- Email: [email protected]
Kainate binding sites in the hippocampal mossy fibers: Localization and plasticity
Neuroscience Vol. 20, No. Printed in Great Britain
3, pp. 739-748,
1987
0306-4522/87 $3.00 + 0.00 Pergamon Journals Ltd 0 1987 IBRO
KAINATE BINDING SITES IN THE HIPPOCAMPAL MOSSY FIBERS: LOCALIZATION AND PLASTICITY A. REPRESA,* E. TREMBLAYand Y. BEN-ARI Laboratoire de Neurobiologie et Neurophysiopathologie du Developpement, 123 Bd de Port-Royal, 75674 Paris Cedex 14, France and LPNl, CNRS, Gif-sur-Yvette 91190, France
INSERM,
U29,
Abstract-The
regional distribution of high affinity binding sites for kainic acid has been determined in rat hippocampi by quantitative autoradiography. Selective lesions were made in order to determine the exact localization of these sites in the mossy fiber system, and to evaluate whether the sprouting and
synaptic reorgani~tion of the mossy fibers are associated with alterations in the dis~bution of these binding sites. The results show that kainate binding sites in the stratum lucidum are more vulnerable to destruction of the granules and their mossy fibers by in~~ipp~mpal colchicine injections than to destruction of the CA31CA4 uvramidal cells bv injection of kainate into the amyadala. This suggests that a substantial
proportibn of -&ekainate bind&g sites is associated with the m&iy fiber termi& (i.e. the presynaptic elements). Furthermore, in keeping with an earlier study, destruction of the pyramidal neurons of CA3 by intracerebral kainate produced a dark Timm positive band in the supragranular zone which is due to the sprouting of mossy fibers. This was associated with an increase in the density of kainate binding sites, which further stresses the parallelism between the distribution of these sites and mossy fiber terminals.
EXPERIMENTAL PROCEDURES mossy fibers (MF) which originate from the dentate granule cells massively project upon Male Wistar rats (22&230 g) were used throughout these the pyramidal neurons of the CA3 region; MF termiexperiments. They had access to food and water ad kibitz nals also innervate cells in the hiiar region and the and were housed in individual cages under diurnal ii~ting conditions, with lights on from 0800 to 2000 h. For lesion supragranular layer of the fascia dentata.10~“~‘*~2’34.33 experiments, animals were anaesthetized with Quithesin The CA3jH3 pyramidal neurons innervated by the [Jensen Salsbury, 3 ml/kg) and submitted to unilateral injecMF are particularly vulnerable to the epilepsies in tions of colchicine in the hippocampus or KA into the man, as well as in the seizures produced, in rats and amygdala. All injections were performed with a microprimates, by administration of kainic acid (KA), the syringe under stereotaxic guidance. With colchicine, two potent excitatory analogue of glutamate (see Ref. 3 injections of 1.5 pg of colchicine (dissolved in 0.6~1 of physiological saline, pH 7.2) were made in the same hemifor a review). Specific binding sites for KA are sphere at the following coordinates according to the Atlas present in the CNS of various species, notably in the of Albe-Fessard;’ A:2.9; L:3.3; H:6.4 and A: 5.4; L: 1.3 and H : 7.0. This procedure was previously found to destroy the hippocampus. 23Autoradiographic studies performed granule cells and their processes within a few days.19 Injecin both rats6,*‘~‘*and humans” indicate that their tion of KA into the amygdala (I.2 pg in 0.3 ~1 of phosphate dist~but~on, in the hippocampus, coincides with the buffer, pH ‘7.4)was made at the following coordinates; A : 6; MF terminaIs, being particularly enriched in the L:4 and H: 2. Animals showing typical KA-induced limbic stratum lucidurn. Ontogenetic studies in rats have motor seizures for at least 2 h upon recovery from anaec shown a good parallelism between the occurrence of thesis were kept for the experiments; these received diazethe seizure/brain damage syndrome produced by pam (Roche, 20mg/kg i.p.) to stop the seizures; such treatment was previously found to produce a seizure-related parenteral KA,*‘s’~the maturation of the MF2 and the lesion of the pyramidal cells in CA3 of the ipsilateral late occurrence of high affinity KA binding sites in hemisphere.4,5 Six or eight days after colchicine or 3-30 days after KA the hippocampus.* These, as well as other observations,‘6’6 suggest that an endogenous EGA-like injections, the animals were anaesthetized with pentobarbital and a modified Timm’s procedure’* was peragent is released from the MF. formed to stain the mossy fibers20 The rats were intraThe aim of the present study was to substantiate cardially perfused first with phosphate buffer containing further the relationship between MF and KA binding 0.37% sodium sulphide (2OOml), then with 0.1% paraformaldehyde (200 ml). This procedure does not affect the sites. First, using relatively specific lesions, we exambinding experiments (see Results). After perfusion, brains ined whether the KA sites are pre- or ~stsynaptically were removed and rapidly frozen in isopentane ( -50°C). located in the MF terminal zone. Secondly, since the Coronal sections (20pm) of the dorsal hip~mpus were MF have been shown to sprout in a variety of cut with a cryostat (-15°C) and thaw-mounted onto expe~men~a~17.2i.27,~.~ gelatin-coated slides for binding experiments. The histoand probably also pathological logical procedures (Timm or Nissl) were made on the same conditions, we examined whether this synaptic reoror alternate sections. ganization is associated with the appearance of KA Visualization of the KA binding sites was done using an binding sites in the reinnervated regions. autoradiographic procedure which enables the high
The hippocampal
*To whom correspondence should be addressed. Abbreuialions: KA, kainic acid; MF, mossy fibers.
affinity-slowly dissociating-KA sites to be preferentially labelled.6~22The sections were preincubated in Tris-acetate buffer (50 mM, pH 7.1) for I h at 3”C, then for 15 min at 739
740
A.
bZT.ESA
30°C to remove
competing endogenous ligands from the tissue. The sections were then incubated at 3°C for 90 min with the same buffer in the presence of 20 nm [VinylidenerH]KA (NEN, 60Ci/mmol). After incubation, the low affinity sites with a fast dissociation rate were rapidly displaced by rinsing the sections for 2 min in the buffer containing 10 PM unlabelled KA. Non-specific binding was routinely evaluated by incubating alternate sections with the radioligand plus an excess of IOpM cold KA. After the procedure, the sections were dried and exposed to 3H-sensitive Ultrofilms (LKB) for 30 davs. simultaneouslv with internal standards. huantification of the density of the binding sites was performed using an image analyser (Imanco Quantimet 720) and relying on the comparison with the optical density of internal standards. At least four sections were used to determine the density of binding sites in each area. Statistical analysis was performed using the Student’s t-test. For Scatchard analysis, hippocampal sections were submitted to the same procedure except that incubation was performed with various concentrations of the radioligand (I-SOOnM); after rinsing, the slices were wiped off on Whatman filters, dried and put in a vial with IO ml of Aquasol, and the radioactivity was counted with a liquid scintillation counter. RESULTS
Eflect of perfusion/fiwation on kainic acid binding Since perfusion/fixation according to Timm’s method was needed to determine precisely the extent of the lesions, we have made preliminary experiments to verify that this procedure does not modify KA binding. The Kd or B,,,,, values evaluated from Scatchard analysis were similar in hippocampal sections from non-perfused (n = 3; 19 + 12 and 40 + 12 fmol/mg protein, respectively, and perfused rats (n = 4; 24 + 9 and 36 + 6 fmol/mg protein, respectively). Furthermore, in autoradiographs from control non-perfused rats (n = 3), the densities of sites measured in CA3a and the supragranular layer (222 and 121 fmol/mg protein, respectively) were similar to those found with perfused rats (Table 1). Decreased binding sites after colchicine injections or kainic acid treatment The autoradiographs presented in Figs 1, 2 and 3 and quantitated in Table 1 illustrate that KA labelling in the stratum lucidum is reduced more extensively and rapidly following unilateral colchicine inTable
rf
(11.
