- Email: [email protected]
Different Drug-Susceptibilities of long-term potentiation in three input systems to the CA3 region of the guinea pig hippocampus in vitro
Neuropharmacology Vol. 29, No. 5, pp. 487492, Printed in Great Britain. All rights reserved
0028-3908/90$3.00+ 0.00 Copyright 0 1990Pergamon Press plc
1990
DIFFERENT DRUG-SUSCEPTIBILITIES OF LONG-TERM POTENTIATION IN THREE INPUT SYSTEMS TO THE CA3 REGION OF THE GUINEA PIG HIPPOCAMPUS IN VITRO K. ISHIHARA, H. KATSUKI, M. SUGIMURA, S. KANEKO and M. SATOH* Department of Pharmacology, Faculty of Pharmaceutical Sciences, Kyoto University, Sakyo-ku, Kyoto 606. Japan (Accepted 22 January 1990)
Summary-The susceptibilities to several drugs of long-term potentiations in the three input systems (mossy, commissural/associational and fimbrial fibres) to CA3 pyramidal neurones were investigated in hippocampal slices from the guinea pig. o-2-Amino-Sphosphonovalerate (D-APV), a selective antagonist at N-methyl-D-aspartate (NMDA) receptors, blocked the long-term potentiations in the commissural/ associational fibre- and fimbrial fibre-CA3 systems, but did not significantly affect that in the mossy fibre-CA3 system. The latter was suppressed by kynurenate, a non-selective glutamate receptor antagonist. On the other hand, naloxone, an opioid antagonist, inhibited and bifemelane, which improves metabolism in brain and has an anti-amnesic action, augmented long-term potentiation in mossy fibre-CA3 system but did not influence those in commissural/associational fibre- and fimbrial fibre-CA3 systems. These findings suggest that the mechanisms, relevant to production of long-term potentiation in the mossy fibre-CA3 system, are different from those in the commissural/associational fibre- and fimbrial fibre-CA3 systems. N-Methyl-D-aspartate receptors are involved in the latter systems, while non-NMDA receptors for L-glutamate and opioid receptors are involved in the former. Further, the mossy fibre-CA3 system is more susceptible to a drug, having an anti-amnesic action, than are the other two systems. Key words-long-term potentiation, CA3 region, guinea pig.
opioid receptor, N-methyl-D-aspartate
It is recognized that the N-methyl-D-aspartate (NMDA) receptor for L-glutamate plays an important role in the production of long-term potentiations in the Schaffer collateral/commissural fibre-CA1 system (Coilingridge, Kehl and McLennan, 1983) and the perforant pathway-dentate granule cell system (Morris, Anderson, Lynch and Baudry, 1986), as these long-term potentiations are blocked by selective antagonists at the NMDA receptor like D-2-amino-5phosphonovalerate (D-APV). However, Harris and Cotman (1986) found that D-APV did not inhibit long-term potentiation in the mossy fibre-CA3 system. Furthermore, Satoh, Ishihara and Katsuki (1988) demonstrated that piracetam and bifemelane which have anti-amnesic action augmented long-term potentiation in the mossy fibre-CA3 system but not that in the Schaffer collateral-CA1 system. Thus, different susceptibilities to some drugs were shown on long-term potentiations in CA1 and CA3 regions. It is known that the pyramidal neurones of the CA3 region receive at least three different inputs, the mossy, commissural/associational and fimbrial fibres (Brodal, 1981). In each of the input systems, longterm potentiation was demonstrated in slice preparations (Yamamoto and Chujo, 1978; Yamamoto and Sawada, 1981; Harris and Cotman, 1986). Harris and Cotman (1986) demonstrated that D-APV *To whom correspondence
receptor, hippocampal slice,
inhibited long-term potentiation in the commissural/ associational fibre-CA3 system but not in the mossy fibre-CA3 one. These observations prompted an investigation of the effects of several drugs, classified into different categories, such as D-APV, naloxone (an opioid antagonist) and bifemelane on long-term potentiations in the three input systems to the CA3 region. METHODS Slices of hippocampus, 400-500 pm thick were prepared from male guinea pigs (200-350 g) either manually or using a microslicer (DTKlOOO, Dosaka E. M. Kyoto, Japan), as described previously (Iwama, Ishihara, Satoh and Takagi, 1986). Slices were incubated for 1 hr at 34-35°C in artificial cerebrospinal fluid (CSF) of the following composition (in mM): NaCl, 124; KCl, 5; KH,PO,, I .24; MgSO,, 1.3; CaCl,, 2.4; NaHCO,, 26; D-glucose, 10; bubbled with 95%0, + 5%CO,. A single slice was transferred into a recording chamber, having a volume of approximately 0.8 ml and perfusion was carried out at a rate of 1.5-2.0 ml/min with the artificial CSF at 34-35°C. The field potential (population spike) was recorded in the pyramidal layer of the CA3 region, with a glass pipette of about 10 pm tip diameter, filled with 0.9% NaCl (< IO Ma). A bipolar stimulating electrode was placed in the granule cell layer of the
should be addressed. 487
K.
488
ISHIHARA et al.
Stim. (com./ana.)
