 1969; Narahashi, 1974). The
purpose of this study was to identify and characterize the ionic currents altered by the
Pt-W wire, and how the effects are mediated, and to assess the potential of Pt-W as a
probe of membrane electrophysiological function.
A preliminary account of this work has been presented in abstract form (Orr &
Lieberman, 1983).
MATERIALS AND METHODS
The medial giant axon of the ventral nerve cord of the crayfish Procambarus
clarkii was dissected and isolated according to the method of Wallin (1967). The
apparatus employed for membrane potential recording, space and current clamping
(Fig. 1) was essentially as previously described (Lieberman & Lane, 1976) with the
following modifications: the electrode used for monitoring membrane potential was a
KCl-filled impaling glass microelectrode; the current-measuring circuit was a virtual
ground system with an additional feedback circuit for voltage-clamping the guard
electrodes and bath to a constant reference value; and two sets of external platinized
Ag-AgCl plate electrodes were used. A set of external stimulating electrodes, placed
near the suboesophageal ganglion, was used to generate propagated action potentials.
Microelectrode tips were bevelled in a swirling solution of 2-5moll" 1 KC1 containing a silica polishing powder (Corson, Goodman & Fein, 1979) to facilitate
impalement of axons. Bevelling had the dual function of sharpening the electrode tip
and lowering its electrical resistance. Best results were obtained with microelectrodes
with resistances between 10 and 20 MQ. The electrodes used to monitor the potential
of the external solution consisted of a glass pipette filled with 2-5moll~1 KC1
suspended in agar gel. Connection between the gel and the amplifier lead was made
with a Ag-AgCl wire in 2-5 mol I"1 KC1 solution. Resistances of the agar electrodes
were between 2 and 10 KQ.
The axial wire current electrode was a 25 /im diameter wire (Pt or W) with an
uninsulated length of 6—8 mm. Electroplating of the wire with Pt was carried out in a
platinizing solution containing 3% PtCl3 and 0-025 % lead acetate in 30mmolP'
HC1. Plating was carried out at a current level of l-5mA for variable periods a^
Cardiac-like action potentials in axons
3
Required to produce the appropriate type of electrode. Current-voltage (I-V)
relationships were determined from the change in membrane potential for a 100- to
300-ms constant-current pulse applied across the membrane and recorded oscillographically using x—y plotting techniques. Membrane resistance, at the resting
membrane potential, was estimated graphically by determining the slope of the I-V
relationship at the zero current intercept. AP traces and membrane potentials (Em)
were permanently recorded using oscillographic methods.
Specific channel blockers were added to the external solution after the Pt-W wire
had begun to take effect. In experiments in which plateau durations were studied,
only large and obvious changes were considered to be significant. This was due to the
dynamic nature of the Pt-W effect. Quantitative measurements and statistical
analysis were therefore not reliable.
Changes in the ionic composition of the crayfish physiological solution (van
Harreveld, 1936), used to study the ionic requirements for the development of the
Fig. 1. Diagram of the current and space clamp system. Membrane potential (E m ) was
monitored with an impaling glass microelectrode (Elrec) referred differentially to a second
glass electrode placed in the bath. The axial wire stimulating electrode (El,) was a glassinsulated 25 (lm diameter platinized tungsten or platinum wire with an exposed length of
7-8 mm. Current was passed between El, and a double set of platinized Ag/AgCl 2 plates
(placed on either side of the axon) serving as guards (Eg) and a centre plate (Elc) with
which membrane current (I m ) was monitored. The two Elc plates represented approximately 20 % of the total surface area of the Elg and Elc surface area. The virtual ground
current monitor utilized a voltage feedback system to maintain the bath and the electrodes
at a uniform voltage (Elref) to ensure a constant partition of current between the plate
electrodes, based on their relative surface area. Current through El, was generated by an
operational amplifier voltage to constant current converter (I—> V) driven by a Textronix
TMS06 pulse generator.
4
L. A. ORR AND E. M.
LIEBERMAN
Pt-W effect, were accomplished by substituting Tris-HCl for NaCl, KC1 or
Osmolarity was maintained with Tris buffer (pH 7-4). For Cl-free solutions all salts
were made with isethionate as the anion. Ca2+ was slightly increased to compensate
for the decrease in Ca2+ activity that occurs with isethionate (Lieberman, 1979).
RESULTS
Preliminary observations on the elongated action potential
Changes in the time course of the crayfish giant axon action potential occurred
within 15 min of cannulation with the Pt-W wire. The initial changes involved a
prolonging of the AP falling phase, a slight slowing of the rising phase and a small
reduction in AP amplitude (Fig. 2A). The falling phase continued to slow, developing into a plateau typically within 10 min of the start of AP elongation (Fig. 2B).
AP durations increased rapidly from this point, reaching durations of 20 ms to 4 s.
Oscillations in the Em were often seen near the end of longer-duration action
potentials (Fig. 2C).