jections in the hippocampus than unilateral injection of KA into the amygdala. Thus, 6 days after colchicine, almost the entire granule-MF system was destroyed, only restricted spots of the Timm’s deposits still being visible (not illustrated). At this delay, the loss of KA binding sites was conspicuous and almost maximal (Table I). Two days later, the complete disappearance of Timm’s staining (Fig. 2) was associated with a further small reduction of the density of sites in CA3a and the supragranular zone (Table 1). In keeping with earlier studies,” the CA3 pyramidal neurons were relatively spared by the procedure (Fig. 2). In spite of some variability, the gradient of destruction produced by colchicine was: granules of the fascia dentata>> CA1 >>CA3. Clearly, the restricted damage of the CA3 neurons cannot explain the colchicine-induced loss of sites since, after KA and similar short delays of survival, the complete destruction of the CA3 neurons was not accompanied by a significant change in binding sites (see below). In keeping with our earlier observations,4x5 KA injection into amygdala produced a selective and rapid loss of pyramidal neurons of CA3 in the ipsilateral Ammon’s horn. This was already present 3 days after the injection, notably in CA3a, which is the most vulnerable brain region to KA.’ 5 At longer delays (8 days or more, cf. Fig. 3A and B), the entire CA3 region was largely destroyed. In contrast to the postsynaptic elements, the presynaptic ones (i.e. the MF terminals) were largely intact since Timm’s stain was present 3-30 days after KA (Fig 3D). The main difference with the control cases concerned the organization of the dark silver deposits which were not aligned vertically to the pyramidal layer (Fig. 3D), as they usually are (see in Fig. 2, the contralateral side). This is due to the destruction of the dendrites of the CA3 pyramidal neurons. Both visual inspection of the KA autoradiographs and quantitative densitometry showed the lack of a significant change in KA binding sites in KA-treated cases with 3-8 day survival periods (Fig. 1, Table 1) i.e. at a time when the pyramidal neurons of CA3 were already extensively destroyed (Fig. 3). With longer survival times (12-30 days), there was a
1. Effects of kainic acid and colchicine treatments on the densities in the hippocampus
Treatment
n
CA3c
Control KA (3 d) KA (8 d) KA (12 d) KA (30 d) Colchicine (6 d) Colchicine (8 d)
4 2 2 2 4 2 2
IL3 +9 121 * 10 115+ 11 109 + 14 94* 5* 70*21** 70+8**
CA3a 226 + 225 * 180 f 161 + 119+ 141 + 135 +
of kainic acid binding CA1
5 5 13 2** 11** 23* 13**
40 * 36 + 50+ 31*7 44* 5s* 53 *
sites
sg 5 3 10 10 17 3
124+5 126 * 10 123+ 11 I51 +4* 165 k 6** 64* 11** 57 If- 10**
Mean densities of KA binding sites (in fmoljmg protein) + SEM, in various subfields, in controls and treated cases. The survival time in.,days (d) and the number of rats (n) are indicated. CA3a and CA3c, stratum lucidum of the CA3a and CA3c regions; sg, supragranular layer of the fascia dentata. *.** Values statistically different from control ones for P < 00.5 and 00. I, respectively (Student’s test).
Kainate
binding
sites in the mossy
741
fibers
Control
KA(8d)
KA(30d)
Colch. (8d) Fig. 1. Autoradiographs
depicting
the distribution of high affinity KA binding sites in the hippocampus (Colch.) treated cases. Dark triangles indicate the side ipsilateral to injection. In perfused controls, the KA labelling is also oonfined to the supragranular layer of the fascia dentata (FD) and the stratum lucidum of the CA3 region. Note the progressive loss of labelling from the stratum lucidum after KA and the extensive and rapid loss after colchicine. (d) Survival delay in days.
in (perfused) controls, KA or colchicine
progressive loss of KA sites in the stratum lucidum on the side of injection. This involved first the CA3a region, where a significant 29% reduction was already present with 12 days survival; at longer delay (30 days), the loss in CA3 was more conspicuous (Table 1). This delayed loss of sites (see also Ref.7) probably reflects a protracted clearance of the damaged CA3 dendrites (see Discussion). As indicated in Table 1, the densities of sites in the other hippocampal regions are not altered (i.e. in CAl) or even enhanced (see below). Increased kainic acid binding sites In addition to the reduction in KA binding sites described above, KA treatment produced a rise in
binding sites in the ipsilateral supragranular layer of the fascia dentata (Table 1) and in the contralateral side in both the supragranular and infrapyramidal zones. Quantitative autoradiography and systematic comparison with the Timm’s stained sections clearly indicated that this rise is associated with the synaptic rearrangements known to occur in these conditions. Thus, in keeping with the recent studies of Nadler et a1.,27 with long survival periods, intracerebral KA produced a conspicuous band of dark deposits in the supragranular layer (Fig. 4A and C). This corresponds to the autoinnervation of the granules which is observed after a variety of lesion-induced deafferentation of the granules.‘7.2’,27 In the four cases
A. REPRESAet al
742
in which a rise in Timm’s stain was found in the ipsilateral supragranular layer, there was also a significant increase in KA binding sites (Table I). In all the other cases (with shorter survival delays), neither Timm’s staining nor KA binding sites were enhanced. Furthermore, enhanced Timm’s staining was found in one case (30 days survival) in the contralateral supragranular layer (not illustrated) and the infrapyramidal CA3 band (Fig. 5A). This enhancement, which probably reflects a sprouting of the MF (see Discussion), was associated with increased densities of KA binding sites in these layers (164 and 269 fmol/mg protein, respectively; also see Fig. 5B). This further stresses the parallelism between Timm’s staining of MF terminals and KA binding sites. Enhanced KA labelling was also found in the contralateral hippocampus after colchicine treatment. The significance of this increase is not clear since the increase (corresponding to approximately 27% of the control values) affected all hippocampal subfields and was already present 6 days after the injections. This increased KA labelling is therefore unlikely to reflect a selective reinnervation process. DISCUSSION
In their earlier study, Monaghan and Cotman, using intrahippocampal colchicine or ICV injections of KA, concluded that both treatments produce a similar loss of KA binding sites in the stratum lucidum. These authors, however, used only one survival delay (30 days) and did not perform quantitative densitometry. In the present study, using quantitative densitometry, we found that the loss of KA sites after colchicine was conspicuous even within 6 days and concerned the entire extent of the regions innervated by the MF (i.e. the stratum lucidum and supragranular layer of the fascia dentata). In contrast, after KA injection, the reduction of sites was more delayed and only involved the stratum lucidurn. Histological examinations of the same or alternate sections from
