Functional changes in somatostatin and neuropeptide Y containing neurons in the rat hippocampus in chronic models of limbic seizures

N

ELSEVIER

EPILEPSY SEARCH Epilepsy Research 26 {]996) 267-27~J

Functional changes in somatostatin and neuropeptide Y containing neurons in the rat hippocampus in chronic models of fimbic seizures A. Vezzani a.~ Co Schwarzer b E.W. Lothman ~,t, j. Wi|liamson c, G. Sperk b a lstitulo di Ricerche Parmacologiche Mario Negri. Milan, Italy h Department ~fPharmacology, Unicersi O" ~flnnsbruck, lmtsbruck, Austria Departmetlt oj'Neaiwlogy, U~icersily oj Virginia, Charlottescil!e, VA. USA

Received 28 September 1995; accepted 4 March 1996

Abstract Using immunocytochemistry and in situ hybridization analysis of mRNA, we investigated the changes in the expression of somatostatin and neuropeptide Y (NPY) in the rat hippocampal principal neurons in kindling or after electrically induced status epilepticus (SE), two models of timbic epilepsy associated with different chronic sequelae of seizures and seizure-related neuropathology. At the preconvulsive stage 2 of kindling and after three consecutive tonic-clonic seizures (stage 5) but not after a single after-discharge (AD), somatostatin and NPY immunoreactivity (|R) were markedly increased in interneurons of the deep hilus and the polymorphic cell layer and their presumed projections to the outer molecular layer nf the de~.t~.te gyrus. Increased mRNA levels were observed in the same neurons. NPY IR and mRNA were highly expressed in pyramidal-shaped basket cells at both stages of kindling. IR was simiDu"two days after stages 2 or 5 of "kindling while less pronounced effects were observed one week after kindling completion. Peptide-containing neurons in the hilus appeared well preserved in spite of an average 24% reduction of Nissl stained cells ( p < 0.01) in the stimulated and contralateral hippocampus at stage 5. No sprouting of mossy fibres in the inner molecular layer was found as assessed by Timm staining. Thirty days after SE, somatostatin IR was slightly reduced or similar to controls in the ventral dentate gyrus and molecular layer in four out of six rats (SE-I group) while in the two other post-SE rats (SE-II), somatostatin IR was lost. These ehangc~ were associated with a different extent of neurodegeneration as assessed by cell counting of Nissl stained sections. In the granule cells/mossy fibres NPY-IR was transiently expressed at stage 2 and after a single AD. Differently, NPY-IR was persistently enhanced in ti',e mossy fibres of all post-SE rats particularly in the SE-II group. In these rats, NPY immunoreactive fibres were detected in the iufrapyramidal region of the stratum oriens CA3 and in the inner molecular layer of the dentate gyrus very likely labeling sprouted mossy fibres. In the hippocampus proper of kindled rats, somatostatin and NPY IR were respectively ev~aneed in the stratum lacunosum moleculare, the subiculum and in the alveus while no significant changes were observed after SE. Changes in peptide expression were bilateral and involved both the dorsal and

' Corresponding author. Tel.: + 39 (2) 3901 4410; fax: + 39 (2) 354 6277. i This work is dedicated to Dr. Eric W. Lothman. a dear friend and an outstanding scientist who passed away on April 15 1995. 0920-1211/96/$15.00 Copyright © 1996 Elsevier Science B.V. All fights reserved. Pll S0920-12 1 1(96)00059-9

268

A. Vezzani et al./ Epilepsy Research 26 tt996) 267-279

the ventral hippocampus.The lasting modificationsin peptides IR and mRNA expression in distinct neuronal populations of the hippocampus may reflect functional modifications in the respective neurons and play a role in limbic epileptogenesis. Keywords: Limbic system; Hippocam"us; Kindling; Mossy fibres; Status epilepticus; Sprouting; Neurodegeneration; Neuronal plasticity;

Seizures

1. Introduction

Temporal lobe epilepsy (TLE) is a form of limbic epilepsy with a largely unknown etiology and particularly refractory to classical anticonvulsant treatment. Studying the morphological and biochemical modifications in brain tissue that may be significantly involved in the pathogenesis of limbic epilepsy, DeLanerolle et al. [8] found extensive changes in the expression of neuropeptide Y (NPY) and somatostatin in neurons of the hippocampal formation of TLE patients [38]. These changes involve the principal neuronal populations of the hippocampus and are very similar to those occurring in the rat hippocampus after experimental limbic convulsions [2,11,14,18,27,31,35,42-44,46,47,55]. In particular, peptides-immunoreactivity (IR) and their mRNA expression are modified in CA1-CA3 pyramidal neurons and in interneurons of the hippocampus proper after kainic acid (KA)-induced seizures [15,27,47]. The granule cell/mossy fibre system shows an enhanced expression of several peptides, including NPY, after iimbic seizures induced by KA [ 18,42,4'/] or sustained stimulation of the perforant path [44,46]. The fact that these cells do not contain NPY constitutively is of particular interest since it suggests that seizure activity may induce gene expression and potentially alter neurotransmitter phenotypes in neurons [ 11]. In the hilar region, a subset of GABAergic cells containing somatostatin and NPY are highly vulnerable to seizures-induced cell damage [ 18,31,42,47,56], whereas other GABA-containing peptidergic neurons in the po!ymorphic cell layer or close to the stratum granulosum are resistant and show a lasting enhanced expression of somatostatin and NPY after seizures [47]. The changes in peptides expression are likely associated to functional alterations in the respective

neuronal systems [16,17]. Thus, the release of somatostatin and NPY is significantly enhanced in the hippocampus of KA-treated rats [34,54] or during and after kindling [37,53]. Electrophysiological and pharmacological studies suggest that somatostatin and NPY exert excitatory or inhibitory actions on synaptic transmission depending on their site of release [41] and their interaction with different receptor subtypes [5,7,34]. Thus, information on the type of neurons expressing the peptides and where they are released following seizures is important for establishing their functional significance. Various models of limbic seizures have been developed in rodents using electrical stimuli [12,25,36,44] or convulsant drugs [3,48] to mimic as closely as possible the neuropathology of TLE. Among them two experimental conditions, e.g. electrical kindling [12,36] and self-sustaining limbic SE [25] induced by electrical stimulation of the rat hippocampus appear particularly interesting. Thus, seizures involve primarily the hippocampus which seems to be a crucial site of seizure initiation and propagation in humans [3,25]. In particular, kindling allows the study of epileptogenesis at various defined points in its evolution, therefore it may help in understanding the mechanisms involved in the establishment of an epileptic focus or aimed at counteracting it [29]. However, spontaneous seizures and widespread neuropathology commonly associated with human limbic epilepsy are not a common feature of classical kindling whereas they typically occur after self-sustaining lirnbic SE [26,29]. Using immunocytochemistry and in situ hybridization analysis of mRNA, we investigated the hippocampal regions and cell populations where somatostatin and NPY are expressed in these two models of limbic epilepsy associated with different chronic sequelae of seizures and seizure-related neurodegeneration.

