- Email: [email protected]
A transient outward cationic current activated by Na+ influx into frog visual neurones in vitro
ELSEVIER
Neuroscience Letters 205 (I 996) 5-8
NEUROSCIENC[ LETTERS
A transient outward cationic current activated by Na ÷ influx into frog visual neurones in vitro A. Zaykin ~, A. Nistri* Biophysics Laboratory, International School for Advanced Studies (S.LS.S.A.), 34013 Trieste, Italy
Received 14 October 1995; revised version received 12 December 1995; accepted 18 December 1995
Abstract
Whole cell patch clamp recording from neurones of a slice preparation of the frog optic tectum showed that depolarizing step commands from -70 mV holding potential generated a fast inward current always followed by a fast outward current. The fast outward current was blocked by tetrodotoxin (TFX) or 4-aminopyridine (4-AP), or by replacing external Na÷ with Li÷. When the patch pipette contained Cs+ instead of K÷ the outward current was fully preserved, suggesting that the membrane channels responsible for this response were relatively non-selective in their permeability properties. This current (termed /cat) is a novel example of a non-specific cationic current dependen~t on influx of Na÷ and presumably important to control the firing characteristics of these cells. Keywords: Sodium current; Potassium current; Patch clamp; Cationic current; Optic rectum
Among the large family of K ÷ currents which control the excitability of nerve cells, considerable interest has recently been attracted by one such a current (termed IK(Na)) which is gated by Na ÷ acting on the internal side of the membrane [2]. Pharmacologically, IK(Na/ is blocked by external Li ÷ or tetrodotoxin [3]. Since the concentration of intracellular Na ÷ needed to activate IK(Na) is >30 mM, it may seem unlikely that this current could be activated under physiological conditions [2], even if its control over neuronal firing has been proposed [3]. In the course of our preliminary experinaents on primary relay neurones of the frog optic tectum we observed a fast outward current with characteristics similar to those of IK(Na), except that it was readily activated by voltage-gated Na ÷ influx, suggesting that such a current might be operative each time an action potential was generated [6]. The present report provides a first description of some basic properties of this novel current. Experiments were carried out on neurones from layers VI and IX [5]of a slice preparation of the frog (Rana temporaria or esculenta) optic tectum in vitro. Under tricaine * Corresponding author. Tel.: +39 40 3787228; fax: +39 40 377528; e-mail: [email protected]. I Permanent address: Institute of Cell Biophysics, Russian Academy of Sciences, Pushchino, Moscow Region, Russian Federation.
anaesthesia the brain was removed and transverse sections ( 2 0 0 - 3 0 0 ~ m ) of the optic tectum were cut with a vibratome at 4°C. After at least 1 h equilibration at room temperature, individual slices were transferred to a recording chamber and continuously superfused with gassed (95% 02/5% CO2) Ringer solution (in mM: NaCI, 100; KCI, 2.5; CaCI 2, 1.5; NaH2PO4, 0.1; MgCI 2, 1.5; NaHCO 3, 17; glucose, 4; pH 7.35). Slices remained viable for up to 24 h. Neurones were recorded under whole cell patch clamp conditions with electrodes (6-8 Mfl) filled (in mM) with K-gluconate (or CsSO4), 100; NaC1, 5; EGTA, 10; MgC12, 4; HEPES, 2.5; ATP, 3 (pH 7.1) after first establishing gigaohm seals. Cell input resistance was 2 5 0 - 3 5 0 M f l ; uncompensated series resistance was always <10% of this value and constantly monitored throughout the experiment. Cell capacitance was calculated to be 0.3-3.2 pF on the basis of 3-10/2m cell body diameter [5]. Responses, elicited by voltage commands (usually 4 - 3 0 ms long: incrementing in 5 mV steps), were recorded via a List EPC-7 amplifier and analyzed with a commercial software (pCLAMP 5.5; Axon Instruments). Online leak current subtraction was performed using the P/4 method of the p C L A M P program. Drugs were applied via the bathing solution. The database of the present study comprises 40 neurones clamped at holding potentials between - 5 0 and
0304-3940/96/$12.00 © 1996 Elsevier Science Ireland Ltd. All rights reserved SSDI 0304-3940(95) 12330-Q
6
A. Zaykin, A. Nistri/Neuroscience Letters 205 (1996) 5-8
A
0.8
onset of TTX effect
2-
B
control / - , - ' ~ .
