Differential effects of scopolamine on working and reference memory of rats in the radial maze

Pharmacology Biochemistry & Behavior, Vol. 20, pp. 659--662, 1984. ©Ankho International Inc. Printed in the U.S.A.

0091-3057/84 $3.00 + .00

Differential Effects of Scopolamine on Working and Reference Memory of Rats in the Radial Maze B. A. W I R S C H I N G , * R I C H A R D J. BENINGER,* K. J H A M A N D A S , t R. J. B O E G M A N t

A N D S. R. E L - D E F R A W Y t

Departments o f * P s y c h o l o g y and tPharmacology, Queen's University Kingston Ontario, K7L 3N6, Canada R e c e i v e d 3 O c t o b e r 1983 WIRSCHING, B. A., R. J. BENINGER, K. JHAMANDAS, R. J. BOEGMAN AND S. R. EL-DEFRAWY. Differential effects of scopolamine on working and reference memory of rats in the radial maze. PHARMACOL BIOCHEM BEHAV 20(5) 659--662, 1984.--Anticholinergics have often been found to impair choice accuracy in the radial maze. Some researchers have suggested that this indicates involvement of cholinergically innervated structures in cognitive mapping while others argue that these structures mediate working memory. However, most results are open to either interpretation since the baiting method did not allow a distinction between reference and working memory errors. To further test these hypotheses this study examined the effects of systemic scopolamine on radial maze performance, using a 4-out-of-8 baiting procedure. Food-deprived Wistar rats were pretrained until working memory choice accuracy stabilized to a criterion of 87% or better. Scopolamine (0.1, 0.4 and 0.8 mg/kg, IP, 30 min before a session) significantly increased the number of working memory errors (re-entries into baited arms) whereas reference memory errors (entries into never baited arms) did not change significantly. Observed deficits appeared not to be attributable to a drug-induced disruption of motivational systems. Results confirm the behavioural similarities between the memorial effects of hippocampectomy and anticholinergics, and implicate cholinergically innervated structures in working memory. Working memory

Reference memory

Radial maze

IT has generally been accepted that the hippocampus plays a critical role in human memory. With the repeated observation that hippocampal lesions severely disrupt accuracy in the radial-arm maze this view has been extended to include animal memory [8,15]. However, in the animal literature, there is little consensus regarding the type of memory systems with which the hippocampus is involved [11, 12, 13, 14,

Scopolamine

Acetylcholine

Cholinergic neurons

trials. The baited arms, on the Other hand, formed a working memory component since the information concerning which of the baited arms had already been visited was only useful for that particular trial. It was reasoned that hippocampal rats would err on both the baited and unbaited arms if the impairment reflected a cognitive deficit. On the other hand, if the impairment was of working memory, animals would re-enter the baited arms from which food had already been eaten. Results indicated that rats with fimbria-fornix lesions performed at chance levels on the working memory component of the task while reference memory was unimpaired. These results support the hypothesis that the hippocampal system may be selectively involved in working memory. Since the hippocampal region is strongly innervated with cholinergic fibres [3,9] studies utilizing anticholinergics such as scopolamine have also attempted to evaluate the cognitive mapping versus working memory hypotheses. Although anticholinergics produce behavioural effects on radial maze performance that closely parallel those arising from hippocampal ablations [2, 5, 17, 18], it has not been possible to conclusively argue in favour of either hypothesis, since the baiting method employed did not allow a distinction between working and reference memory. The present study was undertaken to further test the two hypotheses by examining the effects of scopolamine on performance in a partially baited radial maze, with only 4 of the 8 arms baited. According to the cognitive mapping theory drugged rats should err

15]. O ' K e e f e and associates [11,12] proposed a cognitive mapping theory suggesting that the hippocampus processes and stores spatial information. This system provides the animal with a mapping strategy which enables it to locate itself in a geographic environment as well as generate place hypotheses about that environment. In support of this notion it has been found that hippocampal damage leads to a marked deficit in ability to perform maze tasks, particularly those requiring place learning [11]. In contrast to this spatial interpretation is the working memory hypothesis [13, 14, 15] suggesting that behavioural impairments observed in maze performance following hippocampal ablations stem from an inability to solve the working memory component. For instance, Olton and Papas [15] showed a differential involvement of the hippocampal system in reference versus working memory by only baiting 8 of 17 arms in a radial maze. The unbaited set formed a reference memory component of the task since the information that no food was to be found on these arms was useful for all

659

W I R S C H I N G ET AL.

