Ventral and dorsal striatal cholinergic neurons have different sensitivities to kainic acid

~ Pergamon

Neurochern. Int. Vol. 31, No. 5, pp. 723-730, 1997 © 1997 ElsevierScienceLtd PII: S0197-0186(97)00003-X Printed in Great Britain. All rights reserved 01974)186/97 $17.00+0.00

VENTRAL A N D DORSAL STRIATAL CHOLINERGIC N E U R O N S HAVE D I F F E R E N T SENSITIVITIES TO KAINIC ACID B E A T R I Z H. G U E V A R A , I R E N E S. H O F F M A N N and L U I G I X. C U B E D D U * Neuropharmacology Unit, Department of Pharmacology, School of Pharmacy, Central University of Venezuela, Caracas, Venezuela (Received 19 August 1996) Abstract--The present study was conducted to investigate the sensitivity of the cholinergic elements of ventral and dorsal striatal regions of the rat brain to the neurotoxin kainic acid (KA). Cholinergic activity was assessed by determining choline-acetyltransferase activity (CAT) and by measurements of acetylcholine (Ach) release from slices prelabeled with [3H]-choline. Direct stereotaxic injections of high-dose KA (4#g/2#1) into specific brain regions, reduced CAT in caudate putamen (CP) by 91+ 1%, in nucleus accumbens (Nac) by 71 ___6%, but CAT in the olfactory tubercle (OT) was not affected by KA. The effects of KA on CP CAT were dose- and volume-dependent. In the OT, KA failed to affect CAT at low, moderate or high doses. Slices obtained from CP injected with KA (3 days prior) showed a 90% reduction in the electrically evoked release of [3HI-transmitter release; however, KA had no effect on transmitter release from OT. These results indicate that KA spares the cholinergic elements of the OT, and reveal the existence of marked differences in excitotoxic action of KA between ventral and dorsal striatal regions and among regions of the ventral striatum. Kainic acid preferentially damages neuronal cell bodies, dendrites and terminals intrinsic within the structures injected, with little or no effect on afferent axons and terminal boutons. Therefore, we propose that most of the Ach present in the OT may be within afferent axons and axon terminals. In the CP and NAc, KA lesions reflect loss of intrinsic cholinergic neurons. In addition, variable levels of excitatory inputs and of excitatory receptors, of the mechanisms available to reduce elevated intracellular calcium concentrations and of the levels of free-radical scavenging resources, also could account for the differences in KA neurotoxicty between OT and CP. © 1997 Elsevier Science Ltd

(Moore and Bloom, 1978; M c G e e r et al., 1982; Phelps and Vaughn, 1986). However, the cholinergic neurons of both striatal regions seem to have different neurochemical and anatomical characteristics. M o s t C A T containing neurons of the ventral striatum are of a smaller size than those found in the CP (Phelps and Vaughn, 1986). In addition, Ach turnover and release from the CP are reduced markedly by treatment with dopamine (DA) and D A agonists (Scatton, 1982; Hertting et al., 1980; Cubeddu et al., 1983); whereas these agents induce negligible inhibitory effects on cholinergic function in ventral striatal regions (Consolo et al., 1977; Bianchi et al., 1979; Salama et al., 1986; Suarez-Roca et al., 1987; Guevara et al., 1996). Anatomically, almost all Ach in the dorsal striatum seems to be within interneurons. Histochemical and immunocytochemical studies with monoclonal anti*To whom all correspondence should be addressed at: bodies to C A T suggested that, in addition to interEsquina Avenida Andres Bello y las Acacias, Res. Jardin Florida, Torre B, Apto 11 A, La Florida, Caracas, Vene- neurons and to cell bodies of projecting neurons, the ventral striatum possesses abundant incoming cholzuela. Tel. and fax: 011-582-746647. 723

The part of the forebrain that is ventral and medial to the caudate putamen (CP) has been named as the ventral striatum. Cytological and histochemical observations, as well as neuronal connectivity studies, provided the basis for this subdivision (Heimer and Wilson, 1975; Heimer, 1978; Phelps and Vaughn, 1986). The ventral striatum appears to be a brain region where the limbic and striatal systems could interact directly (Newman and Winans, 1980). The nucleus accumbens (NAc) and the olfactory tubercle (OT) are important parts of the ventral striatum (Switzer et al., 1985; Heimer et al., 1982). Both the dorsal striatum (CP) and the ventral striatum regions have high acetylcholine (Ach), choline-acetyltransferase (CAT) and acetylcholinesterase concentrations

724

B.H. Guevara et al.

inergic nerve terminals (McGeer et al., 1982; Shute and Lewis, 1967; Cuello and Sofroniew, 1984; Phelps and Vaughn, 1986). The exact proportion of the different cholinergic elements on each of the regions composing the ventral striatum has not been established. The present study was designed to provide additional information on the neurochemical characteristics of the cholinergic neurons of the ventral striatum, with focus on the OT, by studying their sensitivity to the neurotoxin, kainic acid (KA). This excitotoxic amino acid preferentially destroys dendrites and pericarya, sparing axons of passage and afferent nerve terminals (see Coyle, 1983, for review). In this work, we compared the sensitivity of the OT, NAc and CP-cholinergic elements to KA. The effects of KA on cholinergic neurons were assessed by performing measurements of CAT activity, a well-established marker for cholinergic structures, combined with measurements of Ach release from superfused slice preparations. The basal and the electrically stimulated release of Ach from KA-treated and untreated OT and CP regions were compared.