Fig. 1. Schematic drawing of the hippocampal slice of the guinea pig. The recording electrode (Rec.) was placed in the pyramidal layer of the CA3 region to record the population spike. The stimulating electrode (Stim.) was positioned in the granule cell layer of the dentate gyrus to stimulate the mossy fibres (mossy),
in the stratum radiatum, near the CA2 region, to stimulate the commissural/associational fibres (com./asso.) or in the fimbria to stimulate the fimbria fibres (fimbria). dentate gyrus, in the stratum radiatum near the CA2 region or in the fimbria for stimulation of the mossy fibres, commissural/associational fibres or fimbrial fibres, respectively (Fig. 1). Rectangular pulses (0.1 msec in duration) were delivered at 0.2 Hz through the respective electrodes, as test stimulation, with a strength (up to 10 V) that gave 70-90% of the maximum response. Population spikes, induced by 10 successive stimulations, were averaged and recorded every 5 min. Three or 4 averaged population spikes were taken before tetanic stimulation and the mean amplitude was used as a basal value for calculating the increase of the population spike. The amplitude of the population spike was measured from the negative to positive peak, as the most accessible measure of the global effects of drugs on the response. tetanic stimulation, with the same strength and pulse duration as those of test stimulation, was applied at 33 Hz for 5 set, using the same stimulating electrode in the respective systems. In some experiments, one second instead of 5 set was used as a duration of tetanic stimulation of commissural/associational or fimbrial pathways. Thereafter, the stimulus frequency reverted to 0.2 Hz and the averaged population spikes were again recorded every 5 min for 1 hr or more. Long-term potentiation was produced in one of the three input systems, in each preparation and was expressed as the percentage increase in the amplitude of averaged population spikes after tetanic stimulation over the mean amplitude before tetanic stimulation. The drugs used were D-APV (Funakoshi), kynurenate (Nacalai tesque), naloxone HCl (a gift from
E. I. DuPont, Garden City, New York, U.S.A.) and bifemelane HCl (a gift from Mitsubishi Chemical Ind. Ltd, Yokohama, Japan). Kynurenate was dissolved in weak NaOH solution and adjusted pH to 7-8 with HCl solution and other drugs were dissolved in distilled water at 1000-fold concentration of the final one used. Those were added to artificial CSF and perfused over the slice for 15 min in the case of bifemelane on mossy fibreCA3 system and 25 min in other cases. Tetanic stimulation was applied at 11 or 21 min after the start of the perfusion of drug. Then, the application of drug was stopped 4min after the tetanic stimulation. The magnitude of long-term potentiation was expressed by calculating the area between the zero line and the curve of the time course of long-term potentiation, from 12 to 62 min after the tetanic stimulation. An example is shown in Fig. 2, in which the area of the non-drug treated group is shown by the dotted area (A) and that of the kynurenate-treated group by the hatched area (B). Effects of drugs were determined by comparison between these areas. In the present experiments, concentrations of the drugs tested were chosen such that population spikes in the absence of tetanic stimulation were not significantly affected in any of the input systems examined. Statistical significance of differences were calculated according to Student’s t-test. RESULTS A single shock to the mossy fibres or fimbria induced, in the CA3 pyramidal layer, a biphasic field
Long-term potentiation
in CA3 region
489
Fig. 2. Time course of long-term potentiation in the non-drug treated group or kynurenate (10m6M)treated group. Ordinate: percentage increase in the amplitude of the population spike (PSA). Abscissa: time after tetanic stimulation (TS) in min (minus sign: time before tetanic stimulation). 0-0 and 0-O indicate non-drug treated group (n = 7) and kynurenate (10e6M) treated group (n = 3), respectively. Each point represents the mean with SEM. Kynurenate was perfused between -2l-4min, shown with open column. The magnitude of long-term potentiation was estimated by calculating the area between zero line and each curve of the time course of the long-term potentiation, from 12 to 62 min after the tetanic stimulation. For example, the non-drug treated group is represented by the dotted area (A) and kynurenate (10e6M) treatment group by the hatched area (B). *P < 0.05, **P < 0.01, Statistically different from non-drug treated group (Student’s r-test). PSA = Population spike amplitude. TS = Tetanic stimulation.
potential starting with a negative phase, the latency of which was 3.5-7 msec or 3.5-6 msec, respectively. Test stimulation of the commissural/associational fibres elicited a field potential, consisting of a negative deflection (latency was less than 2 msec), followed by a biphasic field potential with latency of 4-7 msec. Tetanic stimulation of the mossy, commissural/ associational or fimbrial fibres induced long-term potentiation of the population spike in the CA3 region of the hippocampus. When no drug was added to the artificial CSF, the percentage increase in amplitude of the test responses (population spikes) after tetanic stimulation was 35-45% (n = 7), 200-225% (n = 3) or 225-280% (n = 3) of the mean amplitude before tetanic stimulation in the mossy fibre, commissural/associational fibre or fimbrial pathway, respectively. Each increase persisted for 1 hr or more after tetanic stimulation. Examples of the long-term potentiation in the fimbrial fibre-CA3 system are shown in Fig. 3. The results of drugactions on long-term potentiation in the three input systems to the CA3 region are summarized in Table 1.
of 10m5M, was 0.06 times as large as that in the non-drug treated group (Table 1). An example of population spikes which failed to show long-term potentiation in the fimbrial pathway in a D-APVtreated slice, is shown in Fig. 3. On the other hand, long-term potentiation in the mossy fibre-CA3 system was resistant to D-APV: the means of the magnitudes of long-term potentiation in the presence of 10-j M and 10e4 M of D-APV were 0.94 and 0.92 times, respectively, as large as that in the non-drug treated group. However, kynurenate, a non-selective glutamate receptor antagonist, significantly suppressed the long-term potentiation of the mossy fibre-CA3 system, at a concentration of 10m6M (Fig. 2). The mean of magnitude of long-term potentiation in kynurenate-treated slices, was 0.23 times as large as that in the non-drug treated group (Table 1). Both D-APV and kynurenate, at the concentrations used, did not significantly change the amplitude of the population spike before the tetanic stimulation, in any of the input systems examined.
Effect of D-APV on long-term potentiation
In the commissural/associational or fimbrial pathway to the CA3 region, long-term potentiation was not significantly affected by naloxone, at a concentration of 10m6M: the mean of magnitudes of longterm potentiation was 1.02 or 1.23 times as large as that in the non-drug treated group, respectively (Table 1). An example of long-term potentiation in the fimbrial fibre-CA3 system in the naloxone-treated slice is shown in Fig. 3. In the mossy fibre pathway to the CA3 region, long-term potentiation was
Long-term potentiation in the commissural/ associational fibreCA3 system was significantly suppressed by D-APV: the mean of the magnitudes of long-term potentiation, at a concentration of 10S5 M, was 0.10 times as large as that in the non-drug treated group (Table 1). o-2-Amino-5-phosphonovalerate also significantly suppressed long-term potentiation in the fimbrial fibre-CA3 system: the mean of magnitudes of long-term potentiation, at the concentration
Efect of naloxone on long-term potentiation
K.
