Brain Research EuNerin,

0361-9230190 $3.00 + .ocl

Vol. 24, pp. 313-319. D PergamonPress plc, 1990.Printedin the U.S.A

MK80 1 Prevents Quinolinic Acid-Induced Behavioral Deficits and Neurotoxicity in the Striatum MAGDA

GIORDANO,* LISA M. FORD,? JOAN L. BRAUCKMANN,” ANDMW B. NORMAN* AND PAUL R. SANBERG**

*Division of Neuroscience, Departments of Psychiatry, Psychology, Anatomy, Physiology, and Neurosurgery ~~~versi~ of ein~i~n~ti College of medicine, C~nci~~ti, OH 45267-05.59 l3epartment of Pediatric Neurology, Children’s Hospital, Cincinnati, OH 45229 Received

22 January

1990

GIORDANO, M., L. M. FORD, J. L. BRAUCKMANN, A. B. NORMAN AND P. R. SANBERG. IWKBOIprevents quinolinic de$cirs and ~euroioxici~ in the striatum. BRAIN RES BULL 24(3) 3 13-3 19, 1990. -Bilateral intrastriatal injections of quinolinic acid (QA) (180 nmoles) induced weight loss and neurologic and behavioral deficits including convulsions, decreased catalepsy response to haloperidol, increased nocturnal locomotor activity, and abnormal feeding behavior in adult male Sprague-Dawley rats. Pretreatment with the noncompetitive N-methyl-D-aspartate (NMDA) antagonist, MK801 (4 mg/kg IF) 30 min prior to stereotaxic surgery prevented the appearance of all QA-induced behavioral abnormalities and prevented weight loss. Twelve weeks after surgery the QA-lesioned animals recovered to sham levels on feeding behavior and nocturnal locomotor activity, but showed persistent suction in halo~~dol-induced catalepsy. ~stologic~ ex~nation of the QA-lesioned brains showed extensive lesions of the dorsolateral striatum and frontoparietal cortex. MKSOI pretreatment protected against these lesions. These results confirm that MK801 treatment prevents the appearance of nemopathological damage after QA neurotoxicity, and further show that neuronal protection with MK801 is correlated with the absence of QA-induced behavioral deficits. acid-induced ~~vjoFai

MI@01

Quinolinic acid

Catalepsy

Locomotor behavior

Huntin~on’s disease

N-me~yl-D-~p~~

(NMDA)

Glutamate

ers (2,lO). MK801 also prevents the neu~egene~tion observed in animal models of hypoxia and ischemia (9,12), and kindling (11). However, few studies have explored the behavioral consequences of QA neurotoxicity in conjunction with MK801 treatment. Glutamate analogues such as KA and IA when injected directly into the striatum induce a variety of long-term behavioral deficits. Among them are increased spontaneous nocturnal locomotor activity (32) and abnormalities in feeding behavior (24,30) and learning (8,25). Although there is no rodent model which can precisely mimic the functional deficits observed in HD, previous studies in this laboratory have shown that QA-induced lesions of the dorsal striatum result in increased nocturnal locomotor activity (28) which may be analogous to some aspects of the exaggerated movements observed in HD (29). The present study examined the ability of MK801 to protect against behavioral deficits induced by intrastriatal QA lesions. The behaviors analyzed included spontaneous nocturnal locomotor activity, h~o~~dol-induct catalepsy, and feeding behavior. All behaviors were examined at 4 and 12 weeks postlesion.

SINCE the proposal of the excitotoxin hypothesis of ~untington’s disease (HD) (6,19) various glutamate analogues have been used to generate animal models of the disorder. Recently Beal et al. (1) proposed what they consider an appropriate model of HD using the endogenous excitotoxin quinolinic acid (2,3-pyridine dicarboxylic acid) (QA). QA injections into the s~amm replicated many of the neurotoxic effects induced by other excitotoxins such as ibotenic acid (IA) and kainic acid (KA) (35) but have been reported to produce relative sparing of specific neuronal populations which are also spared in HD (2). However, recent studies have questioned this selective sparing of somatostatin and neuropeptide-Y containing neurons by QA (7, 4, 33). QA acts via the N-methyl-D-aspartate (NMDA) receptor and it has been suggested that blockade of this receptor could prevent QA-induced neurodegeneration (2). Recently a systemically acting anticonvulsant (+) - 5 - methyl - lo,11 - dihydro - 5H - dibenzo [a,d]cyclohepten - 5,lO - imine maleate (MK801) has been characterized as a potent and selective noncompetitive NMDA receptor antagonist (36). Systemic ad~nis~ation of MEL801 has been shown to protect in a dose-related fashion against QA-induced neurotoxicity as measured by histological and biochemical mark-

*Requests for reprints should be addressed to Paul R. Sanberg, Ph.D., Division of Neuroscience, Department of Psychiatry, University of Cincinnati, College of Medicine, 231 Bethesda Avenue, Cincinnati, OH 45267-0559.

