Behavioural Brain Research, 51 (1992) 127-137 9 1992 Elsevier Science Publishers B.V. All rights reserved. 0166-4328/92/$05.00

127

BBR01361

Regulatory behaviour, exploration and locomotion following N M D A or 6-OHDA lesions in the rat nucleus accumbens R u t h W e i s s e n b o r n and Philip W i n n Department of Psychology, University of St. Andrews. St. Andrews, Fife (UK) (Received 3 January 1992) (Revised version received 22 July 1992) (Accepted 31 July 1992)

Key words: Nucleus accumbens; 6-Hydroxydopamine; N-Methyl-D-aspartate; Exploration; Locomotor activity; Dopamine agonist

The effects of bilateral, N-mcthyl-D-aspartate (NMDA)-induced lesions of the nucleus accumbens (N.Acc.) on regulatory and behavioural responding were studied in rats and compared with the effects of bilateral 6-hydroxydopamine (6-OHDA) lesions. After postoperative body weight, food and water intake had been monitored for a period of 4 weeks, rats were tested in an exploration-choice box. Spontaneous locomotion and the locomotor and stereotypy responses to different doses ofdopaminergic agonists were measured subsequently. Detailed assessment of NMDA-induced lesion volumes showed that on average 81.53~ of total N.Acc. area was damaged, depending on excitotoxin dose. Tyrosine hydroxylase immunohistochemistry was used to confirm loss of mesolimbic dopamine neurones following 6-OHDA. Analysis ofthe behavioural data showed that NMDA N.Acc. lesions significantly enhanced exploratory behaviour, spontaneous locomotor activity and the locomotor response to a low dose of D-amphetamine. By comparison, 6-OHDA lesions did not affect exploration and spontaneous locomotion but significantly attenuated the locomotor response to a low dose of D-amphetamine. Regulatory responses were unaffected 28 days after surgery, although NMDA-lesioncd rats took longer to recover from postoperative hypodipsia. The results suggest that NMDA N.Acc. lesions induce a deficit in the control of general locomotor output and are consistent with the hypothesis that the N.Acc. functions as an interface between sensory input and locomotor output and that it is needed to channel activity levels appropriately.

INTRODUCTION

The nucleus accumbens (N.Acc.) forms an important part of the ventral striatal circuitry, where limbic inputs influence behavioural responding by controlling N.Acc. output and feedback mechanisms to brainstem locomotor systems, substantia nigra, limbic systems and the V T A 1'3"18. Innervated by midbrain dopamine (DA)-containing neurones in the ventral tegmental area (VTA-A10), the N.Acc. also receives extensive limbic inputs from the amygdala36, hippocampus 13"34"35 and cortical areas associated with the limbic system3. One of the behavioural effects of manipulations of DA activity in the N.Acc. which has been investigated in detail is locomotor activity. Intra-accumbens administration of DA or its agonists z~'22, as well as electro-

Correspondence: P. Winn, Department of Psychology, University of St. Andrews, St. Andrews, Fife, KYI6 9JU, UK.

lytic N.Acc. lesions 14 increase levels of locomotion, whereas 6-hydroxydopamine (6-OHDA) lesions that deplete DA levels in the N.Acc. reduce spontaneous locomotor activity4'~6, attenuate the locomotor response induced by the indirect DA agonist D-amphetamine and potentiate locomotion following administration of low doses of the direct DA receptor agonist apomorphine8'9'15"~7. Normal activity levels and responses to D-amphetamine and apomorphine have been reported to be restored 4-12 weeks after surgery 15"3z.The stereotyped responding induced by higher doses of D-amphetamine is unaffected by N.Acc. lesions, supporting the differentiation drawn between dorsal and ventral DA systems ~2"~5.Some authors have suggested that the locomotion induced by stimulation of the N.Ace. has in particular an exploratory nature. Fink and Smith 9 observed reduced exploratory locomotion (measured in a novel open field) that could be restored with low doses of apomorphine after 6-OHDA lesions of the N.Acc. However, deficits in exploration

