Changes in field excitatory postsynaptic potential shape induced by tetanization in the ca1 region of the guinea-pig hippocampal slice

Neuro.wknce Vol. 37. No. I. pp. 6169. Printed in Great Britain

03064522/90

1990

$3.00 + 0.00

Pergamon Press plc Q 1990IBRO

CHANGES IN FIELD EXCITATORY POSTSYNAPTIC POTENTIAL SHAPE INDUCED BY TETANIZATION IN THE CA1 REGION OF THE GUINEA-PIG HIPPOCAMPAL SLICE G. HEW* and B. GusTAFssoNt Department of Physiology, University of Giiteborg, P.O. Box 33031, S-400 33 Goteborg, Sweden Abstract-The

present paper contains a description of a prolonged potentiation of the field excitatory postsynaptic potential in the CA1 region of the hippocampal slice preparation following afferent tetanization. In contrast to long-term potentiation, this novel potentiation is not specific to the activated synapses, and manifests itself as a change in the shape of the field excitatory postsynaptic potential with a prolongation of the rising phase and an increased peak amplitude. The potentiation is fully developed within minutes after tetanization and shows no decrement for at least an hour. Although it can appear together with long-term potentiation following tetanization at moderate strength (single volley excitatory postsynaptic potential below threshold for spike initiation), it is more readily seen following tetanization at higher strengths. The N-methyl-u-aspartate receptor antagonist 2-amino-5-phosphonovalerate prevents the induction but not the maintenance of the shape modification. The potentiation is observed in the presence of the GABA, antagonist picrotoxin (100 PM) and is thus not secondary to changes in postsynaptic inhibition. 4-Aminopyridine (50-100 PM) produced changes in the field excitatory postsynaptic potential resembling the shape modification produced by afferent tetanization, suggesting that the potentiation may be due to a blockade of potassium channels, pre- or postsynaptically located. The potentiation is also found to be associated with an increase in the population spike for a given initial slope of the field excitatory postsynaptic potential, and may thus contribute to the excitatory postsynaptic potential-spike potentiation that can be observed following afferent tetanization.

(LTP) in the hippocampus as an enduring increase of the excitatory postsynaptic potential (EPSP) and/or population spike following brief high-frequency afferent tetanization. LTP defined in the above manner may, however, not represent a unitary phenomenon. For example, when the slope of the potentiated field EPSP has been returned to pre-tetanization levels, the EPSP is often associated with a larger population spike than before [EPSP-spike (E-S) potentiation].2.3 According to one hypothesis put forward to account for E-S potentiation, I9 this potentiation may, however, only be an indirect consequence of the fact that tetanization induces a relatively greater potentiation of the monosynaptic excitatory pathway than of the disynaptic inhibitory one. This idea has recently received support from the findings that E-S potentiation is associated with a change in the shape of the field potential suggesting a relatively greater potentiation of the EPSP than of the inhibitory postsynaptic potential (IPSP), and is much reduced following blockade of inhibitory transmission.’ It was also

found, under this latter condition, that the EPSP by LTP-inducing time-course was unaffected tetanization, suggesting that tetanization results in a change in excitatory transmission without change of the temporal properties of the current flow at the excitatory synapses or in passive or active membrane properties that affect the conduction of the EPSP along the dendrites. According to these results E-S potentiation is thus not a separate potentiation phenomena from that measured as a change in the EPSP initial slope. On the other hand, it has been suggested that E-S potentiation may represent a modification of the postsynaptic membrane properties, i.e. that tetanization may give rise to rather local dendritic changes that enhance the synaptic efficacy.” The question then arises as to what extent the appropriate conditions for inducing and/or detecting such changes have been met in experiments in which the EPSP time-course following tetanization was examined. Therefore, in the present study, we have studied, under conditions of blockade of GABA,-mediated inhibition, the time-course of field EPSPs following tetanization at various strengths.

Long-term potentiation is generally described

*Present address: Institute of Zoology, Jagellonian University, Krakow, Poland. tTo whom correspondence should be addressed. Abbreoiarions: APV, 2-amino-5-phosphonovalerate; EPSP, excitatory postsynaptic potential; E-S, EPSP-spike: IPSP, inhibitory postsynaptic potential; LTP, long-term potentiation; NMDA, N-methyl-o-aspartate.