490 Before
No
TS
27min
after
ISHIHARA et al.
TS
drug
D-APV
No loxone
A-.
.r‘_ I
Fig. 3. Examples of long-term potentiation of the population spike, recorded in the pyramidal layer of CA3 region of slices of hippocampus, by simulating fimbrial fibres. Upper traces: non-drug-treated slice. Middle traces: D-APV (IOe5M)-treated slice. Lower traces: naloxone (10e6M)treated slice. Ten responses were averaged in each trace. Test stimulation was delivered at each dot. Each left and right traces are the responses before tetanic stimulation (TS) and 27 min after tetanic stimulation (TS) in the same slice. Calibration: 1mV; 10 msec. significantly suppressed by naloxone at 10e6 M: the mean of magnitudes of long-term potentiation was 0.31 times as large as that in the non-drug treated group (Table 1). Naloxone, at 10d6 M, did not significantly alter the amplitude of the population spikes, before tetanic stimulation, in any of the input systems to the CA3 region. EfSect of bifemelane Bifemelane in the mossy Table
on long-term
potentiation
augmented the long-term potentiation fibre-CA3 system, as described before
I. Effects of o-APV
(IO ~J M), kynurenate (10m6 M), naloxone mossv fibre-. commissuraliassociational
Mossy
DISCUSSION
In the present experiments, the effects of drugs tested on long-term potentiation of population spikes in the three input systems to the hippocampal CA3 region, were produced with no significant effects on the population spikes, in the absence of tetanic stimulation. This would suggest that those drugs specifically act on mechanisms causing long-term potentiation at the concentrations used. Harris and Cotman (1986) reported that D-APV blocked long-term potentiation in the commissural/ associational fibreXA3 system but not that in the mossy fibreCA3 system, using hippocampal slices of the guinea pig. Their observation has been confirmed (10m6 M) and bifemelane (10m6 M) on long-term fibre- and fimbrial-CA3 systems
33 Hz, 5sec
D-APV
1917 f 238” (7) I799 + 239 (3)
10708 f 1734(3) 1056 + 488 “(3)
Kvnurenate Naloxone Bifemelane
443 + I I4 “(3) 589 ; 343 “i4j 5992 + 722 x.$(4)
10896 + 1762 (3) 12085 f 4352 (3)
33 Hz, I set 2533 f 446 (3)
21821375
potentiation
in the
Fimbrial
Commissural/associational
33 Hz, 5 set No drug
(Satoh et al., 1988): the mean of magnitudes of long-term potentiation at 10m6 M was 3.13 times larger than that in the non-drug treated group (Table 1). In contrast, bifemelane did not augment the long-term potentiation of the commissural/ associational or fimbrial pathway to the CA3 region at 10m6M: the mean of magnitudes of long-term potentiation was 1.13 or 0.90 times as large as that in the non-drug treated group, respectively (Table 1). In the commissural/associational fibre- or fimbrial fibreCA3 systems, degrees of the long-term potentiation in non-drug treated slices were large, when 33 Hz for 5 set, standard parameters in this study, were used as tetanic stimulation. Since this might contribute to the failure of an augmenting effect of bifemelane, in some experiments 1 set, instead of 5 set, was used as the duration of tetanic stimulation in these systems. The increase in amplitude of the test response after tetanic stimulation was 40-60% (n = 3) for commissural/ associational or 45-85% (n = 4) for fimbrial stimulation in the non-drug treated group and persisted for 1 hr or more. Even under such a condition, the long-term potentiations in the commissural/ associational fibre- and fimbrial fibreCA3 systems were not affected by bifemelane at 10e6 M: the mean of magnitude of long-term potentiation were 0.86 and 1.12 times as large as those in the non-drug treated group, respectively. Bifemelane, at the concentration used, did not significantly affect the amplitude of population spike before tetanic stimulation in any of the input systems to the CA3 region.
(3)
33 Hz, 5 set
33 Hz, I set
12906*812(3) 753 ?r 576 **(3)
2827 k 784 (4) -
I5811 +2037(3) II588 f 849 (3)
3154+329(3)
YMean f SEM of magnitude of long-term potentiation (area under the graph in arbitrary units). Number in parenthesis of slices tested. -Not tested. **Significant inhibition (P < O.Ol), compared with no-drug group, described in the top of each column. f$Significant augmentation (P < 0.01). compared with no-drug group, described in the top of each column
-
represents the number
Long-term potentiation in CA3 region and further evidence provided for different drugsusceptibilities of long-term potentiations in the three separate input systems to the CA3 region. The drug D-APV (lo-’ M) also inhibited long-term potentiation in the fimbrial fibre-CA3 system, which is suggested to be mediated by NMDA receptors in a similar manner to that in commissural/associational fibre-CA3 system. In the CA3 region, mossy fibres terminate in the stratum lucidum, while commissural/ associational and fimbrial fibres terminate in the stratum radiatum and stratum oriens (Raisman, Cowan and Powell, 1965). Monaghan, Yao and Cotman (1985) reported that the density of NMDA receptors was low and that of kainate receptors was high in the stratum lucidum of the CA3 region but the densities of quisqualate and NMDA receptors were high in the stratum radiatum and stratum oriens. Long-term potentiation in the mossy fibre-CA3 system is probably not mediated by NMDA receptors but by non-NMDA receptors for L-glutamate, since a non-selective antagonist at the excitatory amino acid receptors, kynurenate, suppressed this long-term potentiation, in particular a late phase of it, at a concentration of 10e6M. This concentration of kynurenate is much less than the effective one (2 x or 4 x 10e4 M) for reducing the amplitude of excitatory postsynaptic potentials in the CA3 pyramidal neurone evoked by low frequency (without tetanic) stimulation of the mossy fibres in similar preparations to those employed in this study (Cotman, Flatman, Ganong and Perkins, 1986). In fact, no significant effect of 10e6 M of kynurenate on population