- Email: [email protected]
A voltage-clamp study of the effects of Joro spider toxin and zinc on excitatory synaptic transmission in CA1 pyramidal cells of the guinea pig hippocampal slice
200
Neuroscience Research. 10 ( 1991 ) 2()t,~ 2 i ~,~ ' 1991 Elsevier Scientific Publishers Ireland, Ltd. 0168-0102/91 /$~ 3 ;~
NEURES 00423
A voltage-clamp study of the effects of Joro spider toxin and zinc on excitatory synaptic transmission in CA1 pyramidal cells of the guinea pig hippocampal slice Y o s h i n o r i Sahara, H u g h P.C. R o b i n s o n , A k i k o M i w a a n d N o b u f u m i K a w a i Department of Neurobiology, Tokyo Metropolitan Institute for Neurosciences, Tokyo (Japan) (Received 16 August 1990; Revised version received 15 October 1990; Accepted 23 October 1990)
Key words." Single electrode voltage-clamp; Glutamate receptor; JSTX; Zinc; Hippocampus; Slice
SUMMARY Using the single-electrode voltage-clamp technique, we have examined the effects of a non-N-methyl-Daspartate (NMDA) antagonist, Joro spider toxin (JSTX), and of an NMDA antagonist, zinc, on excitatory postsynaptic currents (EPSCs) evoked by stimulation of stratum radiatum in CA1 pyramidal cells of the guinea-pig hippocampal slice. Pressure application of a synthesized JSTX (JSTX-3) at 10-200 /tM greatly reduced the EPSCs (14/19 cells). The block by JSTX-3 was observed in pyramidal cells where the EPSCs showed linear peak current-voltage (I-V) relations in the control. EPSCs remaining after JSTX-3 application showed non-linear peak I-V relationships (10/14 cells), and were blocked by puff application of the selective NMDA receptor antagonist DL-2-amino-5-phosphonovalerate (APV) at 200/~M (6/10 cells). In the presence of JSTX-3, the decay time constant of the EPSC was increased and was less affected by membrane potential. JSTX-3 had no detectable effects on EPSCs apparently mediated solely by NMDA receptor. These observations suggest that JSTX-3 blocks excitatory synaptic transmission mainly by suppressing non-NMDA-rec,eptor-mediated EPSCs, and that the JSTX-3-insensitive component is mediated at least in part by NMDA receptors in the hippocampal slice. Zinc (100-200 btM) reversibly attenuated EPSCs (6/9 cells) and appeared to block a slower component of the EPSCs, suggesting that mainly NMDA receptor-mediated currents were affected.
INTRODUCTION
Toxins isolated from the venom of araneid spiders, such as Joro spider toxin (JSTX) and argiotoxin or argiopine, have been shown to have pronounced blocking effects on excitatory synaptic transmission in invertebrates u8,22. JSTX has also been shown to block kainate- and quisqualate-activated responses in mammalian hippocampal neurons, with little blocking action on N-methyl-D-aspartate (NMDA) responses 2,28. The greatest impediment to studies on JSTX has been the difficulty of isolation of adequate amounts of material from natural spider venoms. Recently, the chemical structure of JSTX has been determined s, and different components of JSTX (JSTX-1 and -3) as well as Correspondence." Dr. Y. Sahara, Department of Neurobioiogy, Tokyo Metropolitan Institute for Neurosciences, 2-6 Musashidai, Fuchu-city, Tokyo 183, Japan.
201 analogues of JSTX (e.g. 1-naphthylacetyl spermine) have been synthesized 6.14. These synthetic toxins have facilitated biochemical and histochemical studies on n o n - N M D A glutamate receptors of the mammalian brain 29,3o In the present study, we have examined the effects of a synthesized JSTX, JSTX-3, on excitatory synaptic transmission in the mammalian hippocampus, using the single-electrode voltage-clamp technique. It is generally accepted that L-glutamate mediates synaptic transmission in the hippocampal excitatory pathway between Schaffer collateral-commissural fibers and CA1 pyramidal cells. Recent studies using brain slices 3,10,~5 and cultured neurons 8,N have demonstrated that both N M D A receptors and n o n - N M D A receptors are involved. We have also examined the effects on excitatory synaptic transmission of zinc, which has been proposed to modulate excitatory postsynaptic transmission by blocking N M D A receptors ~2,21,2n. A preliminary report of this study has been presented in abstract form 26 MATERIALS AND METHODS
Slice preparation The procedure for preparation of hippocampal slices was similar to that described previously 28. Briefly, the hippocampus was removed from adult guinea pigs (200-250 g) and sliced transversely into 300-400/~m sections with a stainless-steel razor blade. The slices were immediately transferred to the experimental chamber which was perfused with bathing medium at a constant rate of approximately 1 ml/min. The bathing medium contained (mM): NaC1 126, KC1 3, N a H 2 P O 4 1.24, MgSO 4 1.3, CaC12 2.4, N a H C O 3 26, and D-glucose 10. The slices were placed on a lens paper and maintained at the interface between a humidified O2/CO 2 mixture (95%/5%) and the bathing medium. Temperature was maintained at 30-32°C. Recordings were obtained after approximately 1-1.5 h of incubation. Stimulation and recording The stimulating bipolar electrode (Teflon-coated platinum iridium wire, 50 /~m) was placed on the stratum radiatum. The intracellular recording electrode, filled with 4 M potassium acetate (resistance 20-80 MO), was advanced through the pyramidal cell layer of the CA1 region. Extracellular field potentials evoked by stimulation of the stratum radiatum (0.3 Hz, 0.2 ms duration, < 100/~A) were used as a guide to help locate the site of recording in the pyramidal cell layer. After establishing stable intracellular recording from a CA1 pyramidal cell, we attempted to achieve a synaptic response in which there was no apparent inhibitory component evoked by stimulation of the stratum radiatum, by adjusting the intensity of the stimulating current and by applying picrotoxin (5 /~M) to the bathing medium during periods of data acquisition; periods for application of picrotoxin were carefully limited since prolonged application frequently induces epileptiform activity. The cell was then voltage-clamped, using an 'Axoclamp-2A' discontinuous single-electrode voltage-clamp amplifier (Axon Instruments) operating at a switching frequency of 3-11 kHz. The holding potential in the voltage clamp was chosen to be close to the resting potential, and step voltage commands (80-100 ms duration) were applied in 5-mV increments, initially in the hyperpolarizing direction and subsequently in the depolarizing direction, over a membrane potential range of - 1 2 0 to - 4 0 mV. To measure the current-voltage relationship of excitatory postsynaptic currents (EPSCs), 5 or more responses were obtained at each membrane potential.
202 Drug application
Synthesized spider toxin JSTX-3 (10-200/tM), DL-2-amino-5-phosphonovalerate (APV, 100-500/IM, Sigma or Tocris Neuramin), and Zn(CH3CO2) 2 (100-200/~M, Wako Pure Chemical Industries Ltd.), were dissolved in the bathing medium and applied to individual neurons using pressure pulses (20-40 psi.) of 10-30 ms duration, from pipettes (tip diameter 5-10/~m) positioned close to the cell. This application method is limited in that the actual concentration of solutions at the membrane surface is unknown; however, considerable dilution certainly occurs. In two experiments, we confirmed that a nearly 10 times higher concentration of APV was required to obtain the same effect for pressure ejection as produced by bath perfusion. Data analysis
Current and voltage traces were displayed on oscilloscopes and stored on FM magnetic tape (bandwidth: DC-5 kHz). Selected current and voltage data were digitized off-line for analysis, with 12-bit resolution at a sampling rate of 10 kHz. The rise time was measured as the time between 10 and 90% of the peak. The decay time constant was estimated for individual current traces using a least-squares fit to the equation I ( t ) = a • e x p ( - t / r ) + C, where a is the amplitude of relaxation, C is the offset current at t ( ~ ) , and z is the time constant. To study the peak current-voltage relations of EPSCs and the voltage dependence of the decay time constant of EPSCs, the mean and standard deviation of 5-10 measurements at each membrane potential were used. RESULTS Voltage-clamp analysis of CA1 neurons
Intracellular recordings were obtained from 55 cells which had properties characteristic of CA1 pyramidal cells, with resting membrane potentials greater than - 5 0 mV ( - 61.3 + 3.1 mV; mean __+SD). It has been noted that voltage-clamp analysis in mammalian CNS neurons has great limitations because of their extended dendritic arbors, and a simulation study by Johnston and Brown 19 demonstrated that inadequate voltage-clamp of the postsynaptic membrane could lead to underestimation of the measured peak amplitude of postsynaptic potentials (PSPs) and an apparent prolongation of the decay of the synaptic current. Therefore, we adopted the following criteria in selecting data for analysis 9: (1) the decay phases of the EPSCs were reasonably well-fitted by single exponentials at all membrane potentials studied (Fig. 1A); (2) the relationships between the decay time constant of the EPSCs and the membrane potential showed voltage dependence, with the time constant increased by hyperpolarization and decreased by depolarization (Fig. 1B); (3) no significant correlation was seen between the rise time of the EPSCs and the time constant of their decay (Fig. 1C). The following results are based on 28 pyramidal cells which satisfy all of these criteria, and where the effects of JSTX-3 or zinc on EPSCs evoked by stimulation of the stratum radiatum could be properly tested. Effects of J S T X - 3 on synaptic currents
In 14 of 19 cells tested, JSTX-3 at concentrations of 10-200/~M in the pressure pipette greatly reduced the amplitude of the EPSCs. Figure 2A shows superimposed voltage-clamp records with test pulses of - 40 to + 10 mV, relative to the holding potential of - 60 mV, before and after application of JSTX-3. The block by JSTX-3 was generally observed in pyramidal cells where the peak current of the EPSCs increased in an approximately linear manner as the membrane potential was hyperpolarized (Fig. 2B). The peak amplitude of the EPSCs recorded at - 6 0 mV ranged from 300 to 1000 pA in control conditionsl We
203
A
B {a t
517 ms
C
15
~--
z < I-
4
E
lO
:::. :: ' ,... • ,,, .