EXPERIMENTAL

PROCEDURES

The experiments were performed on hippocampal slices from 28 guinea-pigs prepared as previously described.” The slices were kept on a nylon net in a constant flow 61

62

G. HESSand B. GUSTAFSSON

tncubation chamber at 30-32 C. half-submerged by a solution containing (in mM): NaCl 124, KC1 4. CaCl, 4, MgCI: 4. NaHCO, 26. glucose IO, picrotoxin 0.1, and gassed with 95% O,-5% CO-. Concentrations of calcium and magnesium in this solution are higher than normal in order to counteract the excitability increase obtained in the presence of picrotoxin. Thus, spontaneous repetitive field potentials or late discharges following stimulation were only rarely observed in these slices. In some experiments a surgical cut was made between CA1 and CA3 regions to exclude any participation of discharges from the CA3 region in the responses observed. Two stimulating electrodes were placed in the stratum radiatum layer of the CAI region on opposite sides of the recording electrodes to obtain activation of two independent inputs to the same dendritic region.” Stimuli consisted of 0.1 ms negative constant current pulses (20-60pA) delivered through electrolytically sharpened tungsten wires (monopolar stimulation). For tetanic activation, 50 Hz trains were used. Field potentials were recorded from the dendritic and cell body layers using glass micropipettes filled with 3 M NaCl (5-10 MR). After passing through a high impedance preamplifier, the signals were connected to an analogue-digital converter and fed into a Nord-100 minicomputer. The latter was used for on-line signal averaging

test input

A mv/rns 0.75 -

and for measurement of waveform parameters. The field EPSP was measured. using linear regression, as the initial slope of its rising phase, and LTP was measured as percentage increase of this slope. In order to compare the shape of EPSPs before and after an LTP-inducing tetanization. the stimulation strength of the tetanized input was adjusted during the experiment to keep the initial slope of the field EPSP to about the same value as before the tetanization (see Fig. IA). For the final comparison, the averaged records of EPSPs were, if needed, slightly adjusted by multiplication with a constant factor in order to match the initial slopes of the EPSPs (cf. Fig. 4). Malerials

Picrotoxin and 4-aminopyridine were obtained from Sigma Chemical Co. and o-2-amino-5-phosphonovalerate (D-APV) from Tocris Neuramin. APV (5mM) was either disolved in a solution equivalent to that of the perfusion solution and applied as a droplet onto the surface of the slice, or added to the perfusion solution (50 PM). RESULTS

A field EPSP recorded after tetanization in 0.1 mM picrotoxin solution superimposed, after reduction of

B

i L, j

30

60. min

Fig. 1. Effect of tetanization on field EPSP shape. (A and B) Measurements of initial slope of field EPSPs resulting from alternating stimulation (at 0.5 Hz) of test (A) and control (B) inputs are plotted for a series of successive responses. (C and D) Averaged records of iicld EPSPs (n = 30), taken at times indicated in A and B. The test input was tetanized twice with five 24impulac trains (once every 12 s) using 1x and 2 x test strength, respectively, as indicated in A. As in this experiment, stimulus strength was always reduced following tetanization at I x test strength to return the initial slope of the test EPSP to the pretetanization level before tetanization at 2 x test strength.

63

Tetanization-induced field EPSP shape change stimulus strength, on that obtained before tetanization is shown in Fig. lC2. In agreement with an earlier study,’ the field EPSP time-course seems essentially unaltered following this tetanization. This result was obtained in most (24/37) of the cases examined after tetanization with five 20-impulse trains using test strength, these trains evoking LTP varying from 42 to 174% (mean value 93%). An additional tetanization of the above input (but at 2 x test strength) led to little further potentiation of the EPSP initial slope (Fig. lA), but to a substantial change in the EPSP peak amplitude (Fig. lC3). Figure lC4 shows that this potentiation of the EPSP peak amplitude still remains after the stimulation strength was reduced to match the rising phase of the post-tetanus field EPSP with that of the pretetanus one. This modification, which will be the subject of this paper, was clearly observed after tetanization at 1 x test strength in 12 out of the 37 inputs examined, and, after a subsequent tetanization at 2 x test strength, in 21 out of the 25 remaining ones. In three out of these 21 cases, the A 1.0 mV/ms