spikes of the CA3 region evoked by low frequency stimulation of mossy fibres was observed in the absence of tetanic stimulation. These findings suggest that the mechanism(s) of production (and/or maintenance) of long-term potentiation of the population spike, following tetanic stimulation in the mossy fibre-CA3 system, are kynurenate- sensitive and may be mediated by non-NMDA receptors for glutamate. By contrast, naloxone (10m6M) blocked long-term potentiation in the mossy fibre-CA3 system. This finding is consistent with that previously reported by Martin (1983). Further, it was demonstrated that naloxone did not affect long-term potentiations in the other two input systems to the CA3 region. It was revealed, by the immunohistochemical technique, that hippocampal mossy fibres are dense with dynorphin A, an opioid peptide (Khachaturian, Watson, Lewis, Coy, Goldstein and Akil, 1982). Further, Satoh, Ishihara, Katsuki and Sugimura (1989) showed that the opioid peptide, dynorphin A (lo-’ and lo-‘M), augmented long-term potentiation in the mossy fibre-CA3 system and such an augmenting effect was blocked by naloxone (lo-’ M). These findings suggest an involvement of an opioidergic system in the production of long-term potentiation in the mossy fibre-CA3 system, but not in commissural/associational fibre- and fimbrial fibre-CA3
491
systems. On the other hand, bifemelane (10m6M) significantly augmented long-term potentiation in the mossy fibre-CA3 system but not in the other two input systems to the CA3 region. Long-term potentiation in the Schaffer collateral-CA1 system was previously shown not to be affected by the drug (Satoh et al., 1988). From the above findings, long-term potentiations in commissural/associational fibre- and fimbrial fibre-CA3 systems are similar to that in the Schaffer collateral-CA1 system. In this context, Ito, Okada and Sugiyama (1988) reported that intracerebroventricular injections of pertussis toxin blocked long-term potentiation in the mossy fibre-CA3 system but not in Schaffer collateral-CAl. However, Goh and Pennefather (1989) clearly demonstrated an inhibition by the same toxin of long-term potentiation, even in the latter system and pointed out that the apparently disparate effects of pertussis toxin, reported by Ito et al. (1988), were probably due to insufficient penetration of the toxin to the hippocampal CA1 region. It is of interest that the pharmacological characteristics of long-term potentiation in the mossy fibreCA3 system are in marked contrast to those in commissural/associational fibre- and fimbrial fibreCA3 systems: the former was not changed by D-APV but affected by naloxone and bifemelane, while the latter were suppressed by D-APV but not affected by the latter two drugs. From these findings, it is conceivable that even in the hippocampal CA3 region,
long-term potentiations, in three input systems, may be mediated by different mechanisms. The detailed mechanisms of long-term potentiations in the three input systems, including an involvement of pertussis toxin-sensitive guanine nucleotide binding regulatory proteins, remain to be revealed. REFERENCES Brodal A. (1981) The olfactory pathways. The amygdala. The hippocampus. The ‘Limbic system’. In: Neurological Anatomy in Relation to Clinical Medicine, 3rd edn, pp. 640-697. Oxford, New York. Collingridge G. L., Kehl S. J. and McLennan H. (1983) Excitatory amino acids in synaptic transmission in the Schaffer collateral-commissural pathway of the rat hippocampus. J. Physiol., Lond. 334: 33-46. Cotman C. W., Flatman J. A., Ganong A. H. and Perkins M. N. (1986) Effects of excitatory amino acids antagonists on evoked and spontaneous excitatory potentials in guinea-pig hippocampus. J. Physiol., Lo&. j78: 403-415. Goh J. W. and Pennefather P. S. (1989) oertussis toxin. ,A_ sensitive G protein in hippocampal long-term potentiation. Science 244: 980-983. Harris E. W. and Cotman C. W. (1986) Long-term potentiation of guinea pig mossy fiber responses is not blocked by N-methyl o-aspartate antagonists. Neurosci. Lett. 70: 132-137. Ito I., Okada D. and Sugiyama H. (1988) Pertussis toxin suppresses long-term potentiation of hippocampal mossy fiber synapses. Neurosci. Lett. 90: 181-185. Iwama T., Ishihara K., Satoh M. and Takagi H. (1986) Different effect of dynorphin A on in vitro guinea pig hippocampal CA3 pyramidal cells with various degrees of paired-pulse facilitation. Neurosci. Lett. 63: 190-194.
492
K.
ISHIHARA et al.
Khachaturian H., Watson S. J., Lewis M. E., Coy D., Goldstein A. and Akil A. (1982) Dynorphin immunocytochemistry in the rat central nervous system. Peptides 3: 941-954.
Martin M. R. (1983) Naloxone and long-term potentiation of hippocampal CA3 field potentials in oitro. Neuropeptides 4: 45-50.
Monaghan D. T., Yao D. and Cotman C. W. (1985) L-[‘HJGlutamate binds to kainate-, NMDA- and AMPAsensitive binding sites: an autoradiographic analysis. Brain Res. 340: 378-383.
Morris R. G. M., Anderson E., Lynch G. S. and Baudry M. (1986) Selective impairment of learning and blockade of long-term potentiation by an N-methyl-oaspartate receptor antagonist, AP5. Nature 319: 774-776.
Raisman G., Cowan W. M. and Powell T. P. S. (1965) The extrinsic afferent, commissural and association fibers of the hippocampus. Brain 88: 963-996. Satoh M., Ishihara K. and Katsuki H. (1988) Different susceptibilities of long-term potentiations in CA3 and CA1 regions of guinea pig hippocampal slices to nootropic drugs. Nkosci. Lett. 93:.236-241. Satoh M.. Ishihara K.. Katsuki H. and Suaimura M. (1989) Opioidkrgic modulation of long-term pgtentiation in the guinea pig hippocampus in vitro. Adv. Biosci. 75: 193-196. Yamamoto C. and Chujo T. (1978) Long-term potentiation in thin hippocampal sections studied by intracellular and extracellular recordings. Expl Neural. 58: 242-250. Yamamoto C. and Sawada S. (1981) Important factors in induction of long-term potentiation in thin hippocampal sections. Expl Neurol. 74: 122-130.
