- Email: [email protected]
A specific effect of morphine on evoked activity in the rat hippocampal slice
Brain Research, 192 (1980) 227-238 © Elsevier/North-Holland Biomedical Press
227
A SPECIFIC EFFECT OF MORPHINE ON EVOKED ACTIVITY IN THE RAT HIPPOCAMPAL SLICE
W. A. CORRIGALL and M. A. LINSEMAN Neurobiology Section, Addiction Research Foundation, Toronto (Canada)
(Accepted December 6th, 1979) Key words: morphine -- hippocampal slice -- opiate -- limbic system
SUMMARY The effect of morphine (0.5-50 #M) was examined on CA1 field potentials in the transverse hippocampal slice. Morphine consistently produced an augmentation of evoked activity manifest as (i) a decrease in the threshold for generation of a population spike and (ii) generation of an additional population spike(s) whose amplitude was proportional to the position of the sampled response on its input/output curve. Both of these opiate effects were stereospecific and naloxone-reversible. Additional population spikes occurred in opiate medium with either orthodromic or antidromic activation of the pyramidal cells, and the antidromic effect was abolished when synaptic transmission was blocked, suggesting that morphine did not act directly upon the pyramidal cells. Recordings of population EPSPs in the dendrites of the pyramidal cells showed no changes due to opiate exposure near threshold. Opiate effects were mimicked by the y-aminobutyric acid (GABA) antagonist picrotoxin, and were partially to fully reversed by GABA itself, suggesting that disinhibition of pyramidal cells might be involved as a mechanism in this opiate effect. The data are evidence for a specific primary effect of morphine within the hippocampus in spite of the low numbers of opiate receptors in this brain region.
INTRODUCTION Previous recording studies have shown that administration of opiates affects the activity of many areas of the brain is, including the hippocampusa,10,11,17, z0. Nonetheless, in this area of the limbic system, there is still considerable question as to the direction and stereospecificity of the effects. Nicoll et al. 17 and Chou and Wang 3 reported consistent increases in hippocampal unit activity following opiate administration which were antagonized by naloxone; Sega120 and Fry et al. 10 reported that
228 only a small percentage of hippocampal units were affected by opiates, some of which increased while others decreased in rate, and these effects were generally not naloxonereversible. Fry et al. TM further reported no differentiation of responses to levorphanol and dextrorphan, the active and inactive isomers respectively, and concluded that the effects they observed may not have been mediated via specific opiate receptors. Since some of the studies mentioned above involved systemic administration of drugs, it is not clear whether the observed effects are a result of a primary action of morphine on the hippocampus. In other studies, the drugs have been applied iontophoretically whereby the local concentrations of drugs are unknown. Consequently, we studied the effect of morphine on the isolated hippocampal slice using perfusion of solutions of known drug concentrations, so as to be sure of both primacy of effect and the relation of effective drug concentrations to those of known pharmacological effectiveness. As reported here, the hippocampal slice also represents a first in vitro model of opiate action within the brain itself. The advantage of multiple models has recently been shown by the demonstration of differing potencies of opiate agonists on the guinea pig ileum and the mouse vas deferens, thereby the implication of different types of opiate receptors 15. A preliminary report of some of these experiments has been presented 5. METHODS Transverse hippocampal slices were prepared and maintained as previously described19, 22. Male Sprague-Dawley rats (200-350 g) were decapitated and the brain rapidly removed and received in medium at ambient temperature. The hippocampus was dissected free and sliced into 400 #m sections. The slices were received in medium also at ambient temperature which was then immediately transferred to a water bath at 35 °C. Slices were preincubated a minimum of 30 min prior to use. Composition of the medium in mM was as follows: NaCI 124, KC1 5, KH2PO4 1.25, MgSO4"7H20 2, CaCIz.2H20 2, NaHCOa 26, dextrose 10. Medium was equilibrated with 95 % 02/5 ~o CO2. To block synaptic transmission, some experiments were carried out in medium having 0.5 mM Ca 2+ and 10 mM Mg 2+ (in this medium C I - was 16 mM greater than the control value of 133 mM ; all other constituents were unchanged). For electrophysiology, slices were transferred singly to a recording chamber where they were perfused with the same medium as above at 33-35 °C under an atmosphere of warmed, moistened 95% 02/5 % CO2. Conventional electrophysiological methods were used for stimulating and recording. Field potentials were recorded with micropipettes filled with either 4 M sodium chloride or 4 M potassium acetate (5-15 M ~). Stimulus pulses of 0.1 msec duration were delivered at 0.2 Hz via bipolar 62 #m nichrome wire, and were isolated with either current or voltage constant. Signals were AC coupled (10 Hz-1 kHz) and were averaged on-line for further analyses. Depending upon the protocol of the particular experiment, field potentials were recorded at the level of the CA1 cell body layer (stratum pyramidale), the apical dendritic (str. radiatum) or basal dendritic (str. oriens) regions. Pyramidal cells were driven ortho-
229 dromically by stimulation of the stratum radiatum or stratum oriens, or antidromically by stimulation of the alveus. (See Fig. 3A for schematic of CA1 organization.) The possibility of stimulus spread from electrodes in one pathway to another was excluded by recording activity at the level of the synapses on the dendrites of the pyramids in the non-stimulated pathway, and determining that no population EPSP occurred from stimulation of the experimental pathway. A typical experiment involved recording from stratum pyramidale of CA1 and stimulating stratum radiatum. While responses stabilized in control medium during 0.2 Hz stimulation, average samples (n = 8 stimulus presentations) were taken every 5 min. At this point the size of the stimulus (input, I) was altered to determine the range of responses from the tissue (output, O). Responses were sampled at various points between threshold and asymptote in order to construct an input/output or I/O curve from averaged potentials (n = 4 at each stimulus intensity). Following generation of the baseline I/O curve an additional sample was taken to assure that the response at a given voltage was unchanged, and the tissue perfusion was then switched to medium containing a particular drug. Responses were sampled every 5 min until they showed no further change, up to a maximum of 30 min, whereupon a second I/O curve was generated. At this point the perfusate could again be changed as required, and the same recording protocol repeated. Population spike amplitudes used in I/O curves were calculated as the average of maximum initial positivity to peak spike negativity and peak spike negativity to following maximum positivity (see Fig. 1). The following drugs were used: morphine sulfate (BDH); naloxone (Endo); picrotoxin and y-aminobutyric acid (GABA, Sigma); levorphanol tartrate and dextrorphan tartrate (Hoffman La Roche). RESULTS
Effect of opiates on radiatum-evoked field potentials in CA 1 The control response recorded in stratum pyramidale following stimulation of the radiatum consisted of a slow positive potential upon which, at sufficiently intense stimulation, a negative population spike was superimposed (Fig. 1). The slow potential has been ascribed to a population EPSP generated in the apical dendrites while the negative spike has been demonstrated to be summation of synchronously firing unit spikes 1. Morphine (0.5-50 ~M) had two effects upon the field potential recorded in the cell body layer (str. pyramidale). The more obvious effect was the dose-related production of a second (Fig. 1) and, at highest concentrations, additional population spikes. This effect of morphine developed slowly, in part due to the rate of perfusion (approx. 2 ml/min), and because the tissue was not fully submersed in perfusion medium but rather was supported on a nylon net and was bathed from below with only a meniscus of fluid covering its surface. Morphine effects were resistant to washout with control medium, but changing the perfusate to one containing naloxone (0.1-10/~M) resulted in rapid reversal of the opiate-induced augmentation. Furthermore, in 5 out of 5 experiments, levorphanol (5/zM) was effective in producing this
23O control
-
~_q
morphine IOpM
l/~/"~'-kt
5rain
,j
P' S
-~11
/
~'JI , --
[
F
IOmin -
-
25min
! i
,j naloxone 2 uM
/r~jr_ ~'JS
~ "
5rain _ IOmin
J
_~.