660 TABLE 1 DRUG DESIGN Order 1 2 3 4 5 6

Baseline BL BL BL BL BL

Saline Saline Saline Saline Saline Saline

1 1 l 1 1 1

BL BL BL BL BL BL

0.1" 0.! 0.4 0.4 0.8 0.8

BL BE BL BL BL BL

0.4* 0.8 0.8 0.1 0.4 0.1

BL BL BL BL BL BL

0.8* 0.4 0.1 0.8 0.1 0.4

BL BL BL BL BL BL

Saline Saline Saline Saline Saline Saline

2 2 2 2 2 2

*Dose of scopolamine (mg/kg). TABLE 2 COMPARISON OF SALINE WITH PRECEDING BASELINE BL

S~

BL

S,,

Working Memory Errors Mean (_+SEM)

0.08 (+0.08)

0.33 (+0.19)

0.67 (_+0.22)

0.58 (_+0.23)

Reference Memory Errors Mean (_+SEM)

1.8 (+0.11)

2.4 (_+0.36)

0.33 (_+0.14)

0.58 (_+0.23)

n=12.

on both the unbaited and baited arms. The working memory h y p o t h e s i s , o n the o t h e r h a n d , w o u l d p r e d i c t d i s r u p t e d perf o r m a n c e o n only t h e b a i t e d a r m s . METHOD

Subjects E i g h t e e n e x p e r i m e n t a l l y n a i v e male a l b i n o rats o f the W i s t a r strain were individually h o u s e d in a climatically controlled r o o m o n a 12 h r light/dark cycle. Initial free feeding weights of 200 to 270 g w e r e d e c r e a s e d to 80% ( a d j u s t e d for g r o w t h ) b y daily feeding with m e a s u r e d rations.

Apparatus T h e radial m a z e , e l e v a t e d 50 c m a b o v e the floor, consisted o f a n o c t a g o n a l central p l a t f o r m (30 c m wide) surr o u n d e d b y 8 equally s p a c e d radial a r m s (65 c m l o n g × 10 cm wide). F o o d wells, located 1 c m f r o m the e n d o f e a c h a r m w e r e 1.0 c m d e e p a n d 1.5 c m in d i a m e t e r . T e s t i n g w a s c a r r i e d o u t in a w h i t e p a i n t e d r o o m lit by 70 W f l u o r e s c e n t t u b e s . S e v e r a l visually d i s t i n c t c u e s (e.g., door, shelf) w e r e p r e s e n t in the r o o m a n d r e m a i n e d in the s a m e p o s i t i o n with r e s p e c t to the m a z e .

Procedure D e p r i v a t i o n was b e g u n o n e w e e k p r i o r to testing. D u r i n g this p e r i o d e a c h rat was h a n d l e d daily for a p p r o x i m a t e l y o n e min. O n d a y 5 of d e p r i v a t i o n a n i m a l s were fed a small q u a n tity of F r o o t L o o p s cereal in t h e i r h o m e cage as small pieces s u b s e q u e n t l y were utilized as reinforcers. Pretraining. O n d a y 8 o f d e p r i v a t i o n a n i m a l s were p l a c e d for 10 min in pairs o n the c e n t r a l h u b o f the m a z e w i t h F r o o t L o o p pieces s c a t t e r e d o n the p l a t f o r m a n d a r m s . O n d a y 10 rats w e r e p l a c e d singly on the m a z e . Again food was scatt e r e d o n the p l a t f o r m a n d along a r a n d o m l y p r e d e t e r m i n e d