which the autoradiographs were obtained showed that colchicine injections in the hippocampus rapidly destroyed the MF system with relatively restricted effects on the pyramidal neurons,” whereas injections of KA into the amygdala rapidly produced a virtually complete neuronal loss in CA3, with restricted effects on the MF system,4*5 as assessed with both Nissl’s and Timm’s stains. These results, in keeping with the excellent parallelism already noted between the distribution and development of the MF and KA binding sites (see Introduction), suggest that a large proportion of these sites are located on the MF. A presynaptic localization of the KA binding sites was also suggested in earlier biochemical studies in which the toxin produces a release of endogenous excitatory amino acids in the hippocampus;i5 electrophysiological studies also suggest a presynaptic localization of these sites.16 However, a partial localization of KA sites on the postsynaptic elements cannot be excluded. As a matter of fact, even with a virtual complete destruction of the MF by colchicine, a significant number of binding sites remained in this region. It is tempting to suggest that these are located on the pyramidal neurons, in keeping with other neurotransmitter systems in which receptors are both pre- and postsynaptically located. The time course of events noted after KA is in keeping with this hypothesis. A recent biochemical study performed following intraamygdaloid injection of KA also showed a delayed and progressive decline in KA binding in membranes prepared from the microdissected CA3 region;’ this was due to a progressive decline of B,,,,, without a change in the affinity constant.’ This is important since seizures, which are not associated with brain damage (kindling), can also produce a reduction of sites for KA in the hippocampus3’ The progressive decline of binding sites might be due to the slow phagocytosis of intact synaptic complexes, i.e. including fragments of the postsynaptic membrane remaining attached to the presynaptic bouton. Such complexes have been found,
Fig. 2. Photomicrographs depicting the changes produced by injections of cochicine in the hippocampus. (A) and (B) Sections from the same case, ipsilaterally (A) and contralaterally (B) to injection, treated for Timm’s stain and counterstained with Nissl’s stain. Note the loss of granule cells (g) in the injected side (dark triangles); in contrast, the pyramidal cells in CA3 are largely present. Also note in (B) the normal radial disposition of the silver deposits in the stratum lucidum (Lu) as revealed by Timm’s stain, and in (A) the complete loss of staining. Ra, stratum radiatum. The boxed area illustrates the KA autoradiograph obtained from the same case: note the extensive loss of labelling in the side of injection. Fig. 3. Photomicrographs depicting the histopathological changes in hippocampus 8 (A) and 30 (B) days after KA injection into amygdala. There is a loss of pyramidal neurons associated with gliosis in the inferior Ammon’s horn up to the CA3/CA2 boundary zone (arrow); the granule cells (g) are intact (Nissl stain). (C) KA autoradiograph obtained from the same case as in (B); note the lower labelling in the stratum lucidum of the lesioned side (right part of the picture). In (D), an alternate section from the same case stained with Timm’s method shows packing of silver deposits in the stratum lucidum (cf. Fig. 3C). Fig. 4. Photomicrographs illustrating the sprouting of mossy fibers at long delays (30 days) after KA treatment. Note in (A) the intense Timm’s staining ipsilateral to injection, in the supragranular zone (indicated by arrows); the rectangle in (A) is enlarged in (C). (B) and (D) Control case illustrating the lack of intense Timm’s staining in the same layer.
e
744
745
A.
REPRESA et al.
Fig. 5. Photomicrograph illustrating the associated occurrence of Timm’s staining (A) and KA iabelling (B) in the infrapyramidal layer (arrow). Both sections are from the same case, contralateral to the KA-treated side (survival time = 30 days).
for instance, in the striatum even after destruction, by local KA, of intrinsic neurons; this may account for the progressive disappearance of KA or other binding sites from the striatum.9.‘3.29 To the best of our knowledge, the present study shows for the first time that a synaptic rearrangement, which occurs in the hippocampus following lesions, is associated with an increase of binding sites for an excitatory amino acid (but see also Ref. Isa). The ability of the MF to sprout following a variety of experimental conditions has been clearly established both with electron microcopy and Timm’s staining; this is particularly conspicuous in the supragranular layer of the fascia dentata foliowing suppression of commissural, associational and/or entorhinal afferents to the granules.2’~27 Using a combined Golgi/electron microscopic approach, Frotscher and Zimmer” recently provided compelling evidence that this is due to an autoinnervation of the granules by the MF. Aberrant sprouting of the
MF has also been described in other conditions including intracerebral grafts on hippocampal neuronsM interestingly, after ICV” or intra-amygdaloid KA (this study), the destruction of the pyramidal cells and their associative projections to the granules also produce an aberrant autoinnervation of the MF in the supragranular zone. It is not clear whether the KA binding sites occurring in these conditions in the supragranular zone are directly located on the aberrant MF or whether the latter induce the appearance of these sites on the granules. Whatever the exact location, it is likefy that the increased density of sites for KA in the supragranular zone underlines the increase of the excitability which was observed for the granules in similar conditions.34 This enhanced excitability might play a role in the occurrence of the spontaneous seizures which were observed, in rats, at long delays after KA. “,” It is possible that similar changes play a role in human epilepsy. More generally, it is clear that deafferentation or destruction of
Kainate binding sites in the mossy fibers
neurons in a plastic structure such as the hippocampus*j will produce rearrangements with major changes in the excitability of the circuit.
741
Acknowledgements-The authors are endebted to Dr A. Pompidou and D. Schoevart for the use of the image analyser and to G. Charton, J. P. Bouillot, G. Ghilini and S. Guidasci for technical assistance.
REFERENCES
1. Albe-Fessard D., Stutinski F. and Libouban-le-Touze S. (1970) Atlas Stereotanique du Diencephale du Rat Blanc. CNRS, Paris. 2. Amaral D. G. and Dent J. A. (1981) Development of the mossy fibers of the dentate gurus. I. A light and electron microscopic study of the mossy fibers and their extensions. J. camp. Neurol 200, 57-86. 3. Ben-Ari Y. (1983) Limbic seizures and brain damage produced by kainic acid: mechanisms and relevance to human temporal lobe epilepsy. Neuroscience 14, 375403. 4. Ben-Ari Y., Tremblay E. and Ottersen, 0. P. (1980) Injection of kainic acid into the amygdaloid complex of the rat; an electrographic, clinical and histopathological study in relation to the pathology of epilepsy. Neuroscience J,515-528. 5. Ben-Ari Y., Tremblay E., Ottersen 0. P. and Meldrum B. S. (1980) The role of epileptic activity in hippocampal and “remote” cerebral lesions induced by KA. Brain Res. 91, 79-97. 6. Berger M. and Ben-Ari Y. (1983) Autoradiographic visualization of [jH]kainic acid receptor subtypes in the rat hippocampus. Neurosci. Lett. 39, 337-342. 7. Berger M., Charton G. and Ben-Ari Y. (1986) Effects of seizures induced by intra-amygdaloid kainic acid on kainic acid binding sites in rat hippocampus and amygdala. J. Neurochem. 47, 72&727. 8. Berger M., Tremblay E., Nitecka L. and Ben-Ari Y. (1984) Maturation of kainic acid seizure-brain damage syndrome in the rat. III. Postal-natal development of kainic acid binding sites in the limbic system. Neuroscience 13, 1095-I 104. 9. Biziere K. and Coyle J. T. (1978) Influence of cortico-striatal afferents on striatal kainic acid neurotoxicity. Neurosci. Lett. 8, 303-310.