METHODS

Kainic acid injections

Sprague-Dawley rats (250-280 g) were used in all the experiments. All stereotaxic surgery was performed under intraperitoneal sodium thiopental (50 mg/kg) anesthesia. The K A was dissolved in 0.1 M phosphate-buffered saline (PBS, vehicle), final pH 7.4, and always was prepared fresh. In most experiments, a similar volume of vehicle was injected in the contralateral structure. The following stereotaxic coordinates (in mm) from bregma (according to Pellegrino et al., 1981) were employed for injection of vehicle or KA: for CP: A = 2 . 0 , L=3.0, H = 5.5; for OT: A = 3.6, L=2.5, H=9.2; for NAc: A = 3 . 2 , L=2.0, H=7.5; and for globus pallidus (GP): A = 0.8, L = 3.2, H = 6.9. The needle was removed 2 min after the end of the injection. CA T activity

The animals were sacrificed 3 or 7 days after the injection of vehicle or KA, and the structures removed and processed for CAT, as described by Fonnum (1975). Brain structures were dissected and homogenized (1:40, w/v) in 50mM Tris buffer, Na2EDTA 5 mM, pH 7.6. The homogenates were activated with Triton X-100 (0.5% final concentration) to ensure total release of enzyme activity. Samples were cen-

trifuged at 10 000g for 15 min, and 15 ktl of the supernatant (CAT) were employed per tube. Each tissue sample was processed in duplicate. The assay consisted in incubating [3H]-acetyl-CoA (SA, 200mCi/ mmol; New England Nuclear Corp., Boston, MA, U.S.A.), choline, physostigmine and EDTA in 50 mM buffer phosphate (pH 7.4). The labeled acetylCoA was diluted with the unlabeled compound. The following final concentrations were employed: [3H] acetyl-CoA (50 000 dpm/tube), 0.4mM acetyl CoA, 300raM NaC1, 50mM sodium phosphate buffer (pH 7.4), 8 mM choline bromide, 20 mM Na2EDTA (pH 7.4) and l mM physostigmine. The tubes were incubated for 20 min at 37"C. The reaction was stopped by placing the tubes on ice-cold water, and by the addition of a 10 x larger volume (300 #1) of 50 mM cold buffer phosphate, pH 7.4. Next, the formed product (Ach) was isolated by liquid cation-exchange: 300pl of sodium tetraphenylboron (10mg/ml) dissolved in 3heptanone were added to each tube. The samples were mixed and centrifuged. The product was extracted directly into the biphasic aqueous:toluene scintillation mixture. This was achieved by adding 200 pl of the organic phase supernatant to scintillation vials containing 10ml of the scintillation mixture. The CAT activity was expressed in nmoles of substrate formed/h/mg of tissue or as nmoles/h/mg of protein. Proteins were determined by the method of Lowry et al. (1951). The Ach release

In release experiments, the animals were sacrificed by decapitation. The brains were removed rapidly and the OT and CP subsequently were dissected on ice. The OTs were dissected using as landmarks the diagonal band of Broca (caudal and medial limit), the beginning of the olfactory tract (rostral limit), the lateral olfactory tract (lateral limit) and the medial forebrain bundle (dorsal limit). The CP was dissected as described by Starke et al. (1978). Next, the tissues were chopped into 0.4-mm slices by means of a McI1wain tissue chopper. The OT and CP slices were incubated separately for 30min at 37°C in 4ml of superfusion medium containing 0.1 #M of [3H]-choline (S.A., 80 Ci/mmol). The composition of the superfusion medium was as follows (mM): NaC1, 118; KCI, 4.8; CaC12, 1.3; MgSO4, 1.2; NaHCO3, 25; KH2PO4, 1.2; disodium EDTA, 0.03; ascorbic acid, 0.57 and glucose, 11. The medium was gassed continually with 95% 02-5% CO2 and a pH of 7.4 was maintained. After incubation (labeling), the slices were transferred to superfusion chambers and positioned between two platinum electrodes. The slices were

