Working memory deficits following muscarinic blockade combined with depletion of brain somatostatin in rats

348

Brain Research, 610 (1993) 348-353 ,~3 1993 Elsevier Science Publishers B.V. All rights reserved 0006-8993/93/$06.00

BRES 25641

Working memory deficits following muscarinic blockade combined with depletion of brain somatostatin in rats Masuo Ohno, Shigenobu Shibata, Tsuneyuki Yamamoto and Shigenori Watanabe Department of Pharmacology, Faculty of Pharmaceutical Sciences, Kyushu Unil,ersity, Fukuoka (Japan) (Accepted 2 February 1993)

Key words." Cysteamine; Somatostatin: Scopolamine; Muscarinic receptor; Working memory; Reference memory; Immunohistochemistry; Rat

In a working memory task with three-panel runway paradigm, cysteamine, a depletor of somatostatin, at 100 or 200 m g / k g i.p. given 24 h before testing, had no effect on the number of errors (attempts to pass through two incorrect panels of the three panel-gates at four choice points). Cysteamine at 100 m g / k g caused a significant reduction in somatostatin-like immunoreactivity in the rat brain, including the hippocampus and cerebral cortex. Working memory errors were significantly increased by scopolamine, a muscarinic receptor antagonist, at 0.32 m g / k g i.p. given 20 min before testing, whereas errors were not affected by the 0.1 m g / k g dose. Combined administration of 100 m g / k g cysteamine and 0.1 m g / k g scopolamine significantly increased the number of working memory errors. However, cysteamine at 100 m g / k g and scopolamine at 0.1 m g / k g had no effect on reference memory errors, whether they were administered alone or in combination. These results suggest that depletion of brain somatostatin aggravates working memory deficits induced by blockade of muscarinic receptors.

It has been reported that somatostatin-like immunoreactivity (SL1) and somatostatin receptors are significantly reduced in the cortex and hippocampus in patients with Alzheimer's disease 2"4"2j'22. It is, however, unclear whether the somatostatinergic deficit contributes to memory impairment in this pathological state. Animal studies indicate that cysteamine, a cornpound that produces a rapid, relatively selective and reversible depletion of somatostatin without affecting other neuropeptides in the brain ~219'23, causes a significant impairment of passive avoidance learning in rats ~'~L~2. There have been few reports concerning the effects of cysteamine on memory processes assessed with tasks other than the passive avoidance paradigm, It has been proposed that there are two different types of memory function in experimental animals, i.e. working memory and reference memoryl4'ls. Working memory allows animals to remember information that is useful for a single session of an experiment but not for subsequent sessions, whereas reference memory is defined as the holding of information that is of continued value throughout all sessions. We previously re-

ported that a three-panel runway task is a useful method for studying learning and memory in rats ~'24, particularly because it allows us to distinguish between working and reference memoryl6'~7. In this study, we investigated the effects of cysteamine on working and reference memory, as assessed with the three-panel runway task. Even if somatostatinergic deficits fail to affect the memory function, it is possible that they interact synergistically with cholinergic deficits to impair learning and memory. Scopolamine, a muscarinic receptor antagonist, has been shown to significantly impair the performance of rats on the three-panel runway task w'24. We also examined whether cysteamine aggravates the scopolamine-induced disruption of memory performance in this task. Eight- to ten-week-old male rats of the Wistar strain were obtained from Nippon SLC. The animals weighed between 230-250 g at the start of the experiment; they were then placed on a deprivation schedule to maintain their weights at approximately 80% of the freefeeding level. The rats were housed in groups of four per cage under constant temperature (23 + 2°C) and a

Correspondence." M. Ohno, Department of Pharmacology, Faculty of Pharmaceutical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812, Japan. Fax: (81) (92) 632-2752.