128 (measured in the Carlsson box) were not reported 4 weeks or 1 week after surgery23'3z. Significant changes in normal regulatory behaviours such as feeding and drinking do not occur as a consequence of 6-OHDA N.Acc. lesions 15'32. N.Acc. DA has further been ascribed an important role in mediating more complex processes such as reinforcement and reward. Self-administration of intravenous stimulant drugs 25 and conditioned placepreference28 are both attenuated by 6-OHDA lesions of the nucleus, whereas behavioural control by conditioned reinforcement is enhanced after intra-accumbens D-amphetaminez9"3~ Further, excessive drinking elicited by intermittent food-reinforcement (SIP) is abolished by 6-OHDA N.Acc. lesions 16"24. The neurotoxin most frequently used in studies of N.Acc. functions has been 6-OHDA which is taken up into terminal fields of DA-containing neurones where it is degraded, metabolites causing neuronal degeneration and depleting the nucleus of DA, while leaving non-DA containing neurones intact. In order to examine the functional role of the N.Acc. per se, it is essential also to look at the consequences of damage selective to neurones intrinsic to the nucleus, since the behavioural effects of such lesions may differ from those observed following DA depletion. Fibre-sparing lesions can be made using excitotoxins such as kainate, N-methyl-Daspartate (NMDA), and ibotenate. These toxins exert their effects by binding to excitatory amino acid receptors on the dendritic portions of neurones and promoting the uncontrolled entry of Ca z+, which has toxic consequences 33. While studies examining the neurotoxic effects of the glutamate receptor agonist A M P A have identified a 'delayed' degenerative mechanism depending on intracellular Ca 2+ (ref. 11), a Ca z+ overload appears to be responsible for rapid neuronal death following administration of N M D A to specific brain sites ~~ Studies published previously using ibotenic acid 2 or quisqualic acid 6 rather than 6-OHDA to induce N.Acc. lesions report impairments in spatial learning and the abolition of place preference conditioning. However, both studies fail to report the quality of damage induced and individual lesion volumes. Given that excitotoxic ventral striatal lesions have at present not been widely examined, it is important to make detailed assessments of type and extent of damageinduced. Such methods also allow correlations of lesion volumes with behavioural responding to be calculated. The present series of experiments was aimed at evaluating the regulatory and behavioural consequences of N M D A lesions selectively destroying neurones within the N.Acc.. Assessments were made of post-lesion

home cage food and water intake, body weight, exploration, spontaneous locomotion, and the locomotor and stereotypy responses to systemic administration of D-amphetamine and apomorphine. In addition, the behavioural effects of neuronal loss following N M D A lesions were compared directly with DA depletion induced by 6-OHDA. MATERIALS A N D M E T H O D S

Anhnals 33 male Lister hooded rats (bred in-house) were caged individually under a 12 h light/dark cycle (lights on 8.00 a.m.) and maintained on a diet o f a d libitum lab chow and tap water throughout the experiment. Daily measurements were made (to the nearest 0.1 g) of (i) body weight; (ii) weight of food remaining in the cage hopper; (iii)food spillage (collected on foil sheets beneath the food hopper) and (iv)amount of water remaining in the water bottle. Mean body weight at time of surgery was 289.0 g (S.D. = + 25.85).