control
'~ 20 s TTX ** 30 s TTX
,
nA
1 nA 0 kSJ
0
ms
7 d
ms
55
7
-0.8
-I
D
nA o n s e t o f Li ÷ e f f e c t 0.6 ~ c o n t r o l J'
-a
~ - - .
C 0.8 nA 0
-0.~
-0.60.55 control
0 551 ~ 0 i
recovery
ms
10
Fig. 1. Current responses of frog optic tectum neurone. (A) Step depolarization to +10 mV from - 7 0 mV holding potential elicits, in control solution, an early inward current followed by slower outward current. After 1 min application of TTX step commands are applied at 10 s intervals. Note progressive decrease in inward and outward currents which unmasks a slower, TTX-insensitive outward component. Tracings are superimposed to aid comparison. (B) TTX-sensitive currents are obtained by subtracting the TTX-resistant current from the control one. Asterisks indicate at which times after TTX application data are recorded. Note concomitant depression of inward and outward components without shift in their peak. (C) The responses after 10 and 20 s are scaled to the size of the control one and superimposed to the latter to show their identical shape. (D) Currents induced by step command to +30 mV from - 7 0 mV holding potential in control solution and following gradual replacement of extracellular Na + with Li + (after 1 min application of the Li + solution test pulses are delivered at 20 s intervals). Note selective and progressive block of the outward fast current without change in the inward current (top tracings). After 10 rain washout of Li + solution the outward current recovers almost completely (bottom tracings). Cell in (D) different from (A-C).
- 7 0 m V . Fig. 1A shows that a depolarizing step to +30 mV from - 7 0 mV holding potential elicited a biphasic current (recorded with a K-gluconate filled pipette), which was first inward and then turned outward (see trace labelled as control). In the absence of permeable intracellular anions, the fast outward current was presumably due to efflux of K ÷. This type of response was also observed when external Ca 2÷ was replaced by Co 2÷. Fig. 1A also shows superimposed tracings obtained after the start of tetrodotoxin (TTX; 1/tM) superfusion with gradual depression of both current components and unmasking of a residual slower and shallower outward current. The TTXresistant current was then subtracted from the control current to demonstrate the isolated, TTX-sensitive currents (Fig. 1B). Further analysis indicated that, after scaling and superimposition, the currents recorded during the onset of TTX block had identical time-course as the control currents (Fig. 1C), demonstrating the adequacy of the voltage clamp. In summary then, it appears that qTX induced an analogous reduction in fast currents, regardless their polarity (n = 10). Fig. 1D shows another pharmacological characteristic of the transient outward current of frog visual neurones, namely its selective block by replacing external Na ÷ with Li ÷. Li ÷ gradually depressed the outward current component without affecting the peak or onset of the fast inward current. Full recovery was attained after 10 min washout with control Ringer. Similar data were
observed in nine neurones. Further analysis of the fast outward current was, however, hampered by the underlying presence of the distinct outward K ÷ current (insensitive to TTX or Li ÷ as shown in Fig. IA,D). It was therefore attempted to block the slow current by replacing internal K ÷ with Cs ÷ which is known to attenuate delayer rectifier currents [4]. Against a full block of the slow outward current by intracellular Cs ÷, the fast outward current was insensitive to such a block (n = 20) as demonstrated by the representative tracings obtained from a cell clamped at - 6 0 mV following application of depolarizing step commands (Fig. 2A). The fast inward current peak was always followed by the development of a transient outward current with characteristics similar to those observed with a standard K-gluconate filled electrode. Since the transient outward current was present when K ÷ or Cs ÷ was the main internal cation, it was considered as a relatively non-selective cationic current (termed /cat)TTX (0.5-1 btM) fully blocked both inward and outward currents, indicating that they retained the same pharmacological properties observed with intracellular K ÷. Equimolar replacement of external Na ÷ with K ÷ completely suppressed the inward current and prevented the appearance of/cat e v e n following depolarizing steps to +100 mV (data not shown), confirming that extracellular Na ÷ was required for/cat generation. Fig. 2B shows the tail current obtained by returning to the holding potential (-60 mV)
A. Zaykin, A. Nistri / Neuroscience Letters 205 (1996) 5-8
i.c. Cs +
500-
A
B
pA
f
~
o
0
-320~ . ~ t ~ current 0 ms 14 +6
-40C
mV
-
o
~
C
530[ PA I
-100
280 pAq
."