© (..) UJ ~ F-
".. s
..... • .. :•'...'-
t/)
w
3
• ..-.~•: • *...'°
~_ ,° .~
5
• :
|°
U.I rr
.
2 1
O t.u 123
0
i -100
-150
, -50
t 0
0
0
J 2
i 4
, 6
, 8
1J0
~ 500 pA MEMBRANE POTENTIAL ( mV ) DECAY TIME CONSTANT ( m s ) 10 ms Fig. 1. Example of EPSCs which satisfied the criteria used for the present study. All records were obtained from the same cell using an electrode filled with 4 M potassium acetate and with 5 /~M picrotoxin in the bathing medium. (A) The decay time constants of EPSCs were measured for individual records using a least-square fit to an exponential function, and were well-fitted to single exponentials. Exponential fits to the decay phase of the EPSCs are superimposed on the data at different membrane potentials ( - 5 1 , - 7 1 , - 9 1 mV). (B) The relationship between the decay time constant and the membrane potential. (C) Plots of the decay time constant of the EPSCs against their rise time demonstrate no correlation between these two parameters.
A
500 pA
+JSTX3 10 ms
V B
MEMBRANE
-150
POTENTIAL ( mV )
-100
I
-50
@@@@@@@
@@@l
~0"""
+ JSTX-3
¢oj."
0 0
I
"'~"
-13 m -I
-2
C 20 "11 m Z -4
CONTROL
-3 Fig. 2. Effect of JSTX-3 on EPSCs evoked by stimulation of stratum radiatum in CA1 pyramidal cells. (A) Superimposed voltage-clamp records at 4 different membrane potentials ( - 50, - 60, - 70, - 80 mV) before and 10 min after puff-application of JSTX-3 (200 #M); each trace is the average of 5 successive records. (B) The peak I - V relations of the EPSCs from the cell in Fig. 2A. Vertical bars indicate the standard deviation where this is greater than the size of the symbol, and the dashed line represents the linear regression fit.
204
A
B MEMBRANE POTENTIAL ( mV )
CONTROL
a
-150 F - -
:
- 100
0
-50
0
+ APV
"12 m
+ APV 0 C
b
RECOVERY
-I oo m 7, --I
CONTROL
RECOVERY
C
-200
H,~ ~
""J
I 200 pA 10 ms
Fig. 3. EPSCs with non-linear I-V relationships in a CA1 pyramidal cell. Records were obtained during appfication of 5/~M picrotoxin. (A) Effect of APV on the EPSCs; control response (a), immediately after (b), and 15 rain after (c) puff application of 200 #M APV at a membrane potential of -70 mV. Five successive records were averaged in each case. (B) Peak synaptic current (ordinate) is plotted against membrane potential (abscissa); each data point is the mean of 5 or more measurements. Vertical bars indicate the standard deviation where this is greater than the size of the symbol. In this case APV completely suppressed the EPSC, which implies that the EPSC was a purely NMDA-mediated response. In this cell JSTX-3 had little blocking action. mainly used JSTX-3 at concentrations of 100-200 #M, where clear reductions of peak amplitude were observed. A detectable reduction of peak amplitude could be seen at concentrations as low as 10 /xM. The onset of the blocking action of pressure-applied JSTX-3 took about 5 - 2 0 min, and recovery from the block was not observed even after 1 h. JSTX-3 produced no change in the resting membrane potential or membrane conductance at concentrations between 10 and 500 #M. In the remaining 5 of 19 cells, JSTX-3 (100-200/~M) had little blocking action on the EPSCs. The I - V relationship of these peak EPSCs showed non-linearity between - 1 2 0 and - 5 5 mV (Fig. 3B), and pressure ejection of APV at 100-500 # M reduced the peak current of the EPSCs. In 3 cases, puff-applied (100-200 #M) or bath-applied (20-30 #M) APV completely suppressed the EPSC, but recovery from its effects was rapid (Fig. 3A). These APV-sensitive EPSCs have smaller peak amplitude (100-200 pA) at the resting potential, and appeared to be evoked by lower threshold stimulation, and to have short, fixed latencies to onset. This suggests that JSTX-3 has little effect on EPSCs mediated by N M D A receptors. JSTX-3-insensitive synaptic currents The degree of suppression of EPSCs by JSTX-3 at 100-200/zM appeared to vary from neuron to neuron. In 4 of 14 neurons, JSTX-3 suppressed the EPSCs almost completely (Fig. 2), whereas in 10 other neurons JSTX-3 reduced the EPSCs by 30-80% of the peak. Figure 4A shows an example of the I - V relationship in the latter case. The peak currents of the JSTX-3-insensitive EPSCs (closed circles) did not increase progressively with hyperpolarization and the I - V relation of the peak EPSCs showed non-linearity. However, EPSC amplitudes at 20 ms after the peak (triangles) were not ~ c a n t l y different before and after application of JSTX-3. This implies that the EPSCs remaining after
205
MEMBRANE POTENTIAL ~ mV -150
-100
-50
0 0
I
A
0 0 00, ,, "/ -0.5
O
• .•~•
~ 0 (~{'~J"
+ JSTX-3
m z
s d° CONTROL
0 C
-1.0
f s J~ ./
B
-I .5
CONTROL
+ JSTX-3
JSTX-3 + APV
~=8.31ms
I 100pA 10 ms
15
C
E v
I-Z <
10 O.W
•
i
t •
1°o61 O r
cO Z
O
[
f
'1
+JSTX-3
1
O iO
CONTROL O•
O uJ
0
t
-120
-80
I
-40
I
0
MEMBRANE POTENTIAL ( m V ) Fig. 4. JSTX-3 insensitive EPSCs. (A) The I - V relationships of EPSCs before (open symbols) and 15 min after applying JSTX-3 (closed symbols). Currents were measured at the peak of EPSCs (circles) and at 20 ms after the peak (triangles) as shown in the inset. Vertical bars indicate the standard deviation where this is greater than the size of the symbol. The dashed line represents the linear regression fit. (B) Effects of JSTX-3 and APV on EPSCs. EPSCs recorded before (left), after (center) application of 100/.tM JSTX-3, and a subsequent application of 200 ~M APV (right), at a membrane potential of - 7 0 mV. Each trace is the average of 3 successive records. (C) Decay time constant is plotted against membrane potential. After JSTX-3 application (closed circles), the decay time constant is significantly increased and is less voltage-sensitive than that in the control (open circles). The vertical bars indicate the standard deviation where this is greater than the size of the symbol.
206 application of JSTX-3 are partly mediated by N M D A receptors, although there could also be a n o n - N M D A component from synapses out of range of the local application of JSTX-3.