test input

Duration of the change in shape of the field excitatory postsynaptic potential

In the experiment illustrated in Fig. 2, the stimulation strength was reduced 34min following the tetanization (with 2 x test strength) to match the field EPSP initial slope to that of the pre-tetanus one, and was adjusted to allow for a comparison of EPSP shapes for up to an hour. As shown by the averaged records of field EPSPs taken at different times following tetanization (Fig. 2C) shown superimposed on the pre-tetanus field EPSP, the shape modification was practically fully developed a few minutes after the tetanization and was thereafter maintained to much the same extent for over an hour. B

: :

i

tetanization event using 2 x test strength had been preceded by an additional tetanization event using 1 x test strength. This second tetanization (at 1 x test strength) was ineffective in producing the shape modification, thus indicating that mere repetition of LTPinducing tetanization was not sufficient to evoke the effect.

mV/ms

I

control input I

I

i0

C

a+b

$0

tin

30 D

60

min

a*b -

Fig. 2. Duration of elect of tetanization on field EPSP shape. (A-D) Same as in Fig. I. It should be noted that the test input had been tetanized, and thus been nearly saturated with respect to LTP, with five 20-impulse trains at I x test strength prior to the tetanization at 2 x test strength indicated in A. It should also be noted that since no long-lasting potentiation was evoked in the control input (B) these afferents were not activated a1 the 2 x test strength tetanizaation of the test input, i.e. there was no overlap between the two inputs. The amplitudes of the field EPSPs in C and D, obtained after tetanization, were slightly adjusted to match their rising phase to that of the pre-tetanus EPSP.

64

G. HESSand B. GUSTAFSSON oriens

Fig. 3. Spatial restriction of the nonspecific effect of tetanization on field EPSP shape. Averaged records (a = 30) of field EPSPs recorded in stratum oriens (a-d) and stratum radiatum (e and f). In this experiment, two pairs of stimulating and recording electrodes were used, placed in stratum oriens and stratum radiatum, respectively. It can be noted that tetanization in stratum oriens led to no change in shape of the field EPSP evoked and recorded in stratum radiatum.

This long-lasting nature of the modification was found in all five cases examined in this manner. Test of input specificity of the change in shape of the field excitatory postsynaptic potential In contrast to the potentiation of the initial slope of the field EPSP, the modification of the time-course was not restricted to the tetanized synapses. In most experiments, a second input projecting to the same dendritic region was used, and this untetanized field EPSP was also generally modified (21/26 cases). In the experiment illustrated in Fig. 1, the shape change was substantially less in the control input (Fig. lD2) than in the tetanized one (Fig. lC4). However, in other cases the change in the control input was as large as in the tetanized one (Fig. 5B), and, on average, the change in peak amplitude of the control field EPSP was 73% of that of the test EPSP. When two pairs of stimulating and recording electrodes were positioned in different dendritic layers, no nonspecific effect was, however, observed (n = 4). Figure 3 illustrates that tetanization in the basal dendritic layer (stratum oriens) leads to a substantial shape modification of the field EPSP recorded in that layer (Fig. 3b, d) but to none in the untetanized input situated in stratum radiatum (Fig. 3e, f). It can be noted that in this case tetanization at test strength was sufficient to induce a shape modification in the test input. Subsequent tetanization of the stratum radiatum input then gave rise to a shape modification for that input (not shown). Magnitude of the change in shape of the jeld excitatory postsynaptic potential As shown in Figs 1 and 2, the increase in field EPSP peak amplitude (for a given initial slope) following five 20-impulse trains (at 2 x test strength) was around IO-15%, and similar values were also obtained in the other cases. It should be realized, however, that the quantitative assessment of the effect was hampered by the fact that the measured magnitude of the change in peak amplitude was