/
,.~" •- ~ , ' ~
'
mV
~'~%~'~
0j,
D
cont,ol v
~.
/~_ .~-~-~-.~,
100 -560 0
ms
32
-530 L Fig. 2. Currents recorded after the patch electrode had loaded the cell with intracellular Cs + (i.c. Cs+). (A) Representative, superimposed tracings obtained with incrementing step commands (0.1 Hz) in the range -45 mV (single asterisks) to +45 mV (double asterisks) from - 6 0 mV holding potential. Intermediate responses are evoked by steps to -35, -30, -25, -20, -5, + 10 and +30 mV. Traces are averages from eight responses. Note that with i.c. Cs + the inward current is always followed by an outward current. (B) A step to +60 mV induces the inward/outward current followed by a sharp tail current upon repolarization to holding potential. Single response is responsible for noisier record. (C) I/V relation for irtward (open symbols) and outward (filled symbols) currents of the cell shown in (A,B). Arrow indicates data point (asterisk) obtained from tail current protocol (as shown in (B)). Dashed line fitted to filled symbols indicates apparent reversal level (-20 mV) for outward current. (D) Effect of 4-AP on outward current elicited by step to -15 mV from - 6 0 mV holding potential. Note selective depression of outward component (different cell from A-C).
from a depolarizing step to +60 mV. The peak of the tail response was inward (see arrow), almost as if a slow component of the voltage-gated inward current persisted despite voltage-depende, nt inactivation. Nevertheless, the tail current peak was ob:~erved at 7 ms, i.e. at a time when the isolated inward current had fully dissipated (see for instance Fig. 1D with sl:ep to +15 mV). Furthermore, the tail current amplitude was linearly related to the current/voltage (1/V) curve for/'cat (see arrow pointing to asterisk in Fig. 2C), suggesting that permeation of Cs ÷ was perhaps responsible for it. Thus, it was possible to calculate by extrapolation (see dashed line) the apparent reversal potential of/'cat as --20 mV (activation threshold was 45 mV). This negative reversal value suggests that the membrane channels responsible for /'cat possessed mixed permeability to Na t and Cs ÷. The I/V plot (constructed with 5 mV step increments from - 6 0 mV holding poten-
tial; Fig. 2A) shows that the inward current had a peak amplitude at - 2 0 mV from a threshold o f - 4 5 m V and apparently reversed at +80 mV, a value close to the calculated Na ÷ equilibrium potential (+86 mV). Even if/cat w a s not selectively mediated by K ÷, it did retain some pharmacological characteristics of a K ÷ current, namely its sensitivity to a low concentration (0.2 mM) of 4-aminopyridine (4-AP) which in five cells readily blocked Cs ÷mediated Icat (Fig. 2D) without altering the peak of the fast inward current. These observations favour the interpretation that 4-AP mainly acted by removing Jcat rather than by interfering with the fast inward current. The main finding of the present study is the demonstration that visual neurones of the frog brain possess a fast biphasic response generated by depolarization, consisting of an inward current followed by the transient outward current /'cat" Since the inward current was suppressed by T F x or when K ÷ replaced Na ÷, maintained in the absence of Ca 2÷ or when Li ÷ replaced external Na ÷, and had an apparent reversal similar to the Na ÷ equilibrium potential, there is little doubt that the initial inward current was due to depolarization-activated influx of Na ÷. Such an influx triggered /'cat which displayed some features typical of/'K(Na), such as sensitivity to T T X and Li ÷ [3], but it was evidently a