s u b s e t o f only 4 a r m s , r e f e r r e d to as the baited a r m s . T h e baiting p a t t e r n r e m a i n e d the same t h r o u g h o u t the experim e n t b u t varied from rat to rat. During d a y s 10 to 13 the 4 a r m s w e r e r e b a i t e d until the rat l e a r n e d to r u n to the e n d a n d collect the food in 10 min o r less. Type of arm e n t r y (baited or u n b a i t e d ) was r e c o r d e d . A n a r m e n t r y was defined as crossing a line 10 c m into e a c h arm. A f t e r e a c h trial the m a z e was c l e a n e d with a 2.5% c i d a r v i n e g a r solution. Formal training. E a c h rat r e c e i v e d o n e s e s s i o n p e r day, 7 d a y s a week. At t h e start o f e a c h session, the 4 p r e d e t e r m i n e d a r m s were baited at t h e i r distal e n d (note t h a t a r m s w e r e not r e b a i t e d within a session). E a c h rat w a s p l a c e d on the p l a t f o r m and left until all 4 baits were collected, 14 c h o i c e s were m a d e or 10 min had elapsed, w h i c h e v e r c a m e first. Training c o n t i n u e d until c h o i c e a c c u r a c y stabilized o v e r 4 days to a n a v e r a g e c r i t e r i o n o f 87% or b e t t e r ; thus a score o f 100% (4 out o f 4 u n r e p e a t e d baited a r m e n t r i e s within the first 4 c h o i c e s ) on at least 2 o f the 4 days w a s required. Drug testing. This p h a s e c o n s i s t e d of 5 4-day b l o c k s of drug t r e a t m e n t with o n e session p e r day. E a c h t r e a t m e n t b l o c k was followed b y n o n - d r u g b a s e l i n e s e s s i o n s that cont i n u e d until t h e 4-day c r i t e r i o n o f 87% or b e t t e r was ree s t a b l i s h e d . E a c h s e s s i o n o f the first a n d fifth t r e a t m e n t b l o c k was p r e c e d e d 30 rain b y an IP injection o f saline (1 ml/kg). S e s s i o n s o f t h e s e c o n d , third a n d f o u r t h t r e a t m e n t b l o c k s were p r e c e d e d 30 min b y a n IP injection of s c o p o l a m i n e h y d r o b r o m i d e (Sigma C h e m i c a l Co.) dissolved in distilled water, at d o s e s o f 0.1, 0.4 or 0.8 mg/kg. R a t s were r a n d o m l y a s s i g n e d to o n e of the six p o s s i b l e d o s a g e o r d e r s (see T a b l e 1). Dependent measures. Type o f e r r o r a n d r e i n f o r c e m e n t r e c e i p t s w e r e r e c o r d e d . T h e first e n t r y into a baited a r m r e g a r d l e s s o f w h e t h e r or not the bait was collected was s c o r e d as a c o r r e c t choice, while a r e - e n t r y into t h a t a r m was s c o r e d as a w o r k i n g m e m o r y error. E n t r i e s into n e v e r b a i t e d

E F F E C T S OF S C O P O L A M I N E ON M E M O R Y (/3

oc

661 4'-

4

0 r~

rr

hl

,,

'" " 0

5

0

tlr LI.I an

w

Z

Z

÷!

+1

Z

Z

"'

w

0

<

BLI

BL2

BI_3

CHRONOLOGICAL

BL4

BL5

3~ i 2

0

arms were scored as reference memory errors. The scoring system for working memory may have a problem. Namely, if a rat entered one of the baited arms and did not collect the food a re-entry could be considered correct. Therefore another scoring system for working memory errors was employed: re-entries into an arm that still contained the bait were scored as correct choices. RESULTS

Data from animals that died for reasons unrelated to the experimental design (n=2), failed to learn the task (n=2) or did not run on at least 2 of the 4 drug sessions (n=2) were excluded from statistical analyses. One way repeated measures analyses of variance were performed on two dependent variables: total number of working memory errors (both scoring systems) and total number of reference memory errors. Scores were summed over each 4-day testing block and were based on choices within the first 4 arm entries. Data from the first and fifth 4-day non-drug baseline blocks were compared to saline 1 (S1) and saline 2 ($2), respectively. There was no significant effect of saline in either case for the number of working memory errors or reference memory errors (see Table 2). Non-drug baseline blocks were examined for chronological changes in performance. Figure 1 shows the mean ( _ S E M ) type of error within the first four choices for the chronological baseline blocks 1 to 5. Reference memory errors significantly decreased during the course of baseline testing, F(4,44)= 15.07, p<0.0001. Although working memory errors showed a marginal increase the change was not statistically reliable, F(4,44)=2.4, 0.050.10. These data suggest that low doses of scopolamine (0.1 and 0.4 mg/kg) selectively disrupted the working memory component of the radial maze task. Subsequent post-hoc comparisons with Scheffe's F test conducted