sensitive to small variations in the matching of the initial parts of the EPSPs obtained before and after tetanization. For example, estimation of the initial slope by linear regression in a time interval as indicated in Fig. 4A, often led (in particular for the tetanized input, compare Fig. 4A and C) to the post-tetanus EPSP being slightly shifted to the right, and to a smaller change in EPSP peak amplitude than when the two EPSPs were adjusted as indicated in Fig. 4B. This apparent latency shift could be a true one, but may also arise if the process underlying the shape modification has an onset early enough to affect the initial slope as measured in Fig. 4A. In the experiment shown in Fig. SA, tetanization with five 20-impulse tetani (at 2 x test strength) was later followed by two tetanization episodes containing 20 and 50 20-impulse tetani, respectively, leading to further increases in field EPSP peak amplitude. In this case, the total increase in peak amplitude was 49% and 33% for the test and control input, respectively, and similar figures were reached in three other such experiments. The same type of experiment is shown in Fig. 5B as in Fig. 5A, but using five-impulse trains instead of 20-impulse ones. It can be seen that these briefer tetani also induced a considerable shape modification. Relation between the change in shape of the field excitatory postsynaptic potential and the magnitude of the field excitatory postsynaptic potential In the experiment illustrated in Fig. 6A-D, the two recording electrodes were positioned in the dendritic and pyramidal layer, respectively, and the stimulus strength was varied to produce field EPSPs of magnitudes from well below to just above threshold for somatic spike responses. As shown in Fig. 6C, field EPSPs varying from about 20% to 75% of threshold magnitude could be well superimposed on each other indicating no change in time-course. The larger field EPSPs deviate, however, in shape from the smaller ones (Fig. 6D) such that those closer to threshold display a progressively smaller peak

Tetanization-induced A

teal

input

B

65

field EPSP shape change

test input

C

contml input

I

Fig. 4. Estimation of the magnitude of the field EPSP shape change. (A) Averaged records (n = 30) of field EPSPs obtained before and after tetanixation using five 20-impulse trains (at 2 x test strength) are shown superimposed at two different time scales. The input had been tetanized with five 20-impulse trains at 1 x test strength prior to that at 2 x test strength. The closed triangles above the upper records indicate the time window for estimation of the EPSP initial slope using linear regression; these values for initial slope were used as a basis for the normalization of the EPSPs (cf. Experimental Procedures). It can be noted that the post-tetanus field EPSP is slightly shifted to the right as compared with the pre-tetanus one. (B) Same records as in A, but using a different time window. This earlier time window often gave a better overall match between the EPSP rising phases than that shown in A. (C) Averaged records (n = 30) from another experiment showing field EPSPs of the control input before and after 20 20-impulse trains (at 2 x test strength) to the test input. It can be noted that a time window similar to that in A gives a good match between the EPSP rising phases.

amplitude relative to the initial slope. This behaviour was observed in all six cases examined in this manner, but with some variation of the field EPSP magnitude (relative to threshold for somatic firing) when a shape change was observed (SO-70% of threshold magnitude). To examine to what extent this variation of field EPSP shape was related to, or affected, the shape modification induced by tetanization, test EPSPs were averaged at different stimulation strengths before and after tetanization. As shown in Fig. 6E, where field EPSPs with the same initial slopes obtained before and after five 20-impulse trains (at 2 x test strength) are shown superimposed, small EPSPs well below threshold for somatic firing were affected in shape to much the same extent as the larger EPSPs close to threshold. A shape modification of field EPSPs of an amplitude less than 50% of the threshold EPSP was observed in all 13 cases examined. Al

Effect of an N-methyl-D-aspartate receptor antagonist on the change in shape of the field excitatory postsynaptic potential

As indicated from the above results, the conditions for inducing the field EPSP shape modification are distinct from those for inducing the synapse-specific potentiation of the EPSP initial slope. In fact, in nine experiments, tetanization (using 1 x test strength) produced increases of the initial slope of lOO-150% (10-15 min following tetanization) with no visible shape change. To further examine the possible relation between the induction of LTP and the shape modification, in two experiments (two slices) tetanization was performed after bath-application of 50 PM APV. This drug, which blocks LTP induction,‘vr6 was also found to prevent the appearance of the shape modification (not shown), this modification then appearing after tetanization following wash-out of the drug. An A2

test

input

control inpui

Fig. 5. Change in field EPSP shape following repeated tetanixation. (Al) averaged records (n = 30) of geld EPSPs obtained after tetanixation using five 20-impulse trains at 1 x test strength (smallest EPSP) and after five, 20 and 50 20-impulse trains at 2 x test strength are shown superimposed after adjustment of their initial slopes. (A2) Same, but for the control input. (B) Same as in A, but from another experiment in which five-impulse trains were used. A surgical cut between the CA I and CA3 regions had been made in these slices.