The membrane potential showed changes that were typical of those shown in
Fig. 2D. There was a hyperpolarization of about 5 mV prior to the beginning of AP
elongation. Concurrent with plateau formation there was a slow depolarization that
continued until the axon lost its excitability. Holding the Em at 80 mV allowed the
axon to remain excitable for 20 min to 1 h until membrane resistance fell precipitously, resulting in a complete and irreversible loss of excitability.
The ability of the Pt-W wires to generate the elongated AP was very sensitive to the
level and duration of the current used to electroplate the W wire with Pt. Any large
deviation from the 5 s at 1-SmA protocol resulted in a wire which had too thick or
thin a coat of Pt to produce the effect. Apparently, both metals must be exposed in
approximately equal amounts in order for the maximal effect to occur. Plateau
amplitudes (measured from the resting membrane potential) ranged between 20 and
80 mV and averaged 65—70 mV. Each Pt-W wire had an effective lifespan, relative to
action potential elongation, of up to 3 weeks. Electrodes could often be 'recharged' by
an additional plating with Pt.
The rate of increase of AP duration during the formation of a plateau was found to
be affected by three major factors. Significantly increasing the rate of AP production
temporarily decreased the duration of an elongating plateau. This was followed by
continued elongation at a slower rate. Similar results were obtained by increasing the
flow rate of the external bathing solution. Larger axons (with greater volume of
dilution) took longer to exhibit the effect.
In the normal crayfish axon, a single, long-duration depolarizing current pulse
normally produced a single AP. Occasionally several APs were produced by a single
pulse in a Pt-W-altered axon. The repetitive firing response was seen both before and
during AP elongation.
The Pt-W effect was found to be reversible upon removal of the wire from the axon
during the early stages of AP elongation. If the AP duration was less than 50 ms
Cardiac-like action potentials in axons
A
Axon091481Aa
B
Axon 042282Ab
100 mV
100 mV
1 ms
5 ms
Axon 042282Ab
A xc n 0.(1USIAva
L)
-10
I
i
i
!
i
lOOmV
I
500 ms
•
k
• m•
1
-
t
-50
•1
-90
5 min
Fig. 2. The time-dependent development of the elongated action potential following
cannulation of the axon with a platini2ed tungsten wire. In A-C action potentials were
generated at a distance from the position of the Pt-W wire and propagated through the
cannulated region. Action potentials generated by space and current-clamp pulses were
not different from propagated action potentials. (A) Superimposed action potential
traces showing the initial change in action potential kinetics. A small slowing of the rising
phase and small reduction in amplitude occur. (B) Superimposed traces demonstrating
the characteristic plateau formation over a 10-min period. Note change in time scale.
(C) A typical long-duration action potential. The oscillations at the end of the plateau
commonly occurred in long-duration plateaus and appeared to be small, 5-10 ms
duration, action potentials. (D) Strip chart recording of the membrane potential following the cannulation of the axon with the Pt-W wire. The small depolarization near the
beginning of the trace marks the time the axon was cannulated.
removal of the wire usually caused the duration to return to normal within 1 min.
Removal of the wire after the AP duration had reached 100 ms usually had no effect
on reversal of the AP elongation and membrane depolarization.
It was found that hyperpolarizing or depolarizing the axon by passing a continuous
current through the axial wire caused a decrease in duration and plateau amplitude of
an elongated AP. It appeared that the steady-state resting potential of the axon was
the optimal potential for the Pt-W effect.
Components of the Pt-W wire responsible for the elongated action potential
A series of experiments was conducted to determine which component or comftbination of components of the Pt-W wire was responsible for the effect. Axons were
6
L. A. ORR AND E. M. LIEBERMAN
cannulated with a plain W wire for up to 2-5 h with no change in AP kinetics.
Injection of a K + isethionate solution (artificial axoplasm) containing 1 mmol 1~
Na2WO4 into the axon also had no effect. The influence of lead in the platinizing
solution was investigated by cannulating axons with a W wire which had been plated
in a 30 mmol I"1 HC1 solution containing 0-025 % lead acetate as the sole solute. No
change in AP kinetics was observed. Pt wires and platinized Pt wires have been used
for decades with no unusual effects, suggesting that the combination of both W and
Pt was necessary for the action potential elongation to occur. This idea was tested
with a W wire electroplated in a 30 mmol P 1 HC1 solution containing 3 % PtCL^ but
no lead acetate. When cannulated into an axon, this wire was successful in producing
elongated action potentials.
Pt-W amalgams are used extensively in chemical processes as inhomogeneous
catalysts and as such may generate H + or free radicals. The possibility that the
electrode generated H + was tested by injecting the pH indicator, phenol red, into the
axoplasm in sufficient quantity to dye the axoplasm clearly red ( p H > 7 ) . On
cannulating the axon with a Pt-W electrode the axoplasm slowly turned yellow
(pH < 6-6) over a period of 5-10 min. The action potential began to elongate as the
colour change became visible. On occasion, the action potential of a Pt-W-treated
axon alternated between a slightly elongated (5—10 ms) and a fully elongated AP
(>50ms). In one extraordinary experiment this occurred in an axon injected with
phenol red. As the action potential oscillated between long and short durations the
colour oscillated between yellow (pH<6 - 6) and red ( p H > 7 ) , respectively, providing clear evidence of a relationship between pH, and the elongation of the action
potential.