relatively non-selective cationic current since Cs ÷ could readily replace K + as the charge carrier. Hence, it was possible to observe the inward current in isolation (with external Li ÷ or 4-AP), while it was not possible to detect /'cat without the preceding inward current. For the latter reason it was difficult to ascertain the precise reversal potential of/'cat using a standard protocol for tail current reversal: this difficulty was also due in part to possible contamination of Ica t tail by reactivation of the preceding fast inward current (via voltagedependent de-inactivation of the fast inward current). Even if tails were studied at a time interval sufficient for full dissipation of the fast inward current, data should therefore be interpreted with caution. Notwithstanding this complication, it is apparent that the tail was inward rather than outward at a potential much less negative than the K + equilibrium potential, a finding which is consistent with its identification as a response mediated by nonselective cationic channel activity. Depolarization alone was insufficient to gate /'cat, as indicated by the experiments in the absence of external Na +. Since at - 6 0 mV holding potential (with 5 mM Na ÷ present in the patch pipette) /'cat w a s absent, it could not contribute to the resting potential. The mechanism linking activation of the Na t current to generation of ,/cat is presently unclear: since /'cat was blocked when Li ÷ replaced external Na ÷, it seems that /'cat was due to influx of Na t via voltage-gated channels, and it would make it a novel representative of a recently-characterized class of nonspecific cationic currents dependent on Na ÷ [7]. Certainly, the ability of a single inward Na ÷ current to elicit/'cat contrasts this response with the more traditional /'K(Na) which
8
A. Zaykin, A. Nistri/Neuroscience Letters 205 (1996) 5-8
normally requires very high intracellular Na + concentrations to be activated. Thus, it seems feasible to suppose that/cat was i n v o l v e d in the control of the spike duration and repetitive firing of frog visual neurones. So far currents analogous to/cat have not been observed in vertebrate brain neurones, although a voltage-gated nonspecific, slow cationic current has been recently reported in m a m m a l i a n brain neurones [1]. It is tempting to suggest that/cat might be a new, important system to control the short-term excitability of a m p h i b i a n neurones to integrate rapid visual signals. W e wish to thank Prof. Franco Conti for helpful discussion of these data. This work was supported by INFM, C N R and M U R S T .
[1] Alzheimer, C., A novel voltage-dependent cation current in rat neocortical neurones, J. Physiol., 479 (1994) 199-205. 12] Dryer, S.E., Na+-activated K+ channels: a new family of largeconductance ion channels, Trends Neurosci., 17 (1995) 155-160. [3] Dryer, S.E., Fuji, J.T. and Martin, A.R., A Na+-activated K+ current in cultured brain stem neurones from chicks, J. Physiol., 410 (1989) 283-296. [4] Hille, B., Ionic Channels of Excitable Membranes, Sinauer, Sunderland, MA, USA, 1992, p. 131. [5] Szdkely, G. and Lazar, G., Cellular and synaptic architecture of the optic rectum. In R. Llin~ and W. Precht (Eds.), Frog Neurobiology, Springer, New York, 1976, pp. 407-434. [6] Zaykin, A. and Nistri, A., A transient Na+-dependent K+ current of neurones of the frog (Rana temporaria) optic tectum in vitro, J. Physiol., 477 (1994) 49P. [7] Zhainazarov, A.B. and Ache, B.W., Na+-activated non-selective cation channels in primary olfactory neurons, J. Neurophysiol., 73 (1995) 1774-1781.