Surgical procedures Rats were anaesthetized with 10 ml/kg Avertin i.p. (Avertin concentrate: 100g 2,2,2 tri-bromo-ethanol/ 62 ml tertiary-amyl-alcohol; 1.25 ml of this concentrate then added to 5 ml absolute alcohol and 62.5 ml 0.9~o saline). Excitotoxic lesions were made by bilateral stereotaxic infusions of either 1 F1 0.06 M N M D A (Sigma Chemicals) (60 nmol, n = 6) or 1 FI 0.09 M N M D A (90 nmol, n = 6). Seven rats received injections of the vehicle. To obtain a 90 nmol solution of N M D A , 13.24 mg were dissolved in I ml phosphate buffer (pH 7.4) and if necessary the pH was further adjusted with 2 M NaOH; final pH was 7.4 6-OHDA lesions were induced by infusion of 8 Fg/2 FI 6-hydroxydopamine hydrobromide (Sigma Chemicals) into the medial N.Acc. (n = 9). Five rats received ascorbate saline vehicle. 6-OHDA was prepared as the free base in 0.1 mg/ml ascorbate saline and kept on ice throughout. Rats receiving 6-OHDA infusions or the vehicle were pretreated with 15mg/kg pargyline hydrochloride (Sigma Chemicals) (to inhibit MAO activity) and 15 mg/kg desipramine hydrochloride (Sigma Chemicals) (to protect noradrenergic neurones) 30 min prior to surgery. Coordinates used were: anterior-posterior, + 2.0 mm from bregma; lateral, + 1.5 mm from midline; vertical, -7.0 mm from skull surface with level skull 19 for N M D A lesions; anterior-posterior, + 3 . 4 m m from bregma; lateral, + 1.7mm from midline; vertical, -6.2 mm from dura with the incisor bar set 5.0 mm above the intraural line 2~ for 6-OHDA lesions.

129 Injections were made using stereotaxically mounted 30-ga. stainless steel cannulae connecied via polyethylene tubing to 10 Ill SGE syringes driven by a Harvard infusion pump. Injection volumes were 1 Id/2 min for N M D A and its vehicle and 2/11/4 min for 6-OHDA and its vehicle. After infusion of the drug, cannulae were left in situ for 2 and 4 rain respectively to allow for diffusion of the drug away from the cannula tip.

Exploratory behaviour Four weeks after surgery, exploration was tested using an exploration choice box (Carlsson box32). This was a large rectangular box (120 w x 45 1 x 50 h cm) with 3 aluminium walls, a metal grid floor with an aluminium tray underneath, 2 removable tops, and a transparent perspex front. The box was divided into 2 halves by a removable black partition. A perspex startbox (15 w x 15 1 x 8 h c m ) attached to the centre of the front wall gave equal access to both halves of the box. The grid floor was marked off into 4 equal quadrants on each side of the divider. Each animal was placed into one half of the box (familiar) for 1 h, following which it was removed and placed in the startbox. The centre partition was then replaced by one with a hole (8 w x 7 h cm) in the bottom allowing the rat to move from one side of the box to the other. Latency and side (novel or familiar) of emergence from the startbox, number of quadrant crossings and rears, as well as the total amount of time spent on either side of the box were recorded continuously over a 10-min period. A 'quadrant crossing' was recorded when the animal had crossed one of the transition lines with all four paws, but not necessarily the tail. A 'rear' was recorded when the rat was in a position where the two front paws were lifted above the ground and the back at an angle greater than 45 ~ to the ground. Equal numbers of rats were familiarized to the right and left hand sides of the test environment. Spontaneous locomotor activity Following the exploration tests, spontaneous locomotor activity during a 60-min period was recorded daily for 2 weeks. Animals were placed individually into a battery of 18 w x 24 1 • 37 h cm wire activity cages with 2 photoelectric cells 2 cm above the grid floor. Interruptions of infrared lightbeams were recorded by a BB C microprocessor with a Eurorack extension unit and a Spider interface system using 'Spider' BASIC programming language (Paul Fray Ltd.). The programme discounted sequential activation of 1 beam only, requiring the 2 beams to be interrupted one after another.