Cholinergic neurons and kainic acid washed with pre-warmed superfusion medium containing 10#m hemicholinium-3 throughout the experiment, at a rate of 0.5 ml/min. Hemicholinium was used to prevent the re-uptake of [3H]-choline. Sample collection started after 75 min of superfusion with drug-free medium (other than hemicholinium); 5 min samples were collected. In each experiment, the slices were subjected to two similar periods of field stimulation (S 1 and $2) after 80 (S 1) and after 120 min ($2) of superfusion. Field stimulations consisted of 120 supramaximal, unipolar, rectangular pulses (20mA, 2ms) delivered at a frequency of 0.3 Hz. At the end of the superfusion, the amount of radioactivity present in the slices was quantitated. The radioactivity in the superfusate samples and in the tissue slices were determined by liquid scintillation spectrometry. The basal [3H] efflux was expressed either in absolute amounts (nCi/5 min sample) or as the fractional rate of loss from the tissue using the ratio [3H] in the medium sample/[3H] in the tissue at the beginning of sample collection. The stimulation-evoked release of transmitter was calculated by subtracting the prestimulation values from the stimulation and poststimulation samples (three samples), and expressed either in absolute amounts (nCi) or as the percentage of the amount of radioactivity present in tissue at the moment of stimulation.

725

the CP and OT, and lowest activity in the GP (Fig. 1). Part of the CAT present in the GP may derive from cholinergic neurons present in the Ch4 region (nucleus basalis, substantia innominata, part of the vertical and horizontal limb of the diagonal band); since in the rat, the Ch4 is not well differentiated from the GP (Mesulam et al., 1983). No differences in CAT activity were observed between the left and right structures; i.e. between left and fight CP (data not shown). Before studying the effects of KA, control experiments were conducted to determine whether direct stereotaxic injection of the vehicle (PBS) altered CAT activity. In these experiments, the vehicle solution was injected one side (i.e. left OT), and a sham operation with placement of needle was performed on the contralateral side (i.e. fight OT). The CAT activity present in the injected side was compared with the activity present in the contralateral, sham-injected side. The following volumes of vehicle were tested: 0.5, 1, 2, 4 and 8 #1. No effects on OT and CP CAT activity were observed 3 days after injection of 0.5, 1, 2 or 4#1 of vehicle. However, unspecific damage of CAT-containing elements was produced by injections of 8 #1 of vehicle; namely, CAT in OT and CP were reduced by 50 + 4 and 41 _ 4%, respectively.

Statistics 30 --

Results were expressed as means _ SEM; n indicates the number of observations. The statistical significance of the differences of two means was analyzed by Student's or paired t-tests. When several means were compared, an ANOVA test was followed by a Duncan's test.

5__

Z. -

20

< MATERIALS

The substances used in this study and their respective sources were: [3H]-choline (specific activity, 80Ci/mmol) and [3H]-acetylCoA (specific activity: 200mCi/mmol) from New England Nuclear Corp. (North Billerica, MA, U.S.A.). Hemicolinium, choline bromide, physostigmine, sodium tetraphenylboron, 3heptanone and kainic acid from Sigma Chemical Co. (St Louis, MO, U.S.A.). RESULTS

CA T activity in CP (dorsal striatum) and 0 T (ventral striatum) : effects of stereotaxic injection of KA The CAT was measured in the OT, CP, NAc and GP of the rat. Highest activities were encountered in

10 e.

CP GP NAc OT Fig. 1. The CAT activity in caudate putamen, nucleus accumbens, globus pallidus and olfactory tubercle of the rat: effects of KA. Kainic acid was injected stereotaxically either into the right CP, globus paUidus (GP), NAc or OT (4 #g/2 #1); the contralateral side received 2 #1 of vehicle (phosphatebuffered saline). Each rat received only two injections: one of KA (i.e. right CP) and one of vehicle (i.e. left CP). The animals were sacrificed 3 days after KA injection. The CAT activity was determined as described in the Methods section. Results were expressed as nmoles of product/h/mg tissue. Shown are mean values+ SEM of at least six observations per group; *p<0.001. Empty bars, vehicle; stripped bars, KA.

B.H. Guevara et al.

726

The effects of K A injections were studied employing a similar experimental protocol to that described for vehicle injections. When K A was injected on the right OT, the left O T received a similar volume of vehicle (PBS), and vice versa; a similar procedure was employed for each of the structures analyzed. C A T activity present in the KA-injected side was compared with the enzymatic activity present in the vehicleinjected side. The K A injections reduced markedly the C A T activity in CP, N A c and GP, but had no effect on C A T activity in the OT (Table 1 and Fig. 1). The K A effects on CP C A T activity were volume- and dose-dependent. For example, C A T activity in CP was inhibited by 7 2 + 4 , 8 0 + 4 and 5 9 + 8 % when 2/~g (9 nmoles) of K A were injected dissolved in 1, 2 and 4pl of vehicle, respectively (Table 1). Injections of lower doses (2.3 nmoles) and volumes (0.5 #g KA/1 pl) into CP, produced lesser effects (49 + 5% reduction in C A T activity). Higher doses of K A (4 /~g/2 /A), produced greatest reductions in CP C A T ( 9 1 + 1 % ) (Table 1). In the OT, on the other hand, K A had no effect on C A T (Table 1 and Fig. 1). Injections of lower (0.5 #g) or higher (4/~g) doses of KA, dissolved either in low (0.5/~1) or high volumes (4/A), had no effect on C A T activity. Doses and volumes of K A that reduced C P - C A T activity by 90% had no effect on O T - C A T activity, irrespectively of whether enzymatic activity was expressed as nmoles/h/mg tissue or as nmoles/mg of protein (data not shown). The K A injection into GP or N A c reduced C A T activity by 89 and 71%. respectively (Fig. 1).