The possibility of action potential elongation due solely to lowered pH ; was tested
by the creation of an acidic axoplasm using either the NH4 rebound method or
addition of CO2 to a physiological HCO3~ solution (Boron & DeWeer, 1976; Moody,
1980). These methods did not produce AP elongation. Attempts were made to
produce an elongated AP by injecting potassium phosphate solutions (pH4-5)
directly into the axon in sufficient quantity to replace 50 % or more of the axoplasm.
Phenol red was included in the solution to monitor the acidity of the axoplasm.
Action potential duration increased slightly (2 ms) with no change in indicator
colour. Although the evidence showed that a decreased pH ; could not be the sole
mechanism, it is possible that such a decrease is a necessary condition for action
potential elongation. To further investigate the role of pH;, 20 mmol I"1 NH4C1 was
added to the superfusate to counter the Pt-W-induced decrease in pHj. This
procedure caused previously formed plateaus progressively to shorten and disappear.
With the removal of the NH4C1, to generate an acidic axoplasm, the plateau reformed
with durations in excess of those originally seen.
Effect of Pt-W on membrane resistance
A comparison of the I-V relationship for a normal axon with that for a Pt-Wtreated axon, after AP elongation had occurred, reveals an obvious difference in thefl
Cardiac-like action potentials in axons
depolarized potential region (Fig. 3). The control axon exhibits outward rectification whereas the Pt-W-treated axon exhibits an apparent inward rectification. The
large change in E m indicated by the dashed line represents the voltage shift of the AP
plateau. Plateau formation could be explained by three possible mechanisms, alone
or in combination: (1) a large, sustained inward current (i.e. a Na + current with a
delayed inactivation or a sustained Ca + current); (2) a decrease in outward K +
current; (3) a decrease in total K + conductance, relative to rest, similar to that seen
in cardiac muscle (Gettes & Reuter, 1974), in skeletal muscle (Adrian, 1969) and in
oocytes (Hagiwara & Yoshii, 1979). Evidence for these possibilities was sought by
injecting a train of small, negative, constant-current pulses across the membrane as
an elongated AP was generated (Wiedmann, 1951). A typical result from this type of
experiment is shown in Fig. 4, where the voltage deflections are over 200% larger
near the end of the plateau than those seen at the rest potential, suggesting an
increase in membrane resistance (Rm) during the plateau (presence of an inward
rectifying channel). The apparent increase of Rm was variable from axon to axon,
ranging from almost no increase to as much as 300 %. An increase in Rm was usually
not seen in plateaus with durations less than 50 ms. The largest increases occurred in
plateaus with durations greater than 100 ms.
Barium (O'l—lOmmoll"1), which is known to block inwardly rectifying channels
in hyperpolarized membranes (Standen & Stanfield, 1978), was used to test for the
presence of an inward rectifier unmasked by the action of the Pt-W wire. Ba2+ did not
alter the E m or increase the amplitude of membrane potential responses to hyperpolarizing square current pulses, as would be expected if conducting, inwardly
rectifying channels were being blocked.
150
100
50
-50
-100
-150
-200
-150
—100
-50
i
Membrane potential (mV)
Fig. 3. Effect of the Pt-W electrode on current-voltage relationships from a single axon.
An I—V relationship obtained with a platinized Pt electrode (Pt-Pt, closed circles) is
plotted together with that obtained from an axon treated with a platinized W electrode
(Pt-W, open circles) for comparison.
8
L. A. ORR AND E. M. LIEBERMAN
Axon 062382Ab
lOOmV
IVWVWI
100 ms
Fig. 4. Oscilloscope tracing of an action potential with superimposed current pulses
demonstrating an apparent change in membrane resistance during the plateau. Note the
approximately 200 % increase in pulse height at the end of the plateau relative to rest.
Influence of [Na+]0 and the Na+ channel blocker, tetrodotoxin, on plateau
development
+
When the Na concentration in the saline was reduced to Smmoll" 1 , the AP
amplitude of a normal axon was reduced by approximately one-third. After the axon
had been cannulated with a Pt-W wire, there was slight slowing of both the rising and
falling phases of the AP but no plateau was formed. Upon replacement of the control
concentration of Na + (190mmoir'), AP amplitude increased to normal and was
accompanied by the rapid formation of a plateau.
Similar results were obtained with external solutions containing 25, 52, 73 and
97-5mmoll~' Na + . The effects of 25 and 73mmoll~' are shown in Fig. 5. With
increasing external [Na + ] the action potential reached greater final durations with a
maximum observed duration of 15 ms at 97-5mmoll~1 Na + . Addition of control
[Na + ] o rapidly increased the plateaus to 100 ms or greater. I-V plots from Pt-Waltered axons in 5, 25 and 52mmoll~ 1 Na + exhibited normal outward rectification.