10. Blackstad T. W., Brink K., Hem J. and Jeune, B. (1970) Distribution of hippocampal mossy fibers in the rat: an experimental study with silver impregnation methods. J. camp. Neurol. 138, 433450. 11. Cajal S. and Ramon Y. (eds) (1978) The Structure of Ammon’s Horn. C. C. Thomas, Springfield, IL. 12. Cavalheiro E., Riche D. and Le Gal La Salle G. (1983) Long-term effects of intra-hippocampal kainic acid injection in rat: a method for inducing spontaneous recurrent seizures. Electroenceph. clin. Neurophysiol. 262, 201-207. 13. Coyle J. T., Molliver M. E. and Kuhar M. J. (1978) In situ injection of kainic acid: a new method for selectively lesioning the neuronal cell bodies while sparing axons of passage. J. camp. Neurol. 180, 301-324. 14. Crawford I. L. and Connor J. D. (1973) Localisation and release of glutamic acid in relation to the hippocampal mossy fibre pathway. Nature 244, 442443. 15. Ferkany J. W., Zaczek R. and Coyle J. T. (1982) Kainic acid stimulates excitatory amino-acid neurotransmitter release at presynaptic receptors. Nature 298, 757-759. 16. Fisher R. S. and Alger B. E. (1984) Electrophysiological mechanisms of kainic acid induced epileptiform activity in the rat hippocampal slices. J. Neurosci. 4, 1312-1323. 17. Frotscher- M. and Zimmer J. (1983) Lesion induced mossy fibers to the molecular layer of the rat fascia dentata; identification of postsynaptic granule cells by the Golgi-EM technique. J. camp. Neurol. 215, 299-311. 18. Gaarskjaer F. B. (1978) Organisation of the mossy fiber system of the rat studied in extended hippocampi. II. Experimental analysis of fiber distribution with silver impregnation methods. J. camp. Neurol. 178, 73-88. 18a. Geddes J. W., Monaghan D. T. and Cotman C. W. (1985) Plasticity of hippocampal circuitry in Alzheimer’s disease. Science 230, 1179-1181. 19. Goldschmidt R. B. and Steward 0. (1980) Preferential neurotoxicity of colchicine for granule cells of the dentate gyrus of the rat. Proc. natn. Acad. Sci. U.S.A 77, 3047-3051. 20. Haug F. M. S. (1973) A light microscopic study in the rat with Timm’s sulphide silver method. Methodological considerations and cytological and regional staining patterns. Adu. Anat. Embryol. cell. Biol. 47, l-71. 21. Laurberg S. and Zimmer J. (1981) Lesion induced sprouting of hippocampal mossy fiber collaterals to the fascia dentata in developing and adult rats. J. camp. Neurol. 200, 433459. 22. London E. D. and Coyle J. T. (1979) Specific binding of [ ‘Hlkainic acid to receptor sites in rat brain. Molec. Pharmacol. 15, 492-505. 23. London E. D., Klemm N. and Coyle J. T. (1980) Phylogenic distribution of [)H]kainic acid receptor binding sites in neuronal tissue. Brain Res. 192, 463-476. 24. Lorente de No R. (1934) Studies on the structure of the cerebral cortex. II. Continuation of the study of the ammonic system. J. Psychol. Neurol. 46, 113-117. 25. Lynch G. and Cotman C. W. (1975) The hippocampus as a model for studying anatomical plasticity in the adult brain. In The Hippocampus. I. Structure and Development (eds Isaacson R. L. and Prisbam K. H.), pp. 123-154. Plenum Press, New York. 26. Monaghan D. T. and Cotman C. W. (1982) The distribution of [3H]kainic acid binding sites as determined by autoradiography. Brain Res. 252, 91-100. 27. Nadler J. V., Perry B. W., Gentry C. and Cotman C. W. (1980) Loss and reacquisition of hippocampal synapses after selective destruction of CA3CA4 afferents with kainic acid. Brain Res. 191, 387403. 28. Nitecka L., Tremblay E., Charton G., Bouillot J. P. and Ben-Ari Y. (1984) Maturation of kainic acid seizure-brain damage syndrome in the rat-II. Histopathological sequelae. Neuroscience 13, 107331094. 29. Olney J. W. and de Gubareff T. (1978) The fate of synaptic receptor in the kainate lesioned striatum. Brain Res. 140, 34&343. 30. Raisman G., Lawrence J. M., Zhou C. F. and Lindsay R. M. (1985) Some neuronal, glial and vascular interactions which occur when developing hippocampal primordia are incorporated into adult host hippocampal. In Neural Grafting in the MammaIian CNS (eds Bjorklund A. and Stenevi U.), pp. 125-150. Elsevier, Amsterdam. 31. Savage D. D., Nadler J. V., McNamara J. 0. (1984) Reduced kainic acid binding sites in rat hippocampal formation after limbic kindling. Brain Res. 323, 128-13 1.