j2mV
5reset
Fig. 1. Example of the effect of morphine on the field potential in stratum pyramidale of CA1 evoked by stimulation in stratum radiatum. Reversal occurs quickly followingchange to naloxone-containing perfusate. In this and subsequent figures negativity is downwards; all responses shown are averaged. Stimulus occurred at the interruption in the tracings. In this figure, n = 8 responses]average. additional population spike, while dextrorphan (5 #M) was ineffective in 4 out of 4 (Fisher Probability, P < 0.025). For isomeric tests, 5 # M was chosen since morphine reliably produced an effect at this concentration. The size of the opiate-produced secondary population spike as related to the primary spike depended upon the position of the response on its I/O curve. As can be seen in Fig. 2A, the size of the secondary spike became an increasingly greater percentage of the primary spike as the stimulus voltage was increased to asymptote for the primary spike. The second effect of morphine was on the primary population spike, and is most clearly seen in its I/O curve (Fig. 2B). Morphine produced an increase in the amplitude of the primary spike over the entire range of the I/O curve. This increase was fully reversible by naloxone at the lowest stimulus intensities, partially reversible at slightly higher intensities, but not at all reversible in the middle to upper regions of the I/O curve. Fig. 2C shows an example of this gradation of naloxone reversibility in averaged responses near threshold. This action of morphine on the primary population spike was more apparent with levorphanol, which is approximately 7 times more potent than morphine as measured in the guinea pig ileum 12. While some upward drift of the I/O curve was seen on occasion after dextrorphan, it was not naloxone-antagonized anywhere along the curve. This increase most probably represents small changes in the responses occurring over the long times they were followed, since
231 ~80-
o control n morphine 5~M A naloxone |pM
E
~> 2 0 F: eL,
~
.~
J
-& 15
=e6o. ._ ~,
o control o morphine 5jJM A nolox
. '2
0=4o.
o
I0
xo
0.
o E
A
a.
,'o Stimulus
io
3'o
,'o
0 i
!
i
0
I0
20
intensity (volts)
Stimulus
i
Intensity
i
30 (vo[t"s)
40
D 6 volts
C control
morphine SuM
naloxone I}JM
-- " ~ " ~
-
-
~
-.. " ~ . ~ . . ~ . ~
I0 volts
,2vo,,s
-- ~
7 volts
--
8 volts
~
9 volts
~
10 volts I ImV
5 msec
I0 msec
Fig. 2. A: illustration of the I/O dependence of the morphine effect at 30 min exposure. Graph shows increasing amplitude of the secondary population spike (expressed as a percentage of the primary spike) as a function of stimulus intensity. Note that there was no secondary spike in control and that the effect was totally reversed by naloxone over the full range of the I/O curve. B: I/O curves for the primary spike of the same response as in A. Note that the spike amplitude in morphine is increased over the entire I/O curve leading to a shift in spike threshold to lower stimulus intensity. Naloxone antagonizes this increase only near threshold, resulting in a net shift to lower threshold/higher amplitude responses after opiate exposure. C: sample responses from same experiment at stimulation intensities of 12 V (approximate threshold for population spike in control medium), 10 V (just below threshold) and 14 V Oust above threshold). Morphine produces a primary population spike at and below the control threshold which is antagonized by naloxone, but the spike produced above threshold is not (n = 4/averaB¢). D: superimposed population EPSPs in control and 5/~M morphine media recorded in radiatum in another experiment. Those shown are just above threshold for the EPSP and still subthreshold for spike production in control medium. Morphine has no consistent effect on the EPSP at voltages subthreshold for spike production (6,7,8 V). When spike production occurs in pyramidale (9 and 10 V) it is not possible to determine whether there are changes in the EPSP responsible for spike production (e.g. an increased rate of rise) or if spike production is distorting the EPSP records (n = 4/average).
232
olveus
A
_%uJ /
J --
oriens
pyramidale
radiatum i~ Ok" morphine SuM normal medium]
control
morphine 5~M (Mgf/CoI)
B
stimulotion
--
--
--'~
5mV
5 msec
stimulation _ _ J 5mV 5 msec
Fig. 3. A: schematic of the organization of the CAI region. Interneurons are thought to mediate recurrent activity to pyramidale cells. B: recordings from pyramidale following stimulation of oriens (upper tracings) or alveus (lower tracings) in the same slice. In normal medium, morphine resulted in art I/O-dependent appearance of a secondary spike in both cases. In low calcium/high magnesium medium (0.5 mM/10 mM resp.) the synaptic response to oriens stimulation was abolished as was the secondary spike following alveus stimulation. Note that neither the presynaptic volley (first negative-positive deflection) in the oriens-evoked response nor the primary antidromic spike in the alveus-evoked response were significantly affected by the low calcium/high magnesium. Responses shown here are averaged samples at the top of their respective I/O curves (n = 8/average).
c o n t r o l experiments in which responses were followed in n o r m a l m e d i u m over c o m p a r a b l e p e r i o d s o f time i n d i c a t e d t h a t the response f r o m a slice o n occasion exhibited some u p w a r d drift. This was in spite o f the fact t h a t stable baseline responses were o b t a i n e d p r i o r to d r u g a p p l i c a t i o n (see Methods). A p o r t i o n o f these experiments were d o n e with s i m u l t a n e o u s recordings at the level o f the synapses in r a d i a t u m as well as in p y r a m i d a l e (Fig. 3A), in o r d e r to assess w h e t h e r there were changes in the p o p u l a t i o n EPSP due to o p i a t e exposure. A s shown in Fig. 2D, at stimulus intensities which were s u b t h r e s h o l d for the p r o d u c t i o n o f a
233 population spike there were no measurable differences between the averaged EPSP in control and in morphine-containing media. At higher intensities of stimulation (9 and 10 volts in Fig. 2D) opiate perfusion led to the production of a primary spike where none had existed in control (as shown in Fig. 2B, C) and this spike was also 'seen' by the dendritic electrode, giving rise to a distortion of the population EPSP. However, the fact that there were no changes in the population EPSP at subthreshold stimulus intensities (6, 7, 8 volts in Fig. 2D) argues that morphine did not produce its augmentation of pyramidal population spikes following radiatum stimulation through an increase in afferent excitation.