Axons in 73 and 97-5 mmoU"1 Na + exhibited typical Pt-W-induced 'inward'
rectification.
Addition of lOOmmoll"1 TTX into the bathing solution after a plateau had been
formed reduced the plateau duration. The action potential was then abolished. A
current pulse injection resulting in an Em deflection of the same amplitude and
duration as a normal AP would not initiate a plateau. An I-V plot from an axon
exposed to TTX was produced immediately after the axon had been cannulated with
a Pt-W wire and exhibited normal outward rectification. The outward rectification
seen during depolarizing steps began to decrease as the Pt-W wire took effect. After
30 min the curve became completely linear, but did not go on to rectify in an
apparently inward manner (Fig. 6).
Influence of [Ca2+]0 and the Ca2+ channel blockers verapamil and La3+
Experiments similar to those carried out in low [Na + ] o were performed in one-<
quarter and one-half control [Ca 2+ ] o . Plateaus did not develop in Pt-W-altered axons
Cardiac-like action potentials in axons
9
bathed in one-quarter control [Ca 2+ ] o and action potentials were only slightly
elongated in one-half control [Ca 2+ ] 0 . Raising [Ca 2+ ] o to the control level (13-5
mmol I 1 ) caused the plateau duration and amplitude to increase dramatically.
External [Ca2+] affected the plateau size in a titratable manner and like [Na + ] o was
also required for the blocking of outward rectification.
When the Ca2+ channel blocker verapamil (lO^moll" 1 ) was included in the
bathing solution, action potentials had a maximum duration of 10ms and no
plateaus. Plateaus formed in the absence of verapamil were abolished in its presence.
I-V plots (data not shown) demonstrated an absence of normal outward rectification
but no apparent inward rectification. The result was a straight, ohmic I-V plot
similar to that seen with the Pt-W effect plus TTX.
Superfusion with the Ca2+ channel blocker La 3+ at concentrations of 1 and
5 mmol P 1 , abolished plateaus and prevented the apparent inward rectification in a
manner similar to verapamil.
Lo\v[Na+]
25mmoll" 1 Na+
Axon 111482Ab
190mmoir'Na+
Axon 111482Ab
lOOmV
2 ms
73mmoir1Na+
Axon 1214S2Ab
190mmoir 1 Na
AxonO!2283Aa
lOOmV
2 ms
10ms
Fig. 5. Effect of low [Na + ] o on the development of the Pt-W effect. (A) The maximum
duration AP developed in 25 mmol I" 1 [Na + ] o was approximately 2 ms. (B) Within 2 min
of the re-admission of 190 mmol I" 1 [Na ] o the plateau elongated to 100 or more
milliseconds. (C) In 73 mmol I" 1 [Na ] o plateau development is present but held to
5-7 ms (see shorter AP in D). (D) Following the readmission of 190 mmol 1~' [Na + ] o the
plateau further develops to greater than 25 ms within seconds.
10
L. A. ORR AND E. M. LIEBERMAN
Influence of [K+Jo
In initial experiments to determine the role of K + , the external K + concentration
was raised while using a holding current to maintain the Em at the potential expected
in control [K + ] o . High [K + ] o resulted in a decrease of AP duration of Pt-W-treated
axons. Maintaining the Em at a constant level ensured that the reduction in AP
duration was not due to effects of depolarization on voltage-sensitive channels.
In four times control [K + ] o (Zl^mmoll" 1 ), Rm of an axon cannulated with a
platinized Pt wire fell to one-third of its level in 5-4mmoir' [K + ] o (2136 Qcm2 vs
683 Qcm2) (Fig. 7A). When the same procedure was carried out in a Pt-W-altered
axon, a significant drop in Rm at 21-6mmoir' [K + ] o was not seen (1175 Qcm2 vs
1100 Qcm2), suggesting that steady-state leakage channels normally opened by high
[K + ] o are prevented from doing so in the altered membrane (Fig. 7B). High [K + ] o
abolished the Pt-W-induced 'inward-going' rectification, significantly reduced the
plateau and allowed the voltage-sensitive outwardly rectifying channels to operate
normally (open under depolarization).
Comparisons with other agents that cause action potential elongation
Several agents known to cause AP elongation in a number of nerve preparations
were employed to compare their effects on the action potential of the crayfish giant
180
120
60
-60
-120
-200
-150
-100
-50
Membrane potential (mV)
-180
Fig. 6. The effect of tetrodotoxin (TTX) on a Pt-W-altered axon. Three superimposed
I—V plots from TTX-treated axons showing the reduction of outward rectification during
the onset of the Pt-W effect. Closed circles; I-V curve measured immediately after
cannulation of the axon with the Pt-W electrode. Open circles; I-V curve obtained
approximately 5 min after cannulation. The ohmic I-V plot (triangles) from TTXtreated axon was obtained 30 min after cannulation with Pt-W wire. No further change
was seen with additional exposure. T T X prevented the development of the apparent
inward rectification typical of the prolonged action potential. A developed plateau was
reduced in duration by T T X even though the action potential remained near normal
amplitude for several minutes following the initial reduction in duration.