G. HESSand B. GUSTAFSSOK

66

Fig. 6. Relation between magnitude and shape of the field EPSP. (A) Averaged recordings (n = 30) from the cell body (upper records) and dendritic flower records) layers using various stimulation strengths beiuw threshold for somatic firing are shown superimposed. (B) Same as in A, but using higher stimulus strengths. The largest field EPSP in A is the smallest one in B. (C and D) The same records as in A and B, but after normalization of the EPSP initial slopes. The larger the unnormatized peak amplitude in B, the smaller is the normalized amplitude in D. (E) The effect of tetanization on the shape of field EPSPs of different magnitudes. Averaged records (n = 30) of field EPSPs taken before any tetanization and after five 20-impulse trains at 2 x test strength are shown superimposed (after adjustment of their initial slopes). Before this tetanization the input had been tetanized with five 20-impulse trains at I x test strength.

alternative interpretation (to a block of induction) of the above result is that the shape modi~cation represents opening of N-methyl-D-aspartate (NMDA) receptor channels. However, in none of the six cases in which APV was drop-applied (see Ref. 18) after tetanization, did this drug affect the shape modification induced by tetani~tion.

Effect of 4-aminopyridine on the shape excitatory ~ostsynapti~ potential

ofthe field

A shape modification that is nonspecific could possibly be related to a modi~~tion of the electrical properties of the postsynaptic membrane. Following application of cl-aminopyridine the voltage response to injected current is considerably enhanced due to blockade of a voltage-dependent transient outward current,g and one possible explanation for the shape modification is that tetanization reduces the magnitude of such a transient outward current. To examine this possibility, the effect of bath-application of W-100 PM 4-aminopyridine was studied (three experiments). As shown in Fig. 7, application of this drug led to the development of a change in the field EPSP rather closely resembling that produced by tetanization. However, in the picrotoxin solution (even after a surgical cut between the CA1 and CA3 regions) ~aminopy~dine also led to spontaneous discharges and to stimulation-induced after-discharges which prevented any more detailed analysis (see legend of Fig. 7). Relation between the field excitatory potential and the population spike

postsynaptic

In four experiments recordings obtained using various stimulus strengths were taken simultaneously from the dendritic and cell body regions to correlate

the field EPSP initial slope with the magnitude of the ~pulation spike in the cell body recording. In these experiments tetanization led to qualitatively similar changes in the E-S relation. As shown from one of these experiments in Fig. 8, tetanization at 1 x test strength led to a left-shift of the E-S curve (closed to open circles) that was specific for the tetanized input (compare Fig. 8A and B). Subsequent tetanization at 2 x test strength led to a further change of the E-S relation (open circtes to open triangles) in both test and control inputs. As in this illustrated case, the change after 2 x test strength was more of an increase

A

100 ph4 4-m!hq?yrti

IO

ms

Fig. 7. The effect of Caminopyridine on field EPSP shape. (A-C) Averaged records of field EPSPs taken at three different times (at a few minutes interval) following the application of IOOpM Caminopyridine to the perfusion solution are shown superimposed on the field EPSP obtained before the application (after adjustment of the initial slopes). These records were taken before any stimulation-induced after-discharges had appeared. However, some spontaneous discharges were present during the avemging shown in C. A surgical cut between the CAI and CA3 regions had been made in this slice.

Tetanization-induced

field EPSP shape change

t4el input rnr IO

f

P

test

input

control input 4 mV by.) mV (rod.)

