- Email: [email protected]
Scorpion toxin prolongs an inactivation phase of the voltage-dependent sodium current in rat isolated single hippocampal neurons
Brain Research, 487 (1989) 192-195
192
Elsevier BRE 23505
Scorpion toxin prolongs an inactivation phase of the voltage-dependent sodium current in rat isolated single hippocampal neurons M. Kaneda, Y. Oyama, Y. Ikemoto and N. Akaike Department of Physiology, Facultyof Medicine, Kyushu University, Fukuoka (Japan) (Accepted 31 January 1989)
Key words: Rat hippocampal pyramidal neuron; Scorpion toxin; Voltage-dependent sodium current; Sodium current inactivation
The effects of scorpion toxin on the voltage-dependent sodium current (INa) of CA1 pyramidal neurons isolated from rat hippocampus were studied under the single-electrode voltage-clamp condition using a 'concentration-clamp' technique. The toxin increased the peak amplitude of INa and prolonged its inactivation phase in a time- and dose-dependent manner. Inactivation phase of INa proceeded with two exponential components in the absence (control) and presence of the toxin. In the toxin-treated neurons, both the time constant of slow component and its fractional contribution to the total current increased dose-dependently while the fractional contribution of the fast one decreased in a dose-dependent fashion without changing its time constant. Actions of scorpion toxin on the sodium channels of hippocampal pyramidal neurons were essentially similar to those of peripheral preparations. Therefore, it can be concluded that the sodium channels of mammalian brain neurons have structures and functions similar to peripheral channels.
Scorpion toxin is useful to study the sodium channel structure and function 6. Effects of scorpion toxin on the voltage-dependent sodium current (INa) have been mainly studied in squid axonal membrane 12 and frog myelinated nerve 5'11. Recently, we have successfully developed the isolation technique of single neurons from various regions of mammalian central nervous system (CNS) and confirmed the suitability of these neurons to study the voltagedependent sodium channels of mammalian CNS neurons 9'1°. Until now, there is little information about the property of sodium channels in mammalian CNS neurons. Therefore, in the present study, we have studied the effects of scorpion toxin on the INa of hippocampal pyramidal neurons isolated freshly from young rats and provided additional information on the property of CNS sodium channels. Hippocampal pyramidal neurons (CA1 region) of 2-week-old rats were isolated as described previously 2'3'9'1°. In brief, hippocampal slices were treated with 0.1% collagenase and 0.2% pronase E
for about 15-60 min at the temperature of 37 °C. Thereafter the hippocampal fragments of the CA1 region were separated into single neurons by gentle pipetting using a fine glass tube. Experiments were performed on the neurons perfused with internal solution (in mM): N-methyl-D-glucamine-F 100, NaF 30 and H E P E S 10 (pH 7.2); and external solution (in raM): NaC1 50, CsCI 5, MgCI 2 2, sucrose 200, glucose 10 and H E P E S 10 (pH 7.4), using a 'concentration-clamp' technique 1. The current and voltage were recorded by a patch clamp amplifier (List-Medical, EPC), monitored on a digital storage oscilloscope (National, VP-5730OA) and simultaneously stored on an FM tape recorder ( T E A C , MR-30) for analysis. All electrophysiological experiments were carried out at room temperature (23-
25 oc). Enzymes and toxin used in this study were collagenase (Sigma, type I), pronase E (Kaken Pharmaceutical, actinase E) and scorpion toxin (Sigma, Leiurus Quinquestriatus). The toxin was
Correspondence: N. Akaike, Department of Neurophysiology, Tohoku University School of Medicine, Sendai 980, Japan. 0006-8993/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