313

314

GIORDANO

METHOD

500

1

Stereotaxic Injections and MK801 Treatment 475 1

Four groups of adult male Sprague-Dawley rats (200-250 g) were used. All animals were housed in pairs, had food and water available at all times, and were kept on a 12-hour light/dark cycle (0600-1800 hr). The groups were as follows: Group 1 received bilateral intrastriatal injections of 180 nmoles QA (QA). Group 2 received MK801 (4 mg/kg IP) thirty minutes prior to QA lesions (MQ). Group 3 received MK801 (4 mg/kg IP) thirty minutes prior to sham lesions (MK). Group 4 received sham lesions only (SH). All groups received bilateral intrastriatal injections [AP = + 1.2 mm from bregma; ML = 2 3.1 mm; DV = - 4.7 mm from dura according to the atlas of Paxinos and Watson, (23)] of QA (180 nmoles/side) or vehicle under pentobarbital anesthesia (35 mg/kg IP). Thirty min prior to anesthesia, two of the four groups received one IP injection of 4 mg/kg of MK801 (Merck Sharp and Dohme). QA (Sigma Chemical Co.) was dissolved in a solution of 0.2 M phosphate buffered saline and the pH adjusted to 7.0 with NaOH. QA was delivered into the dorsolateral striatum through a Hamilton syringe (28-gauge needle) in a volume of 1 (*l per striatum over a period of 2 min (35). Animals which lost postsurgery more than 10% of their initial body weight were given 20 ml of Similac by means of an orogastric pediatric feeding tube until they recovered or for a maximum of seven days. All animals had access to a food mash consisting of pulverized commercial rat food pellets and Similac. Weight was monitored daily for twelve days and once a week thereafter. After surgery the animals were continuously monitored until recovery from anesthesia and the presence of convulsive behavior was recorded. Behavioral Testing All behavioral tests were conducted once before surgery, and four and twelve weeks afterwards. The observers were blind to the treatment which the animals had received. For the cataleptic response all groups were randomly divided into two subgroups and drug treatment was counterbalanced in such a way that each animal served as its own control. Animals were tested both two and four hours after injection of haloperidol (1 mg/kg IP) or vehicle (1 ml/kg IP) using the bar-test (27). The animal’s paws were placed on top of a wooden bar 12 cm high and 1 cm in diameter. The catalepsy score was the time rats remained in that position before placing both paws on the floor. The cutoff time was 300 seconds. Nocturnal locomotor activity was recorded using the Digiscan16 Animal Activity Monitor System (Omnitech Electronics, Columbus, OH). This system has been described previously (31). Briefly, animals were placed into acrylic boxes (40 x 40 x 35 cm) at 17:00 with access to food and water. After one hour of habituation, several locomotor parameters were monitored for twelve one-hour intervals. Among those parameters measured were total distance travelled (in cm), movement time (in set), vertical activity (counts), number of vertical movements (counts), and time spent in stereotypical movements (in set). Feeding behavior was evaluated following a modification of the methods presented by Pisa (24). Animals were food deprived 24 hours prior to a 5-min observation period. The subjects were placed in an acrylic box with food pellets and were rated on the amount of time they spent: 1) feeding (physical contact between rat and food); 2) picking up and dropping pellets (picking up a pellet with the mouth and holding it or dropping it on the floor); 3) pushing the pellet (pushing the pellet around the box with the snout); 4) holding the pellet off the floor with paws (holding the pellet off the floor with paws supporting the pellet); 5) holding the

a-•QA O-0StiA~ n -¤MK801 O-0 MK80i

ET AL,.

+ QA

425 i 400-1

375--

I 6

/

/

9

I

)

/ 12

DAYS POSTVLESION

FIG. 1. Effect of MK801 on QA-induced postoperative weight loss. Averages for each group in grams are indicated for day 0, 1, 3, 6, 9, and 12 days postoperative. (0) Indicate quinolinic acid-lesioned; (0) indicate sham lesion; (m) indicate MK801-pretreated and QA-lesioned; (0) indicate the MKSOl-pretreated group. Number of subjects per group was 10 to 11.

pellet against the floor (holding the pellet against the floor or comer of the cage with paws resting on the pellet); and 6) biting (time between mouth contact with pellet and onset of chewing). Data Analysis The ten variables of locomotor activity encompassing the horizontal, vertical and stereotypical dimensions were analyzed by means of split-plot and one-way analysis of variance (ANOVA). Similar analyses were performed for catalepsy and feeding behavior scores. Preplanned comparisons among groups were computed and the significance level set at 5% (14). Histology Approximately two months after the last behavioral test all animals were anesthetized and perfused intracardially with 10% formalin. The tissue was processed and stained with cresyl violet following sectioning using paraffin embedding or a freezing microtome (4Xl p,m). The specimens were microscopically evaluated for presence of necrotic cells, gliosis and cell loss. RESULTS

After surgery there was a clear distinction between the MQ and the QA groups. Ten out of fourteen QA-lesioned animals showed convulsions and/or barrel rolling, while none of the MQ animals showed convulsions. Furthermore, four of the QA animals had to be tube-fed due to significant loss of body weight while none of the MQ animals required tube feeding. MK801 seemed to potentiate the effect of the anesthetic. Even using a low dose of pentobarbital (35 mg/kg) two out of twelve animals in the MK group did not recover from anesthesia whereas all sham animals survived. In total 86% of the animals survived throughout the 20 weeks of the study. With respect to weight loss there was a significant group effect three days, F(3,38)=5.437, pcO.01; six days, F(3,38)=5.81, pcO.01; nine days, F(3,38)=4.687, p