- Email: [email protected]
The slow inhibitory postsynaptic potential in rat hippocampal CA1 neurones is blocked by intracellular injection of QX-314
Neuroscience Letters, 110 (1990) 309-313
309
Elsevier Scientific Publishers Ireland Ltd. NSL 06724
The slow inhibitory postsynaptic potential in rat hippocampal CA1 neurones is blocked by intracellular injection of QX-314 T. Nathan, M.S. Jensen and J.D.C. Lambert PharmaBiotic Research Centre, Institute of Physiology, University of Aarhus, Aarhus (Denmark) (Received 13 September 1989; Accepted 17 October 1989)
Key words: CAI pyramidal cell; QX-314; Lidocaine derivative; Hippocampus; GABAs receptor; Inhibitory postsynaptic potential Intracellular lx~cordingswere made from CAI pyramidal neurones in the rat hippocampus slice preparation. The recording electrodes contained potassium acetate (4 M) with or without the quaternary lidocaine derivative, QX-314 (50 mM). Both fast (f) and slow (s) inhibitory postsynaptic potentials (IPSP) were evoked by low-frequency orthodromic stimulation. The s-IPSP was rapidly reduced by QX-314 injection. It decreased along a similar time course to the d V/dt of the action potential (AP). The f-IPSP and excitatory postsynaptic potential were not significantly reduced in size at a time when the s-IPSP was virtually abolished by QX-314. It is concluded that conductance through the K + channels which are coupled to GABAB receptors is readily blocked by QX-314, while the CI- channels which are coupled to GABA^ receptors and the cation channels coupled to the glutamate receptors are relatively resistant to the local anaesthetic.
Local anaesthetics such as lidocaine and procaine block voltage-sensitive Na + channels and inhibit the generation of the AP. The molecular mechanism of this block involves the cationic form of the molecule acting from the intracellular side of the membrane [9] in a voltage- and use-dependent manner [14]. Quaternary derivatives of lidocaine such as QX-314 are permanently charged and lipophobic. When applied intracellularly, they remain within the impaled neurone and do not affect neuronal functioning in the rest of the tissue. Thus, these agents are useful for blocking fast spiking in cases where tetrodotoxin is contra-indicated, as for example is the case with investigations of synapticaUy-evoked potentials [3-5, 7, 12]. Here we report that intraceUular injection of QX-314 rapidly and selectively reduces the slow inhibitory postsynaptic potential (s-IPSP) in CA1 neurones. Preliminary results have been published in abstract form [10]. The experiments were performed on hippocampal slices (400/an) from 17 Wistar rats (150-190 g). The slices were prepared with a McIllwain chopper [2] and mainCorrespondence: J.D.C. Lambert, Institute of Physiology, University of Aarhus, DK-8000 ,A,rhus, Denmark. 0304-3940/90/$ 03.50 © 1990 Elsevier Scientific Publishers Ireland Ltd.
310
tained at 32-34°C at the interface between oxygenated standard Ringer (in mM: NaC1 124, KCI 3.25, NaHzPO4 1.25, NaHCO3 20, CaC12 2, MgC12 2, D-glucose 10: pH, 7.3) and warm humidified gas (95 % O2, 5 % CO2). Conventional glass microelectrodes filled with 4 M KAc or 4 M KAc + 50 mM QX-314 (40-90 Mr2) were used for intracellular recording. Bipolar glass-insulated platinum wire (50/~m) electrodes were used for orthodromic stimulation of the Schaffer collateral-commissural pathway with constant current pulses of 0.1 ms, 0.05-0.1 Hz. IntraceUular recordings were made with an Axoclamp A-2 amplifier and the results recorded on a video recorder for off-line analysis with a Nicolet (model 4094C) digital oscilloscope. Impalements were made with very brief ( < 1 ms, to avoid bolus injection of QX-314) overcompensation of the amplifier's negative capacity circuitry. Only cells with a resting membrane potential (EM) of > -- 55 mV and membrane resistance (RM) of > 20 M(2 were accepted. The time course of QX-314's action was followed in 9 neurones