Co-actioation of NMDA and non-NMDA receptors In order to examine the contribution of N M D A receptors to the EPSCs, we applied the specific N M D A antagonist APV in combination with JSTX-3. In 6 of 10 cells, APV effectively suppressed EPSCs remaining after application of JSTX-3 at the resting membrane potential. In 4 neurons, the remaining EPSCs were not blocked by APV at any of the membrane potential levels tested. An example of the effect of puff-applied APV (200 /~M) is shown in Figure 4B, where APV entirely blocked the residual EPSCs ( 2 / 6 cells). This suggests that the EPSCs remaining in the presence of JSTX-3 are mediated mainly by N M D A receptors. The rise times and time constants of decay of the EPSCs in these 2 cells were compared before and after application of JSTX-3. The 10-90% rise time at the resting membrane potential significantly changed upon application of JSTX-3, from 2.65 + 0.45 ms (mean + SD; n = 4) to 3.38 + 0.74 ms (n = 8) in one cell, and from 3.18 ± 0.30 ms (n = 4) to 4.95 ± 0.63 ms (n = 4) in another cell (t-test, P < 0.05). The decay time constant at the resting membrane potential also increased significantly (t-test, P < 0.05), from 4.24 + 0.12 ms (n = 15) to 5.70 ± 0.11 ms (n = 8) in one cell, and from 4.63 ± 0.68
A CONTROL
+ ZINC
RECOVERY
I 200 pA MEMBRANE POTENTIAL ( mV )
B
C
-150
-100
-50
10 ms 0
+ ZINC I
-lOO :~.j V
o
C
-n m Z
CONTROL
I
-200
1D
>
1oopA
10 ms /
Fig. 5. The effect Of zinc on EPSCs. (A)EPSCs recorded before (left) and after (center) application of 1013.aM zinc, and after recovery(right), at a membrane potential of -65 mV. Five sucr,essive records were a v e r ~ for each trace. (B) Superimposed voltage-clamp records before and after pressure ejection of zinc (100 .aM) at a membrane potential of -50 mV. (C) Peak synaptic current (ordinate) is plotted against membrane potential (abscissa) before (open circles) and after application (closed circles) of zinc. Vertical bars indicate the Standard deviation where this is greater than the size of the symbol; dashed lines represent the linear regression fit.
207 ms (n = 3) to 6.44 + 0.95 ms (n = 4) in another cell after application of JSTX-3 (Fig. 4B). Figure 4C shows the voltage dependence of the decay time constant, where the decay in the control (open circles) slowed progressively with hyperpolarization, while changes of membrane potential after applying JSTX-3 (closed circles) did not significantly modify the decay time constant. This suggests that the time constant of decay of the residual currents after JSTX-3 application reflects the gating of a population of channels with slower kinetics than those blocked by JSTX-3, as has been shown for NMDA-receptormediated synaptic currents 11.15. However, it is also possible that the synapses are located at different distances from the soma and are therefore filtered to different extents.
Effects of zinc We have further examined whether zinc modulates postsynaptic excitatory responses in CA1 pyramidal cells of the slice preparation. In 6 of 9 cells tested, pressure ejection of zinc (100-200 btM) attenuated the EPSCs in CA1 pyramidal cells. No effects were seen in 3 cells. A change of resting membrane conductance by zinc was not detectable at these concentrations. As shown in Figure 5A, zinc (100/xM) produced a reversible attenuation of the EPSCs. Zinc suppressed the EPSCs by nearly 10-25% of the peak, with a similar amount of block by APV in 3 cells where both compounds were tested. I - V relations of the peak amplitudes of EPSCs showed no obvious voltage dependence of block (Fig. 5C). Zinc did not appear to affect the rise time of the EPSCs (Fig. 5B). We have confirmed that the current remaining after applying JSTX-3 was blocked by zinc (n = 2). This suggests that zinc reduces an NMDA-receptor-mediated component of the EPSCs as in cultured neurons 12.21 DISCUSSION
Effects of JSTX-3 In the present study, we have demonstrated that: (1) pressure ejection of a synthesized spider toxin, JSTX-3 (10-200 /tM), suppresses the EPSCs between Schaffer collateralcommissural fibers and CA1 hippocampal pyramidal cells; (2) block by JSTX-3 is observed in pyramidal cells where the EPSCs show linear peak current-voltage (I-V) relations in the control; (3) JSTX-3 has no effect on EPSCs which appear to be mediated solely by N M D A receptor. It is clear, therefore, that JSTX-3 blocks excitatory synaptic transmission mainly by suppressing non-NMDA-receptor-mediated EPSCs in the CA1 pyramidal cells. This is in good agreement with our previous results using purified natural spider toxin 28, where JSTX preferentially blocked iontophoretically evoked quisqualateand kainate-activated responses on CA1 hippocampal pyramidal cells with much less effect on NMDA-activated responses. A recent biochemical study showed that spermine enhances [3H]glycine binding by increasing the affinity of glycine for the N M D A receptor 25. Because JSTX-3 has a polyamine structure 5, it is possible that JSTX-3 might modulate NMDA-receptor-mediated currents to some extent. However, we were not able to detect any effects of JSTX-3 on those EPSCs which were apparently mediated by N M D A receptor. One possibility is that the concentration of extracellular glycine is sufficiently high that binding to the glycine site is already saturated. The action of JSTX-3 on the N M D A receptor remains to be studied.
NMDA-receptor-mediated synaptic currents The present study has also shown that: (1) EPSCs remaining after JSTX-3 application show non-linear peak I - V relationships, and are blocked by the selective N M D A receptor
208 antagonist, APV; (2) the decay time constant of the EPSC is significantly increased and shows less voltage-dependence in the presence of JSTX-3. This suggests that the JSTX-3insensitive component is mediated in part by N M D A receptors in the hippocampal slice. There is a considerable amount of evidence for an NMDA-receptor-mediated component of synaptic transmission in the presence of a physiological concentration of Mg 2+ The N M D A current has previously been revealed in conditions of reduced synaptic inhibition in the presence of picrotoxin, or by stimulating below threshold for activation of GABAergic neurons, as well as in cells depolarized by current injection. Recent studies using a n o n - N M D A receptor antagonist, 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) 3,10,15, have demonstrated that the synaptically released excitatory neurotransmitter acts simultaneously on both N M D A and non-NMDA receptors in CA1 hippocampal pyramidal cells. Thus, our results are basically consistent with these studies. Voltage-clamp analysis of I - V relationships in the present study showed that the EPSCs remaining after JSTX-3 application showed a rather gradual non-linearity, although the I - V relations of N M D A currents in cultured neurons have a characteristic sharp cut-off in conductance at hyperpolarized membrane potentials caused by voltagedependent block by Mg 2+ 7.20. This could be partly due to technical limitations of insufficient space clamp. Because of the limited current-passing capacity of the high-resistance intracellular microelectrode in the single-electrode voltage-clamp technique, the membrane potential can be adequately controlled only over a limited voltage range. The actual potential achieved at the postsynaptic site could be less hyperpolarized than the commanded potential. Another possibility is that JSTX might have a voltage-dependent blocking action on responses at non-NMDA receptors, with relief of block at hyperpolarized potentials as shown for a spider toxin, argiopine, at the crayfish neuromuscular junction 4. However, JSTX has been reported to block the EPSC in a voltage-independent manner at the lobster neuromuscular synapse 22. The detailed kinetics of the blocking action of JSTX on n o n - N M D A receptors in mammalian neurons are currently under investigation 27 The N M D A component of EPSCs has been demonstrated to have a much slower rise time and decay rate than the non-NMDA component. For example, the decay time constant of