Locomotor and stereoO'py responses to apomorphhte and D-amphetamine After habituation to the activity cages, (8 weeks postsurgery) rats received a series of counterbalanced injections of apomorphine (s.c. in the neck, 0.1, 1.0 and 3.0 mg/kg), D-amphetamine (i.p. 1.5 and 5.0 mg/kg), or vehicle prior to the 60-min test session. Order of administration was determined by a Latin square design to control for the possible additive or potentiating effects of repeated drug administration. Injection days were separated by one rest day. In addition to recording of locomotor activity, stereotypy was measured on a scale similar to that described previously~S: O, asleep or stationary; 1, active; 2, predominantly active bursts of stereotyped sniffing or rearing; 3, stereotyped activity predominantly sniffing and rearing over a larger area of the cage; 4, stereotyped behaviour maintained in one location; 5, stereotyped behaviour in one location with bursts of gnawing or licking; 6, continual gnawing or licking of the cage bars. Statistics Repeated measures ANOVAs were carried out on measurements of normal regulatory behaviour, exploratory behaviour, spontaneous locomotor activity and the locomotor responses to apomorphine and D-amphetamine. Significant effects were further analyzed using Newman-Keuls post-hoe tests 31. Stereotypy ratings in response to drugs were analyzed using the Fisher exact probability tests 27. The Pearson correlation coefficient was calculated for linear relationships between N M D A lesion volumes and behavioural responses. Histology On completion of behavioural testing, rats with N M D A lesions of the N.Acc. were deeply anaesthetized with 'Euthatal' (sodium pentobarbitone, 200 mg/ ml) and perfused transcardially with 0 . 9 ~ saline followed by 10~o phosphate-buffered formalin. Brains were removed and stored in formalin until sectioning. All lesion sites were verified using standard Cresyl violet staining methods (40-1Lm sections every 200/~m). The volumes of NMDA-induced lesions were assessed using a Leitz Diaplan microscope fitted with a drawing tube. Silhouettes of lesions were drawn onto appropriate standardized stereotaxie atlas drawings tg. Using a digitized tablet ('Grafpad') connected to a BBC microprocessor, the area of each lesion on each section was calculated and an estimate of lesion volume made on these standardized sections. Since statistical analysis (ANOVA) showed no significant differences in total lesion volume following in-

130 fusion of N M D A at concentrations of 60 nmol or 90nmol (F1.1o = 1.05) and there were no apparent qualitative differences, the two groups were collapsed into one (n = 12) for analysis of lesion and behavioural data.

bmmmohistochemistry Immunohistochemistry was carried out as described previously26 in rats with 6-OHDA lesions of the N.Acc. Rats were sacrificed by i.p. injection of 1.5 ml 'Euthatal' after completion of behavioural testing. Thirty min prior to injection, each rat was pretreated with heparin (2,500 USP units in 0.5 ml sterile saline i.p.). They were then perfused transcardially with 100 ml of Ca 2+ free Tyrodes solution 26 containing heparin (10 USP units ml) at 28-30 ~ at a rate of 20 ml/min using a Gilson Minipuls 3 pump fitted with i.d. 0.125" tubing. The Tyrodes solution was followed by 300 ml fixative ( 4 ~ paraformaldehyde]0.05~ glutaraldehyde in 0.1 M phosphate buffer) at the same rate of delivery. The brains were removed and postfixed in the same fixative solution for 30 rain at room temperature. The brains were cut into 50-ttm sections on a freezing mierotome and placed into phosphate-buffered saline (PBS). For 6 rats, sections were cut either horizontally or sagittally to show the VTA, substantia nigra and medial forebrain bundle projecting to the ventral striatum. Coronal sections of the N.Acc. were taken for a further 3 rats. In all animals, every third section was stained for tyrosine hydroxylase (TOH) and Nissl substance. Sections were washed with 3 0 ~ sucrose for 30 min and then washed 5 times for 5 min in PBS at room temperature. Incubations with antibodies and other reagents were carried out in 24-well tissue culture plates in a volume of 0.3-0.5 ml. Up to 6 sections were incubated per well. For washing sections were transferred to a container on a flat bed shaker, with the exception of initial incubations with the primary antibodies for TOH, which were at 4 ~ Sections were placed in a blocking solution (20~o normal goat serum, 0.1~o triton X-100 in PBS) for 60 rain and then washed 3 times for 5 min with PBS. They were incubated with anti-TOH (from mouse-mouse hybridomas, Boehringer) 1:50 in antibody diluting solution (ADS) for approximately 15 followed by 5 • 5 min washes with PBS. The ADS used was 0 . 1 ~ normal goat serum and 0.1 ~/o tritron X-100 in PBS. The sections were then incubated with anti-mouse IgG (1:30 in ADS) (sheep, Sera-lab) for I h and washed 5 x 5 min in PBS. Following the wash sections were incubated with monoclonal mouse peroxidase antiperoxidase (1:100 (Sigma) in ADS) for 60min and

washed 5 x 5 min in PBS. The IgG and peroxidase anti-peroxidase incubations were repeated in the same order for the same length of time with PBS washes between each step. Finally, sections were incubated with 0.05~o diaminobenzidine (DAB) for 15 min (1 ml/ well) followed by the addition of 10/d/well 1 ~o H202 and a further incubation of 5-10 min. The sections were washed 5 x 5 min in PBS, mounted on glass slides and air-dried. Coverslips were applied with DPX.