Table 1. Sensitivityto kainic acid of CAT-containingelementsof CP and OT Kainic acid Volume (~1) Dose (ltg) 1 1 2 4 2

0.5 2.0 2.0 2.0 4.0

CAT activity CP OT -49+5* -72±4* -80±4* -59±8* -91±1"

+1+3 -3±3 +3±4 +2+4 --1±3

Kainic acid (KA) was injectedinto OT or CP 3 days prior to sacrifice. The contralateral region receivedan injectionof a similarvolume of the vehicle(i.e. right OT receivedKA, left OT receivedvehicle). The effects of three different volumes and doses of KA were tested. Both injectionvolumeand kainicacid concentration determine CAT reductions in activity. The CAT activity in vehicletreated CP and OT averaged 25.5_+1.8 and 24.6_+1.5nmoles/ hr/mg tissue, respectively.Shown are mean values± SEM for the percentage of change in CAT activityinduced by KA with respect to the vehicle-treatedside. *Significantlydifferent from vehicleside at p<0.001 (n=at least six animals per group).

A ch release from CP (dorsal striatum) and 0 T (ventral striatum) of the rat. effects of stereotaxic injection of KA In addition to studying the sensitivity of CAT-containing neurons of dorsal and ventral striatal structures, we investigated the differential effects of K A on Ach release from two of the regions, the CP and the OT. The Ach release was studied as a measure of cholinergic function, and it was assessed from CP and O T slices which were prelabeled with 3H-choline and superfused, according to a well-established experimental protocol (Hertting et al., 1980; James and Cubeddu, 1987). Slices of untreated O T and CP exhibit comparable degrees of labeling (amount of [3H] present per slice after incubation with [3H]-choline, see below), of tritium retention (amount of [3H] present in the slices at the end of the superfusion experiment, Table 2), and comparable increases in transmitter release evoked by electrical stimulation (0.3 Hz, 120pulses) (Fig. 2 and Table 2). In control slices, [3H] release evoked by the second period of stimulation ($2) was close to that elicited by the first period of stimulation (S1) ($2/S1 ratios equal to or greater than 0.90), indicating good viability of the preparation (Fig. 2 and Table 2).

The e[']ects ~] KA on Ach release jrom CP and 0 T are shown in Fig. 2 and Table 2 Slices obtained from CP, which had been treated with K A 3 days prior to the release experiment, had reduced amounts of radioactivity at the end of the superfusion experiment (70_+ 7% decrease) (Table 2) and showed a marked reduction in the amount of 3Htransmitter release evoked by electrical stimulation (90_+ 4% reduction) (Table 2 and Fig. 2), compared to vehicle-treated slices. Both the absolute amount of [3H] released (in nCi) and the fraction or % of tissue [3H] released by electrical stimulation were markedly reduced by K A treatment. Basal release of [3H] (release in the absence of electrical stimulation) also was reduced by K A (Table 2). In a separate group of experiments, the amount of radioactivity present in the CP slices was quantitated at the end of the incubation with 3H-choline (tissue labeling). The K A treatment diminished labeling from 9.7_+ 0.6 (vehicletreated) to 4.7 +0.9 pmoles/mg protein (kainic acidtreated). A small number (n = 4) of additional experiments were conducted where 3H-transmitter release was quantitated 7days after K A injection (instead of 3 days). In these experiments, stimulation-evoked release was reduced by 98 +_0.7% (p < 0.001). The K A injection into the O T had no effect on the

C h o l i n e r g i c n e u r o n s a n d k a i n i c acid

727

Table 2. Effects of stereotaxic injection of kainic acid on the basal and stimulated release of acetylcholine from superfused slices of CP and OT Stimulationevoked release

Basal release

Tissue 3H content

nCi/5min

%

nCi

%

$2/S1

nCi/slice

Caudate putamen Vehicle Kainic acid

0.5_+0.05 0.1 _+0.01"

1.8_+0.1 1.5_+0.1

1.2-+0.1 0.04_+0.1"

4.0_+0.5 0.62_+0.18"