A. Vezza~i e~ a L / Epiteps3 Research 26 (1996) 267-279

2, Methods

2.1. Experimental animals Malc-Sprague Dawley rats (250-280 g, Charles River, Italy) were used. The animals were housed at constant temperature (23°C) and relative humidity (60%) with a fixed 12 h light-dark cycle and free access m food and water. Procedures involving animals and their care were conducted in conformity with the institutional guidelines that are in compliance with national and international laws and pNicies (EEC Council Directive 86/609, OJ L 358, 1, 12 December 1987; NIH Guide for the Care and Use of Laboratory Animals, NIH Publication No. 85-23, 1985).

2.2 Kindling procedure The electrodes were implanted in the dorsal hippocampus under equithesin anaesthesia (1% pentebarbital/4% (vol/vol) chloral hydrate) according to the following coordinates (mm) from bregma: nose bar - 2 . 5 below the interaural line, AP - 3 . 5 , L _+ 2.3, H 2.9 below dura. Electroencephalographic (EEG) recordings were made using bilateral cortical and hippocampal electrodes in unanesthetized, freely moving animals as previously described [53]. Constant current stimuli were delivered unilaterally to the dorsal hippocampus through a bipolar electrode (recording electrode) twice daily for five days then once daily at intervals of at least 6 h. The stimulation parameters were 50 Hz, 2 ms monophasic rectangular wave pulses for 1 s, the current intensity ranging between 60 and 200 /.LA. Before electrical stimulation, the rats were randomly assigned to two groups and received 12 +.+_1 and 27 _+ 2.5 stimuli (mean +_ SE) to reach respectively stages 2 (stereotypies, occasional retraction of a forelimb; n = 10) and 5 (tonic-elonic seizures with rearing and falling; n = 14) of kindling according to Racine's classification [36]. Animals were considered fully kindled when they experienced at least three consecutive stage 5 seizures. Controls were implanted with electrodes, but were not electrically stimulated (n = 15). Rats kindled at stages 2 and 5 and the corresponding shams were killed respectively two days or two days and one week after the last electrical stimulation. These intervals were chosen on the basis of previous

269

studies showing mRNA and release changes for soo matostatin and NPY related to kindling-induced plasticity [4,37,53] but not to the recent experience of seizure activity [24]. A different group of animals (n = 6) received a singte stimulation inducing an AD and was killed two or seven days later.

2.3. Self-sustaining limbic status epilepticus A different group of rats (n = 6) underwent 'continuous' hippocampal stimulation as previously described [25,26]. Briefly, animals were implanted under ketamine/xylazine anesthesia with bipolar electrodes in the left, posterior ventral hippocampus (in ram: AP: 3.6 behind the bregma; L: + 4.9; 5.0 below dura; incisor bar + 5.0). Animals were entered into the study only if their AD thresholds were < 250 /xA. The stimulus intensity was then set to 400 p~A to oven'ide postictal refractoriness. Animals were then exposed to a 'continuous' electrical stimulation protocol lasting 90 rain. Subsequently they developed SE that abated within 24 h, as documented by behavioral and electroencephalographic criteria [25,26]. Another consequence of SE induced by 'continuous' hippocampal stimulation are recurrent, spontaneous seizures detected with hippocampal electrodes even more than one month after stimulation [26]. Controls (n = 6) were surgically implanted with hippocampa! electrodes but not stimulated. The animals were sacrificed 30 days after hippocampal stimulation.

2.4. bnmunocytochemisoy Epileptic rats and their respective controls were anesthetized with equithesin and perfused through the ascending aorta with 50 ml phosphate buffered saline (PBS, 50 mM, pH 7.4) followed by 200 ml chilled 4% paraformaldehyde in phosphate buffered saline. The brains were postfixed in the same fixative for 90 min at 4°(2 then transferred to 20% sucrose in PBS for 24 h at 4°C. The brains were then immersed in --70°C isopentane for 3 rain and stored in tightly sealed vials at -70°(2. Coronal (septal) and horizontal (temporal) sections (40 ~ m ) were obtained from the dorsal and ventral hippocampus respectively. Irnmunocytochemistry was performed at the same time for somato-

270

A. Vezzani et al. / Epilepo, Research 26 (1996) 267-279

statin and NPY on adjacent free-floating sections. Incubations were done in two separate experiments for the dorsal and ventral hippocampus. The indirect peroxidase-antiperoxidase technique of Sternberger [49] was used as described in detail before [47]. Primary antisera for NPY and somatostatin were raised in rabbits against the synthetic peptides which had been covalently bound to egg albumin [47]. The antisera were characterized by radioimmunoassays and no cross-reactivity was found with various other neuropeptides [47]. The antisera were used at the following dilutions: NPY 1:1000 and somatostatin 1:1500. In each erpedment controls were prepared using the primary antisera preadsorbed with the respective neuropeptide (5 /xM, 24 h, 4°C) and incubating the slices without the primary antisera.