Cardiac-like action potentials in axons
11
axon with the response described for Pt-W. External application of either 1 mmol 1 '
tetraethylammonium (TEA) or 2 mmol I"1 Baz+ did not affect the kinetics of the
normal AP. AP plateaus with durations of up to 350ms were produced by external
application of 0-5 mmolP 1 3,4-diaminopyridine (DAP) (Fig. 8). Addition of
A
Pt-Pt
B
180
Pt-W
180
JJ
120
60
120
1
o
60
c
-
0
•4 mmol l"jK^*--^!or
-60
-60
-120
°~r2h6mmo\\~1K+
-120
K+
-180
-180
-200
-150
-100
-50
0
-150
-100
-50
Membrane potential (mV)
Fig. 7. The effect of [ K + ] o on the I - V relationships of control and Pt-W-treated axons.
(A) Axon cannulated with platinized platinum electrode. High [ K + ] o decreases the
membrane resistance in the hyperpolarizing direction as well as decreasing outward-going
rectification. At E m = — 85 mV the axon treated with high [ K + ] o (open circles) has a
much lower resistance than before K + treatment (closed circles). (B) A typical I - V
relationship for a Pt-W-treated axon is shown ( • ) . The I - V curve of an axon treated with
high [ K + ] o (open circles) compared with an axon in control [ K + ] (closed circles)
demonstrates that Pt-W protects the voltage-sensitive outward K + channels from [ K + ] o .
T h e hyperpolarizing segment of the I - V relationship is unchanged by [ K + ] o . T h e
development of the 'inward' rectification is abolished but opening of the outward K +
channels is allowed in the depolarizing range of voltage by excess [ K + ] o .
lOOmV
2 ms
100 ms
Fig. 8. T h e effect of 3,4-diaminopyridine (DAP) on the duration and waveform of the
crayfish action potential. (A) A normal propagated action potential. (B) The full effect of
DAP is illustrated. In comparison with the effect of Pt-W, the DAP-modified action
potential has a greater initial rise time, much more repetitive activity on the plateau,
especially early, and a much faster decay of the plateau suggestive of a passive discharge
rather than a plateau due to a maintained inward current.
1)
z
u
<v
c
ca
Membr
0
12
L. A. ORR AND E. M. LIEBERMAN
10~ 5 moll~' verapamil or 0-5mmolP' Mn 2+ to the DAP-containing solution did not
modify the effect, indicating that there is no significant Ca2+ component of the DAPinduced plateau, in contrast to the Pt-W-altered APs.
DISCUSSION
The duration of action potentials of the crayfish medial giant axon is increased
from < l m s to >50ms by an intracellular Pt-W wire electrode. This effect is
accompanied by a decrease in axonal pH ; , but cannot be produced by such a
reduction induced by various other treatments. The mechanism for the AP
elongation can be explained by a decrease in outward (delayed) rectification coincident with an increase in inward Ca 2+ current. Comparison of I—V plots before
and after the Pt-W wire has taken effect (Fig. 3) reveals a block of delayed rectification. The AP plateau is inhibited by low [Ca 2+ ] o and Ca2+ channel blockers in a
preparation where Ca2+ influx normally plays an insignificant role in AP kinetics of
the crayfish axon (Yamagishi & Grundfest, 1971). Although the apparent increase in
membrane resistance (inward rectification) (Fig. 4) could as well be explained by
changes in inward currents (Ca2+ and/or Na + ) induced by the pulses used to
estimate membrane resistance, true inward rectification cannot be ruled out without
voltage-clamp studies of currents flowing during the plateau (Goldman & Morad,
1977).
In initial studies to investigate the active component of the wire, it was found that
plain tungsten, sodium tungstate, platinum or lead acetate had no effect on the action
potential. The only effective combination was lightly platinized tungsten metal. The
effectiveness of the Pt-W wire was not dependent on the passage of current through
it. Several characteristics of the Pt-W-induced effect seemed to indicate that the
effective agent was released into the axoplasm from the Pt-W wire to react with the
axonal membrane. The AP elongation took time to develop after cannulation of
the axon, and axons with greater diameters (greater volume of dilution) took longer
to exhibit the effect. The reversibility of the effect in its early stages of development
indicates that the reaction product is of a rather labile nature or rapidly buffered.
This would also explain the results of an experiment in which the injection of
Pt-W-treated axoplasm into an axon had no effect on the AP. A constant source of the
product seems necessary to reach an effective concentration. After a time, the
membrane becomes permanently altered; removal of the wire would not cause a
reversal of the effect.