748
A. REPRESA et al.
32. Sloviter R. S. (1982) A simplified Timm stain procedure compatible with formaldehyde fixation and routine paraffin embedding of rat brain. Brain Res. 8, 111--774. 33. Swanson -1. W., Wyss J. M. and Cowan W. M. (1978) An autoradiographic study of the organisation of intrahippocampal association pathway in the rat. J. camp. Neural. 181, 681-716. 34. Tauck D. L. and Nadler J. V. (1985) Evidence for functional mossy fiber sprouting in hippocampal formation of kainic acid treated rats. J. Neurosci. 5, 10161022. 35. Tremblay E. and Ben-Ari Y. (1984) Usefulness of parenteral kainic acid as a model of temporal lobe epilepsy. Rep. E.EG Neuroph~siol. 14, 241-246. 36. Tremblay E., Nitecka L., Berger M. and Ben-An Y. (1984) Maturation of the kainic acid seizure-brain damage syndrome in the rat--I. Clinic& electrographic and metabolic observations. ~e~r[~s~ie~ce13, 1051-1072. 37. Tremblay E., Represa A. and Ben-Ari Y. (1985) Autoradiographic localisation of kainic acid binding sites in the human hippocampus. Bruin Res. 343, 378-382. 38. Unnerstall J. R. and Wamsley J. K. (1983) Autoradiographic focalisation of high affinity [‘Hjkainic acid binding sites in the rat forebrain. Eur. J. Pharnruc. 86, 361-371. (Accepfed I1 July 1986)
3, pp. 739-748,
1987
0306-4522/87 $3.00 + 0.00 Pergamon Journals Ltd 0 1987 IBRO
KAINATE BINDING SITES IN THE HIPPOCAMPAL MOSSY FIBERS: LOCALIZATION AND PLASTICITY A. REPRESA,* E. TREMBLAYand Y. BEN-ARI Laboratoire de Neurobiologie et Neurophysiopathologie du Developpement, 123 Bd de Port-Royal, 75674 Paris Cedex 14, France and LPNl, CNRS, Gif-sur-Yvette 91190, France
INSERM,
U29,
Abstract-The
regional distribution of high affinity binding sites for kainic acid has been determined in rat hippocampi by quantitative autoradiography. Selective lesions were made in order to determine the exact localization of these sites in the mossy fiber system, and to evaluate whether the sprouting and
synaptic reorgani~tion of the mossy fibers are associated with alterations in the dis~bution of these binding sites. The results show that kainate binding sites in the stratum lucidum are more vulnerable to destruction of the granules and their mossy fibers by in~~ipp~mpal colchicine injections than to destruction of the CA31CA4 uvramidal cells bv injection of kainate into the amyadala. This suggests that a substantial
proportibn of -&ekainate bind&g sites is associated with the m&iy fiber termi& (i.e. the presynaptic elements). Furthermore, in keeping with an earlier study, destruction of the pyramidal neurons of CA3 by intracerebral kainate produced a dark Timm positive band in the supragranular zone which is due to the sprouting of mossy fibers. This was associated with an increase in the density of kainate binding sites, which further stresses the parallelism between the distribution of these sites and mossy fiber terminals.
EXPERIMENTAL PROCEDURES mossy fibers (MF) which originate from the dentate granule cells massively project upon Male Wistar rats (22&230 g) were used throughout these the pyramidal neurons of the CA3 region; MF termiexperiments. They had access to food and water ad kibitz nals also innervate cells in the hiiar region and the and were housed in individual cages under diurnal ii~ting conditions, with lights on from 0800 to 2000 h. For lesion supragranular layer of the fascia dentata.10~“~‘*~2’34.33 experiments, animals were anaesthetized with Quithesin The CA3jH3 pyramidal neurons innervated by the [Jensen Salsbury, 3 ml/kg) and submitted to unilateral injecMF are particularly vulnerable to the epilepsies in tions of colchicine in the hippocampus or KA into the man, as well as in the seizures produced, in rats and amygdala. All injections were performed with a microprimates, by administration of kainic acid (KA), the syringe under stereotaxic guidance. With colchicine, two potent excitatory analogue of glutamate (see Ref. 3 injections of 1.5 pg of colchicine (dissolved in 0.6~1 of physiological saline, pH 7.2) were made in the same hemifor a review). Specific binding sites for KA are sphere at the following coordinates according to the Atlas present in the CNS of various species, notably in the of Albe-Fessard;’ A:2.9; L:3.3; H:6.4 and A: 5.4; L: 1.3 and H : 7.0. This procedure was previously found to destroy the hippocampus. 23Autoradiographic studies performed granule cells and their processes within a few days.19 Injecin both rats6,*‘~‘*and humans” indicate that their tion of KA into the amygdala (I.2 pg in 0.3 ~1 of phosphate dist~but~on, in the hippocampus, coincides with the buffer, pH ‘7.4)was made at the following coordinates; A : 6; MF terminaIs, being particularly enriched in the L:4 and H: 2. Animals showing typical KA-induced limbic stratum lucidurn. Ontogenetic studies in rats have motor seizures for at least 2 h upon recovery from anaec shown a good parallelism between the occurrence of thesis were kept for the experiments; these received diazethe seizure/brain damage syndrome produced by pam (Roche, 20mg/kg i.p.) to stop the seizures; such treatment was previously found to produce a seizure-related parenteral KA,*‘s’~the maturation of the MF2 and the lesion of the pyramidal cells in CA3 of the ipsilateral late occurrence of high affinity KA binding sites in hemisphere.4,5 Six or eight days after colchicine or 3-30 days after KA the hippocampus.* These, as well as other observations,‘6’6 suggest that an endogenous EGA-like injections, the animals were anaesthetized with pentobarbital and a modified Timm’s procedure’* was peragent is released from the MF. formed to stain the mossy fibers20 The rats were intraThe aim of the present study was to substantiate cardially perfused first with phosphate buffer containing further the relationship between MF and KA binding 0.37% sodium sulphide (2OOml), then with 0.1% paraformaldehyde (200 ml). This procedure does not affect the sites. First, using relatively specific lesions, we exambinding experiments (see Results). After perfusion, brains ined whether the KA sites are pre- or ~stsynaptically were removed and rapidly frozen in isopentane ( -50°C). located in the MF terminal zone. Secondly, since the Coronal sections (20pm) of the dorsal hip~mpus were MF have been shown to sprout in a variety of cut with a cryostat (-15°C) and thaw-mounted onto expe~men~a~17.2i.27,~.~ gelatin-coated slides for binding experiments. The histoand probably also pathological logical procedures (Timm or Nissl) were made on the same conditions, we examined whether this synaptic reoror alternate sections. ganization is associated with the appearance of KA Visualization of the KA binding sites was done using an binding sites in the reinnervated regions. autoradiographic procedure which enables the high
The hippocampal
*To whom correspondence should be addressed. Abbreuialions: KA, kainic acid; MF, mossy fibers.
affinity-slowly dissociating-KA sites to be preferentially labelled.6~22The sections were preincubated in Tris-acetate buffer (50 mM, pH 7.1) for I h at 3”C, then for 15 min at 739
740
A.
bZT.ESA
30°C to remove
competing endogenous ligands from the tissue. The sections were then incubated at 3°C for 90 min with the same buffer in the presence of 20 nm [VinylidenerH]KA (NEN, 60Ci/mmol). After incubation, the low affinity sites with a fast dissociation rate were rapidly displaced by rinsing the sections for 2 min in the buffer containing 10 PM unlabelled KA. Non-specific binding was routinely evaluated by incubating alternate sections with the radioligand plus an excess of IOpM cold KA. After the procedure, the sections were dried and exposed to 3H-sensitive Ultrofilms (LKB) for 30 davs. simultaneouslv with internal standards. huantification of the density of the binding sites was performed using an image analyser (Imanco Quantimet 720) and relying on the comparison with the optical density of internal standards. At least four sections were used to determine the density of binding sites in each area. Statistical analysis was performed using the Student’s t-test. For Scatchard analysis, hippocampal sections were submitted to the same procedure except that incubation was performed with various concentrations of the radioligand (I-SOOnM); after rinsing, the slices were wiped off on Whatman filters, dried and put in a vial with IO ml of Aquasol, and the radioactivity was counted with a liquid scintillation counter. RESULTS
Eflect of perfusion/fiwation on kainic acid binding Since perfusion/fixation according to Timm’s method was needed to determine precisely the extent of the lesions, we have made preliminary experiments to verify that this procedure does not modify KA binding. The Kd or B,,,,, values evaluated from Scatchard analysis were similar in hippocampal sections from non-perfused (n = 3; 19 + 12 and 40 + 12 fmol/mg protein, respectively, and perfused rats (n = 4; 24 + 9 and 36 + 6 fmol/mg protein, respectively). Furthermore, in autoradiographs from control non-perfused rats (n = 3), the densities of sites measured in CA3a and the supragranular layer (222 and 121 fmol/mg protein, respectively) were similar to those found with perfused rats (Table 1). Decreased binding sites after colchicine injections or kainic acid treatment The autoradiographs presented in Figs 1, 2 and 3 and quantitated in Table 1 illustrate that KA labelling in the stratum lucidum is reduced more extensively and rapidly following unilateral colchicine inTable
rf
(11.