APV-sensitive NMDA-receptor-mediated EPSCs is about 85 ms (24-25°C) between pairs of cultured neurons 11. In the present study, however, the decay time constants of APV-sensitive EPSCs were between 10 and 20 ms (30-32°C), and were fa~ster than those reported in cultured neurons. This may be explained by differences in experimental conditions, mainly temperature. The slow decay time constant of N M D A receptor-mediated EPSCs has been proposed to result from burst openings of N M D A channels lasting 50-100 ms at room temperature 17, although it is still not clear whether the time course of decay is also determined by the time course of glutamate release. The rates of gating of many channels have temperature coefficients (Q10) of around 3 16. Thus, temperature difference could explain the major part of the discrepancy. It is also possible that, in intact tissues, glial cells might have a role in rapid clearance of transmitter by re-uptake at excitatory synapses. Effects of zinc on synaptic currents Zinc has been reported to reduce both NMDA-evoked currents and slow N M D A receptor-mediated excitatory postsynaptic potentials in cultured hippocampal neurons, and it has been suggested that it modulates excitatory postsynaptic transmission under physiological conditions by blocking N M D A receptors 12,21.24. On the other hand, it has also been shown that zinc has effects on responses to kainate and quisqualate in cultured
209 neurons, producing potentiation at low doses (50 /~M), and antagonism at high doses (100-300/~M) of zinc 21. In the present study, we have demonstrated that: (1) puff-applied zinc (100-200 /~M) reversibly attenuates the EPSCs in CA1 pyramidal cells in the hippocampal slice; (2) zinc appears to affect the slower component of the EPSCs preferentially; (3) zinc blocks the EPSCs which remain after applying JSTX-3. These results support the idea that zinc could modulate postsynaptic neuronal responses to excitatory amino acid transmitters by reducing NMDA-receptor-mediated excitation, although the zinc concentration at the membrane is unknown. The functional significance of zinc in excitatory neurotransmission is currently unknown. Histochemical studies show that zinc is present in vesicles of synaptic terminals throughout the hippocampus 23, and is thought to be released during synaptic transmission." Besides effects on the excitatory amino acid receptors in the postsynaptic membrane, a presynaptic action cannot be ruled out, since zinc is an effective blocker of voltage-dependent calcium channels 13 ACKNOWLEDGEMENTS
We wish to thank Dr. T. Nakajima for the supply of JSTX-3. This work was supported by a Grant-in-Aid (No. 63060006) from the Japanese Ministry of Education, Science and Culture. REFERENCES 1 Abe, T., Kawai, N. and Miwa, A., Effects of a spider toxin on the glutaminergic synapse of lobster muscle, J. Physiol. (Lond.), 339 (1983) 243-252. 2 Akaike, N., Kawai, N., Kiskin, N.I., Kljuchko, E.M., Krishtal, O.A. and Tsyndrenko, A.Ya., Spider toxin blocks excitatory amino acid responses in isolated hippocampal pyramidal neurons, Neurosci. Len., 79 (1987) 326-330. 3 Andreasen, M., Lambert, J.D.C. and Jensen, M.S., Effects of new non-N-methyl-D-aspartate antagonists on synaptic transmission in the in vitro rat hippocampus, J. Physiol. (Lond.), 414 (1989) 317-336. 4 Antonov, S.M., Dudel, J., Franke, C. and Hatt, H., Argiopine blocks glutamate-activated single-channel currents on crayfish muscle by two mechanisms, J. Physiol. (Lond.), 419 (1989) 569-587. 5 Aramaki, Y., Yasuhara, T., Higashijima, T., Yoshioka, M., Miwa, A., Kawai, N. and Nakajima, T., Chemical characterization of spider toxin, JSTX and NSTX, Proc. Jpn. Acad. Ser. B, 62 (1986) 359-362. 6 Asami, T., Kagechika, H., Hashimoto, Y., Shudo, K., Miwa, A., Kawai, N. and Nakajima, T., Acylpolyamines mimic the action of Joro spider toxin (JSTX) on crustacean muscle glutamate receptors, Biomed Res., 10 (1989) 185-189. 7 Ascher, P. and Nowak, L., The role of divalent cations in the N-methyl-D-aspartate responses of mouse central neurones in culture, J. Physiol. (Lond.), 399 (1988) 247-266. 8 Bekkers, J.M. and Stevens, C.F., NMDA and non-NMDA receptors are co-localized at individual excitatory synapses in cultured rat hippocampus, Nature, 341 (1989) 230-233. 9 Brown, T.H. and Johnston, D., Voltage-clamp analysis of mossy fiber synaptic input to hippocampal neurons, J. Neurophysiol., 50 (1983) 487-507. 10 Collingridge, G.L., Herron, C.E. and Lester, R.A.J., Synaptic activation of N-methyl-D-aspartate receptors in the Schaffer collateral-commissural pathway of rat hippocampus, J. Physiol. (Lond.), 399 (1988) 283-300. 11 Forsythe, I.D. and Westbrook, G.L., Slow excitatory postsynaptic currents mediated by N-methyl-D-aspartate receptors on cultured mouse central neurones, J. Physiol. (Lond.), 396 (1988) 515-533. 12 Forsythe, I.D., Westbrook, G.L., and Mayer, M.L., Modulation of excitatory synaptic transmission by glycine and zinc in cultures of mouse hippocampal neurons, J. Neurosci., 8 (1988) 3733-3741. 13 Hagiwara, S. and Takahashi, K., Surface density of calcium ions and calcium spikes in the barnacle muscle fiber membrane, J. Gen. Physiol., 50 (1967) 583-601. 14 Hashimoto, Y., Endo, Y., Shudo, K., Aramaki, Y., Kawai, N. and Nakajima, T., Synthesis of spider toxin (JSTX-3) and its analogs, Tetrahedron Left., 28 (1987) 3511-3514.
210 15 Hestrin, S., Nicoll, R.A., Perkel, D.J. and Sah, P., Analysis of excitatory synaptic action in pyramidal cells using whole-cell recording from rat hippocampal slices, J. Physiol. (Lond.), 422 (1990) 203-225. 16 Hille, B., Ionic Channels of Excitable Membranes, Sinauer Associates Inc., MA, 1984. 17 Howe, J.R., Colquhoun, D. and Cull-Candy, S.G., On the kinetics of large-conductance glutamate-receptor ion channels in rat cerebellar granule neurons, Proc. R. Soc. Lond. B 233 (1988) 407-422. 18 Jackson, H. and Usherwood, P.N.R., Spider toxins as tools for dissecting elements of excitatory amino acid transmission, Trends Neurosci. 11 (1988) 278-283. 19 Johnston, D. and Brown, T.H., Interpretation of voltage-clamp measurements in hippocampal neurons, J. Neurophysiol., 50 (1983) 464-486. 20 Mayer, M.L. and Westbrook. G.L., The action of N-methyl-D-aspartic acid on mouse spinal neurones in culture, J. Physiol. (Lond), 361 (1985) 65-90. 21 Mayer, M.L., Vyklicky, Jr., L. and Westbrook, G.L., Modulation of excitatory amino acid receptors by group liB metal cations in cultured mouse hippocampal neurones, J. Physiol. (Lond.), 415 (1989) 329-350. 22 Miwa, A., Kawai, N., Saito, M., Pan-Hou, H. and Yoshioka, M., Effect of a spider toxin (JSTX) on excitatory postsynaptic current at neuromuscular synapse of spiny lobster, J. Neurophysiol., 58 (1987) 319-326. 23 P6rez-Clausell, J. and Danscher, G., Intravesicular localization of zinc in rat telencephalic boutons. A histochemical study, Brain Res., 337 (1985) 91-98. 24 Peters, S., Koh, J. and Choi, D.W., Zinc selectively blocks the action of N-methyl-D-aspartate on cortical neurons, Science, 236 (1987) 589-593. 25 Sacaan, A.1. and Johnson, K.M., Spermine enhances binding to the glycine site associated with the N-methyl-D-aspartate receptor complex, Mol. Pharmacol., 36 (1989) 836-839. 26 Sahara, Y., Miwa, A. and Kawai, N., NMDA-mediated synaptic transmission in the CA1 pyramidal cells of the guinea-pig hippocampal slice, Soc. Neurosci. Abstr., 14 (1988) 97. 27 Sahara, Y., Robinson, H.P.C, Nakajima, T. and Kawai, N., Blocking mechanisms of a JSTX analogue on kainate and quisqualate activated currents in cultured hippocampal neurons, Jpn. J. Physiol., 40 (1990) Sll0. 28 Saito, M., Sahara, Y., Miwa, A., Shimazaki, K., Nakajima, T. and Kawai, N., Effects of a spider toxin (JSTX) on hippocampal CA1 neurons in vitro, Brain Res., 481 (1989) 16-24. 29 Shimazaki, K., Hagiwara, K., Kawai, N. and Nakajima, T., A spider toxin (JSTX)-binding protein in rat hippocampus, Biomed Res., 10 (1989) 401-403. 30 Shimazaki, K., Hirata, Y., Nakajima, T. and Kawai, N., A histochemical study of glutamate receptor in rat brain using biotinyl spider toxin, Neurosci. Lett., 114 (1990) 1-4.