These observations led to the consideration of ionic hydrogen as a possible active
agent. Phenol red studies revealed that the Pt-W wire decreased pH; with a time
course coincident with the initiation of AP elongation. While a decreased pH; alone
did not result in AP elongation, it was found to be a necessary component of the
effect. The NH4C1 rebound experiment provides evidence for this conclusion. This
method for altering pH; produces an initial alkalinization of the axoplasm on addition
of NH4C1 to the superfusate. During this period, the Pt-W-induced plateaus are
abolished. The plateaus are reformed rapidly on re-acidification of the axoplasm by
Cardiac-like action potentials in axons
13
removal of the external NH4C1. Recent studies in squid axons have shown that
passage of large, long-duration currents through wire electrodes generated H +
(Mullins, Requena & Whittenburg, 1985) which in turn was related to a large influx
of Ca + (J. Requena & L. J. Mullins, personal communication). In addition,
decrease of pH; in squid giant axons has been found to decrease outward K + current
(Wanke, Carbone & Testa, 1979; Carbone, Prir & Wanke, 1981). Other than the
necessary decrease in pH ; , it is not known what other products of the Pt-W/axoplasm
reaction are involved in the AP elongation.
The block of outward rectification can account for the AP elongation caused by
DAP (Fig. 8) (Kirsch & Narahashi, 1978) and part of the elongation caused by
Pt-W. The decrease in outward rectification produced by Pt-W was found to be
dependent on [Ca 2+ ] o . Apparent inward rectification is abolished by both verapamil
treatment or [Ca 2+ ] o depletion. Abolition of the action potential can be a result of the
increase in periaxonal K + generated by the long depolarization represented by the
plateau (Shrager, Starkus, Lo & Peracchia, 1983; Frankenhauser & Hodgkin, 1963),
which serves to unblock the delayed rectifier (Dubois & Bergman, 1977), the
increase in [Ca 2+ ] ; serving to enhance K + efflux and limit further Ca2+ influx
(Eckert, Tillotson & Brehm, 1981; Eckert & Ewald, 1982).
In untreated crayfish axons, a [K+]o-induced depolarization causes Rm to decrease
at normal rest Em (Lieberman, 1979) suggesting that high [K + ] o opens the voltagesensitive K + channels and maintains them in a conducting condition even when the
potential is returned to the level expected in control [K + ] o by an applied current
(Fig. 7). In Pt-W-altered crayfish axons, an increase of [K + ] o causes almost no
change in Rm at resting and hyperpolarized levels although voltage sensitivity of the
outward rectifier is re-established. Under conditions where high [K + ] o would be
expected to open outward rectifying channels (Dubois & Bergman, 1977), they
remain closed under the influence of Pt-W.
In cardiac muscle, a slowly activated Ca2+ current plays a major role in producing
and maintaining the action potential plateau (Rougier et al. 1969; Cranefield,
Aronson & Wit, 1974). It is likely that one effect of the Pt-W product is to increase
Ca2+ influx during AP generation to the point that it makes a significant contribution
to plateau formation. Moody (1980) found that internal acidification of crayfish slow
muscle fibres caused a decrease in outward rectification and an increased voltage
contribution of Ca2+ influx, leading to all-or-none Ca2+ action potentials. TEA also
permitted the generation of all-or-none Ca2+ action potentials, suggesting that
inward Ca + current, normally present but shunted by the voltage-sensitive outward
K + current, was now able to modify the membrane potential. In the Pt-W-altered
axon, the increased inward Ca2+ current contributes to the generation of the AP
plateau with a similar decrease in outward rectification.
Although a prolonged inward current (Ca2+ or Na + ) is a sufficient explanation for
the apparent inward rectification (Fig. 3) and the increased Rm during the plateau
(Fig. 4) induced by Pt-W, the contribution of an inward rectifying channel cannot be
ruled out except with voltage-clamp studies of the currents flowing during the
plateau. External Ba2+, did not result in an increased membrane resistance at resting
14
L. A. ORR AND E. M. LIEBERMAN
or hyperpolarized potentials, as might be expected if inwardly rectifying channel^
were present in the membrane. However, inward rectifiers 'created' by the action of
Pt-W would not necessarily be Ba2+-sensitive.
An aspect of the Pt-W effect not seen in the action potential of cardiac muscle is the
slow but continuous depolarization that usually begins soon after plateau formation.
The most likely explanation for this involves the influence of relative chloride
permeabilities in nerve and muscle membrane. Crayfish axon membrane has a
relatively low chloride permeability (Lieberman & Nosek, 1976; Strickholm &
Clark, 1977) compared to muscle membrane (Adrian, 1969). A K + permeability
which decreases during depolarization (inward rectification) would be a liability in
muscle fibres if it were not damped by a high chloride permeability, because it would
otherwise lead to an unstable resting potential. The Pt-W-altered axon contains K +
channels which fail to increase their permeability during depolarization. Removing
chloride from the external solution of Pt-W-altered axons has no effect on AP kinetics
or on the rate of depolarization, indicating that the chloride permeability of the
Pt-W-altered axon remains relatively small and thus provides a plausible explanation
for the Pt-W-induced depolarization.