0.95_+0.11 0.60-+0.03*

14.9_+0.9 4.6_+0.4*

Olfactory tubercle Vehicle Kainic acid

0.7_+0.06 0.7_+0.06

1.8_+0.1 1.8_+0.1

1.7+0.2 1.6_+0.2

4.2_+0.6 4.3+0.4

0.92+0.08 0.94+0.23

13.2_+0.6 12.1 +0.8

Rat CP and OT slices were incubated with [3H]-choline and then superfused. Slices were stimulated twice (S1 and $2) at 0.3 Hz and 120 pulses. Basal release was expressed as the amount of radioactivity (in nCi) present in superfusate collected for 5 min prior to the first period of stimulation; and as % of the tissue [3H] content at the time of sample collection (% of tissue radioactivity lost in 5 min). Stimulation-evoked release was expressed as the amount of [3H] released by electrical stimulation in nCi and as % of tissue 3H released by first period of electrical stimulation (S1); $2/S1, ratio of release evoked by first (SI) and second ($2) stimulation. Tissue 3H content at the end of the experiment was expressed as nCi/slice. Shown are mean values + SEM of at least six experiments per group. *Significant differences between KA and vehicle at *p < 0.01.

Caudate - Putamen 5.0

m

4.0

=~

t 501 il

t I

3.0-

2.0

4.0

~-

/

I.*' ~

Olfactory Tubercle

30-

,o k

'=

~.~

"~- C'~~

,,T

2.0

_

"

.

2d

~ o~

1.0

~v t

--

0

AK

1.0

I 50

o.~

60 S1

I

I

70 80

I

I

I

90 100 110 120 $2

T i m e in m i n u t e s

I

0 50

60 S1

I

I

70

80

L

i

i

I

I

90 100 110 120 $2

T i m e in m i n u t e s

Fig. 2. The A c h release f r o m C P a n d O T o f the rat: effects o f K a i n i c acid. The K A was injected stereotaxically into the r i g h t C P or the right O T (4/tg/2/~1); the c o n t r a l a t e r a l side received 2 #1 o f vehicle ( p h o s p h a t e buffered saline). E a c h rat received only t w o injections: one o f K A a n d one o f vehicle. The a n i m a l s were sacrificed 3 d a y s after K A injection. Sfices were prelabeled w i t h [3H]-choline a n d t h e n superfused. Release was elicited by electrical s t i m u l a t i o n a t 0.3 H z a n d 120 pulses. S h o w n are m e a n values +__S E M o f a t least nine o b s e r v a t i o n s per g r o u p . O r d i n a t e s , efflux o f [3H] in 5 m i n samples, expressed as % o f tissue tritium; abscissae, time in minutes; S1 a n d $2, first a n d second s t i m u l a t i o n s , respectively.

728

B.H. Guevara et al.

tissue [3H] content, the basal efflux of [3H], the fraction of tissue [3H]-transmitter released by stimulation, and the ratio $2/S1 (Table 2). These results indicate that, in the OT, K A had no effect on choline uptake and retention, and on transmitter release evoked by depolarization. These findings are in agreement with the absence of effects of KA on CAT activity in the OT, and indicate that KA spares the cholinergic elements of the OT DISCUSSION

The results of this work indicate that the cholinergic neurons of the CP, and of the OT, exhibit marked differences in their sensitivity to KA. Cholinergic neurotoxicity induced by K A was characterized by marked reductions in CAT activity and in the electrically evoked released of transmitter. In addition, tissue labeling with [3H]-choline and retention of tritium by the tissues were also reduced. The CAT activity is a well-established marker for cholinergic structures and can be localized in individual neurons and in synaptic terminals (Fonnum, 1975; McGeer et al., 1982; Phelps and Vaughn, 1986). In tissues with high Ach concentrations, labeling with radioactive choline is a well-known procedure for studying Ach release (Somogyi and Szerb, 1972; Richardson and Szerb, 1974; Hertting et al., 1980; James and Cubeddu, 1987). Therefore, our findings indicate that KA severely impairs cholinergic function in the CP, while it spares the cholinergic elements of the OT; indicating the existence of marked differences in neuronal vulnerability to KA between dorsal and ventral striatal cholinergic structures. However, CAT activity in the NAc (a ventral striatal region) also was sensitive to KA injection. Direct injection of KA into the NAc produced a 71% reduction in CAT activity. Our findings support those of Walaas and Fonnum (1979), who demonstrated that K A reduces NAc CAT activity by 76%. These observations clearly indicate that, even within the ventral striatum, there are marked differences in the excitotoxic action of KA. The marked differences observed in vulnerability of cholinergic elements to the excitotoxic action of KA may have several interpretations, since many factors determine neuronal vulnerability to an excitotoxic insult, namely: (a) proportion of intrinsic neurons to afferent axons terminals in a structure; (b) presence of endogenous excitatory inputs and of excitatory receptors; (c) appropriate mechanisms to reduce elevated intracellular calcium concentrations; and (e) free-radical scavenging resources (Schwob et al., 1980; Coyle, 1983; Choi, 1992). It should be indicated, however,