193 dissolved in the external solution just before use. Neurons were voltage-clamped at a holding potential of -100 mV and depolarizing pulses of 30 ms in duration were applied to elicit the INa. The maximum peak amplitude of INa was recorded at the membrane potential o f - 3 0 mV in the absence (control) and presence of the toxin. However, to see the action of scorpion toxin on more rapid inactivation phase of INa, the effect was tested at a higher voltage (-10 mV). Because the neurons were perfused internally and externally with K ÷- and Ca 2÷free solution where K ÷ and Ca2+were replaced with equimolar Cs ÷ and Mg2÷, respectively, there were no potassium and calcium currents under these experimental conditions. Scorpion toxin at the dose of 3 x 10- 6 M increased the peak amplitude of INa and prolonged the inactivation phase as shown in Fig. 1. The increase in the peak amplitude of IN~ reached the steady-state within 5 s, while the effect on the inactivation phase was complete within 3 min after the application. Both effects of the toxin on the IN~ did not recover after removing the toxin from the external solution. The IN~ also completely inactivated with 10 s depolarizing step in the presence of scorpion toxin. Fig. 2 shows the semilogarithmic plots of inactivation phase of IN~ that could be fitted with fast and slow exponential components in absence (control) and presence of the toxin. The amplitude of fast and slow components was estimated from the semilogarithmic plots (curve fitting) of the inactivation phase of INa"
Scorpion t o x i n
The value of slow component was given by extrapolating the line responsible for the slow kinetics to a maximum response time in the curve-fitting line of the INa inactivation phase. The value of the fast component was obtained by subtracting the amplitude of the slow from the total amplitude. Scorpion toxin (3 x 10 -6 M ) prolonged the time constant of the slow component and increased its fractional contribution to the total amplitude of INa. On the other hand, the contribution of fast component was decreased by the toxin treatment without affecting its time constant. The contribution of the slow component became prominent by increasing the toxin concentration while that of the fast one decreased as shown in Fig. 3. The slow time constant b ~ e larger with scorpion toxin concentration, while that of the fast component did not change (Fig. 4). One may argue a possibility that scorpion toxin produces another non-specific ionic channel carrying the additional inward current in isolated mammalian CNS neurons. However it is unlikely because the potential to produce maximum peak amplitude and the reversal potential did not change in the presence of the toxin, indicating that the current is still
1.0. Control
0.1"
3xlO-SM
!
t ............................................................................. :
Cont.
0
0.05
J
0.1
,~ 1.0. corpion toxin n-
0.12 msec Fig. 1. Modification of the voltage-dependent sodium current by scorpion toxin. Superimposed current tracings were recorded before and 5 and 30 s after the start of toxin application. The dotted line over the current tracings indicates the level of a holding current o f - 8 0 mV. The sodium currents were evoked by a depolarizing step to -10 mV. The arrow between the current tracings shows the direction of change in the inactivation phase of sodium current after the start of toxin treatment.
I
0
0.05
1
0,1
TIME (s) Fig. 2. Semilogarithmic plots of the inactivation phase of sodium current with or without scorpion toxin (3 x 10--6 M). Upper panel, the control. Lower panel, the inactivation phase of sodium current in the neuron treated with the toxin.