0028-3908/90$3.00+ 0.00 Copyright 0 1990Pergamon Press plc
1990
DIFFERENT DRUG-SUSCEPTIBILITIES OF LONG-TERM POTENTIATION IN THREE INPUT SYSTEMS TO THE CA3 REGION OF THE GUINEA PIG HIPPOCAMPUS IN VITRO K. ISHIHARA, H. KATSUKI, M. SUGIMURA, S. KANEKO and M. SATOH* Department of Pharmacology, Faculty of Pharmaceutical Sciences, Kyoto University, Sakyo-ku, Kyoto 606. Japan (Accepted 22 January 1990)
Summary-The susceptibilities to several drugs of long-term potentiations in the three input systems (mossy, commissural/associational and fimbrial fibres) to CA3 pyramidal neurones were investigated in hippocampal slices from the guinea pig. o-2-Amino-Sphosphonovalerate (D-APV), a selective antagonist at N-methyl-D-aspartate (NMDA) receptors, blocked the long-term potentiations in the commissural/ associational fibre- and fimbrial fibre-CA3 systems, but did not significantly affect that in the mossy fibre-CA3 system. The latter was suppressed by kynurenate, a non-selective glutamate receptor antagonist. On the other hand, naloxone, an opioid antagonist, inhibited and bifemelane, which improves metabolism in brain and has an anti-amnesic action, augmented long-term potentiation in mossy fibre-CA3 system but did not influence those in commissural/associational fibre- and fimbrial fibre-CA3 systems. These findings suggest that the mechanisms, relevant to production of long-term potentiation in the mossy fibre-CA3 system, are different from those in the commissural/associational fibre- and fimbrial fibre-CA3 systems. N-Methyl-D-aspartate receptors are involved in the latter systems, while non-NMDA receptors for L-glutamate and opioid receptors are involved in the former. Further, the mossy fibre-CA3 system is more susceptible to a drug, having an anti-amnesic action, than are the other two systems. Key words-long-term potentiation, CA3 region, guinea pig.
opioid receptor, N-methyl-D-aspartate
It is recognized that the N-methyl-D-aspartate (NMDA) receptor for L-glutamate plays an important role in the production of long-term potentiations in the Schaffer collateral/commissural fibre-CA1 system (Coilingridge, Kehl and McLennan, 1983) and the perforant pathway-dentate granule cell system (Morris, Anderson, Lynch and Baudry, 1986), as these long-term potentiations are blocked by selective antagonists at the NMDA receptor like D-2-amino-5phosphonovalerate (D-APV). However, Harris and Cotman (1986) found that D-APV did not inhibit long-term potentiation in the mossy fibre-CA3 system. Furthermore, Satoh, Ishihara and Katsuki (1988) demonstrated that piracetam and bifemelane which have anti-amnesic action augmented long-term potentiation in the mossy fibre-CA3 system but not that in the Schaffer collateral-CA1 system. Thus, different susceptibilities to some drugs were shown on long-term potentiations in CA1 and CA3 regions. It is known that the pyramidal neurones of the CA3 region receive at least three different inputs, the mossy, commissural/associational and fimbrial fibres (Brodal, 1981). In each of the input systems, longterm potentiation was demonstrated in slice preparations (Yamamoto and Chujo, 1978; Yamamoto and Sawada, 1981; Harris and Cotman, 1986). Harris and Cotman (1986) demonstrated that D-APV *To whom correspondence
receptor, hippocampal slice,
inhibited long-term potentiation in the commissural/ associational fibre-CA3 system but not in the mossy fibre-CA3 one. These observations prompted an investigation of the effects of several drugs, classified into different categories, such as D-APV, naloxone (an opioid antagonist) and bifemelane on long-term potentiations in the three input systems to the CA3 region. METHODS Slices of hippocampus, 400-500 pm thick were prepared from male guinea pigs (200-350 g) either manually or using a microslicer (DTKlOOO, Dosaka E. M. Kyoto, Japan), as described previously (Iwama, Ishihara, Satoh and Takagi, 1986). Slices were incubated for 1 hr at 34-35°C in artificial cerebrospinal fluid (CSF) of the following composition (in mM): NaCl, 124; KCl, 5; KH,PO,, I .24; MgSO,, 1.3; CaCl,, 2.4; NaHCO,, 26; D-glucose, 10; bubbled with 95%0, + 5%CO,. A single slice was transferred into a recording chamber, having a volume of approximately 0.8 ml and perfusion was carried out at a rate of 1.5-2.0 ml/min with the artificial CSF at 34-35°C. The field potential (population spike) was recorded in the pyramidal layer of the CA3 region, with a glass pipette of about 10 pm tip diameter, filled with 0.9% NaCl (< IO Ma). A bipolar stimulating electrode was placed in the granule cell layer of the
should be addressed. 487
K.
488
ISHIHARA et al.
Stim. (com./ana.)