349 12-h light-dark cycle (light period: 07.00-19.00 h), with water freely available, Working memory and reference memory were assessed with a three-panel runway apparatus, as described previously 16'17. In brief, this apparatus (175 x 36 x 25 cm) was composed of a start box, a goal box and four consecutive choice points intervening between them. Each choice point consisted of a gate with three panels (12 x 25 cm). The rats were prevented from passing through two of the three panels in the gate by front stoppers; they were prevented from returning to the start box or to a previous choice point by rear stoppers affixed to each of the panels in all the gates, When the rats reached the goal box, they received two food pellets (about 50 mg each; Muromachi Kikai) as positive reinforcement, Initially, all the front stoppers were removed so that a rat could pass through any one of the three panelgates at each choice point. The rats were made to run the task repeatedly until the time that elapsed from leaving the start box to reaching the goal box was consistently below 20 s. Once this time was reached, the rats were given six consecutive trials (defined as one session) per day with the removal of the front stopper of only one of the three panel-gates (the correct panel-gate) at each choice point. Trials were run at 2-min intervals, and water was freely available between trials in the home cage. In the working memory procedure, the locations of the correct panel-gates were held constant within a session, but were changed from one session to the next. Twelve different patterns of correct panel-gate locations were used, as described previously 8'24. In the reference memory procedure, the correct panel-gate locations were kept constant within a session and in succeeding sessions. The number of times an animal attempted to pass through an incorrect panel-gate (defined as errors) and the time required for the animal to obtain food pellets (defined as latency) were recorded for each rat in each trial of a session. The criterion of learning was less than 12 and less than 6 errors summed across the six trials of a session in the working and reference memory tasks, respectively. A rat was used in the experiment if it achieved this criterion throughout three consecutive sessions. In the case of the working memory task, the number of errors and the latency summed from the second to the sixth trial of a session were important for evaluating the ability of rats to remember new correct panel-gate locations; therefore, these parameters were presented separately from those recorded in the first trial. In the case of the reference memory task, both these parameters were summed across all six trials of a session since this task was given in order to evaluate

the ability of rats to retain the constant location of correct panel-gates. The presence of a significant difference between the groups was determined by a oneway analysis of variance (ANOVA) followed by Dunnett's test when F ratios reached significance ( P < 0.05). For each experimental session, cysteamine hydrochloride (Sigma Chemical Co.) was freshly prepared in distilled water adjusted to pH 7.0-7.4 with an appropriate amount of 1 N N a O H solution. ( - ) S c o p o l a m i n e hydrobromide (Sigma Chemical Co.) was dissolved in distilled water. Drugs were administered at a volume of 0.1 ml per 100 g body weight and the doses were expressed in terms of the salt. Rats that met the learning criterion were injected i.p. with cysteamine and scopolamine 24 h and 20 min, respectively, before the runway test was given. Twenty four hours after an injection of 100 m g / k g cysteamine or saline, the animals received a lethal dose of anesthesia and were perfused transcardially with 50 ml of 0.9% saline solution, followed by 150 ml of 4% paraformaldehyde in 0.1 M phosphate-buffered saline (PBS). The brains were removed and postfixed for 5 h in fixative, and transferred to 20% and 30% sucrose solutions in 0.1 M PBS, until they sank. All brains were frozen and sectioned on a freezing microtome in the coronal plane at 30 /xm (2.0-4.0 mm posterior to bregma). Sections were collected in ice-cold 0.1 M PBS at 90-/zm intervals and divided into two sets for Nissl stain and immunohistochemistry for somatostatin. The tissue was processed for immunohistochemistry according to the avidin-biotin peroxidase complex (ABC) method of Hsu et al. is. Primary antibodies (anti-somatostatin; Immunonuclear) were diluted 1:1,500 in 0.1 M PBS containing 1.0% normal goat serum in 0.3% Triton X-100. The sections were incubated in primary antibody for 48 h at room temperature. After rinsed with PBS, bound antibody was detected using a Vectastain ABC Kit (Vector Labs. Inc.), as described in our previous report t. In the three-panel runway task, the random performance level was four errors per trial, or 24 errors per session. In the working memory task, the number of errors made from the second to sixth trial (working memory errors) markedly decreased with repeated training, whereas the errors in the first trial remained constant at approximately four errors. About 20-30 training sessions were required for the rats to reach the criterion of less than 12 errors summed across the six trials of a session. Latency was also reduced during repeated sessions and was stable from the 10th session on. In the reference memory task, the number of errors and latency in all six trials of a session decreased with

350 TABLE I

Effects of c7steamine and scopolamine on the number of errors and latency in a test ~/" working memoo' The runway test was given 24 h and 20 rain, respectively, after cysteamine and scopolamine were administered. Values are means ± S.E.M. of errors and latencies recorded in the first trial, and those summed from the second to the sixth trial of a session. The significance of the differences from the saline-treated group was determined by a one-way A N O V A followed by Dunnett's test, ** P < 0.01.