RESULTS

Histology Volumes of intact N.Acc. and ventral striatum as represented on standardized sections were calculated to be approximately 6.465 mm 3 (S.D. = +0.22) and 10.223 mm 3 (S.D. = + 0.62) respectively. Lesion volumes resulting from bilateral infusions of N M D A into the medial N.Acc. were expressed as percentages of total accumbens area damaged. Lesion boundaries were identified on the basis of detectable cell loss or reactive gliosis present. All animals had sustained some damage to the ventral striatum (N.Acc., olfactory tubercle, ventromedial caudate putamen, bed nucleus of stria terminalis), ranging from 13.3~ to 98.31~ (mean: 51.92~, S.E.M. = + 8.25) and were included in the statistical analysis. The medial N.Acc. was substantially damaged in all rats. Average N.Acc. lesion volume obtained by infusion of 60 nmol and 90 nmol N M D A was 81.53~o (S.E.M.= +12.87). Other ventral striatal structures (i.e. ventromedial caudate-putamen, olfactory tubercle and/or the bed nucleus of stria terminalis) were also damaged in some cases. The olfactory tubercle was destroyed in 3 rats, 1 rat had sustained damage to the ventromedial caudate-putamen, and in 1 rat both structures were affected. In addition, the channels of ventral pallidal neurones lying between the N.Acc. and olfactory tubercle were damaged in 4 animals. Fig. 1A and B show the smallest and the largest ventral striatal lesions induced by NMDA. TOH immunohistochemical staining techniques showed an extensive loss of mesolimbic DA neurones in all rats following infusion of 6-OHDA into the N.Acc., while mesostriatal DA neurones in the substantia nigra remained intact (see Fig. 2). Thus, TOHpositive staining confirmed biochemical findings reported previously32 (using high performance liquid chromatography), where terminal lesions using 8 ltg/ 2 #1 6-OHDA in the N.Acc. produced large depletion of DA in this area but had no effect on DA in the caudate-putamen.

131

2.70

2.20

1.70

1.60

1.20

1.00

0.70

0.4B

Fig. 1. Smallest (A) and largcst (B) ventral striatal lesions following infusion of I t d 60 nmol or 90 nmol N-methyl-D-aspartate into the nucleus accumbens. Shaded areas represent the areas from which neurones were lost and in which reactive gliosis was present. Sections drawn from the atlas of Paxinos and Watson (1986); numbers indicate the distance from bregma in ram; A, nucleus aeeumbens; C, caudate-putamen; t, olfactory tubercle; V, channel neurones of the ventral pallidum.

Normal regulatory responses

F2,3o = 1.24; food: F2,3o = 0.88). All rats showed signif-

Repeated measures ANOVA showed a steady increase in body weights, food and water intake in all lesion groups during the 5 days prior to surgery (body weights: F4.12o= 2.47, P < 0.05; food intake: F4,12o=60.14, Pz--~_-.- .....:--:. " - %"'~'-

"-?'":

-

" ='"

I

'

Fig. 2. Photographs of sections through the midbrain and medial forebrain stained for tyrosine hydroxylase (TOH). A and B show the distribution of TOH-positive cells in the ventral tegmental area (VTA) in sagittal sections after sham lesion (A) and after 6-hydroxydopamine (6-OHDA) lesion of the nucleus accumbens (B); C and D show the distribution of TOH-positive neurones in the VTA, substantia nigra and mesostriatal projection sites after sham lesion (C) and after 6-OHDA lesion of the nucleus accumbens (D); dmn, dorsomedial hypothalamic nucleus; ml, medial lemniscus; mt, mammillothalamie tract; snpc, substantia nigra pars compacta; snzr, substantia nigra zona reticulata; vta, ventral tegmental area; 3v, 3rd ventricle; vertical arrows indicate mesostriatal fibre systems. Bars = I mm. rats with sham lesions until the end o f week 3 (week 2: P < 0 . 0 5 ; week 3: P < 0 . 0 0 1 ) .

ration choice box ( P < 0 . 0 0 1 ) . The percentage o f time spent in the novel environment was not affected by lesion type (F2,3o = 2.36).