2.5. h~ situ hybridization Rats kindled at stages 2 (n = 3) and 5 (n = 4) and their respective controls (n = 5) were killed by decapitation and their brains were rapidly removed and frozen in -70°C isopentane. Brain coronal sections (20 /xm) were cut, thaw-mounted on gelatin-coated microscope slides and kept dessicated at -20°C until the in situ hybridization experiment. The procedure of Young [57] was followed as previously described [47]. The following oligo DNA probes were used (custom synthetized by Microsynth, Wiudish, Switzerland): NPY mRNA, a 46-mer oligonucleotide complementary to bases 1686-1731 (46mer) of ppNPY mRNA [22] and somatostatin mRNA, a 48-mer oligonucleotide complementary to bases 1019-1066 of the ppsomatostatin mRNA [51]. Synthetic oligonucleotides (10 pmol) were labeled with [3SS]a-thio-dATP (1300 Ci/mmol, NEN) by reaction with terminal deoxynucleotidyltransferase (Boehringer, Mannheim) and precipitated with ethanol/sodium chloride. At least 3 consecutive sections for each experimental animal and the corresponding control were processed in the same experiment. Controls were prepared by preincubating some sections with excess (1 nmol) unlabeled probe in the same hybridization buffer for 2 h.

2.6. Histological procedures Nissl staining was performed on 40 /zm coronal and horizontal sections from brains prepared for

immunocytochemistry or for Timm's staining using cresyl violet. Timm's staining was performed as previously reported [6] on a separate group of stage 5 kindled rats (n = 3) sacrificed one week after the last generalized convulsion and the con'esponding controis (n = 3). Neuronal cell loss and sprouting were assessed by light microscopy by an investigator unaware of the treatment of the animals. Neuronal cell (20-30 /~m diameter) count was done in the hilus on cresyl violet stained sections adjacent to the electrode track at 200 times magnification.

2.Z Statistical analysis Significant differences in ceil counts in kindling or SSLSE rats versus their respective controls were assessed by two-tailed Student's t-test.

3. Results

3.1. Kindling 3.1.1. Neuropathological changes Light microscopic analysis of Nissl stained sections showed that pyramidal and granule cells were well preserved in fully kindled rats in the dorsal hippocampus at the site of electrical stimulation (Fig. la, b) and contralaterally. Cell counting in the hilus of the dentate gyrus of both hippocampi showed no significant differences at stage 2 compared to shams, whereas a 21 + 1% and 27 + 3% decrease of neurons was found in the stimulated and contralateral hippocampus at stage 5 ( p <0.01 by Student's ttest). No significant sprouting of mossy fibres was observed in the inner molecular layer of the dentate gyrus in stage 5 kindled rats compared to shams as assessed by Timm's staining (Fig. lc, d).

3.1.2. lmmunocytochemical evidence Immunocytochemical analysis was done in the dorsal and ventral hippocampus of stage 2 and 5 kindled rats compared to shams. All changes occurred to a similar extent ipsilaterally and contralaterally to the stimulating electrode and in the dorsal and ventral extension of the hippocampus. Representative sections of the dorsal hippocampus are shown in the figures.

A. Vezzani e~ a L / Epilepsy Research 26 ~t 996~ 267-279

27

+

j ~

J

_

Fig. I. Nissl (A, B) and Timm's (C, D) stained sections of the dorsal hippocampus of sham (A, C) and stage 5 kindled rat (]3+ D). Representative sections ot the stimulated hippocampus are shown in (B) aJ~d(D). No obvious neuropathological cb~anges were observed in CA1 and CA3 sector but neuronal density was lower in the hilus (see text). Timm's stained sections dd not reveal any sprouting of mossy fibre collaterals in the inner molecular layer in kindling (D) compared m sham (C). Scale bar: 100 #m.

Fig. 2 s h o w s the c h a n g e s in somatostatin and N P Y IR in the dorsal h i p p o c a m p u s t w o days after stage 2 and o n e week after stage 5 o f kindling compared to shams. In accordance with previous findings [ 1 , 1 9 , 2 3 , 3 0 , 4 5 ] , somatostatin cell bodies were n u m e r o u s in stratum oriens C A I and C A 3 and in stratum radiatum CA3. A similar pattern o f distri-

bution was f o u n d for N P Y - s t a i n e d cells [9,20] as predicted by their extensive c o - e x i s t e n c e in neurons [21]. In the area dentata, somata stained for both neuropeptides were numerous in the hilus but absent in the granule cells or in the molecular layer o f the dentate gyrus. Various N P Y - i m m u n o r e a c t i v e neurons s [ m i l ~ to type 1 and type 2 basket cells were

+ •

,

Fig. 2. Photomicrographs showing somatostatin- (a-c) and neuropeptide Y (NPY, d-f) in the dorsal hippocampus of sham (a, d), stage 2 (b. e), and stage 5 (c, f) kindled rats. In this and the following figures peptides IR was studied two days and one week after stage 2 and stage 5 of kindling respectively. Somatostatin and NPY-IR was markedly increased in interneurons of the hilus and in the outer molecular layer (arrowheads in b and c) of the dentate gyrus and/or s~ratum lacunosum moleculare (arrows in b and c) of stage 2 (b, e) and stage 5 (c, f) rats. In these animals, somatostatin IR was also visible in the pyramidal cell layer. Note the marked increase of NPY-IR in the terminal field of mossy fibers (hilus and CA3a) at stage 2 (see arrows in e). Scale bar: 500/.tin.

272

A. Vezzani et al. /Epilepsy Research 26 (1996) 267-279

located respectively withir, or close to the inner surface of the granule cell layer (Fig. 3a). Peptidepositive fibres were found mainly in the outer molecular layer and stratum lacunosum moleculare (Fig. 2d). At stages 2 and 5, there was a marked increase in somatostatin IR in hilar interneurons and in their

presumed projections to the outer molecular layer [1,23,30] in the dorsal and ventral hippocampus. Staining was also enhanced in the stratum lacunosum moleculare (Fig. 2b, c). Two clays after completion of kindling, the changes in peptide IR were similar to stage 2 (n = 3 rats; not shown) but were less pronounced after one week (Fig. 2b, c). The NPY immunocytochemical pattern in kindling was similar to that of somatostatin (Fig. 2). However, increased staining was also observed in presumed GABA-containing pyramidal shaped basr..~t ceils in the subgranular region (Fig. 3). NPY was expressed in the mossy fibers terminal field at stage 2 (Fig. 2e and Fig. 3b) and two days (not shown), but not one week after stage 5 (Fig. 2f) indicating that this effect was transient. Somatostatin IR and its mRNA expression were increased in the subiculum of the dorsal l,'ppocampus at stage 2 but no differences were observed in NPY (not shown). After a single AD, somatostatin IR was similar to sham except for a slight activation of somatostatin immunoreactive neurons in the hilus after two days. This effect was less intense that at stage 2 and was not found in the animals killed after one week (not shown). NPY IR was increased in the terminal field of mossy fibres two days but not one week after the electrical stimulus. The increase in staining after two days was indistinguishable from that found at stage 2 (not shown).