jections in the hippocampus than unilateral injection of KA into the amygdala. Thus, 6 days after colchicine, almost the entire granule-MF system was destroyed, only restricted spots of the Timm’s deposits still being visible (not illustrated). At this delay, the loss of KA binding sites was conspicuous and almost maximal (Table I). Two days later, the complete disappearance of Timm’s staining (Fig. 2) was associated with a further small reduction of the density of sites in CA3a and the supragranular zone (Table 1). In keeping with earlier studies,” the CA3 pyramidal neurons were relatively spared by the procedure (Fig. 2). In spite of some variability, the gradient of destruction produced by colchicine was: granules of the fascia dentata>> CA1 >>CA3. Clearly, the restricted damage of the CA3 neurons cannot explain the colchicine-induced loss of sites since, after KA and similar short delays of survival, the complete destruction of the CA3 neurons was not accompanied by a significant change in binding sites (see below). In keeping with our earlier observations,4x5 KA injection into amygdala produced a selective and rapid loss of pyramidal neurons of CA3 in the ipsilateral Ammon’s horn. This was already present 3 days after the injection, notably in CA3a, which is the most vulnerable brain region to KA.’ 5 At longer delays (8 days or more, cf. Fig. 3A and B), the entire CA3 region was largely destroyed. In contrast to the postsynaptic elements, the presynaptic ones (i.e. the MF terminals) were largely intact since Timm’s stain was present 3-30 days after KA (Fig 3D). The main difference with the control cases concerned the organization of the dark silver deposits which were not aligned vertically to the pyramidal layer (Fig. 3D), as they usually are (see in Fig. 2, the contralateral side). This is due to the destruction of the dendrites of the CA3 pyramidal neurons. Both visual inspection of the KA autoradiographs and quantitative densitometry showed the lack of a significant change in KA binding sites in KA-treated cases with 3-8 day survival periods (Fig. 1, Table 1) i.e. at a time when the pyramidal neurons of CA3 were already extensively destroyed (Fig. 3). With longer survival times (12-30 days), there was a
1. Effects of kainic acid and colchicine treatments on the densities in the hippocampus
Treatment
n
CA3c
Control KA (3 d) KA (8 d) KA (12 d) KA (30 d) Colchicine (6 d) Colchicine (8 d)
4 2 2 2 4 2 2
IL3 +9 121 * 10 115+ 11 109 + 14 94* 5* 70*21** 70+8**
CA3a 226 + 225 * 180 f 161 + 119+ 141 + 135 +
of kainic acid binding CA1
5 5 13 2** 11** 23* 13**
40 * 36 + 50+ 31*7 44* 5s* 53 *
sites
sg 5 3 10 10 17 3
124+5 126 * 10 123+ 11 I51 +4* 165 k 6** 64* 11** 57 If- 10**
Mean densities of KA binding sites (in fmoljmg protein) + SEM, in various subfields, in controls and treated cases. The survival time in.,days (d) and the number of rats (n) are indicated. CA3a and CA3c, stratum lucidum of the CA3a and CA3c regions; sg, supragranular layer of the fascia dentata. *.** Values statistically different from control ones for P < 00.5 and 00. I, respectively (Student’s test).
Kainate
binding
sites in the mossy
741
fibers
Control
KA(8d)
KA(30d)
Colch. (8d) Fig. 1. Autoradiographs
depicting
the distribution of high affinity KA binding sites in the hippocampus (Colch.) treated cases. Dark triangles indicate the side ipsilateral to injection. In perfused controls, the KA labelling is also oonfined to the supragranular layer of the fascia dentata (FD) and the stratum lucidum of the CA3 region. Note the progressive loss of labelling from the stratum lucidum after KA and the extensive and rapid loss after colchicine. (d) Survival delay in days.
in (perfused) controls, KA or colchicine
progressive loss of KA sites in the stratum lucidum on the side of injection. This involved first the CA3a region, where a significant 29% reduction was already present with 12 days survival; at longer delay (30 days), the loss in CA3 was more conspicuous (Table 1). This delayed loss of sites (see also Ref.7) probably reflects a protracted clearance of the damaged CA3 dendrites (see Discussion). As indicated in Table 1, the densities of sites in the other hippocampal regions are not altered (i.e. in CAl) or even enhanced (see below). Increased kainic acid binding sites In addition to the reduction in KA binding sites described above, KA treatment produced a rise in
binding sites in the ipsilateral supragranular layer of the fascia dentata (Table 1) and in the contralateral side in both the supragranular and infrapyramidal zones. Quantitative autoradiography and systematic comparison with the Timm’s stained sections clearly indicated that this rise is associated with the synaptic rearrangements known to occur in these conditions. Thus, in keeping with the recent studies of Nadler et a1.,27 with long survival periods, intracerebral KA produced a conspicuous band of dark deposits in the supragranular layer (Fig. 4A and C). This corresponds to the autoinnervation of the granules which is observed after a variety of lesion-induced deafferentation of the granules.‘7.2’,27 In the four cases
A. REPRESAet al
742
in which a rise in Timm’s stain was found in the ipsilateral supragranular layer, there was also a significant increase in KA binding sites (Table I). In all the other cases (with shorter survival delays), neither Timm’s staining nor KA binding sites were enhanced. Furthermore, enhanced Timm’s staining was found in one case (30 days survival) in the contralateral supragranular layer (not illustrated) and the infrapyramidal CA3 band (Fig. 5A). This enhancement, which probably reflects a sprouting of the MF (see Discussion), was associated with increased densities of KA binding sites in these layers (164 and 269 fmol/mg protein, respectively; also see Fig. 5B). This further stresses the parallelism between Timm’s staining of MF terminals and KA binding sites. Enhanced KA labelling was also found in the contralateral hippocampus after colchicine treatment. The significance of this increase is not clear since the increase (corresponding to approximately 27% of the control values) affected all hippocampal subfields and was already present 6 days after the injections. This increased KA labelling is therefore unlikely to reflect a selective reinnervation process. DISCUSSION
In their earlier study, Monaghan and Cotman, using intrahippocampal colchicine or ICV injections of KA, concluded that both treatments produce a similar loss of KA binding sites in the stratum lucidum. These authors, however, used only one survival delay (30 days) and did not perform quantitative densitometry. In the present study, using quantitative densitometry, we found that the loss of KA sites after colchicine was conspicuous even within 6 days and concerned the entire extent of the regions innervated by the MF (i.e. the stratum lucidum and supragranular layer of the fascia dentata). In contrast, after KA injection, the reduction of sites was more delayed and only involved the stratum lucidurn. Histological examinations of the same or alternate sections from