Drug

m g / kg

n

Number o f errors Trial 1

Trial 2 6

Trial 1

Trial 2 - 6

6 6 6 6 6

4.7 + 0.5 4.0 + 0.4 4.5+_0.7 3.5 + 0.6 3.8±0.4

4.2 ± 0.7 4.5 ± 1.3 7.3± 1.6 4.2 + 0.9 20.2+_1.6 **

14.7 ± 2.8 13.3 ± 1.5 12.2± 1.1 12.5 ± 1.8 19.5±3.8

28.3 + 2.0 33.3 + 1.3 32.7+ 2.0 25.2 _+ 1.2 71.7± 10.2 **

6

4.7±0.6

15.7_+1.3"*

11.2+1.3

41.5+ 3 . 2 " *

(i.p.) Saline Cysteamine Scopolamine Cysteamine + Scopolamine

100 200 (1.1 0.32 100

Latency (s)

0.1

the repetition of training. The rats could run the task within the six-error criterion summed across six trials after they had about 10 training sessions, Cysteamine at 100 and 200 mg/kg, given i.p. 24 h before testing, had no significant effect either on the number of errors or on the latency, as assessed by the working memory procedure in the three-panel runway task (Fig. 1 and Table I). Scopolamine at 0.1 and 0.32 mg/kg, given i.p. 20 rain before testing, dose-dependently increased the number of working memory errors (F2,15 = 65.92, P < 0.01), an effect that reached significance only for the 0.32 m g / k g dose, whereas it did not affect the number of errors made in the first trial. Scopolamine at 0.32 m g / k g also prolonged the latency summed from the second to the sixth trial of a session (F2.15 = 18.45, P < 0.01), without affecting that recorded in the first trial. Combined administration of 100 m g / k g cysteamine and 0.1 m g / k g scopolamine, each of which had no individual effect on errors or latency, significantly increased the number of working memory errors (F~.~o = 63.82, P < 0.01) and the latency recorded in trials second to sixth (Fl.~O = 12.13, P < 0.01). The combination did not affect errors or latency recorded in the first trial. In the case of the reference memory task, neither cysteamine at 100 and 200 m g / k g i.p. nor 0.1 m g / k g scopolamine had an effect on the number of errors or on the latency (Table II). When rats were given 100 m g / k g cysteamine together with 0.1 m g / k g scopolamine, they did not show a significant increase in reference memory errors or in the latency. Immunohistochemical analysis revealed a consistent distribution pattern of SLI in all brains taken from control rats. In the temporal cortex a n d h i p p o c a m p u s , there occurred many SLI neurons throughout layers I I - g (Fig. 2C) and the dentate gyrus (Fig. 2B), respectively, whereas in the hypothalamus, numerous SLI fibers were found throughout the median eminence

and arcuate nucleus (Fig. 2A). In the 100 m g / k g cysteamine-treated rats, only a few SLI neurons were observed in the cerebral cortex (Fig. 2F) and hippocampal dentate gyrus (Fig. 2E), and there were a moderate amount of SLI fibers in the median eminence and arcuate nucleus (Fig. 2D). In the present study, systemic administration of 100 or 200 m g / k g cysteamine did not affect working or reference memory performance of rats, whereas 0.32 m g / k g of scopolamine, a muscarinic receptor antagonist, significantly impaired working memory. The 100 m g / k g of cysteamine was sufficient to produce a

5

1 u T

o 4 ~ t~ 3

f •~

~ I ~ ~ [ ~

o "2 E = z

$

]

I \~ , ~ .~,"x ~ 1

0

I 1

I 2

i I 3 4 Trials

I 5

f 6

Fig. 1. Effects of cysteamine and scopolamine on the number of errors in a test of working memory. The runway test was given 24 h and 20 rain, respectively, after 100 m g / k g cysteamine and 0.1 m g / k g scopolaminewere i.p. administered (o, saline; e, cysteamine alone; ~, scopolamine alone; A, cysteamine+scopolamine). Each point represents the mean + S.E.M. of errors and latencies for six animals

recorded in each trial of a session.