3.1.3. In situ hybridization

Fig. 3. High magnification phot0~cro~aphs of NPY-IR in dorsal dentate gyrus of sham (a) and stage 2 (b) kindled rats. Note the increase in NPY-IR in the hilar intemeurons of stage 2 rats (b). NPY positive fibers were intensively stained in the hilus and in the outer molecular layer (b). Note immunostained presumed type 1 basket cells (arrow) below the stratum granulosum (sg, in a and b) and type II basket cells in (a) (double-arrow). Scale bar: 50 /zm.

In accordance with the immunocytochcmical findings, somatostatin and NPY mRNAs wer~ expressed in neurons of the stratum oriens and radiatum in the hippocampus proper and in the hilar interneurons of the dentate gyrus in shams [4,47]. Transcript signal was significantly increased in neurons in the hilus of the dentate gyrus of stages 2 and 5 kindled rats while no signal was detected in granule cells (Fig. 4) or the pyramidal cell layer (not shown). The number of neurons expressing somatostatin mRNA was increased in the subiculum of s:-,ge 2 kindled rats (not shown).

3.2. Self-sustaininig limbic status epilepticus 3.2.1. Neuropathological changes Neuropathological changes in post-SE rats have been extensively described previously [25,26]. These

A. Vezza~i et at/Epilepsy Researck 26 (1996~ 267-279

273

3.2.2. k ~ n w w c v z o c k e m i c J ecide~ue Represer~tative sections of the ventral hippocampus contralaterat to the implanted electrode are shown in each group, in accordance with the neuropathotogical findings, the immtmocytochernical results of the post-SE group were divided into two groups. In four animals, no overt loss in somatostatin IR was found. These animals correspond to SE-I rats. The two other post-SE rats showed extensive loss of somatostatin IR in the hilus of the ventral dentate gyrus and presumed sprouting of mossy fibers [25] as indicated by an enhanced NPY IR in the inner molecular layer. They correspond to SE-II rats. Fig. 6 shows the changes in somatostatin- and NPY IR within the ventral hippocarnpus in controls

rats can be divided into two separate groups accordi~ag to their pattern of neurodegeneration (Fig. 5). Cell counting in the ventral hippocarnpus showed that in four out of six post-SE rats (SE-I group) extensive ,.euronal cell loss occurred in the hilus of the dentate gyms (50 + 0.5% decrease; p < 0.01 by Student's t-test) while a 25 +_ 3% decrease i~ cell density was observed in CA1 pyramidal celt layer ( p < 0.01 by Student's t-test). CA3c pyramidal cells appeared well preserved. In the two other post-SE rats (SE-II group) more severe neuronal cell loss was found in the dentate hilus (75-92%), the CA1 (4143%) and in the CA3c (56-66%) pyramidal layers. No obvious changes in the number of granule cells were observed in either group of animals.

A

'

B

....

.

.

i

Fi g. 4 • Ex Pression of mRNA for somatostatin (a-c) and NPY (d'O in the dorsal hippecampus .of sham 2 (b, e) and stage 5 ~c, ) . . (a, . d), stage 3s kindled rats. High-power bright-field (A) and dark-field (B) photomicrographs of coronal seclmns hybndized with S-labelled ohgcnucleotides. The hybridization signal of somatostatin and NPY increased markedly in hilar neurons. No signal or oniy a minute amount was observed in granule neurons (see d in panel A). Note that the grain density over individual neurons was increased in stage 2 and 5 kindled rats (see b, c e and f in panel A). Scale bar: in (A), 500/zm; in (B), 50/zm.

274

A. Vezze.ni et aL / Epilepsy Research 26 (1996) 267-279

and in both sub-groups of post-SE rats. In SE-! rats somatostatin IR in hilar interneurons was comparable to or less than that of control animals (Fig. 6c) while in SE-II rats somatostatin-iR was virtually lost in the ventral hilus (Fig. 6e). In the dorsal hippocampus of SE-I rats, somatostatin IR was increased in the hilus and outer molecular layer of the dentate gyras similarly to that observed in kindled rats while a reduced IR was observed in SE-II rats (not shown). In SE-I rats, NPY IR was particularly enhanced in the outer molecular layer of the dorsal (not shown) and ventral hippocampus (Fig. 6d). These animals also showed some staining of fibres in the hilus and in the terminal field of mossy fibres. In SE-II rats, no

'

.

.

.

.

'~"~ :"L~.---.'.

' ~

.

~

_

~

' . . . ' . .7::..,~.','.~w,".."~ : ,

.

-

-

,"Try.

';"'~''~";::;':'' " ~'~~'~'22'"""

""e:'¢~"'9~..;

e,

t¢

f

.~."¢:'.

~:'7 "

mr.

%

':

'

Fig. 6. Photomicrographs showing somatostatin- (a. c. e) and NPY-immunoreactivities (b. d. D in the vent~,~i hippocampus of controJs (a, b). SE-| (c. d) and SE-il rats (e. f). Somatostati~ IR is slightly reduced i:~ i.qL~.r~eurons o~" ~hc fiius oi' me dentate gyrt~.,, h~ SE-I rats (c) while it is lost in SE-il rats (e). NPY-IR is enhaiiced in the terminal field of mossy fibers of both SE grouw (d, f) although the staining was more intense in SE-ll rats (l'). Note NPV IR in the inner molecular layer in f (arrov s). Scale b,,rs: 200 p.m.

IAL ?-.'";:" iik~,

"

staining was observed in the dentate molecular layer while intense staining was preserit in the terminal field of mossy fibzes and in the int~er molecular layer of the dentate gyrus (Fig. 6D.

"

,.t

:

•

'

'

',~:,'~,~!,..,..- ?