that the extent and severity of the cellular damage at or near the site of injection of K A is at least approximately related to the amount of the drug injected. Furthermore, if sufficient quantities of K A are injected, virtually every type of neuron may be destroyed (Schwob et al., 1980). Initial studies indicated that K A preferentially damages neuronal cell bodies and dendrites; reductions in CAT activity by KA reflects loss of cell bodies and terminals intrinsic within the structures injected. Axons and terminal boutons originating from cell bodies outside of the injected region remain largely intact after K A injection (Olney et al., 1974; McGeer et al., 1982; Coyle, 1983; Choi, 1992). Therefore, KAinduced reductions in CAT may suggest loss of cholinergic interneurons and/or cholinergic cell bodies of projecting neurons. The well-known sensitivity of CP to KA (Olney et al., 1974; McGeer and McGeer, 1976; Schwarcz and Coyle, 1977) could be due to the fact that interneurons comprise most of the cholinergic elements of the CP. Using this approach, it has been proposed that the ratio of cholinergic cell bodies to afferent nerve terminals in the NAc is lower than that of the CP, with nearly 70% of CAT present in cell bodies and 30% in afferent cholinergic axons (Walaas and Fonnum, 1979; present study). The fact that CAT activity within the NAc was not altered by hemitransection at the level of the GP, nor by lesioning the fornix (Walaas and Fonnum, 1979), further supports the view that the K A lesions of the NAc reflect loss of intrinsic cholinergic cells. Using the same reasoning, the lack of effects of K A on the OT could be interpreted to indicate that afferent axons and terminal boutons comprise most of the cholinergic elements of this region, and that the OT should contain a small proportion of cholinergic interneurons and of cholinergic-projecting neurons. However, immunocytochemical studies with monoclonal antibodies have revealed the existence CAT-positive neurons in the OT (Phelps and Vaughn, 1986). Although neuronal cell bodies and dendrites are, in general, sensitive to KA, some neurons are more resistant than others to KA toxicity. For example, the hippocampal granule cells are not destroyed by direct injection of low doses of K A (Herndon and Coyle, 1977). Therefore, it could be envisaged that the cholinergic neurons of the OT may be not vulnerable to the action of KA. Low density, or absence, of K A receptors could be a determining factor for the lack of neurotoxic effects of K A (see above). However, the possibility of a low density of kainate receptors in the OT, compared to NAc and CP, seems unlikely. Employing quantitative receptor autoradiography in

Cholinergic neurons and kainic acid CP, NAc and OT has demonstrated equal and high density of kainate binding sites (Albin et al., 1992). In addition, N-methyl-d-aspartate (NMDA) and alphaamino-3-hydroxy-5-methylioxazole-4-propionic acid binding sites are in higher proportion in OT and NAc than in the CP (Albin et al., 1992). In addition to the presence of excytotoxic receptors, the magnitude of the neurotoxic effect of low-dose K A on some neurons depends on the existence of glutamatergic input. Removal of the cortical glutamatergic input reduces the neurotoxic effects of K A on cholinergic and GABAergic elements of the CP (McGeer and McGeer, 1976). Dorsal and ventral striatal regions receive different cortical inputs. For example, afferents to dorsal striatum originate primarily from the neocortex (Kemp and Powell, 1970); whereas those to ventral striatum arise from cortical structures associated with the iimbic system, such as olfactory and entorhinal cortices and the hippocampus (Swanson and Cowan, 1975; Newman and Winans, 1980). Therefore, poorer glutamatergic input to the cholinergic structures of the OT could be a contributing factor to the insensitivity of the cholinergic elements of the KA. However, when high doses of K A were injected into the CP, the glutamatergic input played no role in determining K A neurotoxicity. In addition, the above-described, highly resistant hippocampal granule cells also may be destroyed by direct injection of sufficient dose of K A in their vicinity (Nadler et al., 1978). We demonstrated that the cholinergic elements of the rat OT are insensitive to low, as well as to very high, doses of KA; suggesting that differences in glutamatergic input should not explain the resistance of the cholinergic elements to KA. The characteristics of the intracellular calcium homeostatic mechanisms and of free-radical scavenging systems operating in the cholinergic structures of the rat OT are unknown; therefore, at present, it is not possible to determine whether these mechanisms are responsible for the observed lack of neuronal vulnerability to KA. An observation that deserves further comment is the apparently discordant finding of the effects of K A on CAT activity, transmitter release and labeling after incubation with [3]-choline, in the CP. The K A reduced CAT activity and [3H]-transmitter release by 90%, whereas labeling after incubation with [3H]choline was reduced between 50 and 70%. This is disturbing, since an excellent correlation has been reported between CAT and choline uptake into cholinergic nerves (Kuhar et al., 1973). However, labeling after extensive cholinergic neuronal destruction in the CP may be due to accumulation of some [3HI-choline in non-cholinergic structures via the low-affinity, high-