194 dependent on Na ÷ (presumably the same INa)- The sodium channels have at least 4 receptive sites for several neurotoxins. They are classified as the sites 1, 2, 3 and 4 and scorpion toxin is susceptible to the site 3 which also accepts sea anemone toxin 4'6. Present study shows that scorpion toxin increased the peak amplitude of INa in rat hippocampal pyramidal neurons and prolonged its inactivation phase as being similar to those previously reported in squid axon and frog myelinated nerve 5'12. In unpublished results (Kaneda et al.), sea anemone toxin also did so on the INa of this preparation, and tetrodotoxin which binds to site 1 did not change the rate of inactivation phase of INaTM. Therefore, the sodium channels of isolated rat hippocampal pyramidal neurons are also confirmed to have the site 3. If the voltage-dependent inactivation of sodium channels shifts to a more depolarizing direction in the neurons treated with the toxin, the rate of inactivation of the current at the membrane potential employed here should be delayed. However this possibility is also unlikely because the INa even at the threshold potential completely inactivated within 100 ms in the control while the INa of the neurons treated with scorpion toxin persisted for 5 s or more. Therefore, scorpion toxin is thought to directly suppress the inactivation of sodium channels. However, the increase in the amplitude of IN~ by the
toxin treatment is thought not to be simply due to the reduced inactivation of the sodium channels, yet it seems to be largely dependent on the modification of channel inactivation by the toxin. Thus, the effect of scorpion toxin on the peak amplitude became a steady level within 5 s while that on the inactivation progressed during the period of about 3 min after the application of the toxin (3 × 10 -6 M). If the increase in the INa amplitude is simply related to the effect of scorpion toxin on the inactivation phase, it would make similar time courses of both effects. At present, the origin of difference between the time courses remains unknown. Further study will be necessary. The inactivation phase of INa in this preparation can be fitted to the sum of fast and slow exponential components, suggesting the presence of two types of the inactivation or the channel, as previously described in rat brain INa expressed by m - R N A on the membrane of Xenopus oocytes 8. It may be useful for elucidating the action mode of scorpion toxin to analyse the effect of the toxin on respective components. In the present study, scorpion toxin dose-dependently increased both the time constant of the slow component and its fractional contribution to the total current while the fractional contribution of the fast one to the total current was reduced without changing its time constant. From these results the possibility can be proposed that
40 1.5
i
~total
// 30
~1.o
/
E --
/
k-.d
•
slow
/
¢1.
20
<
4
W0-5 > m k10
S UJ ¢¢ 0
0
0 I
10-8
0
0
[ IllilLI
I
10 -7 SCORPION
fast u
1 I II[IH 10 .6
~:D
t
]
T O X I N (M}
Fig. 3. Concentration-dependent changes in the fast and slow time constants in the inactivation of sodium current. Each point represents the mean value and S.E.M. in 4-6 experi-
ments.
i
~
r I 10-8
]
[ [ l/till
I
I t IIIIII
10 -7 SCORPION
TOXIN
fast
]
I I I
10 -6 (M)
Fig. 4. Concentration-dependent changes in the relative amplitude of fast and slow current components and the total current. Each point indicates the mean relative value and S.E.M. in 4-5 experiments. The symbols at the left end are the control values.
195