where these criteria were achieved very quickly. Synaptic responses to orthodromic stimulation and the current-evoked AP were then recorded as soon as possible. Synaptic potentials and the AP were recorded as QX-314 was injected into the cell either by diffusion or by the application of cathodat currents (typically +0.3 nA; 300 ms; 1 Hz). The size of the f-IPSP was measured at the peak of the first hyperpolarizing wave (25-45 ms following stimulation), while the s-IPSP was measured after 150 ms. The s-IPSP was identified on the basis of the following characteristics (see also ref. 1). It is a hyperpolarization of 3-10 mV at typical values of resting EM. It has a latency of about 40 ms and its duration is 200-1000 ms (typically 500 ms) with a peak about 150 ms after the stimulation. Its equilibrium potential is around - 9 0 mV and it does not readily reverse at more negative potentials. It becomes attenuated with stimulating frequencies > 0.1 Hz. Its appearance is irrespective of whether or not an AP is evoked on orthodromic stimulation and it is therefore not an afterhyperpolarization. QX-314 application regularly caused a small depolarization ( < 10 mV) of the EM with an increase in RM. To facilitate a quantitative comparison of the effect of QX314 on the synaptic potentials, EM was therefore maintained at a predetermined level by manual injection of small anodal currents (which tend to retain the cationic QX314). Following impalement with a QX-314 containing electrode, the s-IPSP was seen to decline rapidly. Indeed if the initial penetration was not 'clean', there was little evidence of a s-IPSP by the time recording started. The best overall impression of the effect of QX-314 on the IPSPs is obtained by comparing results from neurone samples recorded with and without QX-314-containing electrodes. Fig. 1 shows super-imposed results from 8 neurones impaled with electrodes containing KAc and 8 different neurones which were injected with QX-314. There was no significant change in the f-IPSP (control: 7.8 + 2.1 mV; QX-314:7.5 m V _ 3.5 mV). At the same time, the s-IPSP was greatly reduced by the QX-314 injection (control: 8.3 + 1.9 mV; QX-314:1.6 ___0.4 mV; P < 0.001). This residual hyperpolarization, which is resistant to large injections of QX-314, does not have a peak at 150 ms. We have hitherto been unable to determine whether this represents a residual component of the s-IPSP or whether it is the tail of the CI- current underlying the f-IPSP.
311
A.
4M KAc
•+
+ []
B.
4M KAc +50mM Q X - 3 1 4
•+
+ []
15mv I
150 ms
6
Mean ± S.D.
4 2
n=8
n=8
Fig. 1. QX-314 specifically attenuates the s-IPSP. A: upper records show superimposed intraceUular recordings (4 M KAc electrodes) of synaptic responses from 8 different CAI neurones. The f-IPSP and the s-IPSP were measured at the times shown at the filled and open box, respectively. These results were averaged and plotted on the histogram below. B: superimposed records from a further 8 neurones (KAc + QX-314 (50 mM) containing electrodes) following injection with QX-314. The records in A and B have been normalized to the resting membrane potentials (stippled line) which were in the range of -58 to -68 mV. Each trace is the average of at least 3 evoked responses. QX-314 had no significant effect on the f-IPSP, but caused a large reduction of the s-IPSP and a separate hyperpolarizing wave with a peak around 150 ms was no longer evident. W e have also c o m p a r e d the time course o f action o f QX-314 on the current-evoked A P and the synaptic events (Fig. 2). As QX-314 diffused into the neurone, the size and d V/dt [3, 4] o f the A P decreased while the threshold increased (Fig. 2A). Ultimately, there was only a small (20-40 mV) regenerative potential which was presumably a Ca2+-dependent A P [3]. There was a close correlation between the time course o f the decline o f the s-IPSP and the d V/dt o f the A P following diffusion o f QX-314 (Fig. 2B). Fig. 2 also shows that maximal effect o f QX-314 applied by diffusion was attained 20-30 min after impalement (which was typical for m o s t other neurones) and that a small further decrease in the s-IPSP could then be achieved by injecting QX-314 with cathodal current pulses. T h u s we have shown that QX-314 potently reduces the s-IPSP and that this action is specific with respect to the other synaptic potentials (Figs. 1 and 2). In m o s t other