which the autoradiographs were obtained showed that colchicine injections in the hippocampus rapidly destroyed the MF system with relatively restricted effects on the pyramidal neurons,” whereas injections of KA into the amygdala rapidly produced a virtually complete neuronal loss in CA3, with restricted effects on the MF system,4*5 as assessed with both Nissl’s and Timm’s stains. These results, in keeping with the excellent parallelism already noted between the distribution and development of the MF and KA binding sites (see Introduction), suggest that a large proportion of these sites are located on the MF. A presynaptic localization of the KA binding sites was also suggested in earlier biochemical studies in which the toxin produces a release of endogenous excitatory amino acids in the hippocampus;i5 electrophysiological studies also suggest a presynaptic localization of these sites.16 However, a partial localization of KA sites on the postsynaptic elements cannot be excluded. As a matter of fact, even with a virtual complete destruction of the MF by colchicine, a significant number of binding sites remained in this region. It is tempting to suggest that these are located on the pyramidal neurons, in keeping with other neurotransmitter systems in which receptors are both pre- and postsynaptically located. The time course of events noted after KA is in keeping with this hypothesis. A recent biochemical study performed following intraamygdaloid injection of KA also showed a delayed and progressive decline in KA binding in membranes prepared from the microdissected CA3 region;’ this was due to a progressive decline of B,,,,, without a change in the affinity constant.’ This is important since seizures, which are not associated with brain damage (kindling), can also produce a reduction of sites for KA in the hippocampus3’ The progressive decline of binding sites might be due to the slow phagocytosis of intact synaptic complexes, i.e. including fragments of the postsynaptic membrane remaining attached to the presynaptic bouton. Such complexes have been found,
Fig. 2. Photomicrographs depicting the changes produced by injections of cochicine in the hippocampus. (A) and (B) Sections from the same case, ipsilaterally (A) and contralaterally (B) to injection, treated for Timm’s stain and counterstained with Nissl’s stain. Note the loss of granule cells (g) in the injected side (dark triangles); in contrast, the pyramidal cells in CA3 are largely present. Also note in (B) the normal radial disposition of the silver deposits in the stratum lucidum (Lu) as revealed by Timm’s stain, and in (A) the complete loss of staining. Ra, stratum radiatum. The boxed area illustrates the KA autoradiograph obtained from the same case: note the extensive loss of labelling in the side of injection. Fig. 3. Photomicrographs depicting the histopathological changes in hippocampus 8 (A) and 30 (B) days after KA injection into amygdala. There is a loss of pyramidal neurons associated with gliosis in the inferior Ammon’s horn up to the CA3/CA2 boundary zone (arrow); the granule cells (g) are intact (Nissl stain). (C) KA autoradiograph obtained from the same case as in (B); note the lower labelling in the stratum lucidum of the lesioned side (right part of the picture). In (D), an alternate section from the same case stained with Timm’s method shows packing of silver deposits in the stratum lucidum (cf. Fig. 3C). Fig. 4. Photomicrographs illustrating the sprouting of mossy fibers at long delays (30 days) after KA treatment. Note in (A) the intense Timm’s staining ipsilateral to injection, in the supragranular zone (indicated by arrows); the rectangle in (A) is enlarged in (C). (B) and (D) Control case illustrating the lack of intense Timm’s staining in the same layer.
e
744
745
A.
REPRESA et al.
Fig. 5. Photomicrograph illustrating the associated occurrence of Timm’s staining (A) and KA iabelling (B) in the infrapyramidal layer (arrow). Both sections are from the same case, contralateral to the KA-treated side (survival time = 30 days).
for instance, in the striatum even after destruction, by local KA, of intrinsic neurons; this may account for the progressive disappearance of KA or other binding sites from the striatum.9.‘3.29 To the best of our knowledge, the present study shows for the first time that a synaptic rearrangement, which occurs in the hippocampus following lesions, is associated with an increase of binding sites for an excitatory amino acid (but see also Ref. Isa). The ability of the MF to sprout following a variety of experimental conditions has been clearly established both with electron microcopy and Timm’s staining; this is particularly conspicuous in the supragranular layer of the fascia dentata foliowing suppression of commissural, associational and/or entorhinal afferents to the granules.2’~27 Using a combined Golgi/electron microscopic approach, Frotscher and Zimmer” recently provided compelling evidence that this is due to an autoinnervation of the granules by the MF. Aberrant sprouting of the
MF has also been described in other conditions including intracerebral grafts on hippocampal neuronsM interestingly, after ICV” or intra-amygdaloid KA (this study), the destruction of the pyramidal cells and their associative projections to the granules also produce an aberrant autoinnervation of the MF in the supragranular zone. It is not clear whether the KA binding sites occurring in these conditions in the supragranular zone are directly located on the aberrant MF or whether the latter induce the appearance of these sites on the granules. Whatever the exact location, it is likefy that the increased density of sites for KA in the supragranular zone underlines the increase of the excitability which was observed for the granules in similar conditions.34 This enhanced excitability might play a role in the occurrence of the spontaneous seizures which were observed, in rats, at long delays after KA. “,” It is possible that similar changes play a role in human epilepsy. More generally, it is clear that deafferentation or destruction of
Kainate binding sites in the mossy fibers
neurons in a plastic structure such as the hippocampus*j will produce rearrangements with major changes in the excitability of the circuit.
741
Acknowledgements-The authors are endebted to Dr A. Pompidou and D. Schoevart for the use of the image analyser and to G. Charton, J. P. Bouillot, G. Ghilini and S. Guidasci for technical assistance.
REFERENCES
1. Albe-Fessard D., Stutinski F. and Libouban-le-Touze S. (1970) Atlas Stereotanique du Diencephale du Rat Blanc. CNRS, Paris. 2. Amaral D. G. and Dent J. A. (1981) Development of the mossy fibers of the dentate gurus. I. A light and electron microscopic study of the mossy fibers and their extensions. J. camp. Neurol 200, 57-86. 3. Ben-Ari Y. (1983) Limbic seizures and brain damage produced by kainic acid: mechanisms and relevance to human temporal lobe epilepsy. Neuroscience 14, 375403. 4. Ben-Ari Y., Tremblay E. and Ottersen, 0. P. (1980) Injection of kainic acid into the amygdaloid complex of the rat; an electrographic, clinical and histopathological study in relation to the pathology of epilepsy. Neuroscience J,515-528. 5. Ben-Ari Y., Tremblay E., Ottersen 0. P. and Meldrum B. S. (1980) The role of epileptic activity in hippocampal and “remote” cerebral lesions induced by KA. Brain Res. 91, 79-97. 6. Berger M. and Ben-Ari Y. (1983) Autoradiographic visualization of [jH]kainic acid receptor subtypes in the rat hippocampus. Neurosci. Lett. 39, 337-342. 7. Berger M., Charton G. and Ben-Ari Y. (1986) Effects of seizures induced by intra-amygdaloid kainic acid on kainic acid binding sites in rat hippocampus and amygdala. J. Neurochem. 47, 72&727. 8. Berger M., Tremblay E., Nitecka L. and Ben-Ari Y. (1984) Maturation of kainic acid seizure-brain damage syndrome in the rat. III. Postal-natal development of kainic acid binding sites in the limbic system. Neuroscience 13, 1095-I 104. 9. Biziere K. and Coyle J. T. (1978) Influence of cortico-striatal afferents on striatal kainic acid neurotoxicity. Neurosci. Lett. 8, 303-310.