Neuroscience Letters 205 (I 996) 5-8
NEUROSCIENC[ LETTERS
A transient outward cationic current activated by Na ÷ influx into frog visual neurones in vitro A. Zaykin ~, A. Nistri* Biophysics Laboratory, International School for Advanced Studies (S.LS.S.A.), 34013 Trieste, Italy
Received 14 October 1995; revised version received 12 December 1995; accepted 18 December 1995
Abstract
Whole cell patch clamp recording from neurones of a slice preparation of the frog optic tectum showed that depolarizing step commands from -70 mV holding potential generated a fast inward current always followed by a fast outward current. The fast outward current was blocked by tetrodotoxin (TFX) or 4-aminopyridine (4-AP), or by replacing external Na÷ with Li÷. When the patch pipette contained Cs+ instead of K÷ the outward current was fully preserved, suggesting that the membrane channels responsible for this response were relatively non-selective in their permeability properties. This current (termed /cat) is a novel example of a non-specific cationic current dependen~t on influx of Na÷ and presumably important to control the firing characteristics of these cells. Keywords: Sodium current; Potassium current; Patch clamp; Cationic current; Optic rectum
Among the large family of K ÷ currents which control the excitability of nerve cells, considerable interest has recently been attracted by one such a current (termed IK(Na)) which is gated by Na ÷ acting on the internal side of the membrane [2]. Pharmacologically, IK(Na/ is blocked by external Li ÷ or tetrodotoxin [3]. Since the concentration of intracellular Na ÷ needed to activate IK(Na) is >30 mM, it may seem unlikely that this current could be activated under physiological conditions [2], even if its control over neuronal firing has been proposed [3]. In the course of our preliminary experinaents on primary relay neurones of the frog optic tectum we observed a fast outward current with characteristics similar to those of IK(Na), except that it was readily activated by voltage-gated Na ÷ influx, suggesting that such a current might be operative each time an action potential was generated [6]. The present report provides a first description of some basic properties of this novel current. Experiments were carried out on neurones from layers VI and IX [5]of a slice preparation of the frog (Rana temporaria or esculenta) optic tectum in vitro. Under tricaine * Corresponding author. Tel.: +39 40 3787228; fax: +39 40 377528; e-mail: [email protected]. I Permanent address: Institute of Cell Biophysics, Russian Academy of Sciences, Pushchino, Moscow Region, Russian Federation.
anaesthesia the brain was removed and transverse sections ( 2 0 0 - 3 0 0 ~ m ) of the optic tectum were cut with a vibratome at 4°C. After at least 1 h equilibration at room temperature, individual slices were transferred to a recording chamber and continuously superfused with gassed (95% 02/5% CO2) Ringer solution (in mM: NaCI, 100; KCI, 2.5; CaCI 2, 1.5; NaH2PO4, 0.1; MgCI 2, 1.5; NaHCO 3, 17; glucose, 4; pH 7.35). Slices remained viable for up to 24 h. Neurones were recorded under whole cell patch clamp conditions with electrodes (6-8 Mfl) filled (in mM) with K-gluconate (or CsSO4), 100; NaC1, 5; EGTA, 10; MgC12, 4; HEPES, 2.5; ATP, 3 (pH 7.1) after first establishing gigaohm seals. Cell input resistance was 2 5 0 - 3 5 0 M f l ; uncompensated series resistance was always <10% of this value and constantly monitored throughout the experiment. Cell capacitance was calculated to be 0.3-3.2 pF on the basis of 3-10/2m cell body diameter [5]. Responses, elicited by voltage commands (usually 4 - 3 0 ms long: incrementing in 5 mV steps), were recorded via a List EPC-7 amplifier and analyzed with a commercial software (pCLAMP 5.5; Axon Instruments). Online leak current subtraction was performed using the P/4 method of the p C L A M P program. Drugs were applied via the bathing solution. The database of the present study comprises 40 neurones clamped at holding potentials between - 5 0 and
0304-3940/96/$12.00 © 1996 Elsevier Science Ireland Ltd. All rights reserved SSDI 0304-3940(95) 12330-Q
6
A. Zaykin, A. Nistri/Neuroscience Letters 205 (1996) 5-8
A
0.8
onset of TTX effect
2-
B
control / - , - ' ~ .