I2

Fig. 8. Relation between field EPSP and population spike following tetanization. (A and B) Relation between the field EPSP initial slope recorded in stratum radiatum and the population spike recorded in stratum pyramidale. Values obtained before tetanization, after tetanization with five 20-impu~ trains at 1x and 2 x test strength are indicated by closed circles, open circles and open triangles, respectively. (C and D) Averaged records (n = 30) obtained in stratum pyramidale (upper) and stratum radiatum (lower) for test and control input, respectively. Records obtained before and after tetanization at 1x test strength are shown in a, and before and after tetanization at 2 x test strength in b and c. A surgical cut between the CA1 and CA3 regions had been made in this slice.

in slope of the E-S curve than a decrease in threshoid for spike initiation. Examination of the field EPSP time-course obtained using subthreshold stimulation strengths indicated that tetanization at 1 x test strength led to an observable shape change in only one of the four cases. On the other hand, as also illustrated in Fig. 8 Cb, C,, D,,, tetanization at 2 x test strength led to shape changes in both test and control inputs. A notable fact was, however, that the larger spikes in these cases were not associated with shorter spike latencies (compare Fig. 8C, and C,). IXSCUSSION

In agreement with previous results,’ it was found that LTP, defined as an EPSP increase that is restricted to the tetanized synapses, takes place without any change in the time characteristics of the

synaptic current. In addition, the present data shows that afferent tetanization in the hippocampal CA1 region can also evoke a long-lasting change in excitatory transmission that is distinct from LTP defined in the above manner. This latter potentiation is not specific to the activated synapses, and manifests itself as a change in field EPSP time-course with a prolongation of the rising phase and an increased peak amplitude. This change in the EPSP is associated with more firing for a given field EPSP initial slope and may thus be a mechanism for E-S ~tentiation that is not secondary to a change in balance between excitation and inhibition following tetanization.‘*6 The present study provides no clear answer to the question of what may underly the observed change in EPSP shape. The experiments were performed in the presence of picrotoxin, at a concentration about 10 times that used by others5 to

67

6X

G. HESS and B. GUSTAFSSON

eliminate GABA,-mediated inhibition excluding that the observed shape change results from a relative decrease of the disynaptic IPSP.’ The contribution of a reduction of the later GABA,-mediated IPSP8 to the shape change was not directly tested, but is not likely, due to the early onset and short duration of the shape modification. Moreover, any influence of the GABA,-mediated IPSP on the held EPSP timecourse should likely have revealed itself following reduction in stimulus strength after tetanization at I x test strength (cf. Ref. 1). Since the potentiated EPSP was unaffected by the application of the NMDA receptor antagonist APV the shape modification is not related to an NMDA receptor mediated component of the EPSP. On the other hand, the addition of a disynaptic EPSP component following tetanization is more difficult to exclude. but there is, as far as we know, no evidence for such a pathway in the CA 1 region. This tentative pathway should also be facilitated to much the same extent for test and control inputs. A modification of the shape of the monosynaptic EPSPper se could be due to both pre- or postsynaptic changes. Since the shape modification was found to be nonspecific for inputs projecting to the same dendritic region, but not for inputs located in different dendritic layers, a location of the underlying process to the postsynaptic membrane is not likely to be somatic but closer to the activated synapses. When increasing the field EPSP magnitude from well below to close to threshold for spike initiation, no shape changes indicating activation of some near-threshold activated inward current were observed. This finding, and the observation that the shape modification was observed to much the same extent for small EPSPs far below threshold as for the larger ones, make it unlikely that the potentiation is caused by an increased activation of such a voltage-dependent process. On the other hand, field EPSPs closer to threshold had a relatively smaller peak amplitude (for a given initial slope) than the smaller ones, possibly related to activation of a voltage-dependent outward current process opposing the EPSP current. Application of 4-aminopyridine at concentrations which reduce a transient outward current in hippocampal pyramidal cells’ produced changes in the field EPSP time-course closely resembling those found after tetanization, raising the possibility that the shape modification results from a tetanization-induced reduction of such a transient outward current located dendrititally. We have, however, no direct evidence that the shape modification of the EPSPs with increasing stimulus strength is due to activation of such a transient outward current, or, for example, simply due to a closer approach to the EPSP reversal potential. Moreover, since the small EPSPs were also modified in shape after tetanization, their timecourse should also have to be controlled by this outward current, which seems at variance with the