Exploratory behaviour D a t a relating to exploration are shown in Fig. 4. Multiple A N O V A indicated that rats in all lesion groups engaged in significantly more exploratory behaviours ( q u a d r a n t crossings and rearing; scores corrected for time spent on novel/familiar side o f the test box) in a novel than in a familiar environment (quadrant crossings: F|,6o=6.74, P < 0 . 0 5 ; rearing: Ft.~o=4.61, P < 0.05). Post-hoc analysis o f significant group effects on quadrant crossings (F2,6o = 5.97, P < 0.01) and rearing (Fz.6o = 13.89, P < 0 . 0 0 1 ) showed that N M D A induced lesions o f the N.Acc. significantly enhanced exploration o f both environments c o m p a r e d to shamlesioned control rats (quadrant crossings: P < 0 . 0 1 ; rearing: P < 0.001). In addition, 6 - O H D A lesions o f the N.Acc. were found to attenuate significantly the number o f quadrant crossings on both sides o f the explo-

Spontmteotts locomotor activity M e a n levels o f spontaneous l o c o m o t o r activity are shown in Fig. 5. Repeated measures A N O V A on the square root transformed photocell counts revealed no significant between-group differences over 2 weeks o f habituation to the locomotor activity cages (/'2,30 = 2.95). Similarly, activity levels did not differ significantly between week 1 and 2 ( F t , 3 o = 1.68). H o w ever, post-hoc analysis o f a group • week interaction (Fz,3o=8.04, P < 0 . 0 1 ) showed that N M D A - l e s i o n e d animals were significantly more active than the shamlesioned group during weeks 1 and 2 ( P < 0 . 0 1 ) . 6 - O H D A lesions o f the N.Acc. did not induce hypoactivity but led to significant increases in l o c o m o t o r activity during the 2nd week o f testing, c o m p a r e d to the 1st week ( P < 0 . 0 0 1 ) .

133 130

6.

~_

120

5

$$

110

4

~

100

t~

r O O)

*

3 O

o N

90 80

!

0! Lesion

!

5

i

I

10 15 Days post lesion

!

!

20

25 0

120

NovelEnvl FamiliarEnv't

Total

NovelEnvl FamiliarEnv't

Total

t, ~ too o ~ 80 60

"6

40

9-=

20

to

E i

0! Lesion

I

5

I

f

10 15 Days post lesion

I

l

20

25

2

1

120 I00

0

~ ,o~ 80 Q

C

~.-

60

E

"6

80

O

40

20

60 0 !

5

Lesion

10

15

20

25

Days post lesion

Lesion group I eVehicle oNMDA &6-OHDA Fig. 3. Percentages of average prelesion body weights, water and food intake measured over 26 days following N-methyl-D-aspartate, 6-hydroxydopamine or sham lesions of the nucleus accumbens.

Locomotor and stereotypy responses to apomorph#ze and D-amphetamhte Fig. 6 shows locomotor activity in response to administration of different doses of apomorphine and D-amphetamine. ANOVA with repeated measures on the square root transformed photocell counts per hour showed a significant effect of administration of different doses of apomorphine on locomotor activity (/73,90 = 17.21, P

Regulatory behaviour, exploration and locomotion following NMDA or 6-OHDA lesions in the rat nucleus accumbens.

The effects of bilateral, N-methyl-D-aspartate (NMDA)-induced lesions of the nucleus accumbens (N.Acc.) on regulatory and behavioural responding were ...
2MB Sizes 0 Downloads 0 Views