Fig. 1. Schematic drawing of the hippocampal slice of the guinea pig. The recording electrode (Rec.) was placed in the pyramidal layer of the CA3 region to record the population spike. The stimulating electrode (Stim.) was positioned in the granule cell layer of the dentate gyrus to stimulate the mossy fibres (mossy),
in the stratum radiatum, near the CA2 region, to stimulate the commissural/associational fibres (com./asso.) or in the fimbria to stimulate the fimbria fibres (fimbria). dentate gyrus, in the stratum radiatum near the CA2 region or in the fimbria for stimulation of the mossy fibres, commissural/associational fibres or fimbrial fibres, respectively (Fig. 1). Rectangular pulses (0.1 msec in duration) were delivered at 0.2 Hz through the respective electrodes, as test stimulation, with a strength (up to 10 V) that gave 70-90% of the maximum response. Population spikes, induced by 10 successive stimulations, were averaged and recorded every 5 min. Three or 4 averaged population spikes were taken before tetanic stimulation and the mean amplitude was used as a basal value for calculating the increase of the population spike. The amplitude of the population spike was measured from the negative to positive peak, as the most accessible measure of the global effects of drugs on the response. tetanic stimulation, with the same strength and pulse duration as those of test stimulation, was applied at 33 Hz for 5 set, using the same stimulating electrode in the respective systems. In some experiments, one second instead of 5 set was used as a duration of tetanic stimulation of commissural/associational or fimbrial pathways. Thereafter, the stimulus frequency reverted to 0.2 Hz and the averaged population spikes were again recorded every 5 min for 1 hr or more. Long-term potentiation was produced in one of the three input systems, in each preparation and was expressed as the percentage increase in the amplitude of averaged population spikes after tetanic stimulation over the mean amplitude before tetanic stimulation. The drugs used were D-APV (Funakoshi), kynurenate (Nacalai tesque), naloxone HCl (a gift from
E. I. DuPont, Garden City, New York, U.S.A.) and bifemelane HCl (a gift from Mitsubishi Chemical Ind. Ltd, Yokohama, Japan). Kynurenate was dissolved in weak NaOH solution and adjusted pH to 7-8 with HCl solution and other drugs were dissolved in distilled water at 1000-fold concentration of the final one used. Those were added to artificial CSF and perfused over the slice for 15 min in the case of bifemelane on mossy fibreCA3 system and 25 min in other cases. Tetanic stimulation was applied at 11 or 21 min after the start of the perfusion of drug. Then, the application of drug was stopped 4min after the tetanic stimulation. The magnitude of long-term potentiation was expressed by calculating the area between the zero line and the curve of the time course of long-term potentiation, from 12 to 62 min after the tetanic stimulation. An example is shown in Fig. 2, in which the area of the non-drug treated group is shown by the dotted area (A) and that of the kynurenate-treated group by the hatched area (B). Effects of drugs were determined by comparison between these areas. In the present experiments, concentrations of the drugs tested were chosen such that population spikes in the absence of tetanic stimulation were not significantly affected in any of the input systems examined. Statistical significance of differences were calculated according to Student’s t-test. RESULTS A single shock to the mossy fibres or fimbria induced, in the CA3 pyramidal layer, a biphasic field
Long-term potentiation
in CA3 region
489
Fig. 2. Time course of long-term potentiation in the non-drug treated group or kynurenate (10m6M)treated group. Ordinate: percentage increase in the amplitude of the population spike (PSA). Abscissa: time after tetanic stimulation (TS) in min (minus sign: time before tetanic stimulation). 0-0 and 0-O indicate non-drug treated group (n = 7) and kynurenate (10e6M) treated group (n = 3), respectively. Each point represents the mean with SEM. Kynurenate was perfused between -2l-4min, shown with open column. The magnitude of long-term potentiation was estimated by calculating the area between zero line and each curve of the time course of the long-term potentiation, from 12 to 62 min after the tetanic stimulation. For example, the non-drug treated group is represented by the dotted area (A) and kynurenate (10e6M) treatment group by the hatched area (B). *P < 0.05, **P < 0.01, Statistically different from non-drug treated group (Student’s r-test). PSA = Population spike amplitude. TS = Tetanic stimulation.
potential starting with a negative phase, the latency of which was 3.5-7 msec or 3.5-6 msec, respectively. Test stimulation of the commissural/associational fibres elicited a field potential, consisting of a negative deflection (latency was less than 2 msec), followed by a biphasic field potential with latency of 4-7 msec. Tetanic stimulation of the mossy, commissural/ associational or fimbrial fibres induced long-term potentiation of the population spike in the CA3 region of the hippocampus. When no drug was added to the artificial CSF, the percentage increase in amplitude of the test responses (population spikes) after tetanic stimulation was 35-45% (n = 7), 200-225% (n = 3) or 225-280% (n = 3) of the mean amplitude before tetanic stimulation in the mossy fibre, commissural/associational fibre or fimbrial pathway, respectively. Each increase persisted for 1 hr or more after tetanic stimulation. Examples of the long-term potentiation in the fimbrial fibre-CA3 system are shown in Fig. 3. The results of drugactions on long-term potentiation in the three input systems to the CA3 region are summarized in Table 1.
of 10m5M, was 0.06 times as large as that in the non-drug treated group (Table 1). An example of population spikes which failed to show long-term potentiation in the fimbrial pathway in a D-APVtreated slice, is shown in Fig. 3. On the other hand, long-term potentiation in the mossy fibre-CA3 system was resistant to D-APV: the means of the magnitudes of long-term potentiation in the presence of 10-j M and 10e4 M of D-APV were 0.94 and 0.92 times, respectively, as large as that in the non-drug treated group. However, kynurenate, a non-selective glutamate receptor antagonist, significantly suppressed the long-term potentiation of the mossy fibre-CA3 system, at a concentration of 10m6M (Fig. 2). The mean of magnitude of long-term potentiation in kynurenate-treated slices, was 0.23 times as large as that in the non-drug treated group (Table 1). Both D-APV and kynurenate, at the concentrations used, did not significantly change the amplitude of the population spike before the tetanic stimulation, in any of the input systems examined.
Effect of D-APV on long-term potentiation
In the commissural/associational or fimbrial pathway to the CA3 region, long-term potentiation was not significantly affected by naloxone, at a concentration of 10m6M: the mean of magnitudes of longterm potentiation was 1.02 or 1.23 times as large as that in the non-drug treated group, respectively (Table 1). An example of long-term potentiation in the fimbrial fibre-CA3 system in the naloxone-treated slice is shown in Fig. 3. In the mossy fibre pathway to the CA3 region, long-term potentiation was
Long-term potentiation in the commissural/ associational fibreCA3 system was significantly suppressed by D-APV: the mean of the magnitudes of long-term potentiation, at a concentration of 10S5 M, was 0.10 times as large as that in the non-drug treated group (Table 1). o-2-Amino-5-phosphonovalerate also significantly suppressed long-term potentiation in the fimbrial fibre-CA3 system: the mean of magnitudes of long-term potentiation, at the concentration
Efect of naloxone on long-term potentiation
K.