scorpion toxin transforms the channels forming the fast exponential c o m p o n e n t into those for the slow c o m p o n e n t of which the time constant is profoundly increased by the toxin. A n increase in the time constant of slow c o m p o n e n t m a y reflect the changes in channel kinetics, an increase in open time, a decrease in closed time or both. Results from single channel recordings will give us this answer. In conclusion, the actions of scorpion toxin on the sodium channels of rat h i p p o c a m p a l pyramidal neu1 Akaike, N., Inoue, M. and Krishtal, O.A., 'Concentration clamp' study of y-aminobutyric-acid-inducedchloride current kinetics of frog sensory neurones, J. Physiol. (Lond.), 379 (1986) 171-185. 2 Akaike, N., Kaneda, M., Hori, N. and Krishtal, O.A., Blockade of N-methyl-o-aspartate response in enzymetreated rat hippocampal neurons, Neurosci. Lett., 87 (1988) 75-79. 3 Akaike, N., Kostyuk, EG. and Oscipchuk, Y.V., Dihydropyridine-sensitive low-threshold calcium channels in isolated rat hypothalamic neurones, J. Physiol. (Lond.), in press. 4 Barchi, R.L., Probing the molecular structure of the sodium channel, Annu. Rev. Neurosci., 11 (1988) 455-495. 5 Benoit, E. and Dubois, J., Properties of maintained sodium current induced by a toxin from Androctonus scorpion in frog node of Ranvier, J. Physiol. (Lond.), 383 (1987) 93-114. 6 Catterall, W.A., Molecular properties of voltage-sensitive sodium channels, Annu. Rev. Biochem., 55 (1986) 953985. 7 Gonoi, T., Hille, B. and Catterail, W.A., Voltage clamp analysis of sodium channels in normal and scorpion toxin-resistant neuroblastoma cells, J. Neurosci., 4 (1984)
rons are essentially similar to those r e p o r t e d in o t h e r p r e p a r a t i o n s 5'7"12. T h e r e f o r e , it is likely that the sodium channels in m a m m a l i a n brain neurons have a structure and function similar to those of peripheral ones. This study was s u p p o r t e d by Grants-in A i d to N. A k a i k e (62870102, 634880107 and 63641526) and to Y. I k e m o t o (63570059) from the J a p a n e s e Ministry of Education, Science and Culture. 2836-2842. 8 Gundersen, C.B., Miledi, R.ER.S. and Parker, I., Voltage-operated channels induced by foreign messenger RNA in Xenopus oocytes, Proc. R. Soc. Lond. Ser. B, 220 (1983) 131-140. 9 Kaneda, M., Nakamura, H. and Akaike, N., Mechanical and enzymatical isolation of mammalian CNS neurons, Neurosci. Res., 5 (1988) 299-315. 10 Kaneda, M., Oomura, Y., Ishibashi, O. and Akaike, N., Permeability to various cations of the voltage-dependent sodium channel of isolated rat hippocampal pyramidal neurons, Neurosci. Lett., 88 (1988) 253-256. 10a Kaneda, M., Oyama, Y., Ikemoto, Y. and Akaike, N., Blockade of the voltage-dependent sodium current by tetrodotoxin and iidocaine, Brain Research, in press. 11 Mozhayeva, G.N., Naumov, A.P., Nosyreva, E.D. and Grishin, E.V., Potential-dependent interaction of toxin venom of the scorpion Buthus eupeus with sodium channels in myelinated fibre, Biochim. Biophys. Acta, 597 (1980) 587-602. 12 Narahashi, T., Shapiro, B.I., Deguchi, T., Scuka, M. and Wang, C.M. Effects of scorpion venom on squid axon membranes, Am. J. Physiol., 222 (1972) 850-857.
192
Elsevier BRE 23505
Scorpion toxin prolongs an inactivation phase of the voltage-dependent sodium current in rat isolated single hippocampal neurons M. Kaneda, Y. Oyama, Y. Ikemoto and N. Akaike Department of Physiology, Facultyof Medicine, Kyushu University, Fukuoka (Japan) (Accepted 31 January 1989)
Key words: Rat hippocampal pyramidal neuron; Scorpion toxin; Voltage-dependent sodium current; Sodium current inactivation
The effects of scorpion toxin on the voltage-dependent sodium current (INa) of CA1 pyramidal neurons isolated from rat hippocampus were studied under the single-electrode voltage-clamp condition using a 'concentration-clamp' technique. The toxin increased the peak amplitude of INa and prolonged its inactivation phase in a time- and dose-dependent manner. Inactivation phase of INa proceeded with two exponential components in the absence (control) and presence of the toxin. In the toxin-treated neurons, both the time constant of slow component and its fractional contribution to the total current increased dose-dependently while the fractional contribution of the fast one decreased in a dose-dependent fashion without changing its time constant. Actions of scorpion toxin on the sodium channels of hippocampal pyramidal neurons were essentially similar to those of peripheral preparations. Therefore, it can be concluded that the sodium channels of mammalian brain neurons have structures and functions similar to peripheral channels.