729

capacity choline uptake (Kuhar et al., 1973), since glial cells are left unchanged after application of K A (Olney et al., 1974; Schwarcz and Coyle, 1977). In fact, in the absence of the high-affinity uptake sites, accumulation via the low-affinity sites is favored. We did not measure the actual amount of [3H]-Ach formed after incubation with [3H]-choline, in control vs KA-treated tissues. However, after cholinergic destruction by KA, only 0.4% of the tissue [3H] was released by electrical stimulation; whereas, in control CP slices (or vehicle-treated slices), 4% of tissue [3HI is released by electrical stimulation. In addition, the ratio $2/S1 was reduced to 0.6 from 0.9 in slices from KA-treated rats. These observations suggest that, after KA, the tissue radioactivity is not located in a releasable neuronal pool. In summary, the cholinergic elements of the OT differ from those of the CP. Previous studies revealed that Ach release and turnover from OT are poorly responsive to inhibition by dopaminergic influences (Consolo et al., 1977; Bianchi et al., 1979; Salama et al., 1986; Suarez-Roca et al., 1987; Guevara et al., 1996). In addition, the cholinergic elements of the OT are completely insensitive to the neurotoxin action of even very high doses of KA. The most likely interpretation of these findings is that most of the Ach in the OT is within afferent nerve terminals, being therefore unresponsive to K A toxicity; other possibilities have been discussed above. If the OT is a functionally and anatomically integral part of the ventral striatum as suggested by Heimer and Wilson (1975); Heimer et al. (1982); Heimer et al. (1985), then our findings indicate the existence of marked differences between the cholinergic neurons of dorsal and ventral striatal regions. In addition, we report the existence of heterogeneity in the sensitivity to K A even within structures of the ventral striatum (NAc vs OT). A large proportion of the CAT-containing (cholinergic) elements of the NAc were vulnerable to K A injection; in contrast, the cholinergic elements of the OT were unresponsive to K A toxicity.

REFERENCES

Albin, R. L., Makowiec, R. L., Hollingsworth, Z. R., Dure, L. S., Penney, J. B. and Young, A. B. (1992) Excitatory amino acid binding sites in the basal ganglia of the rat: a quantitative autoradiographic study. Neuroscience 46, 3548. Bianchi, C., Tanganelli, S. and Beani, L. (1979) Dopamine modulation of acetylcholine release from guinea-pig brain. European Journal of Pharmacology 58, 235-246. Choi, D. W. (1992) Exitotoxic cell death. Journal of Neurobioloyy 23, 1261-1276.

730

B.H. Guevara et al.