490 Before
No
TS
27min
after
ISHIHARA et al.
TS
drug
D-APV
No loxone
A-.
.r‘_ I
Fig. 3. Examples of long-term potentiation of the population spike, recorded in the pyramidal layer of CA3 region of slices of hippocampus, by simulating fimbrial fibres. Upper traces: non-drug-treated slice. Middle traces: D-APV (IOe5M)-treated slice. Lower traces: naloxone (10e6M)treated slice. Ten responses were averaged in each trace. Test stimulation was delivered at each dot. Each left and right traces are the responses before tetanic stimulation (TS) and 27 min after tetanic stimulation (TS) in the same slice. Calibration: 1mV; 10 msec. significantly suppressed by naloxone at 10e6 M: the mean of magnitudes of long-term potentiation was 0.31 times as large as that in the non-drug treated group (Table 1). Naloxone, at 10d6 M, did not significantly alter the amplitude of the population spikes, before tetanic stimulation, in any of the input systems to the CA3 region. EfSect of bifemelane Bifemelane in the mossy Table
on long-term
potentiation
augmented the long-term potentiation fibre-CA3 system, as described before
I. Effects of o-APV
(IO ~J M), kynurenate (10m6 M), naloxone mossv fibre-. commissuraliassociational
Mossy
DISCUSSION
In the present experiments, the effects of drugs tested on long-term potentiation of population spikes in the three input systems to the hippocampal CA3 region, were produced with no significant effects on the population spikes, in the absence of tetanic stimulation. This would suggest that those drugs specifically act on mechanisms causing long-term potentiation at the concentrations used. Harris and Cotman (1986) reported that D-APV blocked long-term potentiation in the commissural/ associational fibreXA3 system but not that in the mossy fibreCA3 system, using hippocampal slices of the guinea pig. Their observation has been confirmed (10m6 M) and bifemelane (10m6 M) on long-term fibre- and fimbrial-CA3 systems
33 Hz, 5sec
D-APV
1917 f 238” (7) I799 + 239 (3)
10708 f 1734(3) 1056 + 488 “(3)
Kvnurenate Naloxone Bifemelane
443 + I I4 “(3) 589 ; 343 “i4j 5992 + 722 x.$(4)
10896 + 1762 (3) 12085 f 4352 (3)
33 Hz, I set 2533 f 446 (3)
21821375
potentiation
in the
Fimbrial
Commissural/associational
33 Hz, 5 set No drug
(Satoh et al., 1988): the mean of magnitudes of long-term potentiation at 10m6 M was 3.13 times larger than that in the non-drug treated group (Table 1). In contrast, bifemelane did not augment the long-term potentiation of the commissural/ associational or fimbrial pathway to the CA3 region at 10m6M: the mean of magnitudes of long-term potentiation was 1.13 or 0.90 times as large as that in the non-drug treated group, respectively (Table 1). In the commissural/associational fibre- or fimbrial fibreCA3 systems, degrees of the long-term potentiation in non-drug treated slices were large, when 33 Hz for 5 set, standard parameters in this study, were used as tetanic stimulation. Since this might contribute to the failure of an augmenting effect of bifemelane, in some experiments 1 set, instead of 5 set, was used as the duration of tetanic stimulation in these systems. The increase in amplitude of the test response after tetanic stimulation was 40-60% (n = 3) for commissural/ associational or 45-85% (n = 4) for fimbrial stimulation in the non-drug treated group and persisted for 1 hr or more. Even under such a condition, the long-term potentiations in the commissural/ associational fibre- and fimbrial fibreCA3 systems were not affected by bifemelane at 10e6 M: the mean of magnitude of long-term potentiation were 0.86 and 1.12 times as large as those in the non-drug treated group, respectively. Bifemelane, at the concentration used, did not significantly affect the amplitude of population spike before tetanic stimulation in any of the input systems to the CA3 region.
(3)
33 Hz, 5 set
33 Hz, I set
12906*812(3) 753 ?r 576 **(3)
2827 k 784 (4) -
I5811 +2037(3) II588 f 849 (3)
3154+329(3)
YMean f SEM of magnitude of long-term potentiation (area under the graph in arbitrary units). Number in parenthesis of slices tested. -Not tested. **Significant inhibition (P < O.Ol), compared with no-drug group, described in the top of each column. f$Significant augmentation (P < 0.01). compared with no-drug group, described in the top of each column
-
represents the number
Long-term potentiation in CA3 region and further evidence provided for different drugsusceptibilities of long-term potentiations in the three separate input systems to the CA3 region. The drug D-APV (lo-’ M) also inhibited long-term potentiation in the fimbrial fibre-CA3 system, which is suggested to be mediated by NMDA receptors in a similar manner to that in commissural/associational fibre-CA3 system. In the CA3 region, mossy fibres terminate in the stratum lucidum, while commissural/ associational and fimbrial fibres terminate in the stratum radiatum and stratum oriens (Raisman, Cowan and Powell, 1965). Monaghan, Yao and Cotman (1985) reported that the density of NMDA receptors was low and that of kainate receptors was high in the stratum lucidum of the CA3 region but the densities of quisqualate and NMDA receptors were high in the stratum radiatum and stratum oriens. Long-term potentiation in the mossy fibre-CA3 system is probably not mediated by NMDA receptors but by non-NMDA receptors for L-glutamate, since a non-selective antagonist at the excitatory amino acid receptors, kynurenate, suppressed this long-term potentiation, in particular a late phase of it, at a concentration of 10e6M. This concentration of kynurenate is much less than the effective one (2 x or 4 x 10e4 M) for reducing the amplitude of excitatory postsynaptic potentials in the CA3 pyramidal neurone evoked by low frequency (without tetanic) stimulation of the mossy fibres in similar preparations to those employed in this study (Cotman, Flatman, Ganong and Perkins, 1986). In fact, no significant effect of 10e6 M of kynurenate on population