Neuroscience Research. 10 ( 1991 ) 2()t,~ 2 i ~,~ ' 1991 Elsevier Scientific Publishers Ireland, Ltd. 0168-0102/91 /$~ 3 ;~
NEURES 00423
A voltage-clamp study of the effects of Joro spider toxin and zinc on excitatory synaptic transmission in CA1 pyramidal cells of the guinea pig hippocampal slice Y o s h i n o r i Sahara, H u g h P.C. R o b i n s o n , A k i k o M i w a a n d N o b u f u m i K a w a i Department of Neurobiology, Tokyo Metropolitan Institute for Neurosciences, Tokyo (Japan) (Received 16 August 1990; Revised version received 15 October 1990; Accepted 23 October 1990)
Key words." Single electrode voltage-clamp; Glutamate receptor; JSTX; Zinc; Hippocampus; Slice
SUMMARY Using the single-electrode voltage-clamp technique, we have examined the effects of a non-N-methyl-Daspartate (NMDA) antagonist, Joro spider toxin (JSTX), and of an NMDA antagonist, zinc, on excitatory postsynaptic currents (EPSCs) evoked by stimulation of stratum radiatum in CA1 pyramidal cells of the guinea-pig hippocampal slice. Pressure application of a synthesized JSTX (JSTX-3) at 10-200 /tM greatly reduced the EPSCs (14/19 cells). The block by JSTX-3 was observed in pyramidal cells where the EPSCs showed linear peak current-voltage (I-V) relations in the control. EPSCs remaining after JSTX-3 application showed non-linear peak I-V relationships (10/14 cells), and were blocked by puff application of the selective NMDA receptor antagonist DL-2-amino-5-phosphonovalerate (APV) at 200/~M (6/10 cells). In the presence of JSTX-3, the decay time constant of the EPSC was increased and was less affected by membrane potential. JSTX-3 had no detectable effects on EPSCs apparently mediated solely by NMDA receptor. These observations suggest that JSTX-3 blocks excitatory synaptic transmission mainly by suppressing non-NMDA-rec,eptor-mediated EPSCs, and that the JSTX-3-insensitive component is mediated at least in part by NMDA receptors in the hippocampal slice. Zinc (100-200 btM) reversibly attenuated EPSCs (6/9 cells) and appeared to block a slower component of the EPSCs, suggesting that mainly NMDA receptor-mediated currents were affected.
INTRODUCTION
Toxins isolated from the venom of araneid spiders, such as Joro spider toxin (JSTX) and argiotoxin or argiopine, have been shown to have pronounced blocking effects on excitatory synaptic transmission in invertebrates u8,22. JSTX has also been shown to block kainate- and quisqualate-activated responses in mammalian hippocampal neurons, with little blocking action on N-methyl-D-aspartate (NMDA) responses 2,28. The greatest impediment to studies on JSTX has been the difficulty of isolation of adequate amounts of material from natural spider venoms. Recently, the chemical structure of JSTX has been determined s, and different components of JSTX (JSTX-1 and -3) as well as Correspondence." Dr. Y. Sahara, Department of Neurobioiogy, Tokyo Metropolitan Institute for Neurosciences, 2-6 Musashidai, Fuchu-city, Tokyo 183, Japan.
201 analogues of JSTX (e.g. 1-naphthylacetyl spermine) have been synthesized 6.14. These synthetic toxins have facilitated biochemical and histochemical studies on n o n - N M D A glutamate receptors of the mammalian brain 29,3o In the present study, we have examined the effects of a synthesized JSTX, JSTX-3, on excitatory synaptic transmission in the mammalian hippocampus, using the single-electrode voltage-clamp technique. It is generally accepted that L-glutamate mediates synaptic transmission in the hippocampal excitatory pathway between Schaffer collateral-commissural fibers and CA1 pyramidal cells. Recent studies using brain slices 3,10,~5 and cultured neurons 8,N have demonstrated that both N M D A receptors and n o n - N M D A receptors are involved. We have also examined the effects on excitatory synaptic transmission of zinc, which has been proposed to modulate excitatory postsynaptic transmission by blocking N M D A receptors ~2,21,2n. A preliminary report of this study has been presented in abstract form 26 MATERIALS AND METHODS
Slice preparation The procedure for preparation of hippocampal slices was similar to that described previously 28. Briefly, the hippocampus was removed from adult guinea pigs (200-250 g) and sliced transversely into 300-400/~m sections with a stainless-steel razor blade. The slices were immediately transferred to the experimental chamber which was perfused with bathing medium at a constant rate of approximately 1 ml/min. The bathing medium contained (mM): NaC1 126, KC1 3, N a H 2 P O 4 1.24, MgSO 4 1.3, CaC12 2.4, N a H C O 3 26, and D-glucose 10. The slices were placed on a lens paper and maintained at the interface between a humidified O2/CO 2 mixture (95%/5%) and the bathing medium. Temperature was maintained at 30-32°C. Recordings were obtained after approximately 1-1.5 h of incubation. Stimulation and recording The stimulating bipolar electrode (Teflon-coated platinum iridium wire, 50 /~m) was placed on the stratum radiatum. The intracellular recording electrode, filled with 4 M potassium acetate (resistance 20-80 MO), was advanced through the pyramidal cell layer of the CA1 region. Extracellular field potentials evoked by stimulation of the stratum radiatum (0.3 Hz, 0.2 ms duration, < 100/~A) were used as a guide to help locate the site of recording in the pyramidal cell layer. After establishing stable intracellular recording from a CA1 pyramidal cell, we attempted to achieve a synaptic response in which there was no apparent inhibitory component evoked by stimulation of the stratum radiatum, by adjusting the intensity of the stimulating current and by applying picrotoxin (5 /~M) to the bathing medium during periods of data acquisition; periods for application of picrotoxin were carefully limited since prolonged application frequently induces epileptiform activity. The cell was then voltage-clamped, using an 'Axoclamp-2A' discontinuous single-electrode voltage-clamp amplifier (Axon Instruments) operating at a switching frequency of 3-11 kHz. The holding potential in the voltage clamp was chosen to be close to the resting potential, and step voltage commands (80-100 ms duration) were applied in 5-mV increments, initially in the hyperpolarizing direction and subsequently in the depolarizing direction, over a membrane potential range of - 1 2 0 to - 4 0 mV. To measure the current-voltage relationship of excitatory postsynaptic currents (EPSCs), 5 or more responses were obtained at each membrane potential.
202 Drug application
Synthesized spider toxin JSTX-3 (10-200/tM), DL-2-amino-5-phosphonovalerate (APV, 100-500/IM, Sigma or Tocris Neuramin), and Zn(CH3CO2) 2 (100-200/~M, Wako Pure Chemical Industries Ltd.), were dissolved in the bathing medium and applied to individual neurons using pressure pulses (20-40 psi.) of 10-30 ms duration, from pipettes (tip diameter 5-10/~m) positioned close to the cell. This application method is limited in that the actual concentration of solutions at the membrane surface is unknown; however, considerable dilution certainly occurs. In two experiments, we confirmed that a nearly 10 times higher concentration of APV was required to obtain the same effect for pressure ejection as produced by bath perfusion. Data analysis
Current and voltage traces were displayed on oscilloscopes and stored on FM magnetic tape (bandwidth: DC-5 kHz). Selected current and voltage data were digitized off-line for analysis, with 12-bit resolution at a sampling rate of 10 kHz. The rise time was measured as the time between 10 and 90% of the peak. The decay time constant was estimated for individual current traces using a least-squares fit to the equation I ( t ) = a • e x p ( - t / r ) + C, where a is the amplitude of relaxation, C is the offset current at t ( ~ ) , and z is the time constant. To study the peak current-voltage relations of EPSCs and the voltage dependence of the decay time constant of EPSCs, the mean and standard deviation of 5-10 measurements at each membrane potential were used. RESULTS Voltage-clamp analysis of CA1 neurons
Intracellular recordings were obtained from 55 cells which had properties characteristic of CA1 pyramidal cells, with resting membrane potentials greater than - 5 0 mV ( - 61.3 + 3.1 mV; mean __+SD). It has been noted that voltage-clamp analysis in mammalian CNS neurons has great limitations because of their extended dendritic arbors, and a simulation study by Johnston and Brown 19 demonstrated that inadequate voltage-clamp of the postsynaptic membrane could lead to underestimation of the measured peak amplitude of postsynaptic potentials (PSPs) and an apparent prolongation of the decay of the synaptic current. Therefore, we adopted the following criteria in selecting data for analysis 9: (1) the decay phases of the EPSCs were reasonably well-fitted by single exponentials at all membrane potentials studied (Fig. 1A); (2) the relationships between the decay time constant of the EPSCs and the membrane potential showed voltage dependence, with the time constant increased by hyperpolarization and decreased by depolarization (Fig. 1B); (3) no significant correlation was seen between the rise time of the EPSCs and the time constant of their decay (Fig. 1C). The following results are based on 28 pyramidal cells which satisfy all of these criteria, and where the effects of JSTX-3 or zinc on EPSCs evoked by stimulation of the stratum radiatum could be properly tested. Effects of J S T X - 3 on synaptic currents
In 14 of 19 cells tested, JSTX-3 at concentrations of 10-200/~M in the pressure pipette greatly reduced the amplitude of the EPSCs. Figure 2A shows superimposed voltage-clamp records with test pulses of - 40 to + 10 mV, relative to the holding potential of - 60 mV, before and after application of JSTX-3. The block by JSTX-3 was generally observed in pyramidal cells where the peak current of the EPSCs increased in an approximately linear manner as the membrane potential was hyperpolarized (Fig. 2B). The peak amplitude of the EPSCs recorded at - 6 0 mV ranged from 300 to 1000 pA in control conditionsl We
203
A
B {a t
517 ms
C
15
~--
z < I-
4
E
lO
:::. :: ' ,... • ,,, .