Consolo, S., Ladinsky, H., Bianchi, S. and Ghezzi, D. (1977) A. I. (1983) Central cholinergic pathways in the rat: an Apparent lack of a dopaminergic-cholinergic link in the overview based on an alternative nomenclature (Chl rat nucleus accumbens septi-tuberculum olfatorium. Brain Ch6). Neuroscience 10, 1185 1201. Moore, R. Y. and Bloom, F. E. (1978) Central catecholamine Research 135, 255-263. Coyle, J. T. (1983) Neurotoxic action of kainic acid. Journal nervous system: anatomy and physiology of the dopaof Neurochemistry 41, 1-1. minergic system. Annual Review of Neuroscience 1, 129 Cubeddu, L. X., Hoffmann, I. S. and James, M. K. (1983) 169. Frequency-dependenteffects of neuronal uptake inhibitors Newman, R. and Winans, S. (1980) An experimental study on the autoreceptor mediated modulation of dopamine of the ventral striatum of the golden hamster. II. Neuronal and acetylcholine release from the rabbit striatum. Journal connections of the olfactory tubercle. Journal of Comof Pharmacology & Experimental Therapy 226, 88 94. parative Neurology 191, 193-212. Cuello, A. C. and Sofroniew, M. V. (1984) The Anatomy of Olney, J. W., Rhee, V. and Ho, O. L. (1974) Kainic acid: a the CNS cholinergic neurons. Trends in Neuroscience 7, powerful neurotoxic analogue of glutamate. Brain 74-83. Research 77, 507 512. Fonnum, F. (1975) A rapid radiochemical method for the Pellegrino, L. J., Pellegrino, A. S. and Cushman, A. J. (1981) determination of choline acetyltransferase. Journal ~?f A Stereotaxic Atlas of the Rat Brain. Plenum Press, New Neurochemistrv 24, 407~,09. York.. Guevara, B. E., Talmaciu, R. K., Hoffmann, 1. S. and Phelps, E. P. and Vaughn, J. E. (1986) Immunocytochemical Cubeddu, L. X. (1996) Dopamine-acetylcholine interlocalization of choline-acetyltransferase activity in rat actions in the ventral and dorsal striatum of rabbit and rat ventral striatum: alight and electron microscopic study. brain. Brain Research 733, 105-107. Journal of Neurocytology 15, 595~17. Heimer, L. (1978) The olfactory cortex and the ventral striRichardson, I. W. and Szerb, J. C. (1974) The release of atum. In The Continuing Evolution of the Limbic System labelled acetylcholine and choline from cerebral cortical Concept, eds K. E. Livingston and O. Hornykiewicz, pp. slices stimulated electrically. British Journal of Pharma95 187. Plenum Press, New York. cology 52, 499-507. Heimer, L., Alheid, G. and Zaborszky, L. (1985) Basal gangSalama, A. I., Czuprina, M. J., Stuimpo, A. and Sailer, lia. In The Rat Nervous System, ed.. Paxinos, Vol. 1, pp. C. F. (1986) Dopamine agonists suppress acetylcholine 37-86. Orlando, FL. release from the striatum but not from the olfactory Heimer, L., Switzer, R. D. and Van Hoesen, G. W. (1982) tubercle. Annals of the New York Academy of Science 463, Ventral striatum and ventral pallidum: components of the 137 139. motor system? TINS 5, 83 87. Scatton, B. (1982) Effect of dopamine agonists and neuroHeimer, L. and Wilson, R. D. (1975) The subcortical proleptic agents on astriatal acetylcholine transmission in the jections of the allocortex: similarities in the neural associrat: evidence against dopamine receptor multiplicity. Jouration of the hippocampus, the piriform cortex and the nal of Pharmacology & Experimental Therapy 220, 197 neocortex. In Centennial Symposium, Proceedings, ed. M. 202. Santini, pp. 177-193. Raven Press, New York. Schwarcz, R. and Coyle, J. T. (1977) Kainic acid: neurotoxic Herndon, R. M. and Coyle, J. T. (1977) Selective destruction effects intraocular injection. Investigations in Ophthalmic of neurons by a selective agonist. Science 198, 71 72. & Visual Science 16, 141-148. Hertting, G., Zumstein, A., Jacksih, T., Hoffmann, I. and Schwob, J. E., Fuller, T., Price, J. L. and Olney, J. W. Starke, K. (1980) Modulation by endogenous dopamine (1980) Widespread patterns of neuronal damage following of the release of acetylcholine in caudate nucleus of the systemic or intracerebral injections of kainic acid; a hisrabbit. Naunyn Sehmiedeberg ~ Archives qf" Pharmacology tological study. Neuroscience 5, 991-1014. 315, 111-117. Somogyi, G. T. and Szerb, J. C. (1972) Demonstration of James, K. K. and Cubeddu, L. X. (1987) Pharmacological acetylcholine release by measuring efflux of labelled cholcharacterization and functional role of muscarinic autoine from cerebral cortical slices. Journal of Neuroehemistry receptors in the rabbit striatum. Journal of Pharmacology 19, 2667-2677. & Experimental Therapy 240, 203-215. Starke, K., Reimann, W., Zumstein, A. and Hertting, G. Kemp, J. M. and Powell, T. P. S. (1970) The cortico-striate (1978) Effect of dopamine receptor agonists on release of projection in the monkey. Brain 93, 525 546. dopamine on caudate nucleus in vitro. Naunyn SehmiedeKuhar, M. J., Sethy, V. H., Roth, R. H. and Aghajanian, T. berg's Archives of Pharmacology 305, 27 36. (1973) Choline: selective accumulation by central chol- Suarez-Roca, H., Lovemberg, T. and Cubeddu, L. X. (1987) inergic neurons. Journal of Neurochemistry 20, 581 593. Comparative dopamine-cholinergic mechanisms in the Lowry, O. H., Rosebrough, H. J., Farr, A. L. and Randal, olfactory tubercle and in the striatum: effects of metoL. J. (1951) Protein measurement with the Folin phenol clopramide. Journal of Pharmacology & Experimental reagent. Journal of Biology & Chemistry 193, 265-275. Therapy 243, 84(~851. McGeer, P. L., Kimura, H., McGeer, E. G. and Peng, J. H. Swanson, L. W. and Cowan, W. M. (1975) A note on the (1982) Cholinergic systems in the CNS. In Neuroconnections and development of the nucleus accumbens. transmitters Interactions and Compartmentation, ed. H. F. Brain Research 92, 324-330. Bradford, pp. 253-289. Plenum Press, New York. Switzer, R. C., de Olmos, J. and Heimer, L. (1985) Olfactoo, McGeer, E. G. and McGeer, P. L. (1976) Duplication of System In The Rat Nervous System. ed. G. Paxins, Vol. 1, biochemical changes of Huntington's chorea by intrapp. 1-36. Academic Press, Orlando, FL.. striatal injections of glutamic and kainic acids. Nature 263, Walaas, I. and Fonnum, F. (1979) The effects of surgical and 517-519. chemical lesions on neurotransmitter candidates in the Mesulam, M. M., Mufson, E. J., Wainer, B. H. and Levey, nucleus accumbens of the rat. Neuroscience 4, 209 216.