~

~

"" ' . . .

. -. .

-=...:

"i

,;!<

,

r

4. Discussion

17;......... ': .....

.

"...

"-".-".,~r.a"

Fig. 5. Nissl stained sections of the dorsal hippocampus 30 days after self-sustaining limbic status epilepticus. (a) control; (b) SE-I rat; (c) SE-II rat. Note neuronal cell loss in the hilus and in CA3c pyramidal layer in (b and c) and the loss of CAI neurons in (b). Scale bar: 200 ~m.

This study shows distinct changes in the expression of somatostatin and NPY in various subfields and cell populations of the rat hippocampus in kindling anu after SE. These changes may reflect functional modifications in peptide-containing neurons induced by seizures. Kindling and SE induced in rat~ by electrical hippocampai stimulation differ in many respects. Thus,, the stimulation protocol used in kindling consists of stimuli delivered every six hours over one (stage 2) to four weeks (stage 5) while SE rats received a continuous stimulation for 90 min. While the kindling paradigm evoked periodic epileptiform discharges and behavioral seizures of graded severity

A. Vc .xmi dr ,a/. .,' £/£/~7).,,'~ £~'.~'~;'ci', 20 i/796,* 2 0 7 2~9

and sh¢,rt dura~,io~., cominu(ms hippocampal stirm~lation induced a self-st~stah~mg limbic SE lasting abom 12 h, Sponta~c+aus seizm'cs were ~?o~ observed i~ ki~d!ed ra~s w!~,ilethey o~cc~.prved~or at teas~ 6 mom,s after SE. Fina|ly, seizure-related cell damage and sprouting occur to a difl~mnt extent in ~he ~wo epilepsy models. These findings will be discussed in reiation to: ( 1) the synoptic connectio~:~s of the neurons expressing the pepfides and their co-exismnce with GABA [t0] and glutamate; (2) the effects of the peptides on synoptic transmission :rod on hippocampal excitability. 4. t. Hitar ~teurm~.s

Hilar internemo~>, which mostly contain ~omatc,statin and NPY are intensively stained in the hippoeampus at the preconvulsive stage 2 and after kindling acquisition, This probaNy results from accumulation of the peptide~, after their ~ncrea:;ed ~,,'nthesis, as indicated by the enhanced mRNA expression in the hilar neurons. An intense immunoreactive band of fibres was observed in the outer molecular layer of the dentate gyrus, very likely representing increased peptides in the mrminal field of the der
275

iayer of the dentate ~%,i~_,, and of i>Irap,~idal° ~eurens ir~

~,~ata ;~(iia~.m~

aqd

~actm,n.~:,m~~ molecular:

[l,9,2iL23,30]. Tht~s the peps:ides re!~sased in these hipp(;campai regions !37,53] may serve to control granule and pyramidal cel~ excitabiTity is kindled ~issue.

The post-SE fats showed a he{erogeneous patter, of JR. A slight incret~se in somatostatin and NPY IR was fot!nd in the hilus and outer molecular layer of SE-I rats similar to kindling. Differently, an almost entire loss of hflar son3atoslafin I[R and in the expressio~ of pre-prosomatosmtin mRNA (not shown) was observed in the ventral hippucampus of SE-II rats. These results suggest flint soma~ostatin-containing r~eurons degenerate in SEdl rats while they are relatively spared or activated ia SE-I ;rod kindled rats respectively. ARh~ugh the incidence of spontaneous seizures was not measured in this study, this may explain the dill:rest extent of neurodegeneration and tb..e loss of samatostatm ~P in SE4 ;elau,-, SE-I~ :'~,,,. Indeed, somatostatin neurons of the dentate gyrus have been shown to have a relative;y !~ow threshold for activation [40] and high susceptibility to sustained epileptiform activity [44.46,4:7]. NPY (but not somatostatin) is also present in pyramidal-shaped basket cells close to the hilar surface of the granule cell layer [9]. These neurons are resistant to epileptic damage and become strongly activated in kindling and alter SE as indicated by their increased NPY IR and mRNA expression. These changes were similar to those observed after KA-induced limbic seizures or pentylenetetrazol kindling [2,11,27,47]. 4.2. Granule celt / mossy fibre system

A noticeablc staining of NPY (but not of somatostalin) was observed in the terminal field of mossy fibers and at the border between the CA2 and CA3 sectors in SE-I rats. This staining was more pronounced in SE-II rats. This suggests an increased synthesis of NPY in granule cells (as also h~dicated by the expression of NPY mRNA in granule cells; not shown) and the accumulation of the pepfide in their terminal projections. Similar changes have been previously described after sustained limNc seizures in rats and in TLE [8,11,18,46,47]. In SE-II rats, NPY-IR was also found in the inner molecul~x layer

276

A. Vezzani e~ al./ Epilepsy Reseatz'h 26 ~1996) 267-279

of the dentate gyrus in the ventral area probably representing accumulation of the peptide in mossy fibres which undergo sprouting as previously assessed by Timm's staining [25]. NPY was transiently expressed in granule cells of kindled rats. This effect does not depend on kindling progression since it occurs after one AD but it is more likely a consequence of the direct excitation of granule neurons. Accordingly, the release of NPY but not of somatostatin was significantly enhanced after a single electrical stimtflus [37,53]. After KA, only rats with sustained motor convulsions ( > 3 h) showed a lasting increase of NPY-IR in mossy fibres [14]. This may be related to the acute seizures activity and/or to the recurrent spontaneous epileptic events observed in these animals [25,26]. Thus, it has been reported that maximal dentate activation occurs only in animal with generalized motor seizures [50]. These observations suggest that the lack of motor convulsions at stage 2 or the inductio~ of brief generalized se,zures at stage 5 ( < 1 rain) ma) be associated with milder granule cell activation in kindling than in rats experiencing SE and this may account for the transient expression of NPY in granule ceils. It is noteworthy that the accumulation of NPY in mossy fibres in SE rats was associated with extensive loss of somatostatin IR in the hilus suggesting that sustained seizure activity, as predicted by the damage of hilar neurons, is necessary to induce NPY synthesis in granule cells. 4.3. Hippocampus proper

The expression of somatostatin and NPY was enhanced in the alveus and/o~ in the stratum lacunosum moleculare of kindled rats probably reflecting peptide accumulation in dendrites of interneurons located in the strata oriens and pyramidale. This may represent activation of peptidergic cells that play a role in contiolling the activity of pyramidal amul'oas and their efferent projections as also supported by ultrastructural evidence of symmetric synaptic contacts of peptide-positive boutons with perikarya and dendrites of pyramidal neurons [1,9,20,23,30]. Increased somatostatin IR and its mRNA was found in the subiculum of kindled rats. This is particularly interesting considering that the fibres connecting the hippocampos to the enthorinal cortex pass through it.