control
'~ 20 s TTX ** 30 s TTX
,
nA
1 nA 0 kSJ
0
ms
7 d
ms
55
7
-0.8
-I
D
nA o n s e t o f Li ÷ e f f e c t 0.6 ~ c o n t r o l J'
-a
~ - - .
C 0.8 nA 0
-0.~
-0.60.55 control
0 551 ~ 0 i
recovery
ms
10
Fig. 1. Current responses of frog optic tectum neurone. (A) Step depolarization to +10 mV from - 7 0 mV holding potential elicits, in control solution, an early inward current followed by slower outward current. After 1 min application of TTX step commands are applied at 10 s intervals. Note progressive decrease in inward and outward currents which unmasks a slower, TTX-insensitive outward component. Tracings are superimposed to aid comparison. (B) TTX-sensitive currents are obtained by subtracting the TTX-resistant current from the control one. Asterisks indicate at which times after TTX application data are recorded. Note concomitant depression of inward and outward components without shift in their peak. (C) The responses after 10 and 20 s are scaled to the size of the control one and superimposed to the latter to show their identical shape. (D) Currents induced by step command to +30 mV from - 7 0 mV holding potential in control solution and following gradual replacement of extracellular Na + with Li + (after 1 min application of the Li + solution test pulses are delivered at 20 s intervals). Note selective and progressive block of the outward fast current without change in the inward current (top tracings). After 10 rain washout of Li + solution the outward current recovers almost completely (bottom tracings). Cell in (D) different from (A-C).
- 7 0 m V . Fig. 1A shows that a depolarizing step to +30 mV from - 7 0 mV holding potential elicited a biphasic current (recorded with a K-gluconate filled pipette), which was first inward and then turned outward (see trace labelled as control). In the absence of permeable intracellular anions, the fast outward current was presumably due to efflux of K ÷. This type of response was also observed when external Ca 2÷ was replaced by Co 2÷. Fig. 1A also shows superimposed tracings obtained after the start of tetrodotoxin (TTX; 1/tM) superfusion with gradual depression of both current components and unmasking of a residual slower and shallower outward current. The TTXresistant current was then subtracted from the control current to demonstrate the isolated, TTX-sensitive currents (Fig. 1B). Further analysis indicated that, after scaling and superimposition, the currents recorded during the onset of TTX block had identical time-course as the control currents (Fig. 1C), demonstrating the adequacy of the voltage clamp. In summary then, it appears that qTX induced an analogous reduction in fast currents, regardless their polarity (n = 10). Fig. 1D shows another pharmacological characteristic of the transient outward current of frog visual neurones, namely its selective block by replacing external Na ÷ with Li ÷. Li ÷ gradually depressed the outward current component without affecting the peak or onset of the fast inward current. Full recovery was attained after 10 min washout with control Ringer. Similar data were
observed in nine neurones. Further analysis of the fast outward current was, however, hampered by the underlying presence of the distinct outward K ÷ current (insensitive to TTX or Li ÷ as shown in Fig. IA,D). It was therefore attempted to block the slow current by replacing internal K ÷ with Cs ÷ which is known to attenuate delayer rectifier currents [4]. Against a full block of the slow outward current by intracellular Cs ÷, the fast outward current was insensitive to such a block (n = 20) as demonstrated by the representative tracings obtained from a cell clamped at - 6 0 mV following application of depolarizing step commands (Fig. 2A). The fast inward current peak was always followed by the development of a transient outward current with characteristics similar to those observed with a standard K-gluconate filled electrode. Since the transient outward current was present when K ÷ or Cs ÷ was the main internal cation, it was considered as a relatively non-selective cationic current (termed /cat)TTX (0.5-1 btM) fully blocked both inward and outward currents, indicating that they retained the same pharmacological properties observed with intracellular K ÷. Equimolar replacement of external Na ÷ with K ÷ completely suppressed the inward current and prevented the appearance of/cat e v e n following depolarizing steps to +100 mV (data not shown), confirming that extracellular Na ÷ was required for/cat generation. Fig. 2B shows the tail current obtained by returning to the holding potential (-60 mV)
A. Zaykin, A. Nistri / Neuroscience Letters 205 (1996) 5-8
i.c. Cs +
500-
A
B
pA
f
~
o
0
-320~ . ~ t ~ current 0 ms 14 +6
-40C
mV
-
o
~
C
530[ PA I
-100
280 pAq
."