Scorpion toxin is useful to study the sodium channel structure and function 6. Effects of scorpion toxin on the voltage-dependent sodium current (INa) have been mainly studied in squid axonal membrane 12 and frog myelinated nerve 5'11. Recently, we have successfully developed the isolation technique of single neurons from various regions of mammalian central nervous system (CNS) and confirmed the suitability of these neurons to study the voltagedependent sodium channels of mammalian CNS neurons 9'1°. Until now, there is little information about the property of sodium channels in mammalian CNS neurons. Therefore, in the present study, we have studied the effects of scorpion toxin on the INa of hippocampal pyramidal neurons isolated freshly from young rats and provided additional information on the property of CNS sodium channels. Hippocampal pyramidal neurons (CA1 region) of 2-week-old rats were isolated as described previously 2'3'9'1°. In brief, hippocampal slices were treated with 0.1% collagenase and 0.2% pronase E
for about 15-60 min at the temperature of 37 °C. Thereafter the hippocampal fragments of the CA1 region were separated into single neurons by gentle pipetting using a fine glass tube. Experiments were performed on the neurons perfused with internal solution (in mM): N-methyl-D-glucamine-F 100, NaF 30 and H E P E S 10 (pH 7.2); and external solution (in raM): NaC1 50, CsCI 5, MgCI 2 2, sucrose 200, glucose 10 and H E P E S 10 (pH 7.4), using a 'concentration-clamp' technique 1. The current and voltage were recorded by a patch clamp amplifier (List-Medical, EPC), monitored on a digital storage oscilloscope (National, VP-5730OA) and simultaneously stored on an FM tape recorder ( T E A C , MR-30) for analysis. All electrophysiological experiments were carried out at room temperature (23-
25 oc). Enzymes and toxin used in this study were collagenase (Sigma, type I), pronase E (Kaken Pharmaceutical, actinase E) and scorpion toxin (Sigma, Leiurus Quinquestriatus). The toxin was
Correspondence: N. Akaike, Department of Neurophysiology, Tohoku University School of Medicine, Sendai 980, Japan. 0006-8993/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
193 dissolved in the external solution just before use. Neurons were voltage-clamped at a holding potential of -100 mV and depolarizing pulses of 30 ms in duration were applied to elicit the INa. The maximum peak amplitude of INa was recorded at the membrane potential o f - 3 0 mV in the absence (control) and presence of the toxin. However, to see the action of scorpion toxin on more rapid inactivation phase of INa, the effect was tested at a higher voltage (-10 mV). Because the neurons were perfused internally and externally with K ÷- and Ca 2÷free solution where K ÷ and Ca2+were replaced with equimolar Cs ÷ and Mg2÷, respectively, there were no potassium and calcium currents under these experimental conditions. Scorpion toxin at the dose of 3 x 10- 6 M increased the peak amplitude of INa and prolonged the inactivation phase as shown in Fig. 1. The increase in the peak amplitude of IN~ reached the steady-state within 5 s, while the effect on the inactivation phase was complete within 3 min after the application. Both effects of the toxin on the IN~ did not recover after removing the toxin from the external solution. The IN~ also completely inactivated with 10 s depolarizing step in the presence of scorpion toxin. Fig. 2 shows the semilogarithmic plots of inactivation phase of IN~ that could be fitted with fast and slow exponential components in absence (control) and presence of the toxin. The amplitude of fast and slow components was estimated from the semilogarithmic plots (curve fitting) of the inactivation phase of INa"
Scorpion t o x i n
The value of slow component was given by extrapolating the line responsible for the slow kinetics to a maximum response time in the curve-fitting line of the INa inactivation phase. The value of the fast component was obtained by subtracting the amplitude of the slow from the total amplitude. Scorpion toxin (3 x 10 -6 M ) prolonged the time constant of the slow component and increased its fractional contribution to the total amplitude of INa. On the other hand, the contribution of fast component was decreased by the toxin treatment without affecting its time constant. The contribution of the slow component became prominent by increasing the toxin concentration while that of the fast one decreased as shown in Fig. 3. The slow time constant b ~ e larger with scorpion toxin concentration, while that of the fast component did not change (Fig. 4). One may argue a possibility that scorpion toxin produces another non-specific ionic channel carrying the additional inward current in isolated mammalian CNS neurons. However it is unlikely because the potential to produce maximum peak amplitude and the reversal potential did not change in the presence of the toxin, indicating that the current is still
1.0. Control
0.1"
3xlO-SM
!
t ............................................................................. :
Cont.