No significant changes in peptide-lR were found in the hippocampus proper of p,~st-SE rats. 4.4. Functional consequences

It appears that peptide expression in the hippocampus is affected differently depending on how seizures are triggered and on the outcome of the chronic sequelae following the acute seizures. Two main differences between kindling and SE are noteworthy, (1) Peptide expression is increased in presumed inhibitory hilar neurons and their projections to the molecular layer of tt~]e dentate gyrus in kindling while these neurons are less intensively stained (or even lost) after SE. The activation in kindled rats of hilar (inhibitory) peptidergic neurons is supported by the iinding of an enhanced release of somatostatin [37] and NPY [53] in the hippocampus of kindled rats and may contribute to lowering the granule cells excitability. It is tempting to speculate that the degeneration of peptide-containing neurons after SE, as suggested by the loss of IR and their mRNA expression, may facilitate seizure spread contributing to the occurrence of spontaneous convulsions [25,26]. Somatostatin has recently been shown to have anticonvulsant effects when applied to the hippocampus [28,M,32] and to exert a tonic inhibitory action on kindling epileptogenesis [33]. The enhanced peptide IR in the hippocampus proper (alveus, subiculum and stratmn lacunosum moleculare) of kindled rats may play a role in controlling the activity of pyramidal neurons and their efferent projections. (2) A lasting expression of NPY is found in the granule cell/mossy fibre system after SE but not after kindling. The peptide is present in granule cells also in KA-treated rats [47]. This indicates that the enhanced release of the peptide in kindling may occur at different synaptic sites compared to SE and consequently affect hippocanapal excitability differently. Thus, NPY has inhibitory or excitatory effects in the hippocampus depending on the type of receptors it stimulates and on the site of its release [5,7,13,32]. The Y2 receptor subtypes located on mossy fibres are increased after KA [39]. This suggests that NPY-mediated neurotransmission is enhanced through Y2 receptors which are known to mediate the inhibitory action of NPY on glutamatergic neurotransmission [7,13]. However, the infusion of a selec-

A. Ve=zani e~ aL ,/£))iiepsy ¢¢eveo,ch 2~5(198%) 267-279

live antibody against. NPY m the dentate gyrus has been reported to lower chronic seizure ~aaceDibili~y to pentylene~etrazol in KAqreated rats [541 suggest, ing a tacilitatory effect of the peptide on generalized seizures. Therefore, the functional significance of the presence of NPY in mossy fibres and the duration of its expression in granule cells remains to be clarified. This evidence together with phm'macological and electrophysiological findings indicates that changes in peptide expression after an epileptic stimulus are assocmted with lasting functional alterations :n peptidergic neurotransmission in the hippocampus as also demonstrated by peptide release and receptor expression studies. The study of the functional consequences of the activation of peptidergic neurons in models of ]imbic epilepsy leading to different chronic sequelae will add important information on their possible role in the establishment of a chronic epileptic focus.

Acknowledgements Mr. E. Kirchmair and Mr. C. Trawoger are gratefully acknowledged. This work was supposed by the National Research Council (CNR, Rome, italy: Corn venzione Psicofarmacologia and Contract No. 93.00876. CT04), by a grant m A.V. from Sandoz Ltd. Basel, and by the Austrian Scientific Research Funds.

References [1] Bakst, I., Avendano, C., Morrison, J.H. and Amaral, D.G., An experimental analysis of the origins of somatostatin-like immunoreacti'vity in the dentate gyms of the rat, .L Neurosci., 6 (1986) 1452-1462. [2] Bei!mann, R, Widmann, R., Olenik, C., Meyer, D.K., Maas, D., Marksteiner, 1. and Sperk G., Enha_,ced rate of expression and biosynthesis of neuropeptide Y after kair,'ic acid induced seizures, J. Nettrochem., 56 (1991) 525-530. [3] Ben-Aft, Y., Limbic seizures and brain damage produced by kainic acid: Mechanisms and relevance to human temporal iobe epilepsy, Neuroscience, 14 (1985) 375-403. [4] Beadotti, C., Vezzani, A., Tarizzo, G. and Samanin, R., Increased expression of GAP-43, somatostatin and neumpeptide Y mRNA in the hippocampus during development of