312
- - ~ ' ~ " ~
200 V/S I 40mV 4 ms
1 25mY B. Relative value
(%)
o
100
o
80
o
• <>
o.,O0 60
• "°':°.°
40
eO
Oe ~
e¢
0
eqbO0 0
0
0
I
1
0 ~°0 %~0000 O°
0 "eo*~
2O
o•
<> 0
i
0
i
10
,
j
20
i
:
i
'i
30
40 rain. Time after impalement Fig. 2. The depression of the dV/dt of the AP and the s-IPSP by QX-314 follows a similar time course. A: the neurone was penetrated with an electrode containing 50 mM QX-314 in 4 M KAc. The membrane potential was manually clamped at - 6 0 inV. The current-evoked AP (pulse: + 1.5 hA; 0.2 Hz; middle row), its first derivative (dV/dt; top row) and synaptically evoked potentials (0.05 Hz; lower row) were recorded at the times shown followingpenetration. As the diffusion of QX-314 progressed, the AP became smaller and broader and its threshold increased. The d V/dt also decreased. B: graph of the results from the experiment depicted in A. The results for both the d V/dt (filledcircles) and the s-IPSP (open diamonds; measured at the arrow in A. 150 ms after the stimulation) were the averl~ of three determinations taken each minute. The values recorded just after impalement have been normalized to 100~. The reduction of the d V/dt and the s-lPSP followed a similar time course and little fu~ier depression was seen after 20 rain. QX-314 was then injected (+0.3 nA; 300 ms; 0.5 Hz; during the period indicated by arrows). This had little further effect on the AP (which now resembled a Ca2+-dependent spike), but caused a further reduction of the s-IPSP. studies, the f-IPSP a n d the m o n o s y n a p t i c E P S P have been s h o w n to b e generally resistant to injection o f Q X substances [3, 5, 7, 8]. Puff a n d Carten, howe,oct, showed that the m o n o s y n a p t i c E P S P in g u i n e a pig CA1 n e u r o n e s was reduced by i n t r a ~ H u lar injections o f QX-222 [12]. W e are aware o f only two other studies where the effect
313
of a QX compound on the s-IPSP has been investigated [13, 15]. In both cases, the s-IPSP was found to be resistant to the anaesthetic. However, the record published by Segal [13] shows a potential which is reminiscent of that seen in the present Fig. 1B, where we have shown that the majority of the s-IPSP has already been removed by QX-314. In Segal's experiments it is possible that the majority of the s-IPSP had been blocked during the interval between penetration and the start of recording. The s-IPSP is mediated by an increase in gK+ [6, 11]. It is possible that QX substances are relatively potent blockers of K + channels and this may be a general feature of quaternary amines. Indeed, methylxylocholine has also been shown to be an effective blocker of voltage-dependent K + currents [4]. The depolarization and an increase in RM following QX-314 injection probably results from a block of resting K + channels (QX-314 being more potent in this respect than QX-222; see refs 4, 5, 12). The action on the s-IPSP probably therefore results from a block of the K + channels comprising the GABAB ionophore. This work was supported by the Danish MRC and the PharmaBiotec biotechnology programme. Astra are thanked for a gift of QX-314 bromide. T.N. was supported by a F~erdigg~relsesstipendium from the Danish Ministry of Education. 1 Alger, B.E., Characteristics of a slow hyperpolarizing synaptic potential in rat hippocampal pyramidal cells in vitro, J. Neurophysiol., 52 (1984) 892-910. 2 Andreasen, M., Lambert, J.D.C. and Jensen, M.S., Effects of new non-NMDA antagonists on synaptic transmission in the in vitro rat hippocampus, J. Physiol. (Lond.), 414 (! 