© (..) UJ ~ F-
".. s
..... • .. :•'...'-
t/)
w
3
• ..-.~•: • *...'°
~_ ,° .~
5
• :
|°
U.I rr
.
2 1
O t.u 123
0
i -100
-150
, -50
t 0
0
0
J 2
i 4
, 6
, 8
1J0
~ 500 pA MEMBRANE POTENTIAL ( mV ) DECAY TIME CONSTANT ( m s ) 10 ms Fig. 1. Example of EPSCs which satisfied the criteria used for the present study. All records were obtained from the same cell using an electrode filled with 4 M potassium acetate and with 5 /~M picrotoxin in the bathing medium. (A) The decay time constants of EPSCs were measured for individual records using a least-square fit to an exponential function, and were well-fitted to single exponentials. Exponential fits to the decay phase of the EPSCs are superimposed on the data at different membrane potentials ( - 5 1 , - 7 1 , - 9 1 mV). (B) The relationship between the decay time constant and the membrane potential. (C) Plots of the decay time constant of the EPSCs against their rise time demonstrate no correlation between these two parameters.
A
500 pA
+JSTX3 10 ms
V B
MEMBRANE
-150
POTENTIAL ( mV )
-100
I
-50
@@@@@@@
@@@l
~0"""
+ JSTX-3
¢oj."
0 0
I
"'~"
-13 m -I
-2
C 20 "11 m Z -4
CONTROL
-3 Fig. 2. Effect of JSTX-3 on EPSCs evoked by stimulation of stratum radiatum in CA1 pyramidal cells. (A) Superimposed voltage-clamp records at 4 different membrane potentials ( - 50, - 60, - 70, - 80 mV) before and 10 min after puff-application of JSTX-3 (200 #M); each trace is the average of 5 successive records. (B) The peak I - V relations of the EPSCs from the cell in Fig. 2A. Vertical bars indicate the standard deviation where this is greater than the size of the symbol, and the dashed line represents the linear regression fit.
204
A
B MEMBRANE POTENTIAL ( mV )
CONTROL
a
-150 F - -
:
- 100
0
-50
0
+ APV
"12 m
+ APV 0 C
b
RECOVERY
-I oo m 7, --I
CONTROL
RECOVERY
C
-200
H,~ ~
""J
I 200 pA 10 ms
Fig. 3. EPSCs with non-linear I-V relationships in a CA1 pyramidal cell. Records were obtained during appfication of 5/~M picrotoxin. (A) Effect of APV on the EPSCs; control response (a), immediately after (b), and 15 rain after (c) puff application of 200 #M APV at a membrane potential of -70 mV. Five successive records were averaged in each case. (B) Peak synaptic current (ordinate) is plotted against membrane potential (abscissa); each data point is the mean of 5 or more measurements. Vertical bars indicate the standard deviation where this is greater than the size of the symbol. In this case APV completely suppressed the EPSC, which implies that the EPSC was a purely NMDA-mediated response. In this cell JSTX-3 had little blocking action. mainly used JSTX-3 at concentrations of 100-200 #M, where clear reductions of peak amplitude were observed. A detectable reduction of peak amplitude could be seen at concentrations as low as 10 /xM. The onset of the blocking action of pressure-applied JSTX-3 took about 5 - 2 0 min, and recovery from the block was not observed even after 1 h. JSTX-3 produced no change in the resting membrane potential or membrane conductance at concentrations between 10 and 500 #M. In the remaining 5 of 19 cells, JSTX-3 (100-200/~M) had little blocking action on the EPSCs. The I - V relationship of these peak EPSCs showed non-linearity between - 1 2 0 and - 5 5 mV (Fig. 3B), and pressure ejection of APV at 100-500 # M reduced the peak current of the EPSCs. In 3 cases, puff-applied (100-200 #M) or bath-applied (20-30 #M) APV completely suppressed the EPSC, but recovery from its effects was rapid (Fig. 3A). These APV-sensitive EPSCs have smaller peak amplitude (100-200 pA) at the resting potential, and appeared to be evoked by lower threshold stimulation, and to have short, fixed latencies to onset. This suggests that JSTX-3 has little effect on EPSCs mediated by N M D A receptors. JSTX-3-insensitive synaptic currents The degree of suppression of EPSCs by JSTX-3 at 100-200/zM appeared to vary from neuron to neuron. In 4 of 14 neurons, JSTX-3 suppressed the EPSCs almost completely (Fig. 2), whereas in 10 other neurons JSTX-3 reduced the EPSCs by 30-80% of the peak. Figure 4A shows an example of the I - V relationship in the latter case. The peak currents of the JSTX-3-insensitive EPSCs (closed circles) did not increase progressively with hyperpolarization and the I - V relation of the peak EPSCs showed non-linearity. However, EPSC amplitudes at 20 ms after the peak (triangles) were not ~ c a n t l y different before and after application of JSTX-3. This implies that the EPSCs remaining after
205
MEMBRANE POTENTIAL ~ mV -150
-100
-50
0 0
I
A
0 0 00, ,, "/ -0.5
O
• .•~•
~ 0 (~{'~J"
+ JSTX-3
m z
s d° CONTROL
0 C
-1.0
f s J~ ./
B
-I .5
CONTROL
+ JSTX-3
JSTX-3 + APV
~=8.31ms
I 100pA 10 ms
15
C
E v
I-Z <
10 O.W
•
i
t •
1°o61 O r
cO Z
O
[
f
'1
+JSTX-3
1
O iO
CONTROL O•
O uJ
0
t
-120
-80
I
-40
I
0
MEMBRANE POTENTIAL ( m V ) Fig. 4. JSTX-3 insensitive EPSCs. (A) The I - V relationships of EPSCs before (open symbols) and 15 min after applying JSTX-3 (closed symbols). Currents were measured at the peak of EPSCs (circles) and at 20 ms after the peak (triangles) as shown in the inset. Vertical bars indicate the standard deviation where this is greater than the size of the symbol. The dashed line represents the linear regression fit. (B) Effects of JSTX-3 and APV on EPSCs. EPSCs recorded before (left), after (center) application of 100/.tM JSTX-3, and a subsequent application of 200 ~M APV (right), at a membrane potential of - 7 0 mV. Each trace is the average of 3 successive records. (C) Decay time constant is plotted against membrane potential. After JSTX-3 application (closed circles), the decay time constant is significantly increased and is less voltage-sensitive than that in the control (open circles). The vertical bars indicate the standard deviation where this is greater than the size of the symbol.
206 application of JSTX-3 are partly mediated by N M D A receptors, although there could also be a n o n - N M D A component from synapses out of range of the local application of JSTX-3.