Mechanism of opiate augmentation In addition to their synaptic input from fibers in the radiatum, the pyramidal cells of CA1 also have synaptic endings on their basal dendrites from fibers in the stratum oriens (Fig. 3A), and stimulation of these fibers produces in pyramidale a field potential similar to that following radiatum stimulation (Fig. 3B). Morphine augmented this response as well (Fig. 3B, upper traces) suggesting that the opiate phenomenon did not depend specifically on excitatory input from radiatum. Pyramidal neurons may also be activated non-synaptically by stimulation of their axons in the alveus. As depicted schematically in Fig. 3A, this activation would still involve any recurrent synaptic systems. Morphine again led to the production of a secondary population spike in the antidromic field potential recorded in pyramidale (Fig. 3B). If however, after the morphine effect was generated, the perfusate was switched control
',j
/ f -~ C . ~ o t - f~
/
oxin 5,uM --~__
5 min
/ J
I0 rain
20rain
5rnsec
Fig. 4. Example of the effect of picrotoxin on the CA1 field potential elicited by stimulation of stratum radiatum (n = 8/average).
234 A
morphine SuM
control
morphine * GABA 5 uM ! mM
naloxone l uM
20 v
~:
i,/
:
30v
_ _j2mV 5msec o control a morphine
I00 - o c o n t r o l
5)JM
c~ m o r p h i n e 5/JM A naloxone IpM • ~ o r p h i n e 5.uM • G A B A I
tt naloxone l,uM • morphine 5HM÷GABA
ImM
> E
80. c~
=
•~. 4
qu
v)
6O
c o
v) c
c~ o o. >,2
a. o 0.
o
40"
o
-o
/
2O
o (n
0
0
I '0 Stimulus
2' 0 Intensity
3'0 (volts]
4'0
g
5//
,'o
Stimulus
2'o
Intensity
3'o
4'o
(volts}
Fig. 5. A: responses obtained at two stimulus voltages as tissue was perfused successively by control medium, morphine, morphine plus GABA, and naloxone (n ~ 4/average). B and C: I/O curves for primary and secondary population spikes respectively. Note again the increase in the primary spike over the entire I/O curve in opiate medium and the antagonism of this increase near threshold by naloxone. The secondary population spike is partially reduced by GABA (C) while the amplitude of the primary spike is increased somewhat in GABA (B).
to one containing morphine in high magnesium/low calcium medium sufficient to block synaptic activity, the morphine effect on the antidromic response was quickly abolished (Fig. 3C). Upon return to morphine in normal medium the opiate effect was reestablished and was subsequently antagonized by naloxone. This evidence suggested that morphine did not produce its effect by a direct action upon pyramidale neurons but rather by an action on other neurons coupled synaptically to them. Of the possible mechanisms, an action of the opiate upon inhibitory interneurons seemed likely. Since these interneurons are thought to use GABA as transmitter 1,z3, the effect of the GABA antagonist picrotoxin on the radiatum-evoked CA1 field potential was examined. Picrotoxin, at least with short-duration exposures,
235 produced changes in the response similar to morphine, including shift of the threshold of the I/O curve to lower stimulus intensities and production of additional population spikes (Fig. 4). Picrotoxin effects were resistant to washout with control medium and often became convulsive-type activity consisting of several to many population spikes (Fig. 4, 20 min, shows the beginning of such activity). Because a GABA-antagonist mimicked the effect of morphine on the hippocampal slice, it was of interest to determine if GABA itself would antagonize the action of morphine. As shown in the example of Fig. 5 it was observed that GABA antagonized the opiate-induced secondary spike while at the same time causing an increase in the primary spike. The balance of the secondary spike not antagonized by GABA was reversed by naloxone (Fig. 5C). While in this particular case the secondary spike was antagonized only partially by GABA, other experiments using the same concentration showed complete block of the secondary spike. Furthermore, GABA reversal of the threshold increase in the primary spike due to morphine was also seen in other experiments, although there was still a GABA-induced increase at higher stimulus intensities. DISCUSSION The effect of morphine on the CA1 pyramidale field potential was consistent augmentation of the response. In no case was depression of the pyramidale field potential observed. (The drug specificity of this effect is demonstrated by the observation that ethanol, another depressant, consistently reduced the amplitude of the primary population spike in the same preparation4.) The augmentation occurred both as an increase in the primary population spike and as the production of a secondary field spike. The latter was stereospecific and naloxone reversible as was the former at and near threshold. The reason for the naloxone insensitivity of increased primary population spike at higher stimulus intensities remains unclear. It may represent a non-specific opiate action which others have reported in the hippocampus1°, 20. If so, it points out the importance of examining the full functional range of the system when investigating pharmacological effects since in the present study we would have observed only this naloxone-insensitive component of the primary spike if we had used suprathreshold stimulation exclusively. The phenomenon may also, however, be a consequence of a specific effect of opiate exposure, for example, some potentiation of the response produced when the threshold is lowered by morphine or levorphanol, which remains after opiate antagonism. If this proves to be so, it would be an important action of opiates since it results in a shift of the 1/O curve to lower threshold/higher amplitude responses. Regarding the question of mechanism, the augmentation of pyramidal cell activity did not depend by which pathway the cells were activated but, in the case of the antidromic responses, did require synaptic transmission to occur, arguing that morphine did not act directly upon the pyramidal cells. Although it was not possible to determine if there were changes in the EPSP at stimulus intensities above spike threshold, which may qualify our interpretation, changes in afferent excitation in the case of
236 the orthodromic response were ruled out on the basis of our extracellular data indicating population EPSPs below spike threshold were unchanged by morphine. Replication of the morphine effects with picrotoxin, and reversal of the opiate effects with GABA provide corroborative evidence to the proposal of Siggins et al. 21 that opiates may act upon inhibitory interneurons in the hippocampus. The GABA evidence is particularly indirect, of course, since GABA at appropriate concentration would be expected to reduce the entire response whether or not its augmentation were due to morphine depression of GABAergic cells or some other mechanism. However, antagonism of the secondary population spike, while there was a concommitant increase in the primary, does suggest some selectivity which may point towards an origin for the secondary spike as reduced GABA at the pyramidal cell following primary spike generation. The mechanism for the GABA-induced increase in the primary spike remains unclear. Our finding of consistent increases are basically in agreement with the spontaneous unit studies of Chou and Wang 3 and Nicoll et al. 17. However, they are inconsistent with data reported recently by Fry et al. 1° and by Segal z°, in tha~ our effects were both stereospecific and naloxone-reversible, and in no case was depression of activity observed. There are several methodological differences between the latter studies and our own that may account for the discrepancy. Both Segal 2° and Fry et al. 1° applied drugs by iontophoresis. This may lead to unknown, and possibly higher local concentrations which might increase the probability of observing non-specific opiate effects18. For example, Dingledine et al. 8 have reported that a 300/~M concentration of naloxone acted directly as a GABA antagonist, and we have observed that 100 #m naloxone caused effects similar, rather than antagonistic, to morphine in the hippocampal slice (unpublished observations). The concentrations of morphine used in our study compare favorably with doses required to produce effects in vivo (see discussion in ref. 6). Furthermore, as opposed to sampling spontaneous unit activity, we recorded an evoked population response which may lead to greater homogeneity of effect. Finally, examination of I/O curves allowed us to assess the effect of the drug over the entire functional range of the neurons. The overall effect of morphine, then, was an increased excitability of the pyramidal cells of hippocampus. Other studies in vivo suggest opiates may function in this region to elicit epileptiform activity. For example, Elazar et al. 9 have reported EEG excitation and convulsive activity at the site of intracerebral injections of enkephalin into the dorsal layer of hippocampus. Linseman and Grupp (in preparation) observed hippocampal spiking in the EEG following i.v. injections of low doses of morphine to rats. Although there is no direct evidence, it is possible that the latter effects may be mediated by a cellular mechanism such as proposed here. In summary, our data are evidence for a specific primary effect of opiates within the hippocampus in spite of the reported low density of opiate receptors in this area 1~. This model, together with those described in tissue culture 2,6,7,16, may be useful in elucidating the mechanisms of opiate action within the central nervous system.