989) 317-336. 3 Connors, B.W. and Prince, D.A., Effects of local anesthetic QX-314 on the membrane properties of hippocampal pyramidal neurons, J. Pharmacol. Exp. Ther., 220 (1982) 476-481. 4 Engberg, I., Flatman, J.A. and Lambert, J.D.C., The response of cat spinal motoneurones to the intracellular application of agents with local anaesthetic action, Br. J. Pharmacol., 81 (1984) 2 ! 5-224. 5 Flatman, J.A., Engberg, I. and Lambert, J.D.C., Reversibility of Ia EPSP investigated with intracellulady iontophoresed QX-222, J. Neurophysiol., 48 (1982) 419-430. 6 Howe, J.R., Sutor, B. and Zieglg~nsberger, W., Characteristics of long-duration inhibitory postsynaptic potentials in rat neocortical neurons in vitro, Cell. Mol. Neurobiol., 7 (1987) 1-18. 7 Kita, T., Kita, H. and Kitai, H. and Kitai, S.T., Local stimulation induced GABAergic response in rat striatal slice preparations: intracellular recordings on QX-314 injected neurons, Brain Res., 360 (1985) 304-310. 8 Mody, 1., Stanton, P.K. and Heinemann, U., Activation of N-methyl-D-aspartate receptors parallels changes in cellular and synaptic properties of dentate gyrus granule cells after kindling, J. Neurophysiol., 59 (1988) 1033-1054. 9 Narahashi, T., Frazier, D.T. and Moore, J.W., Comparison of tertiary and quaternary amine local anesthetics in their ability to depress membrane ionic c,onductances, J. Neurobiol., 3 (1972) 267-276. 10 Nathan, T., Lambert, J.D.C. and Jensen, M.S., Postsynaptic hyperpolarizing responses in CAI neurones are reduced by intracellular application of QX-314, Eur. J. Neurosci., Suppl. 2 (1989) 157. 11 Newberry, N.R. and Nicoll, R.A., A bicuculline-resistant inhibitory post-synaptie potential in rat hippocampal pyramidal ceils in vitro, J. Physiol. (Lond.), 348 (1984) 239-254. 12 Puil, E. and Carlen, P.L., Attenuation of glutamate-action, excitatory postsynaptic potentials, and spikes by intracellular QX 222 in hippocampal neurons, Neuroscience, 11 (1984) 389-398. 13 Segal, M., Effects ofa lidocaine derivative QX-572 on CA1 neuronal responses to electrical and chemical stimuli in a hippocampal slice, Neuroscienee, 27 (1988) 905-909. 14 Strichartz, G.R., The inhibition of sodium currents in myelinated nerve by quaternary derivatives of lidocaine, J. Gen. Physiol., 62 (1973) 37-57. 15 Thalmann, R.H. and Ayala, G.F., A picrotoxin-resistant hyperpolarizing response is elicited by orthodromic stimulation of hippocampal neurons, Soc. Neurosei. Abstr., 6 (I 980) 300-300.
309
Elsevier Scientific Publishers Ireland Ltd. NSL 06724
The slow inhibitory postsynaptic potential in rat hippocampal CA1 neurones is blocked by intracellular injection of QX-314 T. Nathan, M.S. Jensen and J.D.C. Lambert PharmaBiotic Research Centre, Institute of Physiology, University of Aarhus, Aarhus (Denmark) (Received 13 September 1989; Accepted 17 October 1989)
Key words: CAI pyramidal cell; QX-314; Lidocaine derivative; Hippocampus; GABAs receptor; Inhibitory postsynaptic potential Intracellular lx~cordingswere made from CAI pyramidal neurones in the rat hippocampus slice preparation. The recording electrodes contained potassium acetate (4 M) with or without the quaternary lidocaine derivative, QX-314 (50 mM). Both fast (f) and slow (s) inhibitory postsynaptic potentials (IPSP) were evoked by low-frequency orthodromic stimulation. The s-IPSP was rapidly reduced by QX-314 injection. It decreased along a similar time course to the d V/dt of the action potential (AP). The f-IPSP and excitatory postsynaptic potential were not significantly reduced in size at a time when the s-IPSP was virtually abolished by QX-314. It is concluded that conductance through the K + channels which are coupled to GABAB receptors is readily blocked by QX-314, while the CI- channels which are coupled to GABA^ receptors and the cation channels coupled to the glutamate receptors are relatively resistant to the local anaesthetic.