Co-actioation of NMDA and non-NMDA receptors In order to examine the contribution of N M D A receptors to the EPSCs, we applied the specific N M D A antagonist APV in combination with JSTX-3. In 6 of 10 cells, APV effectively suppressed EPSCs remaining after application of JSTX-3 at the resting membrane potential. In 4 neurons, the remaining EPSCs were not blocked by APV at any of the membrane potential levels tested. An example of the effect of puff-applied APV (200 /~M) is shown in Figure 4B, where APV entirely blocked the residual EPSCs ( 2 / 6 cells). This suggests that the EPSCs remaining in the presence of JSTX-3 are mediated mainly by N M D A receptors. The rise times and time constants of decay of the EPSCs in these 2 cells were compared before and after application of JSTX-3. The 10-90% rise time at the resting membrane potential significantly changed upon application of JSTX-3, from 2.65 + 0.45 ms (mean + SD; n = 4) to 3.38 + 0.74 ms (n = 8) in one cell, and from 3.18 ± 0.30 ms (n = 4) to 4.95 ± 0.63 ms (n = 4) in another cell (t-test, P < 0.05). The decay time constant at the resting membrane potential also increased significantly (t-test, P < 0.05), from 4.24 + 0.12 ms (n = 15) to 5.70 ± 0.11 ms (n = 8) in one cell, and from 4.63 ± 0.68
A CONTROL
+ ZINC
RECOVERY
I 200 pA MEMBRANE POTENTIAL ( mV )
B
C
-150
-100
-50
10 ms 0
+ ZINC I
-lOO :~.j V
o
C
-n m Z
CONTROL
I
-200
1D
>
1oopA
10 ms /
Fig. 5. The effect Of zinc on EPSCs. (A)EPSCs recorded before (left) and after (center) application of 1013.aM zinc, and after recovery(right), at a membrane potential of -65 mV. Five sucr,essive records were a v e r ~ for each trace. (B) Superimposed voltage-clamp records before and after pressure ejection of zinc (100 .aM) at a membrane potential of -50 mV. (C) Peak synaptic current (ordinate) is plotted against membrane potential (abscissa) before (open circles) and after application (closed circles) of zinc. Vertical bars indicate the Standard deviation where this is greater than the size of the symbol; dashed lines represent the linear regression fit.
207 ms (n = 3) to 6.44 + 0.95 ms (n = 4) in another cell after application of JSTX-3 (Fig. 4B). Figure 4C shows the voltage dependence of the decay time constant, where the decay in the control (open circles) slowed progressively with hyperpolarization, while changes of membrane potential after applying JSTX-3 (closed circles) did not significantly modify the decay time constant. This suggests that the time constant of decay of the residual currents after JSTX-3 application reflects the gating of a population of channels with slower kinetics than those blocked by JSTX-3, as has been shown for NMDA-receptormediated synaptic currents 11.15. However, it is also possible that the synapses are located at different distances from the soma and are therefore filtered to different extents.
Effects of zinc We have further examined whether zinc modulates postsynaptic excitatory responses in CA1 pyramidal cells of the slice preparation. In 6 of 9 cells tested, pressure ejection of zinc (100-200 btM) attenuated the EPSCs in CA1 pyramidal cells. No effects were seen in 3 cells. A change of resting membrane conductance by zinc was not detectable at these concentrations. As shown in Figure 5A, zinc (100/xM) produced a reversible attenuation of the EPSCs. Zinc suppressed the EPSCs by nearly 10-25% of the peak, with a similar amount of block by APV in 3 cells where both compounds were tested. I - V relations of the peak amplitudes of EPSCs showed no obvious voltage dependence of block (Fig. 5C). Zinc did not appear to affect the rise time of the EPSCs (Fig. 5B). We have confirmed that the current remaining after applying JSTX-3 was blocked by zinc (n = 2). This suggests that zinc reduces an NMDA-receptor-mediated component of the EPSCs as in cultured neurons 12.21 DISCUSSION
Effects of JSTX-3 In the present study, we have demonstrated that: (1) pressure ejection of a synthesized spider toxin, JSTX-3 (10-200 /tM), suppresses the EPSCs between Schaffer collateralcommissural fibers and CA1 hippocampal pyramidal cells; (2) block by JSTX-3 is observed in pyramidal cells where the EPSCs show linear peak current-voltage (I-V) relations in the control; (3) JSTX-3 has no effect on EPSCs which appear to be mediated solely by N M D A receptor. It is clear, therefore, that JSTX-3 blocks excitatory synaptic transmission mainly by suppressing non-NMDA-receptor-mediated EPSCs in the CA1 pyramidal cells. This is in good agreement with our previous results using purified natural spider toxin 28, where JSTX preferentially blocked iontophoretically evoked quisqualateand kainate-activated responses on CA1 hippocampal pyramidal cells with much less effect on NMDA-activated responses. A recent biochemical study showed that spermine enhances [3H]glycine binding by increasing the affinity of glycine for the N M D A receptor 25. Because JSTX-3 has a polyamine structure 5, it is possible that JSTX-3 might modulate NMDA-receptor-mediated currents to some extent. However, we were not able to detect any effects of JSTX-3 on those EPSCs which were apparently mediated by N M D A receptor. One possibility is that the concentration of extracellular glycine is sufficiently high that binding to the glycine site is already saturated. The action of JSTX-3 on the N M D A receptor remains to be studied.
NMDA-receptor-mediated synaptic currents The present study has also shown that: (1) EPSCs remaining after JSTX-3 application show non-linear peak I - V relationships, and are blocked by the selective N M D A receptor
208 antagonist, APV; (2) the decay time constant of the EPSC is significantly increased and shows less voltage-dependence in the presence of JSTX-3. This suggests that the JSTX-3insensitive component is mediated in part by N M D A receptors in the hippocampal slice. There is a considerable amount of evidence for an NMDA-receptor-mediated component of synaptic transmission in the presence of a physiological concentration of Mg 2+ The N M D A current has previously been revealed in conditions of reduced synaptic inhibition in the presence of picrotoxin, or by stimulating below threshold for activation of GABAergic neurons, as well as in cells depolarized by current injection. Recent studies using a n o n - N M D A receptor antagonist, 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) 3,10,15, have demonstrated that the synaptically released excitatory neurotransmitter acts simultaneously on both N M D A and non-NMDA receptors in CA1 hippocampal pyramidal cells. Thus, our results are basically consistent with these studies. Voltage-clamp analysis of I - V relationships in the present study showed that the EPSCs remaining after JSTX-3 application showed a rather gradual non-linearity, although the I - V relations of N M D A currents in cultured neurons have a characteristic sharp cut-off in conductance at hyperpolarized membrane potentials caused by voltagedependent block by Mg 2+ 7.20. This could be partly due to technical limitations of insufficient space clamp. Because of the limited current-passing capacity of the high-resistance intracellular microelectrode in the single-electrode voltage-clamp technique, the membrane potential can be adequately controlled only over a limited voltage range. The actual potential achieved at the postsynaptic site could be less hyperpolarized than the commanded potential. Another possibility is that JSTX might have a voltage-dependent blocking action on responses at non-NMDA receptors, with relief of block at hyperpolarized potentials as shown for a spider toxin, argiopine, at the crayfish neuromuscular junction 4. However, JSTX has been reported to block the EPSC in a voltage-independent manner at the lobster neuromuscular synapse 22. The detailed kinetics of the blocking action of JSTX on n o n - N M D A receptors in mammalian neurons are currently under investigation 27 The N M D A component of EPSCs has been demonstrated to have a much slower rise time and decay rate than the non-NMDA component. For example, the decay time constant of