237 ACKNOWLEDGEMENT We t h a n k H o f f m a n - L a Roche (Canada) for their gift o f levorphanol a n d dextrorphan.
REFERENCES 1 Andersen, P., Organization of hippocampal neurons and their interconnections. In R. L. Isaacson and K. H. Pribram (Eds.), The Hippocampus, VoL 1: Structure and Development, Plenum Press, New York, 1975, pp. 155-175. 2 Barker, J. L., Neale, J. H., Smith, T. G., Jr. and Macdonald, R. L., Opiate peptide modulation of amino acid responses suggests novel form of neuronal communication, Science, 199 (1978) 1451-1453. 3 Chou, D. T. and Wang, S. C., Unit activity of amygdala and hippocampal neurons: Effects of morphine and benzodiazepines, Brain Research, 126 (1977) 427-440. 4 Corrigall, W. A., Carlen, P. L. and Staiman, A. L., Acute effect of ethanol on hippocampal CAI field potentials recorded in vitro, Neurosci. Abstr., 3 (1977) 288. 5 Corrigall, W. A. and Linseman, M. A., Acute effects of morphine on the hippocampus in vitro, Fed. Proc., 38 (1979) 3296. 6 Crain, S. M., Crain, B., Peterson, E. R. and Simon, E. J., Selective depression by opioid peptides of sensory-evoked dorsal-horn network responses in orgtnized spinal cord cultures, Brain Research, 157 (1978) 196-201. 7 Crain, S. M., Peterson, E. R., Crain, B. and Simon, E. J., Selective opiate depression of sensoryevoked synaptic networks in dorsal horn regions of spinal cord cultures, Brain Research, 133 (1977) 162-166. 8 Dingledine, R., Iversen, L. L. and Breuker, E., Naloxone as a GABA antagonist: evidence from iontophoretic, receptor binding and convulsant studies, Europ. J. Pharmacol., 47 (1978) 19-27. 9 Elazar, Z., Motles, E., Ely, Y. and Simantov, R., Acute tolerance to the excitatory effect of enkephalin microinjections into hippocampus, Life ScL, 24 (1979) 541-548. 10 Fry, J. P., Zieglghnsberger, W. and Herz, A., Specific versus non-specific actions of opioids on hippocampal neurons in the rat brain, Brain Research, 163 (1979) 295-305. 11 Hill, R. G., Mitchell, J. F. and Pepper, C. M., The excitation and depression of hippocampal neurones by iontophoretically applied enkephalins, J. Physiol. (Lond.), 272 (1976) 50-51P. 12 Kosterlitz, H. W., Leslie, F. M. and Waterfield, A. A., Rates of onset and offset of action of narcotic analgesics in isolated preparations, Europ. J. PharmacoL, 32 (1975) 10-16. 13 Kuhar, M. J., Pert, C. B. and Snyder, S. H., Regional distribution of opiate receptor binding in monkey and human brain, Nature (Lond.), 245 (1973) 447-450. 14 Linseman, M. A. and Grupp, L., Acute and chronic opiate effects on single units and EEG of medial thalamus and hippocampus: a latency analysis, in preparation. 15 Lord, J. A. H., Waterfield, A. A., Hughes, J. and Kosterlitz, H. W., Endogenous opioid peptides: multiple agonists and receptors, Nature (Lond.), 267 (1977) 495-499. 16 Macdonald, R. L. and Nelson, P. G., Specific opiate-induced depression of transmitter release from dorsal root ganglion cells in culture, Science, 199 (1978) 1449-1451. 17 Nicoll, R. A., Siggins, G. R., Ling, N., Bloom, F. E. and Guillemin, R., Neuronal actions of endorphins and enkephalins among brain regions: a comparative microiontophoretic study, Proc. nat. Acad. Sci. (Wash.), 74 (1977) 2584-2588. 18 North, R. A., Opiates, opioid peptides and single neurones, Life Sci., 24 (1979) !527-1546. 19 Schwartzkroin, P. A. and Wester, K., Long-lasting facilitation of a synaptic potential following tetanization in the in vitro hippocampal slice, Brain Research, 89 (1975) 107-119. 20 Segal, M., Morphine and enkephalin interactions with putative neurotransmitters in rat hippocampus, Neuropharmacology, 16 (1977) 587-592. 21 Siggins, G. R., Zieglg/insberger, W., French, E., Ling, N. and Bloom, F. E., Opiates and opioid peptides may excite hippocampal pyramidal neurons by inhibiting adjacent inhibitory interneurons, NeuroscL Abstr., 4 (1978) 414. 22 Spencer, H. J., Gribkoff, V. K., Cotman, C. W. and Lynch, G. S., GDEE antagonism of iontopho-
238 retic amino acid excitations in the intact hippocampus and in the hippocampal slice preparation, Brain Research, 105 (1976) 471-481. 23 Straughan, D. W., Neurotransmitters and the hippocampus. In R. L. Isaacson and K. H. Pribram (Eds.), The Hippocampus, Vol. 1: Structure and Development, Plenum Press, New York, 1975, pp. 239-267.