spikes of the CA3 region evoked by low frequency stimulation of mossy fibres was observed in the absence of tetanic stimulation. These findings suggest that the mechanism(s) of production (and/or maintenance) of long-term potentiation of the population spike, following tetanic stimulation in the mossy fibre-CA3 system, are kynurenate- sensitive and may be mediated by non-NMDA receptors for glutamate. By contrast, naloxone (10m6M) blocked long-term potentiation in the mossy fibre-CA3 system. This finding is consistent with that previously reported by Martin (1983). Further, it was demonstrated that naloxone did not affect long-term potentiations in the other two input systems to the CA3 region. It was revealed, by the immunohistochemical technique, that hippocampal mossy fibres are dense with dynorphin A, an opioid peptide (Khachaturian, Watson, Lewis, Coy, Goldstein and Akil, 1982). Further, Satoh, Ishihara, Katsuki and Sugimura (1989) showed that the opioid peptide, dynorphin A (lo-’ and lo-‘M), augmented long-term potentiation in the mossy fibre-CA3 system and such an augmenting effect was blocked by naloxone (lo-’ M). These findings suggest an involvement of an opioidergic system in the production of long-term potentiation in the mossy fibre-CA3 system, but not in commissural/associational fibre- and fimbrial fibre-CA3
491
systems. On the other hand, bifemelane (10m6M) significantly augmented long-term potentiation in the mossy fibre-CA3 system but not in the other two input systems to the CA3 region. Long-term potentiation in the Schaffer collateral-CA1 system was previously shown not to be affected by the drug (Satoh et al., 1988). From the above findings, long-term potentiations in commissural/associational fibre- and fimbrial fibre-CA3 systems are similar to that in the Schaffer collateral-CA1 system. In this context, Ito, Okada and Sugiyama (1988) reported that intracerebroventricular injections of pertussis toxin blocked long-term potentiation in the mossy fibre-CA3 system but not in Schaffer collateral-CAl. However, Goh and Pennefather (1989) clearly demonstrated an inhibition by the same toxin of long-term potentiation, even in the latter system and pointed out that the apparently disparate effects of pertussis toxin, reported by Ito et al. (1988), were probably due to insufficient penetration of the toxin to the hippocampal CA1 region. It is of interest that the pharmacological characteristics of long-term potentiation in the mossy fibreCA3 system are in marked contrast to those in commissural/associational fibre- and fimbrial fibreCA3 systems: the former was not changed by D-APV but affected by naloxone and bifemelane, while the latter were suppressed by D-APV but not affected by the latter two drugs. From these findings, it is conceivable that even in the hippocampal CA3 region,
long-term potentiations, in three input systems, may be mediated by different mechanisms. The detailed mechanisms of long-term potentiations in the three input systems, including an involvement of pertussis toxin-sensitive guanine nucleotide binding regulatory proteins, remain to be revealed. REFERENCES Brodal A. (1981) The olfactory pathways. The amygdala. The hippocampus. The ‘Limbic system’. In: Neurological Anatomy in Relation to Clinical Medicine, 3rd edn, pp. 640-697. Oxford, New York. Collingridge G. L., Kehl S. J. and McLennan H. (1983) Excitatory amino acids in synaptic transmission in the Schaffer collateral-commissural pathway of the rat hippocampus. J. Physiol., Lond. 334: 33-46. Cotman C. W., Flatman J. A., Ganong A. H. and Perkins M. N. (1986) Effects of excitatory amino acids antagonists on evoked and spontaneous excitatory potentials in guinea-pig hippocampus. J. Physiol., Lo&. j78: 403-415. Goh J. W. and Pennefather P. S. (1989) oertussis toxin. ,A_ sensitive G protein in hippocampal long-term potentiation. Science 244: 980-983. Harris E. W. and Cotman C. W. (1986) Long-term potentiation of guinea pig mossy fiber responses is not blocked by N-methyl o-aspartate antagonists. Neurosci. Lett. 70: 132-137. Ito I., Okada D. and Sugiyama H. (1988) Pertussis toxin suppresses long-term potentiation of hippocampal mossy fiber synapses. Neurosci. Lett. 90: 181-185. Iwama T., Ishihara K., Satoh M. and Takagi H. (1986) Different effect of dynorphin A on in vitro guinea pig hippocampal CA3 pyramidal cells with various degrees of paired-pulse facilitation. Neurosci. Lett. 63: 190-194.
492
K.
ISHIHARA et al.
Khachaturian H., Watson S. J., Lewis M. E., Coy D., Goldstein A. and Akil A. (1982) Dynorphin immunocytochemistry in the rat central nervous system. Peptides 3: 941-954.
Martin M. R. (1983) Naloxone and long-term potentiation of hippocampal CA3 field potentials in oitro. Neuropeptides 4: 45-50.
Monaghan D. T., Yao D. and Cotman C. W. (1985) L-[‘HJGlutamate binds to kainate-, NMDA- and AMPAsensitive binding sites: an autoradiographic analysis. Brain Res. 340: 378-383.
Morris R. G. M., Anderson E., Lynch G. S. and Baudry M. (1986) Selective impairment of learning and blockade of long-term potentiation by an N-methyl-oaspartate receptor antagonist, AP5. Nature 319: 774-776.
Raisman G., Cowan W. M. and Powell T. P. S. (1965) The extrinsic afferent, commissural and association fibers of the hippocampus. Brain 88: 963-996. Satoh M., Ishihara K. and Katsuki H. (1988) Different susceptibilities of long-term potentiations in CA3 and CA1 regions of guinea pig hippocampal slices to nootropic drugs. Nkosci. Lett. 93:.236-241. Satoh M.. Ishihara K.. Katsuki H. and Suaimura M. (1989) Opioidkrgic modulation of long-term pgtentiation in the guinea pig hippocampus in vitro. Adv. Biosci. 75: 193-196. Yamamoto C. and Chujo T. (1978) Long-term potentiation in thin hippocampal sections studied by intracellular and extracellular recordings. Expl Neural. 58: 242-250. Yamamoto C. and Sawada S. (1981) Important factors in induction of long-term potentiation in thin hippocampal sections. Expl Neurol. 74: 122-130.