Tungsten wire serves as a good substitute for Pt in the construction of axial wire
electrodes for quantitative electrophysiological studies of axons. We are using these
electrodes regularly in this laboratory on crayfish giant axons with appropriate
precautions to prevent their reactivity with axoplasm, as described in this study. The
primary advantage of tungsten is its mechanical strength, as compared to Pt of the
same diameter, allowing the use of smaller wire for studies on axons with diameters
as small as 100 fim.
The technique used by Nussbaumer (1981) to etch tungsten overcomes the
necessity to platinize the tungsten wire, thus avoiding problems of electrode
reactivity. If platinization is desirable a low-current, long-term platinization procedure will provide an even, full coverage of tungsten preventing its reactivity with
axoplasm.
As described in this study, the Pt-W electrode may serve as a useful tool to
generate H + in a controlled manner for studies of H + transport, its effect on
electrical properties of membranes and relationships to biochemical structure. The
advantage of the electrode is that it can be used in intact axons and avoids the
problem of external membrane surface exposure to agents used to change pH; such as
CO2 or NH4"1" or to problems associated with replacement of the axoplasm, in whole
or in part, with artificial solutions.
Finally, the events related to the alteration of a crayfish axon by Pt-W, which
causes conductance changes resembling those expected in cardiac cells, are
schematically represented in Fig. 9 and may provide some insights into the relationship between different types of excitable membranes. It is unlikely that the Pt-W
product creates channels de novo considering the speed of onset of the Pt-W effect.
All channels responsible for the Pt-W effect are therefore assumed to be channels
already present in the membrane but structurally modified or chemically inhibited to
be non-functional. An agent which alters ionic currents in one membrane type so that
Cardiac-like action potentials in axons
[Na+]0+[Ca
Pt-W.
15
gK£
H+
+
X?
TTX
gNa+
Verapamil
Fig. 9. The mode of action of Pt-W leading to action potential elongation. It is uncertain
what the immediate product of the Pt-W and axoplasm reaction is (X?) in addition to H + .
The product appears to reduce the voltage sensitivity of the outward rectifier (gK vt ) in a
manner dependent on [Na + ] and [Ca 2+ ] in the external solution. Depletion of either ion
reduces the block while tetrodotoxin (TTX), verapamil or La 3+ prevents the AP
elongation but not the block of the outward rectifier. In order for the background steadystate K + conductance (gKM) to be reduced both Na + and Ca 2+ fluxes are required. The
Pt-W/axoplasm product appears to prevent opening of a proportion of the so-called
steady-state 'leak* channels carrying K + outwardly under certain conditions (high
external [K + ]). Whether the development of the inward (anomalous) rectifier occurs and
is important to the development of the plateau is unresolved at this time.
they resemble ionic currents in another suggests that excitable membranes possess
similar ionic channels. Evidence exists that unitary Ca2+ currents in nerve of three
different species have similar kinetics (Brown, Camerer, Kunze & Lux, 1982). The
difference between different classes of excitable membranes (i.e. nerve vs muscle)
could be due to modifications or 'masking' influences on the membrane channels.
The products of the reaction of Pt-W with axoplasm may add or remove such a
masking influence from a particular channel type, so that it responds in a manner
characteristic of a different class of membrane.
The authors are appreciative of the technical assistance of J. Pascarella, S. Hassan
and Dr A. M. Butt during the course of this work and the secretarial assistance of
Brenda Elks and Denise Wilson in preparing the manuscript. This work was
supported in part by a grant from Sigma Xi (to LAO) and a grant from the Army
Research Office DAAG 29-82-K-0182 (to EML).
REFERENCES
ADRIAN, R. H. (1969). Rectification in muscle membrane. Prog. Biophys. molec. Biol. 19, 339-369.
BORON, W. F. & DEWEER, P. (1976). Intracellular pH transients in squid giant axons caused by
CO 2 , NH 3 , and metabolic inhibitors. J. gen. Physiol. 67, 91-112.
BROWN, A. M., CAMERER, H., KUNZE, D. L. & Lux, H. D. (1982). Similarity of unitary C a + + in
three different species. Nature, Land. 299, 156-158.
CARBONE, E., PRIR, L. T . & WANKE, E. (1981). Intracellular pH and ionic channels in theLoligo
vulgaris giant axon. Biophys. J. 35, 393-413.
16
L. A. ORR AND E. M. LIEBERMAN
CORSON, D. W., GOODMAN, S. & FEIN, A. (1979). An adaptation of the jet stream microelectrode
beveler. Science 205, 1302.
CRANEFIELD, P. F., ARONSON, R. S. & Wrr, A. L. (1974). Effect of verapamil on the normal action
potential and on a calcium dependent slow response of canine cardiac Purkinje fibers. Circulation
Res. 34, 204-213.
ECKERT, R. & EWALD, D . (1982). Residual calcium ions depress activation of calcium-dependent
current. Science 216, 730—733.
ECKERT, R., TILLOTSON, D . L. & BREHM, P. (1981). Calcium-mediated control of Ca and K
currents. Fedn Proc. Fedn Am. Socs exp. Biol. 40, 2226-2232.