APV-sensitive NMDA-receptor-mediated EPSCs is about 85 ms (24-25°C) between pairs of cultured neurons 11. In the present study, however, the decay time constants of APV-sensitive EPSCs were between 10 and 20 ms (30-32°C), and were fa~ster than those reported in cultured neurons. This may be explained by differences in experimental conditions, mainly temperature. The slow decay time constant of N M D A receptor-mediated EPSCs has been proposed to result from burst openings of N M D A channels lasting 50-100 ms at room temperature 17, although it is still not clear whether the time course of decay is also determined by the time course of glutamate release. The rates of gating of many channels have temperature coefficients (Q10) of around 3 16. Thus, temperature difference could explain the major part of the discrepancy. It is also possible that, in intact tissues, glial cells might have a role in rapid clearance of transmitter by re-uptake at excitatory synapses. Effects of zinc on synaptic currents Zinc has been reported to reduce both NMDA-evoked currents and slow N M D A receptor-mediated excitatory postsynaptic potentials in cultured hippocampal neurons, and it has been suggested that it modulates excitatory postsynaptic transmission under physiological conditions by blocking N M D A receptors 12,21.24. On the other hand, it has also been shown that zinc has effects on responses to kainate and quisqualate in cultured
209 neurons, producing potentiation at low doses (50 /~M), and antagonism at high doses (100-300/~M) of zinc 21. In the present study, we have demonstrated that: (1) puff-applied zinc (100-200 /~M) reversibly attenuates the EPSCs in CA1 pyramidal cells in the hippocampal slice; (2) zinc appears to affect the slower component of the EPSCs preferentially; (3) zinc blocks the EPSCs which remain after applying JSTX-3. These results support the idea that zinc could modulate postsynaptic neuronal responses to excitatory amino acid transmitters by reducing NMDA-receptor-mediated excitation, although the zinc concentration at the membrane is unknown. The functional significance of zinc in excitatory neurotransmission is currently unknown. Histochemical studies show that zinc is present in vesicles of synaptic terminals throughout the hippocampus 23, and is thought to be released during synaptic transmission." Besides effects on the excitatory amino acid receptors in the postsynaptic membrane, a presynaptic action cannot be ruled out, since zinc is an effective blocker of voltage-dependent calcium channels 13 ACKNOWLEDGEMENTS
We wish to thank Dr. T. Nakajima for the supply of JSTX-3. This work was supported by a Grant-in-Aid (No. 63060006) from the Japanese Ministry of Education, Science and Culture. REFERENCES 1 Abe, T., Kawai, N. and Miwa, A., Effects of a spider toxin on the glutaminergic synapse of lobster muscle, J. Physiol. (Lond.), 339 (1983) 243-252. 2 Akaike, N., Kawai, N., Kiskin, N.I., Kljuchko, E.M., Krishtal, O.A. and Tsyndrenko, A.Ya., Spider toxin blocks excitatory amino acid responses in isolated hippocampal pyramidal neurons, Neurosci. Len., 79 (1987) 326-330. 3 Andreasen, M., Lambert, J.D.C. and Jensen, M.S., Effects of new non-N-methyl-D-aspartate antagonists on synaptic transmission in the in vitro rat hippocampus, J. Physiol. (Lond.), 414 (1989) 317-336. 4 Antonov, S.M., Dudel, J., Franke, C. and Hatt, H., Argiopine blocks glutamate-activated single-channel currents on crayfish muscle by two mechanisms, J. Physiol. (Lond.), 419 (1989) 569-587. 5 Aramaki, Y., Yasuhara, T., Higashijima, T., Yoshioka, M., Miwa, A., Kawai, N. and Nakajima, T., Chemical characterization of spider toxin, JSTX and NSTX, Proc. Jpn. Acad. Ser. B, 62 (1986) 359-362. 6 Asami, T., Kagechika, H., Hashimoto, Y., Shudo, K., Miwa, A., Kawai, N. and Nakajima, T., Acylpolyamines mimic the action of Joro spider toxin (JSTX) on crustacean muscle glutamate receptors, Biomed Res., 10 (1989) 185-189. 7 Ascher, P. and Nowak, L., The role of divalent cations in the N-methyl-D-aspartate responses of mouse central neurones in culture, J. Physiol. (Lond.), 399 (1988) 247-266. 8 Bekkers, J.M. and Stevens, C.F., NMDA and non-NMDA receptors are co-localized at individual excitatory synapses in cultured rat hippocampus, Nature, 341 (1989) 230-233. 9 Brown, T.H. and Johnston, D., Voltage-clamp analysis of mossy fiber synaptic input to hippocampal neurons, J. Neurophysiol., 50 (1983) 487-507. 10 Collingridge, G.L., Herron, C.E. and Lester, R.A.J., Synaptic activation of N-methyl-D-aspartate receptors in the Schaffer collateral-commissural pathway of rat hippocampus, J. Physiol. (Lond.), 399 (1988) 283-300. 11 Forsythe, I.D. and Westbrook, G.L., Slow excitatory postsynaptic currents mediated by N-methyl-D-aspartate receptors on cultured mouse central neurones, J. Physiol. (Lond.), 396 (1988) 515-533. 12 Forsythe, I.D., Westbrook, G.L., and Mayer, M.L., Modulation of excitatory synaptic transmission by glycine and zinc in cultures of mouse hippocampal neurons, J. Neurosci., 8 (1988) 3733-3741. 13 Hagiwara, S. and Takahashi, K., Surface density of calcium ions and calcium spikes in the barnacle muscle fiber membrane, J. Gen. Physiol., 50 (1967) 583-601. 14 Hashimoto, Y., Endo, Y., Shudo, K., Aramaki, Y., Kawai, N. and Nakajima, T., Synthesis of spider toxin (JSTX-3) and its analogs, Tetrahedron Left., 28 (1987) 3511-3514.
210 15 Hestrin, S., Nicoll, R.A., Perkel, D.J. and Sah, P., Analysis of excitatory synaptic action in pyramidal cells using whole-cell recording from rat hippocampal slices, J. Physiol. (Lond.), 422 (1990) 203-225. 16 Hille, B., Ionic Channels of Excitable Membranes, Sinauer Associates Inc., MA, 1984. 17 Howe, J.R., Colquhoun, D. and Cull-Candy, S.G., On the kinetics of large-conductance glutamate-receptor ion channels in rat cerebellar granule neurons, Proc. R. Soc. Lond. B 233 (1988) 407-422. 18 Jackson, H. and Usherwood, P.N.R., Spider toxins as tools for dissecting elements of excitatory amino acid transmission, Trends Neurosci. 11 (1988) 278-283. 19 Johnston, D. and Brown, T.H., Interpretation of voltage-clamp measurements in hippocampal neurons, J. Neurophysiol., 50 (1983) 464-486. 20 Mayer, M.L. and Westbrook. G.L., The action of N-methyl-D-aspartic acid on mouse spinal neurones in culture, J. Physiol. (Lond), 361 (1985) 65-90. 21 Mayer, M.L., Vyklicky, Jr., L. and Westbrook, G.L., Modulation of excitatory amino acid receptors by group liB metal cations in cultured mouse hippocampal neurones, J. Physiol. (Lond.), 415 (1989) 329-350. 22 Miwa, A., Kawai, N., Saito, M., Pan-Hou, H. and Yoshioka, M., Effect of a spider toxin (JSTX) on excitatory postsynaptic current at neuromuscular synapse of spiny lobster, J. Neurophysiol., 58 (1987) 319-326. 23 P6rez-Clausell, J. and Danscher, G., Intravesicular localization of zinc in rat telencephalic boutons. A histochemical study, Brain Res., 337 (1985) 91-98. 24 Peters, S., Koh, J. and Choi, D.W., Zinc selectively blocks the action of N-methyl-D-aspartate on cortical neurons, Science, 236 (1987) 589-593. 25 Sacaan, A.1. and Johnson, K.M., Spermine enhances binding to the glycine site associated with the N-methyl-D-aspartate receptor complex, Mol. Pharmacol., 36 (1989) 836-839. 26 Sahara, Y., Miwa, A. and Kawai, N., NMDA-mediated synaptic transmission in the CA1 pyramidal cells of the guinea-pig hippocampal slice, Soc. Neurosci. Abstr., 14 (1988) 97. 27 Sahara, Y., Robinson, H.P.C, Nakajima, T. and Kawai, N., Blocking mechanisms of a JSTX analogue on kainate and quisqualate activated currents in cultured hippocampal neurons, Jpn. J. Physiol., 40 (1990) Sll0. 28 Saito, M., Sahara, Y., Miwa, A., Shimazaki, K., Nakajima, T. and Kawai, N., Effects of a spider toxin (JSTX) on hippocampal CA1 neurons in vitro, Brain Res., 481 (1989) 16-24. 29 Shimazaki, K., Hagiwara, K., Kawai, N. and Nakajima, T., A spider toxin (JSTX)-binding protein in rat hippocampus, Biomed Res., 10 (1989) 401-403. 30 Shimazaki, K., Hirata, Y., Nakajima, T. and Kawai, N., A histochemical study of glutamate receptor in rat brain using biotinyl spider toxin, Neurosci. Lett., 114 (1990) 1-4.