finding that there was no gradually increasing change EPSP shape with increasing stimulus strength. Thus, while a reduction in a dendritically located voltage-dependent transient outward current may be a plausible mechanism underlying the observed shape modification, there is at present no firm evidence for this view. The shape modification may, on the other hand, be based on a presynaptic alteration. For example, substances released from the activated postsynaptic cells may diffuse to nearby active and inactive terminals,4 and there induce changes in the timecourse of transmitter release. It may be noted that the shape modification observed here after application of 4-aminopyridine might be related to a prolongation of the spike in the presynaptic terminal following blockade of presynaptically located potassium channels,“” and a similar mechanism may then underly the shape change produced by the afferent tetanization. However, at the concentrations of 4-aminopyridine presently used, it is uncertain to what extent its presynaptic action is related to such a spike prolongation rather than to a direct effect on the presynaptic calcium channels.14 It thus appears that while the results obtained with 4-aminopyridine may indicate that the observed shape modification is related to a blockade of potassium channels it cannot be decided whether these are pre- or postsynaptically located. Since experiments dealing with LTP are generally performed with intact inhibition, the presently observed potentiation, if elicited, will generally be obscured by the disynaptic IPSP as well as by population spikes. The question to what extent this potentiation has affected previous estimations of LTP is not easily answered. However, whenever LTP is not measured as a change in initial slope of the field EPSP, and whenever a control input to the same dendritic region is not provided, the participation of this potentiation to LTP cannot be excluded. While the threshold for inducing this shape modification following afferent tetanization may be higher than for LTP, this may not be the case in experiments in which potentiation is induced using e.g. drug application. It has not escaped our attention that some potentiations described in connection with LTP appear to refer, at least partly, to a potentiation resembling the present one (e.g. see Refs 10, 13). However, further studies will be necessary to decide under what conditions the presently observed potentiation may contribute to the measured LTP. Application of the NMDA receptor antagonist APV showed that, in common with LTP,‘.” the induction, but not the maintenance, of the shape modification depends on NMDA receptor activation. However, to what extent this result implies a direct participation of the ionic flux through NMDA receptor channels in the induction of the shape modification cannot be evaluated by this single finding. Since APV will reduce the amount of postin

Tetanization-induced

synaptic depolarization obtained during tetanization, this factor may be the more decisive one. Relation

between

potential

and the population

the field

excitatory

postsynaptic

spike

While E-S potentiation was largely absent’ or significantly reduced6 in previous studies carried out in picrotoxin solution, in the present study a quite substantial E-S potentiation was found. Part of this E-S potentiation appeared following tetanization at 1 x test strength that led to no observable shape modification of subthreshold field EPSPs, and was

specific to the tetanized input. We have no explanation why this E-S left-shift was more pronounced in the present study than in the previous ones’.6 or what may underly this potentiation. It can just be noted that in the previous studies, the E-S relation in picrotoxin solution was examined in absolute terms in relation to that found in normal solution (the E-S shift to a very large extent produced by picrotoxin itself), and any changes in the E-S curve induced by tetanization in picrotoxin solution might then easier have been overlooked.

field EPSP shape change

69

While part of the E-S potentiation found presently is thus unaccounted for, the E-S potentiation appearing using higher stimulation strength during tetanization was associated with, and could thus possibly be related to, an EPSP shape modification. A substantial part of the discrepancy between this study and the previous ones might then be related to the higher tetanization strengths presently used. It was, however, noted that the E-S potentiation obtained at higher tetanization strengths was not expressed so much as a left-shift of the E-S curve as an increased slope, at least for the tetanized input. This result, together with the fact that the spike latency was not shortened, shows that the connection between the EPSP shape change and E-S potentiation is not a simple one, and indicates that tetanization at higher strength evokes additional processes to the EPSP shape modification that effect the E-S potentiation. Acknowledgement-This

work was supported by the Swedish Medical Research Council (oroiect 05180) and

Magnus Bergvalls StiAelse. Dr G. Hesswai a grantee of the European Training Programme in Brain and Behaviour Research (1989).