227
A SPECIFIC EFFECT OF MORPHINE ON EVOKED ACTIVITY IN THE RAT HIPPOCAMPAL SLICE
W. A. CORRIGALL and M. A. LINSEMAN Neurobiology Section, Addiction Research Foundation, Toronto (Canada)
(Accepted December 6th, 1979) Key words: morphine -- hippocampal slice -- opiate -- limbic system
SUMMARY The effect of morphine (0.5-50 #M) was examined on CA1 field potentials in the transverse hippocampal slice. Morphine consistently produced an augmentation of evoked activity manifest as (i) a decrease in the threshold for generation of a population spike and (ii) generation of an additional population spike(s) whose amplitude was proportional to the position of the sampled response on its input/output curve. Both of these opiate effects were stereospecific and naloxone-reversible. Additional population spikes occurred in opiate medium with either orthodromic or antidromic activation of the pyramidal cells, and the antidromic effect was abolished when synaptic transmission was blocked, suggesting that morphine did not act directly upon the pyramidal cells. Recordings of population EPSPs in the dendrites of the pyramidal cells showed no changes due to opiate exposure near threshold. Opiate effects were mimicked by the y-aminobutyric acid (GABA) antagonist picrotoxin, and were partially to fully reversed by GABA itself, suggesting that disinhibition of pyramidal cells might be involved as a mechanism in this opiate effect. The data are evidence for a specific primary effect of morphine within the hippocampus in spite of the low numbers of opiate receptors in this brain region.
INTRODUCTION Previous recording studies have shown that administration of opiates affects the activity of many areas of the brain is, including the hippocampusa,10,11,17, z0. Nonetheless, in this area of the limbic system, there is still considerable question as to the direction and stereospecificity of the effects. Nicoll et al. 17 and Chou and Wang 3 reported consistent increases in hippocampal unit activity following opiate administration which were antagonized by naloxone; Sega120 and Fry et al. 10 reported that
228 only a small percentage of hippocampal units were affected by opiates, some of which increased while others decreased in rate, and these effects were generally not naloxonereversible. Fry et al. TM further reported no differentiation of responses to levorphanol and dextrorphan, the active and inactive isomers respectively, and concluded that the effects they observed may not have been mediated via specific opiate receptors. Since some of the studies mentioned above involved systemic administration of drugs, it is not clear whether the observed effects are a result of a primary action of morphine on the hippocampus. In other studies, the drugs have been applied iontophoretically whereby the local concentrations of drugs are unknown. Consequently, we studied the effect of morphine on the isolated hippocampal slice using perfusion of solutions of known drug concentrations, so as to be sure of both primacy of effect and the relation of effective drug concentrations to those of known pharmacological effectiveness. As reported here, the hippocampal slice also represents a first in vitro model of opiate action within the brain itself. The advantage of multiple models has recently been shown by the demonstration of differing potencies of opiate agonists on the guinea pig ileum and the mouse vas deferens, thereby the implication of different types of opiate receptors 15. A preliminary report of some of these experiments has been presented 5. METHODS Transverse hippocampal slices were prepared and maintained as previously described19, 22. Male Sprague-Dawley rats (200-350 g) were decapitated and the brain rapidly removed and received in medium at ambient temperature. The hippocampus was dissected free and sliced into 400 #m sections. The slices were received in medium also at ambient temperature which was then immediately transferred to a water bath at 35 °C. Slices were preincubated a minimum of 30 min prior to use. Composition of the medium in mM was as follows: NaCI 124, KC1 5, KH2PO4 1.25, MgSO4"7H20 2, CaCIz.2H20 2, NaHCOa 26, dextrose 10. Medium was equilibrated with 95 % 02/5 ~o CO2. To block synaptic transmission, some experiments were carried out in medium having 0.5 mM Ca 2+ and 10 mM Mg 2+ (in this medium C I - was 16 mM greater than the control value of 133 mM ; all other constituents were unchanged). For electrophysiology, slices were transferred singly to a recording chamber where they were perfused with the same medium as above at 33-35 °C under an atmosphere of warmed, moistened 95% 02/5 % CO2. Conventional electrophysiological methods were used for stimulating and recording. Field potentials were recorded with micropipettes filled with either 4 M sodium chloride or 4 M potassium acetate (5-15 M ~). Stimulus pulses of 0.1 msec duration were delivered at 0.2 Hz via bipolar 62 #m nichrome wire, and were isolated with either current or voltage constant. Signals were AC coupled (10 Hz-1 kHz) and were averaged on-line for further analyses. Depending upon the protocol of the particular experiment, field potentials were recorded at the level of the CA1 cell body layer (stratum pyramidale), the apical dendritic (str. radiatum) or basal dendritic (str. oriens) regions. Pyramidal cells were driven ortho-
229 dromically by stimulation of the stratum radiatum or stratum oriens, or antidromically by stimulation of the alveus. (See Fig. 3A for schematic of CA1 organization.) The possibility of stimulus spread from electrodes in one pathway to another was excluded by recording activity at the level of the synapses on the dendrites of the pyramids in the non-stimulated pathway, and determining that no population EPSP occurred from stimulation of the experimental pathway. A typical experiment involved recording from stratum pyramidale of CA1 and stimulating stratum radiatum. While responses stabilized in control medium during 0.2 Hz stimulation, average samples (n = 8 stimulus presentations) were taken every 5 min. At this point the size of the stimulus (input, I) was altered to determine the range of responses from the tissue (output, O). Responses were sampled at various points between threshold and asymptote in order to construct an input/output or I/O curve from averaged potentials (n = 4 at each stimulus intensity). Following generation of the baseline I/O curve an additional sample was taken to assure that the response at a given voltage was unchanged, and the tissue perfusion was then switched to medium containing a particular drug. Responses were sampled every 5 min until they showed no further change, up to a maximum of 30 min, whereupon a second I/O curve was generated. At this point the perfusate could again be changed as required, and the same recording protocol repeated. Population spike amplitudes used in I/O curves were calculated as the average of maximum initial positivity to peak spike negativity and peak spike negativity to following maximum positivity (see Fig. 1). The following drugs were used: morphine sulfate (BDH); naloxone (Endo); picrotoxin and y-aminobutyric acid (GABA, Sigma); levorphanol tartrate and dextrorphan tartrate (Hoffman La Roche). RESULTS
Effect of opiates on radiatum-evoked field potentials in CA 1 The control response recorded in stratum pyramidale following stimulation of the radiatum consisted of a slow positive potential upon which, at sufficiently intense stimulation, a negative population spike was superimposed (Fig. 1). The slow potential has been ascribed to a population EPSP generated in the apical dendrites while the negative spike has been demonstrated to be summation of synchronously firing unit spikes 1. Morphine (0.5-50 ~M) had two effects upon the field potential recorded in the cell body layer (str. pyramidale). The more obvious effect was the dose-related production of a second (Fig. 1) and, at highest concentrations, additional population spikes. This effect of morphine developed slowly, in part due to the rate of perfusion (approx. 2 ml/min), and because the tissue was not fully submersed in perfusion medium but rather was supported on a nylon net and was bathed from below with only a meniscus of fluid covering its surface. Morphine effects were resistant to washout with control medium, but changing the perfusate to one containing naloxone (0.1-10/~M) resulted in rapid reversal of the opiate-induced augmentation. Furthermore, in 5 out of 5 experiments, levorphanol (5/zM) was effective in producing this
23O control
-
~_q
morphine IOpM
l/~/"~'-kt
5rain
,j
P' S
-~11
/
~'JI , --
[
F
IOmin -
-
25min
! i
,j naloxone 2 uM
/r~jr_ ~'JS
~ "
5rain _ IOmin
J
_~.