/
,.~" •- ~ , ' ~
'
mV
~'~%~'~
0j,
D
cont,ol v
~.
/~_ .~-~-~-.~,
100 -560 0
ms
32
-530 L Fig. 2. Currents recorded after the patch electrode had loaded the cell with intracellular Cs + (i.c. Cs+). (A) Representative, superimposed tracings obtained with incrementing step commands (0.1 Hz) in the range -45 mV (single asterisks) to +45 mV (double asterisks) from - 6 0 mV holding potential. Intermediate responses are evoked by steps to -35, -30, -25, -20, -5, + 10 and +30 mV. Traces are averages from eight responses. Note that with i.c. Cs + the inward current is always followed by an outward current. (B) A step to +60 mV induces the inward/outward current followed by a sharp tail current upon repolarization to holding potential. Single response is responsible for noisier record. (C) I/V relation for irtward (open symbols) and outward (filled symbols) currents of the cell shown in (A,B). Arrow indicates data point (asterisk) obtained from tail current protocol (as shown in (B)). Dashed line fitted to filled symbols indicates apparent reversal level (-20 mV) for outward current. (D) Effect of 4-AP on outward current elicited by step to -15 mV from - 6 0 mV holding potential. Note selective depression of outward component (different cell from A-C).
from a depolarizing step to +60 mV. The peak of the tail response was inward (see arrow), almost as if a slow component of the voltage-gated inward current persisted despite voltage-depende, nt inactivation. Nevertheless, the tail current peak was ob:~erved at 7 ms, i.e. at a time when the isolated inward current had fully dissipated (see for instance Fig. 1D with sl:ep to +15 mV). Furthermore, the tail current amplitude was linearly related to the current/voltage (1/V) curve for/'cat (see arrow pointing to asterisk in Fig. 2C), suggesting that permeation of Cs ÷ was perhaps responsible for it. Thus, it was possible to calculate by extrapolation (see dashed line) the apparent reversal potential of/'cat as --20 mV (activation threshold was 45 mV). This negative reversal value suggests that the membrane channels responsible for /'cat possessed mixed permeability to Na t and Cs ÷. The I/V plot (constructed with 5 mV step increments from - 6 0 mV holding poten-
tial; Fig. 2A) shows that the inward current had a peak amplitude at - 2 0 mV from a threshold o f - 4 5 m V and apparently reversed at +80 mV, a value close to the calculated Na ÷ equilibrium potential (+86 mV). Even if/cat w a s not selectively mediated by K ÷, it did retain some pharmacological characteristics of a K ÷ current, namely its sensitivity to a low concentration (0.2 mM) of 4-aminopyridine (4-AP) which in five cells readily blocked Cs ÷mediated Icat (Fig. 2D) without altering the peak of the fast inward current. These observations favour the interpretation that 4-AP mainly acted by removing Jcat rather than by interfering with the fast inward current. The main finding of the present study is the demonstration that visual neurones of the frog brain possess a fast biphasic response generated by depolarization, consisting of an inward current followed by the transient outward current /'cat" Since the inward current was suppressed by T F x or when K ÷ replaced Na ÷, maintained in the absence of Ca 2÷ or when Li ÷ replaced external Na ÷, and had an apparent reversal similar to the Na ÷ equilibrium potential, there is little doubt that the initial inward current was due to depolarization-activated influx of Na ÷. Such an influx triggered /'cat which displayed some features typical of/'K(Na), such as sensitivity to T T X and Li ÷ [3], but it was evidently a