REFERENCES

1. Abraham W. C., Gustafsson B. and Wigstriim H. (1987) Long-term potentiation involves enhanced synaptic excitation relative to synaptic inhibition in guinea-pig hippocampus. J. Physiol. 394, 367-380. 2. Andersen P., Sundberg S. H., Sveen O., Swann J. W. and Wigstrijm H. (1980) Possible mechanisms for long-lasting potentiation of synaptic transmission in hippocampal slices from guinea-pigs. J. Physiol. 302, 463-482. 3. Bliss T. V. P. and Limo T. (1973) Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized rabbit following stimulation of the perforant path. J. Physiol. 232, 331-356. 4. Bevan S. and Wood J. N. (1987) Arachidonic-acid metabolites as second messengers. Nature 328, 20. 5. Brown T. H. and Johnston D. (1983) Voltage-clamp analysis of mossy fiber synaptic input to hippocampal neurons. J. Neurophysiol. 50, 487-507.

6. Chavez-Noriega

L. E., Bliss T. V. P. and Halliwell J. V. (1989) The EPSP-spike (E-S) component of long-term potentiation in the rat hippocampal slice is modulated by GABAergic but not chlolinergic mechanisms. Neurosci. L..err. 104.58-64. 7. Collingridge G. L., Kehl S. J. and McLennan H. (1983) Excitatory amino acids in synaptic transmission in the Schaffer collateral~ommissural pathway of the rat hippocampus. J. Physiol. 334, 3346. 8. Dutar P. and Nicoll R. A. (1988) A physiological role for GABA, receptors in the central nervous system. Nurure 332, 156-158. 9. Gustafsson B., Galvan M., Grafe P. and Wigstriim H. (1982) A transient outward current in a mammalian neurone blocked by 4-aminopyridine. Nature 299, 252-254. 10. Kauer J. A., Malenka R. C. and Nicoll R. A. (1988) NMDA application potentiates synaptic transmission in the hippocampus. Nature 334, 250-252. Il. Llinas R., Steinberg 1. Z. and Walton K. (1981) Presynaptic calcium currents in squid giant synapse. Biophys. J. 33, 289-322. 12. Llinas R., Steinberg I. Z. and Walton K. (1981) Relation between presynaptic calcium current and postsynaptic potential in squid giant synapse. Biophys. J. 33, 323-351.

13. Mulkeen D., Anwyl R. and Rowan M. J. (1987) Enhancement of long-term potentiation by the calcium channel agonist Bayer KS644 in CA1 of the rat hippocampus in vitro. Neurosci. Letr. 80, 351-355. 14. Rogawski M. A. and Barker J. L. (1983) Effects of Caminopyridine on calcium action potentials and calcium current under voltage clamp in spinal neurons. Brain Res. 280, 180-185. P. A. (1988) Mechanisms of long-term potentiation: EPSP/spike dissociation, intradendritic recordings and glutamate sensitivity. J. Neurosci. 8, 1632-1644. 16. Wigstriim H. and Gustafsson B. (1984) A possible correlate of the postsynaptic condition for long-lasting potentiation in the guinea-pig hippocampus in oitro. Neurosci. Lpft. 44, 327-332. 17. Wigstrijm H. and Gustafsson B. (1985) Facilitation of hippocampal long-lasting potentiation by GABA antagonists. Acla physiol. stand. 125, 159-172. 18. Wigstriim H., Gustafsson B. and Huang Y.-Y. (1986) Mode of action of excitatory amino acid receptor antagonists .. .. . on mppocampal long-lastmg potenllatlon. Neuroscience 17, 1105-l 115. 19. Wilson R. C., Levy W. B. and Steward 0. (1981) Changes in translation of synaptic excitation to dentate granule cell discharge accompanying long-term potentiation. II. An evaluation of mechanisms utilizing dentate gyrus dually innervated by surviving ipsilateral and sprouted crossed temporodentate inputs. J. Neurophysiol. 46, 339-355.

15. Taube J. S. and Schwartzkroin

(Accepted 5 February 1990)