I

I

I

S

.I

.4

.8

DOSE OF S C O P O L A M I N E ( m g / k g )

BASEUNE

FIG. 1. Mean (±sem) type of error within the first 4 choices as a function of chronological baseline. Reference memory errors (©----), working memory errors (Q--), n= 12.

I

BL

FIG. 2. Mean (±sem) type of error within the first 4 choices at various doses of scopolamine. Baseline (BL), saline (S), reference memory errors (O---), working memory errors (O--), n= 12.

~o 0

0

m

"'

z

0

~

z

w

0

4

3

I

i

I

I

1

BL

S

.I

.4

.8

DOSE OF SCOPOLAMINE (mg/kg)

FIG. 3. Mean (±sem) total number of working memory errors from the alternate scoring system (see Procedure section in text). Points show results for baseline (BL), saline (S) and various doses of scopolamine. (n= 12). on working memory errors showed that the combined means of baseline and saline differed significantly from the combined means of the 3 drug doses (.o<0.05). Scopolamine also increased the number of working memory errors relative to the saline condition alone (p <0.05). Although administration of 0.1 mg/kg of scopolamine produced the largest increase in working memory errors, this dose did not differ significantly from 0.4 or 0.8 mg/kg. Similarly there was no significant difference between 0.4 and 0.8 mg/kg. Results from the alternate scoring system of working memory errors are depicted in Fig. 3. In contrast to the scores shown in Fig. 2, re-entries into arms of the baited set where the food was not eaten were scored as correct rather than incorrect choices (see Procedure section). Comparison of Figs. 2 and 3 suggests that failures to eat occurred most often under the higher doses (0.4 and 0.8 mg/kg) of scopolamine. Nevertheless, even when re-entries into still baited arms were scored as correct working memory choices, scopolamine administration significantly increased the mean number of working memory errors, F(4,44)=6.1, p<0.001. Hence, the results from both scoring systems were similar.

662

WlRSCHING ET AL. DISCUSSION

Results of the present experiment are consistent with previous reports showing that anticholinergic drug treatments, like hippocampectomy, severely impair choice accuracy in the radial-arm maze [2, 17, 18]. Eckerman et al. [2], for example, showed that 0.1 mg/kg of scopolamine, like hippocampal lesions reduced continuous choice accuracy in the 8-arm maze. More importantly the present findings demonstrated that low doses of scopolamine (0.1 and 0.4) selectively disrupted the working memory component of this task, a finding consistent with the effects of hippocampectomy [8,15], and cholinergic blockade [10]. Within a session scopolamine-treated animals had difficulty remembering which of the arms had already been visited. This result is in agreement with Olton and associates' [13, 14, 15] working memory proposal. It does not, however, support O 'K eef e and colleagues' [11,12] cognitive mapping theory since scopolamine administration did not significantly increase the number of responses to the unbaited radial arms; spatial abilities apparently were not deficient. Scopolamine thus appears to have deleterious and selective effects on working memory. One possible explanation for this selective action lies in the degree of interference to which each type of memory system is susceptible. Reference memory is composed of permanent representations and is relatively immune to disruption, while working memory is composed of newly formed representations and is vulnerable to disruption [6,14]. If animals with disrupted cholinergic function are more distractable or reactive to environmental stimuli, interference effects might be enhanced resulting in the selective impairment of working memory. This possibility gains some