relatively non-selective cationic current since Cs ÷ could readily replace K + as the charge carrier. Hence, it was possible to observe the inward current in isolation (with external Li ÷ or 4-AP), while it was not possible to detect /'cat without the preceding inward current. For the latter reason it was difficult to ascertain the precise reversal potential of/'cat using a standard protocol for tail current reversal: this difficulty was also due in part to possible contamination of Ica t tail by reactivation of the preceding fast inward current (via voltagedependent de-inactivation of the fast inward current). Even if tails were studied at a time interval sufficient for full dissipation of the fast inward current, data should therefore be interpreted with caution. Notwithstanding this complication, it is apparent that the tail was inward rather than outward at a potential much less negative than the K + equilibrium potential, a finding which is consistent with its identification as a response mediated by nonselective cationic channel activity. Depolarization alone was insufficient to gate /'cat, as indicated by the experiments in the absence of external Na +. Since at - 6 0 mV holding potential (with 5 mM Na ÷ present in the patch pipette) /'cat w a s absent, it could not contribute to the resting potential. The mechanism linking activation of the Na t current to generation of ,/cat is presently unclear: since /'cat was blocked when Li ÷ replaced external Na ÷, it seems that /'cat was due to influx of Na t via voltage-gated channels, and it would make it a novel representative of a recently-characterized class of nonspecific cationic currents dependent on Na ÷ [7]. Certainly, the ability of a single inward Na ÷ current to elicit/'cat contrasts this response with the more traditional /'K(Na) which
8
A. Zaykin, A. Nistri/Neuroscience Letters 205 (1996) 5-8
normally requires very high intracellular Na + concentrations to be activated. Thus, it seems feasible to suppose that/cat was i n v o l v e d in the control of the spike duration and repetitive firing of frog visual neurones. So far currents analogous to/cat have not been observed in vertebrate brain neurones, although a voltage-gated nonspecific, slow cationic current has been recently reported in m a m m a l i a n brain neurones [1]. It is tempting to suggest that/cat might be a new, important system to control the short-term excitability of a m p h i b i a n neurones to integrate rapid visual signals. W e wish to thank Prof. Franco Conti for helpful discussion of these data. This work was supported by INFM, C N R and M U R S T .
[1] Alzheimer, C., A novel voltage-dependent cation current in rat neocortical neurones, J. Physiol., 479 (1994) 199-205. 12] Dryer, S.E., Na+-activated K+ channels: a new family of largeconductance ion channels, Trends Neurosci., 17 (1995) 155-160. [3] Dryer, S.E., Fuji, J.T. and Martin, A.R., A Na+-activated K+ current in cultured brain stem neurones from chicks, J. Physiol., 410 (1989) 283-296. [4] Hille, B., Ionic Channels of Excitable Membranes, Sinauer, Sunderland, MA, USA, 1992, p. 131. [5] Szdkely, G. and Lazar, G., Cellular and synaptic architecture of the optic rectum. In R. Llin~ and W. Precht (Eds.), Frog Neurobiology, Springer, New York, 1976, pp. 407-434. [6] Zaykin, A. and Nistri, A., A transient Na+-dependent K+ current of neurones of the frog (Rana temporaria) optic tectum in vitro, J. Physiol., 477 (1994) 49P. [7] Zhainazarov, A.B. and Ache, B.W., Na+-activated non-selective cation channels in primary olfactory neurons, J. Neurophysiol., 73 (1995) 1774-1781.