0
0.05
J
0.1
,~ 1.0. corpion toxin n-
0.12 msec Fig. 1. Modification of the voltage-dependent sodium current by scorpion toxin. Superimposed current tracings were recorded before and 5 and 30 s after the start of toxin application. The dotted line over the current tracings indicates the level of a holding current o f - 8 0 mV. The sodium currents were evoked by a depolarizing step to -10 mV. The arrow between the current tracings shows the direction of change in the inactivation phase of sodium current after the start of toxin treatment.
I
0
0.05
1
0,1
TIME (s) Fig. 2. Semilogarithmic plots of the inactivation phase of sodium current with or without scorpion toxin (3 x 10--6 M). Upper panel, the control. Lower panel, the inactivation phase of sodium current in the neuron treated with the toxin.
194 dependent on Na ÷ (presumably the same INa)- The sodium channels have at least 4 receptive sites for several neurotoxins. They are classified as the sites 1, 2, 3 and 4 and scorpion toxin is susceptible to the site 3 which also accepts sea anemone toxin 4'6. Present study shows that scorpion toxin increased the peak amplitude of INa in rat hippocampal pyramidal neurons and prolonged its inactivation phase as being similar to those previously reported in squid axon and frog myelinated nerve 5'12. In unpublished results (Kaneda et al.), sea anemone toxin also did so on the INa of this preparation, and tetrodotoxin which binds to site 1 did not change the rate of inactivation phase of INaTM. Therefore, the sodium channels of isolated rat hippocampal pyramidal neurons are also confirmed to have the site 3. If the voltage-dependent inactivation of sodium channels shifts to a more depolarizing direction in the neurons treated with the toxin, the rate of inactivation of the current at the membrane potential employed here should be delayed. However this possibility is also unlikely because the INa even at the threshold potential completely inactivated within 100 ms in the control while the INa of the neurons treated with scorpion toxin persisted for 5 s or more. Therefore, scorpion toxin is thought to directly suppress the inactivation of sodium channels. However, the increase in the amplitude of IN~ by the
toxin treatment is thought not to be simply due to the reduced inactivation of the sodium channels, yet it seems to be largely dependent on the modification of channel inactivation by the toxin. Thus, the effect of scorpion toxin on the peak amplitude became a steady level within 5 s while that on the inactivation progressed during the period of about 3 min after the application of the toxin (3 × 10 -6 M). If the increase in the INa amplitude is simply related to the effect of scorpion toxin on the inactivation phase, it would make similar time courses of both effects. At present, the origin of difference between the time courses remains unknown. Further study will be necessary. The inactivation phase of INa in this preparation can be fitted to the sum of fast and slow exponential components, suggesting the presence of two types of the inactivation or the channel, as previously described in rat brain INa expressed by m - R N A on the membrane of Xenopus oocytes 8. It may be useful for elucidating the action mode of scorpion toxin to analyse the effect of the toxin on respective components. In the present study, scorpion toxin dose-dependently increased both the time constant of the slow component and its fractional contribution to the total current while the fractional contribution of the fast one to the total current was reduced without changing its time constant. From these results the possibility can be proposed that
40 1.5
i
~total
// 30
~1.o
/
E --
/
k-.d
•
slow
/
¢1.
20
<
4
W0-5 > m k10
S UJ ¢¢ 0
0
0 I
10-8
0
0
[ IllilLI
I
10 -7 SCORPION
fast u
1 I II[IH 10 .6
~:D
t
]
T O X I N (M}
Fig. 3. Concentration-dependent changes in the fast and slow time constants in the inactivation of sodium current. Each point represents the mean value and S.E.M. in 4-6 experi-
ments.
i
~
r I 10-8
]
[ [ l/till
I
I t IIIIII
10 -7 SCORPION
TOXIN
fast
]
I I I
10 -6 (M)
Fig. 4. Concentration-dependent changes in the relative amplitude of fast and slow current components and the total current. Each point indicates the mean relative value and S.E.M. in 4-5 experiments. The symbols at the left end are the control values.