Local anaesthetics such as lidocaine and procaine block voltage-sensitive Na + channels and inhibit the generation of the AP. The molecular mechanism of this block involves the cationic form of the molecule acting from the intracellular side of the membrane [9] in a voltage- and use-dependent manner [14]. Quaternary derivatives of lidocaine such as QX-314 are permanently charged and lipophobic. When applied intracellularly, they remain within the impaled neurone and do not affect neuronal functioning in the rest of the tissue. Thus, these agents are useful for blocking fast spiking in cases where tetrodotoxin is contra-indicated, as for example is the case with investigations of synapticaUy-evoked potentials [3-5, 7, 12]. Here we report that intraceUular injection of QX-314 rapidly and selectively reduces the slow inhibitory postsynaptic potential (s-IPSP) in CA1 neurones. Preliminary results have been published in abstract form [10]. The experiments were performed on hippocampal slices (400/an) from 17 Wistar rats (150-190 g). The slices were prepared with a McIllwain chopper [2] and mainCorrespondence: J.D.C. Lambert, Institute of Physiology, University of Aarhus, DK-8000 ,A,rhus, Denmark. 0304-3940/90/$ 03.50 © 1990 Elsevier Scientific Publishers Ireland Ltd.
310
tained at 32-34°C at the interface between oxygenated standard Ringer (in mM: NaC1 124, KCI 3.25, NaHzPO4 1.25, NaHCO3 20, CaC12 2, MgC12 2, D-glucose 10: pH, 7.3) and warm humidified gas (95 % O2, 5 % CO2). Conventional glass microelectrodes filled with 4 M KAc or 4 M KAc + 50 mM QX-314 (40-90 Mr2) were used for intracellular recording. Bipolar glass-insulated platinum wire (50/~m) electrodes were used for orthodromic stimulation of the Schaffer collateral-commissural pathway with constant current pulses of 0.1 ms, 0.05-0.1 Hz. IntraceUular recordings were made with an Axoclamp A-2 amplifier and the results recorded on a video recorder for off-line analysis with a Nicolet (model 4094C) digital oscilloscope. Impalements were made with very brief ( < 1 ms, to avoid bolus injection of QX-314) overcompensation of the amplifier's negative capacity circuitry. Only cells with a resting membrane potential (EM) of > -- 55 mV and membrane resistance (RM) of > 20 M(2 were accepted. The time course of QX-314's action was followed in 9 neurones where these criteria were achieved very quickly. Synaptic responses to orthodromic stimulation and the current-evoked AP were then recorded as soon as possible. Synaptic potentials and the AP were recorded as QX-314 was injected into the cell either by diffusion or by the application of cathodat currents (typically +0.3 nA; 300 ms; 1 Hz). The size of the f-IPSP was measured at the peak of the first hyperpolarizing wave (25-45 ms following stimulation), while the s-IPSP was measured after 150 ms. The s-IPSP was identified on the basis of the following characteristics (see also ref. 1). It is a hyperpolarization of 3-10 mV at typical values of resting EM. It has a latency of about 40 ms and its duration is 200-1000 ms (typically 500 ms) with a peak about 150 ms after the stimulation. Its equilibrium potential is around - 9 0 mV and it does not readily reverse at more negative potentials. It becomes attenuated with stimulating frequencies > 0.1 Hz. Its appearance is irrespective of whether or not an AP is evoked on orthodromic stimulation and it is therefore not an afterhyperpolarization. QX-314 application regularly caused a small depolarization ( < 10 mV) of the EM with an increase in RM. To facilitate a quantitative comparison of the effect of QX314 on the synaptic potentials, EM was therefore maintained at a predetermined level by manual injection of small anodal currents (which tend to retain the cationic QX314). Following impalement with a QX-314 containing electrode, the s-IPSP was seen to decline rapidly. Indeed if the initial penetration was not 'clean', there was little evidence of a s-IPSP by the time recording started. The best overall impression of the effect of QX-314 on the IPSPs is obtained by comparing results from neurone samples recorded with and without QX-314-containing electrodes. Fig. 1 shows super-imposed results from 8 neurones impaled with electrodes containing KAc and 8 different neurones which were injected with QX-314. There was no significant change in the f-IPSP (control: 7.8 + 2.1 mV; QX-314:7.5 m V _ 3.5 mV). At the same time, the s-IPSP was greatly reduced by the QX-314 injection (control: 8.3 + 1.9 mV; QX-314:1.6 ___0.4 mV; P < 0.001). This residual hyperpolarization, which is resistant to large injections of QX-314, does not have a peak at 150 ms. We have hitherto been unable to determine whether this represents a residual component of the s-IPSP or whether it is the tail of the CI- current underlying the f-IPSP.
311
A.
4M KAc
•+
+ []
B.