DUBOIS, J. M. & BERGMAN, C. (1977). The steady-state potassium conductance of the Ranvier
node at varius external K-concentrations. Pflugers Arch. ges. Physiol. 370, 185-194.
FRANKENHAUSER, B. & HODGKJN, A. L. (19S6). The after-effects of impulses in the giant nerve
fibers oiLoligo.J. Physiol., Land. 131, 341-376.
GETTES, L. & REUTER, H. (1974). Slow recovery from inactivation of inward currents in
mammalian myocardial fibres. .7. Physiol., Land. 240, 703-724.
GOLDMAN, Y. &MORAD, M. (1977). Ionic conductance during the time course of the cardiac action
potential. J. Physiol., Lond. 268, 655-695.
HAGIWARA, S. & YOSHII, M. (1979). Effects of internal potassium and sodium on the anomalous
rectification of the starfish egg as examined by internal perfusion. J. Physiol., Lond. 292,
251-265.
KIRSCH, G. E. & NARAHASHI, T . (1978). 3,4-Diaminopyridine, a potent new potassium channel
blocker. Biophys.J. 22, 507-512.
LIEBERMAN, E. M. (1979). Effect of external potassium on the coupled sodium:potassium
transport ratio of axons. Pflugers Arch, ges. Physiol. 378, 243-249.
LIEBERMAN, E. M. & LANE, T. G. (1976). The influence of cardioactive steroids, metabolic
inhibitors, temperature and sodium on membrane conductance and potential of crayfish giant
axons. Pflugers Arch. ges. Physiol. 366, 189-193.
LIEBERMAN, E. M. & NOSEK, T . M. (1976). The influence of chloride on the ouabain-sensitive
membrane potential and conductance of crayfish giant axons. Pflugers Arch. ges. Physiol. 366,
195-202.
MOODY, W. JR (1980). Appearance of calcium action potentials in crayfish slow muscle fibres under
conditions of low intracellular pH. J. Physiol., Lond. 302, 335-346.
MuLLINS, L. J., REQUENA, J. & WHITTENBURG, J. (1985). Ca entry in squid axons during voltage
clamp pulses is mainly Na/Ca exchange. Proc. natn. Acad. Sci. U.SA. 82, 1847-1851.
NARAHASHI, T . (1974). Chemicals as tools in the study of excitable membranes. Physiol. Rev. 54,
813-889.
NUSSBAUMER, J. C. (1981). A new low-resistance tungsten microelectrode..7. Neuwsci. Melh. 3,
247-250.
ORR, L. A. & LIEBERMAN, E. M. (1983). The ionic basis for elongated action potentials in crayfish
giant axons induced by platinized-tungsten metal. Biophys.J. 41, 387a.
REUTER, H. (1967). The dependence of the slow inward current on external calcium concentration
in Purkinje fibres. J. Physiol., Lond. 192, 479-492.
ROUGIER, O., VASSORT, G., GARNIER, D., GARGOUGH, Y. M. & CORABOEUF, E. (1969). Existence
and role of a slow inward current during frog atrial action potential. Pflugers Arch ges. Physiol.
308,91-110.
SHRAGER, P. A., STARKUS, J. C , Lo, M.-V. & PERACCHIA, C. (1983). The periaxonal space of
crayfish giant axons. J. gen. Physiol. 82, 221-244.
SHRAGER, P. G., MACEY, R. I. & STRICKHOLM, A. (1969). Internal perfusion of crayfish axons:
Effects.of tannic acid, D D T , and TEA. J . cell. Physiol. 74, 77-90.
SHRAGER, P. G., STRICKHOLM, A. & MACEY, R. I. (1969). Chemical modification of crayfish axons
by protein crosslinking aldehydes, J. cell. Physiol. 74, 91-100.
STANDEN, N. B. & STANFIELD, P. R. (1978). Inward rectification in skeletal muscle: A blocking
particle model. Pflugers Arch. ges. Physiol. 378, 173-176.
STRICKHOLM, A. & CLARK, H. R. (1977). Ionic permeability of K, Na, and Cl in crayfish nerve.
Biophys.J. 19,29-48.
VAN HARREVELD, A. (1936). A physiological solution for freshwater crustaceans. Proc. Soc. exp.
Biol. 34, 428-433.
Cardiac-like action potentials in axons
17
'WALUN, B. G. (1967). Intracellular ion concentrations in single crayfish axons. Ada phvsiol.
scand. 70, 419-430.
WANKE, E., CARBONE, E. & TESTA, P. L. (1979). K + conductance modified by a titratable group
accessible to protons from the intracellular side of the squid axon membrane. Biophvs. J. 26,
319-324.
WIEDMANN, S. (1951). Effect of current flow on the membrane potential of cardiac muscle.
J. Physiol., Land. 115, 227-236.
YAMAGISHI, S. & GRUNDFEST, H. (1971). Contributions of various ions to the resting and action
potentials of crayfish medial giant axons. J. Membr. Biol. 5, 345-365.