277

hippocampM kiad~i~g i~ rats, K,r. J. /V~,~e,.}ci., 5 (1993) ~312-]320. [5] Brooks, P.A., Kelly, .I.S., A!!en, J.M., Smith, D.A.S. and Sto~e, T.W.. Direct excilamry effects of neuropeptide Y {NPY) on ra~ hippocampal aeam~s in vim~. Broi~ Re~., 4(}8 { !'~7} 295-298. [6] Cavazos, LE., Golarai, G. and Sutu~a, T.P., Mossy fibre synaptic reorganization induced by kindling: Time course of development, progression, and permane~ce, Z Neur,*sci., 11 ( 199 ~) 2795-2803. [7t Colmers, W.F., Klapstein, G.J.. Fournier, A., St-Pierre. S. and Treherne, K.A.. Presynaptic hlhibition by neuropepfide Y in rat hippocampJ slice in vitro is mediated by a Y, receptor, Br. J. PharmacoL, 102 (]99~) 41-Z4. I81 DeLanerolle, N.C., Kim, J.H. Rubbings, R.J. and Spencer, D.D., Hippocampal interneuron loss and plasticity in 7 umaa temporal lobe epiBepsy, Brain Res., 495 (R989) 387-395. [9] De]ler, T. and Leranth. C., Synaptic covmections of neuropepitide Y (NPY) immunorcactive neurons tn the hilar ~ea of ~he rat hippocam~,us, J. Cutup. Ne,rol., 3tD (19~am 433447. [10] Esclapez, M. and Houser, C.R., Somatostatin neurons are a subpopulafion of GABA neur, ns in the rat dentate gyrus: Evidence from colocalizatio_n of pre-prosomatosmtin and glutamate decarboxylase messenger RNAs, Neuro:cience, 64 (1995) 339-355. [11] Gall, C., Lautcrborn, J., lsackson. P. and White, J.. Seizures. neuropeptide regulation, and mRNA expression m tile hippocampus. Prog. Brain Res.. 83 (1990) 371-390. [12] Goddard, G.V., Mclntyre, D.C. and Leech, C K., A permanent change in brain function resulting from ,daily electrical stimulation, EWI. Ne,n,l.. 25 (1969) 295-3,~0. [13] Greber, S., Schwarzer. C. and Sperk, G.. Ne0_ropeptide Y inhibits potassium-stimulated glu,tamate release through Y, receptors in rat hippocampaJ siices m vitro. Br. J. Ph,trmacoL, i13 {1994) 737-740. [141 Gruber, B,, Greber, S., Rupp, E. and Sperk, G,, Differential NPY mRNA expression in granule cells and imerneurons of the rat dentate gyms alter kainic acid, Hippocampus, 4 (! 994) i -9. [15] I-iashimoto, T, and Obata, K., Induction of somatostatin by kainic acid in pyramidal and granule cells of the rat hippocampus, Neurosci. l/ett., 12 (1991) 514-527. [161 H6kfelt, T., Johansson. O,, Ljungdahl, A., Lundberg, J.M. and Schultzberg, M., Peptidergic neurones, Nat, re, 284 (1980) 515-521. [17] HiJkfelt, T., Neuropeptides in perspective: The last ten years, Neuron, 7 (1991) 867-879. [18] Johansen, F.F., Tonder, N., Bruhn. T. and Diemer, N.H., Neuropeptide Y immunoreactivity is lost in rat hippocam ms after cerebral ischemia and kainic acid seizures, /z,r 3. Neurosci., Suppl. 2 (198'9) 13. [19] KiJhler, C., Chan-Patay, V., Somatostati,-like immunoreactire neurons in ~rhe hippacampus: An immunocymchemical study in the rat, Neurosci. Lett., 34 (1982) 259-264. [20] Kiihler, C., Eriksson. L.. Davies, S. and Chan-Palay, V.,

278

A. Vezzani et al. / Epilepsy Research 26 (1996) 267-279

Neuropeptide Y innervation of the hippocampai region in the rat and monkey brain, J. Comp. Neurol., 244 (1986) 384-400. [211 Ktihler, C., Eriksson, L.G., Davies, S. and Chan-Palay, V., Co-localization of neuropeptide tyrosine and somatostatin immunoreactivity in neurons of individual subfields of the rat hippocampai region, Neurosci. Lett., 78 (1987) 1-6. [22] Larhammar, t)., Ericsson, A. and Persson, H., Structure and expression of the rat neur~t, eptide Y gene, P,'oc. Natl. Acad. Sci. USA, 84 (1987) 2068-2072. [23] Leranth, C., Malcolm, AJ, and Frotscher, M., Afferent and efferent synaptic connections of somatostatin-immunoreactire neurons in the rat fascia dentata, J. Comp. Neurol., 295 (1990) I 11-122. [24] Leung, L.S., Spontaneous tfippocampal interictal spikes following local kindling: Time course of changes and relation to behavioral seizures, Brain Res., 513 (1990) 308-314. [25] Lothman, E.W., Bertram, IlL E.H. and Stringer, J.L., Functional anatomy of hippocampal seizures, Prog. Neurobiol., 37 (1991)!-81. [26] Lothman, E.W. and Bertram, Iil, E.H,, Epileptogenic effects of status epilepticus, Epilepsia, 3£1 (1993) $59-$70. [27] Manfridi, A., Forioni, G.L., Vezzani, A., Fodritto. F., De Situoni, M.G., Functional and histological consequences of quinolinic and kainic acid-induced seizures on hippocampal somatostatin neurons, IVeuro~'ciente, d t ( 1991 ) ! 27- i 35. [2/,] Mazaran, A.M., and Ieledgy, G., Etfec~:~ of somatostatin and anti-somatostatin serum on picrotoxin-kindled seizures, Neuropharmaeology, 31 (1992) 793-797. [29] McNamara, J.O., Kindling: An animal model of comlAex partial epilepsy, Ann. Neurol., 16 (Suppl.) (1~)84) $72-$76. [30] Milner, T.A., Bacon, C.E., Ultrastructural localization of somatostatin-like immunoreactivity in the rat dentate gyrus, J. eomp. Ne,rol.. 290 (1989)544-560. [3 I] Mitchell, J., Gatherer, M. and Sundstrom, L.E., Loss of hilar somatostatin neurons following tetanus [oxin-induced seizures, Aem Neuropathol., 89 (1995) 425-430. [32] Monnet, F.P., Fournier, A., Debonnel, G. and de Montigny, C., Neuropeptide Y potentiate,,; response in the CA3 dorsal hippocampus. I. Involvement of an atypical neuropeptide Y receptor, J. Pharmacol. Exp. Ther., 263 (1992) 1212-1218. [33] Monno, A., Rizzi, M., Samanin, R., and Vezzani, A., Antisomatostatin antibody enhances the rate of hippocampal kindling in rats, Brain Res., 602 (1993) 148-152. [34] Per~z, .L, Vezzani, A. Civenni, G., Tutka, P., Rizzi, M., Schupbach, E. and Hoyer, D., Functional effects of D-Phe-c(Cys-Tyr-D-trp-Lys-VaI-Cys)-Trp-NH 2 and differential changes in somatostatin receptor messenger RNAs, binding sites and somatostatin release in kainic acid-treated rats, Neuroscience, 65 (1995) 1087-1097. [35] Pitkanen, A., Sirvio, J., Jolkkonen, J. and Riekkinen, P., Somatostatin-like immunoreactivity and somatostatin receptor binding in rat brain before and after pentylenetetrazole-induced convulsion, Neuropeptides, 7 (1986) 63-71. [36] Racine, R.J., Modification of seizure activity by electrical stimulatir~n. It. Motor seizure, Electroenceph. clin. Neurophysiol. 32 (1972) 281-294, [37] Rizzi, IVi., Monno, A., Samanin, R., Sperk, G. and Vezzani,