4M KAc +50mM Q X - 3 1 4
•+
+ []
15mv I
150 ms
6
Mean ± S.D.
4 2
n=8
n=8
Fig. 1. QX-314 specifically attenuates the s-IPSP. A: upper records show superimposed intraceUular recordings (4 M KAc electrodes) of synaptic responses from 8 different CAI neurones. The f-IPSP and the s-IPSP were measured at the times shown at the filled and open box, respectively. These results were averaged and plotted on the histogram below. B: superimposed records from a further 8 neurones (KAc + QX-314 (50 mM) containing electrodes) following injection with QX-314. The records in A and B have been normalized to the resting membrane potentials (stippled line) which were in the range of -58 to -68 mV. Each trace is the average of at least 3 evoked responses. QX-314 had no significant effect on the f-IPSP, but caused a large reduction of the s-IPSP and a separate hyperpolarizing wave with a peak around 150 ms was no longer evident. W e have also c o m p a r e d the time course o f action o f QX-314 on the current-evoked A P and the synaptic events (Fig. 2). As QX-314 diffused into the neurone, the size and d V/dt [3, 4] o f the A P decreased while the threshold increased (Fig. 2A). Ultimately, there was only a small (20-40 mV) regenerative potential which was presumably a Ca2+-dependent A P [3]. There was a close correlation between the time course o f the decline o f the s-IPSP and the d V/dt o f the A P following diffusion o f QX-314 (Fig. 2B). Fig. 2 also shows that maximal effect o f QX-314 applied by diffusion was attained 20-30 min after impalement (which was typical for m o s t other neurones) and that a small further decrease in the s-IPSP could then be achieved by injecting QX-314 with cathodal current pulses. T h u s we have shown that QX-314 potently reduces the s-IPSP and that this action is specific with respect to the other synaptic potentials (Figs. 1 and 2). In m o s t other
312
- - ~ ' ~ " ~
200 V/S I 40mV 4 ms
1 25mY B. Relative value
(%)
o
100
o
80
o
• <>
o.,O0 60
• "°':°.°
40
eO
Oe ~
e¢
0
eqbO0 0
0
0
I
1
0 ~°0 %~0000 O°
0 "eo*~
2O
o•
<> 0
i
0
i
10
,
j
20
i
:
i
'i
30
40 rain. Time after impalement Fig. 2. The depression of the dV/dt of the AP and the s-IPSP by QX-314 follows a similar time course. A: the neurone was penetrated with an electrode containing 50 mM QX-314 in 4 M KAc. The membrane potential was manually clamped at - 6 0 inV. The current-evoked AP (pulse: + 1.5 hA; 0.2 Hz; middle row), its first derivative (dV/dt; top row) and synaptically evoked potentials (0.05 Hz; lower row) were recorded at the times shown followingpenetration. As the diffusion of QX-314 progressed, the AP became smaller and broader and its threshold increased. The d V/dt also decreased. B: graph of the results from the experiment depicted in A. The results for both the d V/dt (filledcircles) and the s-IPSP (open diamonds; measured at the arrow in A. 150 ms after the stimulation) were the averl~ of three determinations taken each minute. The values recorded just after impalement have been normalized to 100~. The reduction of the d V/dt and the s-lPSP followed a similar time course and little fu~ier depression was seen after 20 rain. QX-314 was then injected (+0.3 nA; 300 ms; 0.5 Hz; during the period indicated by arrows). This had little further effect on the AP (which now resembled a Ca2+-dependent spike), but caused a further reduction of the s-IPSP. studies, the f-IPSP a n d the m o n o s y n a p t i c E P S P have been s h o w n to b e generally resistant to injection o f Q X substances [3, 5, 7, 8]. Puff a n d Carten, howe,oct, showed that the m o n o s y n a p t i c E P S P in g u i n e a pig CA1 n e u r o n e s was reduced by i n t r a ~ H u lar injections o f QX-222 [12]. W e are aware o f only two other studies where the effect
313