j2mV
5reset
Fig. 1. Example of the effect of morphine on the field potential in stratum pyramidale of CA1 evoked by stimulation in stratum radiatum. Reversal occurs quickly followingchange to naloxone-containing perfusate. In this and subsequent figures negativity is downwards; all responses shown are averaged. Stimulus occurred at the interruption in the tracings. In this figure, n = 8 responses]average. additional population spike, while dextrorphan (5 #M) was ineffective in 4 out of 4 (Fisher Probability, P < 0.025). For isomeric tests, 5 # M was chosen since morphine reliably produced an effect at this concentration. The size of the opiate-produced secondary population spike as related to the primary spike depended upon the position of the response on its I/O curve. As can be seen in Fig. 2A, the size of the secondary spike became an increasingly greater percentage of the primary spike as the stimulus voltage was increased to asymptote for the primary spike. The second effect of morphine was on the primary population spike, and is most clearly seen in its I/O curve (Fig. 2B). Morphine produced an increase in the amplitude of the primary spike over the entire range of the I/O curve. This increase was fully reversible by naloxone at the lowest stimulus intensities, partially reversible at slightly higher intensities, but not at all reversible in the middle to upper regions of the I/O curve. Fig. 2C shows an example of this gradation of naloxone reversibility in averaged responses near threshold. This action of morphine on the primary population spike was more apparent with levorphanol, which is approximately 7 times more potent than morphine as measured in the guinea pig ileum 12. While some upward drift of the I/O curve was seen on occasion after dextrorphan, it was not naloxone-antagonized anywhere along the curve. This increase most probably represents small changes in the responses occurring over the long times they were followed, since
231 ~80-
o control n morphine 5~M A naloxone |pM
E
~> 2 0 F: eL,
~
.~
J
-& 15
=e6o. ._ ~,
o control o morphine 5jJM A nolox
. '2
0=4o.
o
I0
xo
0.
o E
A
a.
,'o Stimulus
io
3'o
,'o
0 i
!
i
0
I0
20
intensity (volts)
Stimulus
i
Intensity
i
30 (vo[t"s)
40
D 6 volts
C control
morphine SuM
naloxone I}JM
-- " ~ " ~
-
-
~
-.. " ~ . ~ . . ~ . ~
I0 volts
,2vo,,s
-- ~
7 volts
--
8 volts
~
9 volts
~
10 volts I ImV
5 msec
I0 msec
Fig. 2. A: illustration of the I/O dependence of the morphine effect at 30 min exposure. Graph shows increasing amplitude of the secondary population spike (expressed as a percentage of the primary spike) as a function of stimulus intensity. Note that there was no secondary spike in control and that the effect was totally reversed by naloxone over the full range of the I/O curve. B: I/O curves for the primary spike of the same response as in A. Note that the spike amplitude in morphine is increased over the entire I/O curve leading to a shift in spike threshold to lower stimulus intensity. Naloxone antagonizes this increase only near threshold, resulting in a net shift to lower threshold/higher amplitude responses after opiate exposure. C: sample responses from same experiment at stimulation intensities of 12 V (approximate threshold for population spike in control medium), 10 V (just below threshold) and 14 V Oust above threshold). Morphine produces a primary population spike at and below the control threshold which is antagonized by naloxone, but the spike produced above threshold is not (n = 4/averaB¢). D: superimposed population EPSPs in control and 5/~M morphine media recorded in radiatum in another experiment. Those shown are just above threshold for the EPSP and still subthreshold for spike production in control medium. Morphine has no consistent effect on the EPSP at voltages subthreshold for spike production (6,7,8 V). When spike production occurs in pyramidale (9 and 10 V) it is not possible to determine whether there are changes in the EPSP responsible for spike production (e.g. an increased rate of rise) or if spike production is distorting the EPSP records (n = 4/average).
232
olveus
A
_%uJ /
J --
oriens
pyramidale
radiatum i~ Ok" morphine SuM normal medium]
control
morphine 5~M (Mgf/CoI)
B
stimulotion
--
--
--'~
5mV
5 msec
stimulation _ _ J 5mV 5 msec
Fig. 3. A: schematic of the organization of the CAI region. Interneurons are thought to mediate recurrent activity to pyramidale cells. B: recordings from pyramidale following stimulation of oriens (upper tracings) or alveus (lower tracings) in the same slice. In normal medium, morphine resulted in art I/O-dependent appearance of a secondary spike in both cases. In low calcium/high magnesium medium (0.5 mM/10 mM resp.) the synaptic response to oriens stimulation was abolished as was the secondary spike following alveus stimulation. Note that neither the presynaptic volley (first negative-positive deflection) in the oriens-evoked response nor the primary antidromic spike in the alveus-evoked response were significantly affected by the low calcium/high magnesium. Responses shown here are averaged samples at the top of their respective I/O curves (n = 8/average).
c o n t r o l experiments in which responses were followed in n o r m a l m e d i u m over c o m p a r a b l e p e r i o d s o f time i n d i c a t e d t h a t the response f r o m a slice o n occasion exhibited some u p w a r d drift. This was in spite o f the fact t h a t stable baseline responses were o b t a i n e d p r i o r to d r u g a p p l i c a t i o n (see Methods). A p o r t i o n o f these experiments were d o n e with s i m u l t a n e o u s recordings at the level o f the synapses in r a d i a t u m as well as in p y r a m i d a l e (Fig. 3A), in o r d e r to assess w h e t h e r there were changes in the p o p u l a t i o n EPSP due to o p i a t e exposure. A s shown in Fig. 2D, at stimulus intensities which were s u b t h r e s h o l d for the p r o d u c t i o n o f a
233 population spike there were no measurable differences between the averaged EPSP in control and in morphine-containing media. At higher intensities of stimulation (9 and 10 volts in Fig. 2D) opiate perfusion led to the production of a primary spike where none had existed in control (as shown in Fig. 2B, C) and this spike was also 'seen' by the dendritic electrode, giving rise to a distortion of the population EPSP. However, the fact that there were no changes in the population EPSP at subthreshold stimulus intensities (6, 7, 8 volts in Fig. 2D) argues that morphine did not produce its augmentation of pyramidal population spikes following radiatum stimulation through an increase in afferent excitation.