support from Carlton's [1] suggestion that cholinergic systems have an inhibitory effect on behaviour and the observation that anticholinergics lead to prolonged reactivity to sensory stimuli [7]. It could be argued that the scopolamine-induced deficit reported here may be attributed to a disruption of motivational systems rather than a selective interference with the processes essential for working memory since anticholinergics produce dry mouth effects possibly making food unpalatable [4]. Although refusal of the food was observed during the course of drug testing especially at the higher doses, analysis of the alternate scoring system revealed that when these instances were excluded the mean number of working memory errors still increased significantly. These results are consistent with previous reports that neither scopolamine nor peripherally acting methylscopolamine altered reinforcer effectiveness [2,16] and support the notion that scopolamine may have selective effects on memorial processes. In conclusion, scopolamine like hippocampectomy, produced a selective impairment of working memory. This might suggest that cholinergic neurons projecting to the hippocampus play a major role in this memory process. Work is presently under way to test this hypothesis by selectively depleting the hippocampus of cholinergic afferents with localized injections of neurotoxins.

ACKNOWLEDGEMENTS We wish to thank R. G. Weisman for his critical reading of this manuscript. This research was supported by grants from the Ontario Mental Health Foundation to K. Jhamandas, R. J. Boegman and R. J. Beninger and the Ontario Ministry of Health to R. J. Beninger.

REFERENCES

1. Carlton, P. L. Brain-acetylcholine and inhibition. In: Reinforcement and Behavior, edited by J. T. Tapp. New York: Academic Press, 1969, pp. 286-327. 2. Eckerman, D. A., W. A. Gordon, J. D. Edwards, R. C. MacPhail and M. I. Gage. Effects of scopolamine, pentobarbital and amphetamine on radial maze performance in the rat. Pharmacol Biochem Behav 12: 595-602, 1980. 3. Fibiger, H. C. The organization and some projections of cholinergic neurons of the mammalian forebrain. Brain Res Rev 4: 327-388, 1982. 4. Gilman, A. G., J. S. Goodman and A. Gilman. The Pharmacological Basis o f Therapeutics, (6th ed). New York: MacMillian Press, 1980. 5. Godding, P. R., J. R. Rush and W. W. Beatty. Scopolamine does not disrupt spatial working memory in rats. Pharmacol Biochem Behav 16: 91%923, 1982. 6. Honig, W. K. Studies on working memory in the pigeon. In: Cognitive Processes in Animal Behavior, edited by S. H. Hulse, H. Fowler and W. K. Honig. Hillsdale, NJ: Lawrence Erlbaum Press, 1978, pp. 211-248. 7. Hughes, R. N. A review of atropinic drug effects on exploratory choice behavior in laboratory rodents. Behav Neural Biol 34: 5-41, 1982. 8. Jarrard, L. E. Selective hippocampal lesions and spatial discrimination in the rat. Soc Neurosci Abstr 3: 22, 1978.

9. Lewis, P. R. and C. C. D. Shute. Cholinergic pathways in CNS. In: Handbook o f Psychopharmacology, edited by L. L. Iversen and S. H. Snyder. New York: Plenum Press, 1978, pp. 315-355. 10. Lydon, R. Effects of scopolamine on the performance of a radial maze task in the rat. Honours Bachelors Thesis, Dalhousie University, 1983. l 1. O'Keefe, J. and D. H. Conway. On the trail of the hippocampal engram. Physiol Psychol 8: 22%238, 1980. 12. O'Keefe, J. and L. Nadel. The Hippocampus as a Cognitive Map. Oxford: The Clarendon Press, 1978. 13. Olton, D. S., J. T. Becker and G. E. Handelmann. Hippocampus, space, and memory. Behav Brain Sci 2: 313-365, 1979. 14. Olton, D. S., J. T. Becker and G. E. Handelmann. Hippocampal function: Working memory or cognitive mapping? Physiol Psychol 8: 23%246, 1980. 15. Olton, D. S. and B. C. Papas. Spatial memory and hippocampal function. Neuropsychologia 17: 66%682, 1979. 16. Robustelli, F., S. D. Glick, T. L. Goldfarb, A. Geller and M. E. Jarvik. A further analysis of scopolamine impairment of delayed matching with monkeys. Commun Behav Biol 3: 101-109, 1969. 17. Stevens, R. Scopolamine impairs spatial maze performance in rats. Physiol Behav 27: 385-386, 1981. 18. Watts, J., R. Stevens and C. Robinson. Effects of scopolamine on radial performance in rats. Physiol Behav 26: 845-851, 1981.