of a QX compound on the s-IPSP has been investigated [13, 15]. In both cases, the s-IPSP was found to be resistant to the anaesthetic. However, the record published by Segal [13] shows a potential which is reminiscent of that seen in the present Fig. 1B, where we have shown that the majority of the s-IPSP has already been removed by QX-314. In Segal's experiments it is possible that the majority of the s-IPSP had been blocked during the interval between penetration and the start of recording. The s-IPSP is mediated by an increase in gK+ [6, 11]. It is possible that QX substances are relatively potent blockers of K + channels and this may be a general feature of quaternary amines. Indeed, methylxylocholine has also been shown to be an effective blocker of voltage-dependent K + currents [4]. The depolarization and an increase in RM following QX-314 injection probably results from a block of resting K + channels (QX-314 being more potent in this respect than QX-222; see refs 4, 5, 12). The action on the s-IPSP probably therefore results from a block of the K + channels comprising the GABAB ionophore. This work was supported by the Danish MRC and the PharmaBiotec biotechnology programme. Astra are thanked for a gift of QX-314 bromide. T.N. was supported by a F~erdigg~relsesstipendium from the Danish Ministry of Education. 1 Alger, B.E., Characteristics of a slow hyperpolarizing synaptic potential in rat hippocampal pyramidal cells in vitro, J. Neurophysiol., 52 (1984) 892-910. 2 Andreasen, M., Lambert, J.D.C. and Jensen, M.S., Effects of new non-NMDA antagonists on synaptic transmission in the in vitro rat hippocampus, J. Physiol. (Lond.), 414 (! 989) 317-336. 3 Connors, B.W. and Prince, D.A., Effects of local anesthetic QX-314 on the membrane properties of hippocampal pyramidal neurons, J. Pharmacol. Exp. Ther., 220 (1982) 476-481. 4 Engberg, I., Flatman, J.A. and Lambert, J.D.C., The response of cat spinal motoneurones to the intracellular application of agents with local anaesthetic action, Br. J. Pharmacol., 81 (1984) 2 ! 5-224. 5 Flatman, J.A., Engberg, I. and Lambert, J.D.C., Reversibility of Ia EPSP investigated with intracellulady iontophoresed QX-222, J. Neurophysiol., 48 (1982) 419-430. 6 Howe, J.R., Sutor, B. and Zieglg~nsberger, W., Characteristics of long-duration inhibitory postsynaptic potentials in rat neocortical neurons in vitro, Cell. Mol. Neurobiol., 7 (1987) 1-18. 7 Kita, T., Kita, H. and Kitai, H. and Kitai, S.T., Local stimulation induced GABAergic response in rat striatal slice preparations: intracellular recordings on QX-314 injected neurons, Brain Res., 360 (1985) 304-310. 8 Mody, 1., Stanton, P.K. and Heinemann, U., Activation of N-methyl-D-aspartate receptors parallels changes in cellular and synaptic properties of dentate gyrus granule cells after kindling, J. Neurophysiol., 59 (1988) 1033-1054. 9 Narahashi, T., Frazier, D.T. and Moore, J.W., Comparison of tertiary and quaternary amine local anesthetics in their ability to depress membrane ionic c,onductances, J. Neurobiol., 3 (1972) 267-276. 10 Nathan, T., Lambert, J.D.C. and Jensen, M.S., Postsynaptic hyperpolarizing responses in CAI neurones are reduced by intracellular application of QX-314, Eur. J. Neurosci., Suppl. 2 (1989) 157. 11 Newberry, N.R. and Nicoll, R.A., A bicuculline-resistant inhibitory post-synaptie potential in rat hippocampal pyramidal ceils in vitro, J. Physiol. (Lond.), 348 (1984) 239-254. 12 Puil, E. and Carlen, P.L., Attenuation of glutamate-action, excitatory postsynaptic potentials, and spikes by intracellular QX 222 in hippocampal neurons, Neuroscience, 11 (1984) 389-398. 13 Segal, M., Effects ofa lidocaine derivative QX-572 on CA1 neuronal responses to electrical and chemical stimuli in a hippocampal slice, Neuroscienee, 27 (1988) 905-909. 14 Strichartz, G.R., The inhibition of sodium currents in myelinated nerve by quaternary derivatives of lidocaine, J. Gen. Physiol., 62 (1973) 37-57. 15 Thalmann, R.H. and Ayala, G.F., A picrotoxin-resistant hyperpolarizing response is elicited by orthodromic stimulation of hippocampal neurons, Soc. Neurosei. Abstr., 6 (I 980) 300-300.