Mechanism of opiate augmentation In addition to their synaptic input from fibers in the radiatum, the pyramidal cells of CA1 also have synaptic endings on their basal dendrites from fibers in the stratum oriens (Fig. 3A), and stimulation of these fibers produces in pyramidale a field potential similar to that following radiatum stimulation (Fig. 3B). Morphine augmented this response as well (Fig. 3B, upper traces) suggesting that the opiate phenomenon did not depend specifically on excitatory input from radiatum. Pyramidal neurons may also be activated non-synaptically by stimulation of their axons in the alveus. As depicted schematically in Fig. 3A, this activation would still involve any recurrent synaptic systems. Morphine again led to the production of a secondary population spike in the antidromic field potential recorded in pyramidale (Fig. 3B). If however, after the morphine effect was generated, the perfusate was switched control
',j
/ f -~ C . ~ o t - f~
/
oxin 5,uM --~__
5 min
/ J
I0 rain
20rain
5rnsec
Fig. 4. Example of the effect of picrotoxin on the CA1 field potential elicited by stimulation of stratum radiatum (n = 8/average).
234 A
morphine SuM
control
morphine * GABA 5 uM ! mM
naloxone l uM
20 v
~:
i,/
:
30v
_ _j2mV 5msec o control a morphine
I00 - o c o n t r o l
5)JM
c~ m o r p h i n e 5/JM A naloxone IpM • ~ o r p h i n e 5.uM • G A B A I
tt naloxone l,uM • morphine 5HM÷GABA
ImM
> E
80. c~
=
•~. 4
qu
v)
6O
c o
v) c
c~ o o. >,2
a. o 0.
o
40"
o
-o
/
2O
o (n
0
0
I '0 Stimulus
2' 0 Intensity
3'0 (volts]
4'0
g
5//
,'o
Stimulus
2'o
Intensity
3'o
4'o
(volts}
Fig. 5. A: responses obtained at two stimulus voltages as tissue was perfused successively by control medium, morphine, morphine plus GABA, and naloxone (n ~ 4/average). B and C: I/O curves for primary and secondary population spikes respectively. Note again the increase in the primary spike over the entire I/O curve in opiate medium and the antagonism of this increase near threshold by naloxone. The secondary population spike is partially reduced by GABA (C) while the amplitude of the primary spike is increased somewhat in GABA (B).
to one containing morphine in high magnesium/low calcium medium sufficient to block synaptic activity, the morphine effect on the antidromic response was quickly abolished (Fig. 3C). Upon return to morphine in normal medium the opiate effect was reestablished and was subsequently antagonized by naloxone. This evidence suggested that morphine did not produce its effect by a direct action upon pyramidale neurons but rather by an action on other neurons coupled synaptically to them. Of the possible mechanisms, an action of the opiate upon inhibitory interneurons seemed likely. Since these interneurons are thought to use GABA as transmitter 1,z3, the effect of the GABA antagonist picrotoxin on the radiatum-evoked CA1 field potential was examined. Picrotoxin, at least with short-duration exposures,
235 produced changes in the response similar to morphine, including shift of the threshold of the I/O curve to lower stimulus intensities and production of additional population spikes (Fig. 4). Picrotoxin effects were resistant to washout with control medium and often became convulsive-type activity consisting of several to many population spikes (Fig. 4, 20 min, shows the beginning of such activity). Because a GABA-antagonist mimicked the effect of morphine on the hippocampal slice, it was of interest to determine if GABA itself would antagonize the action of morphine. As shown in the example of Fig. 5 it was observed that GABA antagonized the opiate-induced secondary spike while at the same time causing an increase in the primary spike. The balance of the secondary spike not antagonized by GABA was reversed by naloxone (Fig. 5C). While in this particular case the secondary spike was antagonized only partially by GABA, other experiments using the same concentration showed complete block of the secondary spike. Furthermore, GABA reversal of the threshold increase in the primary spike due to morphine was also seen in other experiments, although there was still a GABA-induced increase at higher stimulus intensities. DISCUSSION The effect of morphine on the CA1 pyramidale field potential was consistent augmentation of the response. In no case was depression of the pyramidale field potential observed. (The drug specificity of this effect is demonstrated by the observation that ethanol, another depressant, consistently reduced the amplitude of the primary population spike in the same preparation4.) The augmentation occurred both as an increase in the primary population spike and as the production of a secondary field spike. The latter was stereospecific and naloxone reversible as was the former at and near threshold. The reason for the naloxone insensitivity of increased primary population spike at higher stimulus intensities remains unclear. It may represent a non-specific opiate action which others have reported in the hippocampus1°, 20. If so, it points out the importance of examining the full functional range of the system when investigating pharmacological effects since in the present study we would have observed only this naloxone-insensitive component of the primary spike if we had used suprathreshold stimulation exclusively. The phenomenon may also, however, be a consequence of a specific effect of opiate exposure, for example, some potentiation of the response produced when the threshold is lowered by morphine or levorphanol, which remains after opiate antagonism. If this proves to be so, it would be an important action of opiates since it results in a shift of the 1/O curve to lower threshold/higher amplitude responses. Regarding the question of mechanism, the augmentation of pyramidal cell activity did not depend by which pathway the cells were activated but, in the case of the antidromic responses, did require synaptic transmission to occur, arguing that morphine did not act directly upon the pyramidal cells. Although it was not possible to determine if there were changes in the EPSP at stimulus intensities above spike threshold, which may qualify our interpretation, changes in afferent excitation in the case of
236 the orthodromic response were ruled out on the basis of our extracellular data indicating population EPSPs below spike threshold were unchanged by morphine. Replication of the morphine effects with picrotoxin, and reversal of the opiate effects with GABA provide corroborative evidence to the proposal of Siggins et al. 21 that opiates may act upon inhibitory interneurons in the hippocampus. The GABA evidence is particularly indirect, of course, since GABA at appropriate concentration would be expected to reduce the entire response whether or not its augmentation were due to morphine depression of GABAergic cells or some other mechanism. However, antagonism of the secondary population spike, while there was a concommitant increase in the primary, does suggest some selectivity which may point towards an origin for the secondary spike as reduced GABA at the pyramidal cell following primary spike generation. The mechanism for the GABA-induced increase in the primary spike remains unclear. Our finding of consistent increases are basically in agreement with the spontaneous unit studies of Chou and Wang 3 and Nicoll et al. 17. However, they are inconsistent with data reported recently by Fry et al. 1° and by Segal z°, in tha~ our effects were both stereospecific and naloxone-reversible, and in no case was depression of activity observed. There are several methodological differences between the latter studies and our own that may account for the discrepancy. Both Segal 2° and Fry et al. 1° applied drugs by iontophoresis. This may lead to unknown, and possibly higher local concentrations which might increase the probability of observing non-specific opiate effects18. For example, Dingledine et al. 8 have reported that a 300/~M concentration of naloxone acted directly as a GABA antagonist, and we have observed that 100 #m naloxone caused effects similar, rather than antagonistic, to morphine in the hippocampal slice (unpublished observations). The concentrations of morphine used in our study compare favorably with doses required to produce effects in vivo (see discussion in ref. 6). Furthermore, as opposed to sampling spontaneous unit activity, we recorded an evoked population response which may lead to greater homogeneity of effect. Finally, examination of I/O curves allowed us to assess the effect of the drug over the entire functional range of the neurons. The overall effect of morphine, then, was an increased excitability of the pyramidal cells of hippocampus. Other studies in vivo suggest opiates may function in this region to elicit epileptiform activity. For example, Elazar et al. 9 have reported EEG excitation and convulsive activity at the site of intracerebral injections of enkephalin into the dorsal layer of hippocampus. Linseman and Grupp (in preparation) observed hippocampal spiking in the EEG following i.v. injections of low doses of morphine to rats. Although there is no direct evidence, it is possible that the latter effects may be mediated by a cellular mechanism such as proposed here. In summary, our data are evidence for a specific primary effect of opiates within the hippocampus in spite of the reported low density of opiate receptors in this area 1~. This model, together with those described in tissue culture 2,6,7,16, may be useful in elucidating the mechanisms of opiate action within the central nervous system.