A., Electrical kindling of ~he hippocam~3us ;s ~,:sociated ~ ith functional activation of neuropeptide Y-cot,t~init~g newons. Eur. J. Neurosci., 5 (1993) 1534-1538. [38] Ro0bins, R.],. Brines, M.L., 7,,i,~n¢ LH., Adrian, 1., deLanerolle, N., Welsh, M.S. und 50enter. D.D., A sc]ccti~.e loss of somatostatin in the hippocampas of patients with temporal lobe epik, psy~ Ann. Neur~,l., 29 (1991) 325~332. [39] Roder, C., Schwarzer, C., Vezzani, A, Gobbi, M., Mennini, T. and Sperk, G., Auloradiographic analysis of neuropeptide Y bindin- sites in ~he rat hippocampu~ afte~ kainic acid-induced limbie seizures, Netovscience, in press. [40] Scharfman, H.E. and Schwartzkroin, P.A., Responses of eelis of the rat fascia dentata to prolonged stimulation of the pefforant path: Sensitivity of hilar cells and ei'~(nges in granule cell exe~.tabifity, Neu~oscieme, 35 (1990) 491--5('~. [4i] Scharfinan, H.E., Presynaptic and postsynaptic action, s of somatostatin in a~ea CA I and the dentate gyms o~" lot and rabbit hippocampal slices. In: T.V. Dunwiddie and D.M. Lovinger (Eds.). Presynaptic Receptors in the Mamma/ta~ Brain, Birkhauser, Boston, 191~30pp. 42-70. [42] Schwarzer, C., Williamson, 3.M., Lothman, E.W, Vezzani A. and Sperk, G., Somatostatin. neurope~jtide Y. neurokinin B and choleeystokinil) immunoreactivity in two chronic models of temporal lobe epilepsy, Neuroscience, 69 (~995) 831845. [43] Shinoda, H., Nadi, N.S., Schwartz, J.P., Alterations i.~ somatostatin and proenkephalin mRNA in response to a single amygdaloid stimulation versus kindling, Mol. Bruin Res.~ l I (1991) 221-226. [44] Sloviter, R.S., Decreased hippocampal inhibition and a selective loss of interneurons in experimental epilepsy, Science, 235 (1987) 73-76. [451 Sloviler, R.S. and Niklver, G., hnmu1~ocytochemical localization of GABA- cholecystokinin-, vasoactive intestinal polypepfide-, and somatostatin-like immunoreactivity in the area dentata and hippocampus of the rat, J. comp. Neurol., 256 (1987) 42-60, [46] Sloviter, R.S. and Lowenstein, D.H., Heat shock protein expression in vulnerable cells of the rat hippocampus as an indicator of excitation-induced neuronal stress, J. Neurosci., 12 (1992) 3004-3009. [47] Sperk, G., Marksteiner, J., Gruber, B., Bellmann, R.. Mahata, M, and Ortler, M., Functional changes in neuropeptide Y and somatostatin containing neurons induced by limbic seizures in the rat, Neuroscience, 50 (1992) 831-846. [48] Sperk, G., Kainie acid seizures in the rat, Prog. Neurobiol., 42 (1994) 1-32. [49] Sternberger, L., Immunohistochemlsto,, 2nd ed., Wiley, New York, 1979. [50] Stringer, J. anO Lothman, E.W., Bilateral maximal dentate activation is critical tbr the appearance of an after-discharge in the dentate gyms, Neuroscience, 46 (1992) 309-314. [51] Tavianini, M.A., Hayes. T.E., Magazin, M.D., Mintb, C.D. and Dixon, J.E., Isolation, characterization, and DNA sequence of the rat somatostatin gene, J. biol. Chem., 259 (198,:1.) 11798-11803.

[52] Vezzani, A., Serafini, R., Stasi, M.A., Vigano', G., Rizzi, M.

A, Ve;7~mi e', al. ,/gFih, ps3 Research 26 ¢19961 267-279 and Samanin R , Soma~ostath~ m~d i[s oclapeptid¢ anak:~g SMS 20!-995 n~<~d~!iat¢ seizt~res h~duced by quh~oiirUc and kaiaic acids differei~tly in ~he ~at htpD~campu~, Ne~:roplmrmac(dog_< 3{t { 10~;1) ~,45-352. [53] Vezz:mi, A., Momva, A., Rizzi. M., Ga!Ji, A., Barrios, M. a~d Samanin, R., So:namstatia rdea;e [,~; m~hanced in the hippocampus of partially and fu!iy kindled ra!s, Neuroscience, 5 ~ ([992) 41-46. [541 Vezzam, A., Civenm, G., Rizzi, M , Mormo, A , Messali, S. and Samani~. R., Enhanced neuropepUd,- Y ~dease in the hippocamptt~ is associated whh chrvmic ,,eizure suscepffbi!ity in kainic acid trealed rats. Brain [ges., 6,e:l (1994) t 38- !43.

279

[55] Wanscber, B . Kragh~ J., Ban-y, D.L, Bolwig. T. and ZJmmer, J., |ncreased somatosl, afin and enkep~-~aiin-iikc immunoreactiviLy in 1he ra~ hippoca~~pus }bi[owh~g hippocampal kindling, NeuroagL keN., I !8 (~990) 33-.36. [56] Y*inen, A., Valjakka, A., Lahth~en. H,, Mieuine..'u R,, Freund, T.F. a~d RieM:men, P., Vigaba~.rin pre-treatmem prevents hi]ar somamstalin cell loss and the development of inteficta] spiking activity fo|]owing sustained stimulation of ~he pertbrant pa~_b, Neuropepzides, 19 (K99I) 205-211. [57] Young, W.S. Ilt, ~n situ hybridization. Hisfochemical de~ec~itm of nemopeptide mRNA using DNA and RNA probcs, Meghods EnzymoL, 168 (~989) 702-709.