10. Blackstad T. W., Brink K., Hem J. and Jeune, B. (1970) Distribution of hippocampal mossy fibers in the rat: an experimental study with silver impregnation methods. J. camp. Neurol. 138, 433450. 11. Cajal S. and Ramon Y. (eds) (1978) The Structure of Ammon’s Horn. C. C. Thomas, Springfield, IL. 12. Cavalheiro E., Riche D. and Le Gal La Salle G. (1983) Long-term effects of intra-hippocampal kainic acid injection in rat: a method for inducing spontaneous recurrent seizures. Electroenceph. clin. Neurophysiol. 262, 201-207. 13. Coyle J. T., Molliver M. E. and Kuhar M. J. (1978) In situ injection of kainic acid: a new method for selectively lesioning the neuronal cell bodies while sparing axons of passage. J. camp. Neurol. 180, 301-324. 14. Crawford I. L. and Connor J. D. (1973) Localisation and release of glutamic acid in relation to the hippocampal mossy fibre pathway. Nature 244, 442443. 15. Ferkany J. W., Zaczek R. and Coyle J. T. (1982) Kainic acid stimulates excitatory amino-acid neurotransmitter release at presynaptic receptors. Nature 298, 757-759. 16. Fisher R. S. and Alger B. E. (1984) Electrophysiological mechanisms of kainic acid induced epileptiform activity in the rat hippocampal slices. J. Neurosci. 4, 1312-1323. 17. Frotscher- M. and Zimmer J. (1983) Lesion induced mossy fibers to the molecular layer of the rat fascia dentata; identification of postsynaptic granule cells by the Golgi-EM technique. J. camp. Neurol. 215, 299-311. 18. Gaarskjaer F. B. (1978) Organisation of the mossy fiber system of the rat studied in extended hippocampi. II. Experimental analysis of fiber distribution with silver impregnation methods. J. camp. Neurol. 178, 73-88. 18a. Geddes J. W., Monaghan D. T. and Cotman C. W. (1985) Plasticity of hippocampal circuitry in Alzheimer’s disease. Science 230, 1179-1181. 19. Goldschmidt R. B. and Steward 0. (1980) Preferential neurotoxicity of colchicine for granule cells of the dentate gyrus of the rat. Proc. natn. Acad. Sci. U.S.A 77, 3047-3051. 20. Haug F. M. S. (1973) A light microscopic study in the rat with Timm’s sulphide silver method. Methodological considerations and cytological and regional staining patterns. Adu. Anat. Embryol. cell. Biol. 47, l-71. 21. Laurberg S. and Zimmer J. (1981) Lesion induced sprouting of hippocampal mossy fiber collaterals to the fascia dentata in developing and adult rats. J. camp. Neurol. 200, 433459. 22. London E. D. and Coyle J. T. (1979) Specific binding of [ ‘Hlkainic acid to receptor sites in rat brain. Molec. Pharmacol. 15, 492-505. 23. London E. D., Klemm N. and Coyle J. T. (1980) Phylogenic distribution of [)H]kainic acid receptor binding sites in neuronal tissue. Brain Res. 192, 463-476. 24. Lorente de No R. (1934) Studies on the structure of the cerebral cortex. II. Continuation of the study of the ammonic system. J. Psychol. Neurol. 46, 113-117. 25. Lynch G. and Cotman C. W. (1975) The hippocampus as a model for studying anatomical plasticity in the adult brain. In The Hippocampus. I. Structure and Development (eds Isaacson R. L. and Prisbam K. H.), pp. 123-154. Plenum Press, New York. 26. Monaghan D. T. and Cotman C. W. (1982) The distribution of [3H]kainic acid binding sites as determined by autoradiography. Brain Res. 252, 91-100. 27. Nadler J. V., Perry B. W., Gentry C. and Cotman C. W. (1980) Loss and reacquisition of hippocampal synapses after selective destruction of CA3CA4 afferents with kainic acid. Brain Res. 191, 387403. 28. Nitecka L., Tremblay E., Charton G., Bouillot J. P. and Ben-Ari Y. (1984) Maturation of kainic acid seizure-brain damage syndrome in the rat-II. Histopathological sequelae. Neuroscience 13, 107331094. 29. Olney J. W. and de Gubareff T. (1978) The fate of synaptic receptor in the kainate lesioned striatum. Brain Res. 140, 34&343. 30. Raisman G., Lawrence J. M., Zhou C. F. and Lindsay R. M. (1985) Some neuronal, glial and vascular interactions which occur when developing hippocampal primordia are incorporated into adult host hippocampal. In Neural Grafting in the MammaIian CNS (eds Bjorklund A. and Stenevi U.), pp. 125-150. Elsevier, Amsterdam. 31. Savage D. D., Nadler J. V., McNamara J. 0. (1984) Reduced kainic acid binding sites in rat hippocampal formation after limbic kindling. Brain Res. 323, 128-13 1.
748
A. REPRESA et al.
32. Sloviter R. S. (1982) A simplified Timm stain procedure compatible with formaldehyde fixation and routine paraffin embedding of rat brain. Brain Res. 8, 111--774. 33. Swanson -1. W., Wyss J. M. and Cowan W. M. (1978) An autoradiographic study of the organisation of intrahippocampal association pathway in the rat. J. camp. Neural. 181, 681-716. 34. Tauck D. L. and Nadler J. V. (1985) Evidence for functional mossy fiber sprouting in hippocampal formation of kainic acid treated rats. J. Neurosci. 5, 10161022. 35. Tremblay E. and Ben-Ari Y. (1984) Usefulness of parenteral kainic acid as a model of temporal lobe epilepsy. Rep. E.EG Neuroph~siol. 14, 241-246. 36. Tremblay E., Nitecka L., Berger M. and Ben-An Y. (1984) Maturation of the kainic acid seizure-brain damage syndrome in the rat--I. Clinic& electrographic and metabolic observations. ~e~r[~s~ie~ce13, 1051-1072. 37. Tremblay E., Represa A. and Ben-Ari Y. (1985) Autoradiographic localisation of kainic acid binding sites in the human hippocampus. Bruin Res. 343, 378-382. 38. Unnerstall J. R. and Wamsley J. K. (1983) Autoradiographic focalisation of high affinity [‘Hjkainic acid binding sites in the rat forebrain. Eur. J. Pharnruc. 86, 361-371. (Accepfed I1 July 1986)