237 ACKNOWLEDGEMENT We t h a n k H o f f m a n - L a Roche (Canada) for their gift o f levorphanol a n d dextrorphan.
REFERENCES 1 Andersen, P., Organization of hippocampal neurons and their interconnections. In R. L. Isaacson and K. H. Pribram (Eds.), The Hippocampus, VoL 1: Structure and Development, Plenum Press, New York, 1975, pp. 155-175. 2 Barker, J. L., Neale, J. H., Smith, T. G., Jr. and Macdonald, R. L., Opiate peptide modulation of amino acid responses suggests novel form of neuronal communication, Science, 199 (1978) 1451-1453. 3 Chou, D. T. and Wang, S. C., Unit activity of amygdala and hippocampal neurons: Effects of morphine and benzodiazepines, Brain Research, 126 (1977) 427-440. 4 Corrigall, W. A., Carlen, P. L. and Staiman, A. L., Acute effect of ethanol on hippocampal CAI field potentials recorded in vitro, Neurosci. Abstr., 3 (1977) 288. 5 Corrigall, W. A. and Linseman, M. A., Acute effects of morphine on the hippocampus in vitro, Fed. Proc., 38 (1979) 3296. 6 Crain, S. M., Crain, B., Peterson, E. R. and Simon, E. J., Selective depression by opioid peptides of sensory-evoked dorsal-horn network responses in orgtnized spinal cord cultures, Brain Research, 157 (1978) 196-201. 7 Crain, S. M., Peterson, E. R., Crain, B. and Simon, E. J., Selective opiate depression of sensoryevoked synaptic networks in dorsal horn regions of spinal cord cultures, Brain Research, 133 (1977) 162-166. 8 Dingledine, R., Iversen, L. L. and Breuker, E., Naloxone as a GABA antagonist: evidence from iontophoretic, receptor binding and convulsant studies, Europ. J. Pharmacol., 47 (1978) 19-27. 9 Elazar, Z., Motles, E., Ely, Y. and Simantov, R., Acute tolerance to the excitatory effect of enkephalin microinjections into hippocampus, Life ScL, 24 (1979) 541-548. 10 Fry, J. P., Zieglghnsberger, W. and Herz, A., Specific versus non-specific actions of opioids on hippocampal neurons in the rat brain, Brain Research, 163 (1979) 295-305. 11 Hill, R. G., Mitchell, J. F. and Pepper, C. M., The excitation and depression of hippocampal neurones by iontophoretically applied enkephalins, J. Physiol. (Lond.), 272 (1976) 50-51P. 12 Kosterlitz, H. W., Leslie, F. M. and Waterfield, A. A., Rates of onset and offset of action of narcotic analgesics in isolated preparations, Europ. J. PharmacoL, 32 (1975) 10-16. 13 Kuhar, M. J., Pert, C. B. and Snyder, S. H., Regional distribution of opiate receptor binding in monkey and human brain, Nature (Lond.), 245 (1973) 447-450. 14 Linseman, M. A. and Grupp, L., Acute and chronic opiate effects on single units and EEG of medial thalamus and hippocampus: a latency analysis, in preparation. 15 Lord, J. A. H., Waterfield, A. A., Hughes, J. and Kosterlitz, H. W., Endogenous opioid peptides: multiple agonists and receptors, Nature (Lond.), 267 (1977) 495-499. 16 Macdonald, R. L. and Nelson, P. G., Specific opiate-induced depression of transmitter release from dorsal root ganglion cells in culture, Science, 199 (1978) 1449-1451. 17 Nicoll, R. A., Siggins, G. R., Ling, N., Bloom, F. E. and Guillemin, R., Neuronal actions of endorphins and enkephalins among brain regions: a comparative microiontophoretic study, Proc. nat. Acad. Sci. (Wash.), 74 (1977) 2584-2588. 18 North, R. A., Opiates, opioid peptides and single neurones, Life Sci., 24 (1979) !527-1546. 19 Schwartzkroin, P. A. and Wester, K., Long-lasting facilitation of a synaptic potential following tetanization in the in vitro hippocampal slice, Brain Research, 89 (1975) 107-119. 20 Segal, M., Morphine and enkephalin interactions with putative neurotransmitters in rat hippocampus, Neuropharmacology, 16 (1977) 587-592. 21 Siggins, G. R., Zieglg/insberger, W., French, E., Ling, N. and Bloom, F. E., Opiates and opioid peptides may excite hippocampal pyramidal neurons by inhibiting adjacent inhibitory interneurons, NeuroscL Abstr., 4 (1978) 414. 22 Spencer, H. J., Gribkoff, V. K., Cotman, C. W. and Lynch, G. S., GDEE antagonism of iontopho-
238 retic amino acid excitations in the intact hippocampus and in the hippocampal slice preparation, Brain Research, 105 (1976) 471-481. 23 Straughan, D. W., Neurotransmitters and the hippocampus. In R. L. Isaacson and K. H. Pribram (Eds.), The Hippocampus, Vol. 1: Structure and Development, Plenum Press, New York, 1975, pp. 239-267.