195
scorpion toxin transforms the channels forming the fast exponential c o m p o n e n t into those for the slow c o m p o n e n t of which the time constant is profoundly increased by the toxin. A n increase in the time constant of slow c o m p o n e n t m a y reflect the changes in channel kinetics, an increase in open time, a decrease in closed time or both. Results from single channel recordings will give us this answer. In conclusion, the actions of scorpion toxin on the sodium channels of rat h i p p o c a m p a l pyramidal neu1 Akaike, N., Inoue, M. and Krishtal, O.A., 'Concentration clamp' study of y-aminobutyric-acid-inducedchloride current kinetics of frog sensory neurones, J. Physiol. (Lond.), 379 (1986) 171-185. 2 Akaike, N., Kaneda, M., Hori, N. and Krishtal, O.A., Blockade of N-methyl-o-aspartate response in enzymetreated rat hippocampal neurons, Neurosci. Lett., 87 (1988) 75-79. 3 Akaike, N., Kostyuk, EG. and Oscipchuk, Y.V., Dihydropyridine-sensitive low-threshold calcium channels in isolated rat hypothalamic neurones, J. Physiol. (Lond.), in press. 4 Barchi, R.L., Probing the molecular structure of the sodium channel, Annu. Rev. Neurosci., 11 (1988) 455-495. 5 Benoit, E. and Dubois, J., Properties of maintained sodium current induced by a toxin from Androctonus scorpion in frog node of Ranvier, J. Physiol. (Lond.), 383 (1987) 93-114. 6 Catterall, W.A., Molecular properties of voltage-sensitive sodium channels, Annu. Rev. Biochem., 55 (1986) 953985. 7 Gonoi, T., Hille, B. and Catterail, W.A., Voltage clamp analysis of sodium channels in normal and scorpion toxin-resistant neuroblastoma cells, J. Neurosci., 4 (1984)
rons are essentially similar to those r e p o r t e d in o t h e r p r e p a r a t i o n s 5'7"12. T h e r e f o r e , it is likely that the sodium channels in m a m m a l i a n brain neurons have a structure and function similar to those of peripheral ones. This study was s u p p o r t e d by Grants-in A i d to N. A k a i k e (62870102, 634880107 and 63641526) and to Y. I k e m o t o (63570059) from the J a p a n e s e Ministry of Education, Science and Culture. 2836-2842. 8 Gundersen, C.B., Miledi, R.ER.S. and Parker, I., Voltage-operated channels induced by foreign messenger RNA in Xenopus oocytes, Proc. R. Soc. Lond. Ser. B, 220 (1983) 131-140. 9 Kaneda, M., Nakamura, H. and Akaike, N., Mechanical and enzymatical isolation of mammalian CNS neurons, Neurosci. Res., 5 (1988) 299-315. 10 Kaneda, M., Oomura, Y., Ishibashi, O. and Akaike, N., Permeability to various cations of the voltage-dependent sodium channel of isolated rat hippocampal pyramidal neurons, Neurosci. Lett., 88 (1988) 253-256. 10a Kaneda, M., Oyama, Y., Ikemoto, Y. and Akaike, N., Blockade of the voltage-dependent sodium current by tetrodotoxin and iidocaine, Brain Research, in press. 11 Mozhayeva, G.N., Naumov, A.P., Nosyreva, E.D. and Grishin, E.V., Potential-dependent interaction of toxin venom of the scorpion Buthus eupeus with sodium channels in myelinated fibre, Biochim. Biophys. Acta, 597 (1980) 587-602. 12 Narahashi, T., Shapiro, B.I., Deguchi, T., Scuka, M. and Wang, C.M. Effects of scorpion venom on squid axon membranes, Am. J. Physiol., 222 (1972) 850-857.