351 m a r k e d r e d u c t i o n o f S L I n e u r o n s in t h e h i p p o c a m p u s a n d c e r e b r a l c o r t e x a n d o f S L I fibers in t h e h y p o t h a l a mus. C y s t e a m i n e at d o s e s h i g h e r t h a n 150 m g / k g n o t only d e p l e t e s s o m a t o s t a t i n , b u t also d e c r e a s e s cortical n o r a d r e n a l i n e a n d i n c r e a s e s cortical d o p a m i n e levels ~2, by acting as a d o p a m i n e - / 3 - h y d r o x y l a s e i n h i b i t o r 6. D o s e s o f 50 a n d 100 m g / k g

cysteamine, capable of

TABLE II Effects of cysteamine and scopolamine on the number of errors and latency in a test of reference memory

The runway test was given 24 h and 20 min, respectively, after cysteamine and scopolamine were administered. Values are means_ S.E.M. of errors and latencies summed across all six trials of a session. Drug

i n d u c i n g specific s o m a t o s t a t i n d e p l e t i o n , have b e e n

mg / kg

n

Number of errors Trial 1-6

Latency (s) Trial 1-6

5 5 5 5

2.0_+0.9 1.0-+0.6 1.8-+0.4 2.25:0.4

33.6+_2.2 38.4+3.4 45.0+3.9 34.0-+2.8

5

1.0-+0.5

37,0+3.3

(i.p.)

r e p o r t e d to i m p a i r r e t e n t i o n o f single trial passive avoidance when administered 1 h before or immedi-

Saline Cysteamine

ately a f t e r t r a i n i n g session 5'~2. H o w e v e r , relatively w e a k initial t r a i n i n g a p p e a r s to b e r e q u i r e d to d e m o n s t r a t e t h e d i s r u p t i v e effect o f c y s t e a m i n e on the passive a v o i d a n c e r e s p o n s e , i.e. this effect is a t t e n u a t e d w h e n

Cysteamine + Scopolamine

Scopolamine

-

100 200 0.1 100 0.1

~m Fig. 2. Photomicrographs showing somatostatin-like immunoreactive neurons and fibers in the brain of control rats (left, A-C) and 100 mg/kg cysteamine-treated rats (right, D-F). Immunoreactivity was assessed 24 h after cysteamine or saline was administered. A, D, hypothalamus; B, E, hippocampus; C, F, cerebral cortex; Bar = 100 ~m.

352 shock intensity is i n c r e a s e d 5'12. W i t h r e g a r d to l e a r n i n g tasks o t h e r t h a n the passive a v o i d a n c e p a r a d i g m , cyst e a m i n e fails to have significant effects on the M o r r i s w a t e r m a z e or d e l a y e d a l t e r n a t i o n task 5'7. T a k e n to-

statinergic deficits e v i d e n c e d by r e d u c t i o n s in SLI a n d s o m a t o s t a t i n r e c e p t o r s have, for e x a m p l e , b e e n d e m o n s t r a t e d as o n e of the most c o n s i s t e n t c h a n g e s o b s e r v e d in the A D b r a i n s 2'4'21'22. Thus, the p r e s e n t finding that

gether, t h e s e findings suggest t h a t the role of b r a i n s o m a t o s t a t i n in m e m o r y p r o c e s s e s is m o r e limited a n d trivial t h a n that of the cholinergic system, which is known to be critical for l e a r n i n g a n d m e m o r y . T h e m a x i m a l d e p l e t i o n of s o m a t o s t a t i n i n d u c e d by cyst e a m i n e is rarely g r e a t e r t h a n 50% of control levels, even if doses h i g h e r t h a n 100 m g / k g a r e administ e r e d 7'11A2'23 Thus, we can not rule out the possibility • that d e p l e t i o n of m o r e t h a n 50% of b r a i n s o m a t o s t a t i n , if possible, may affect m e m o r y processes. T h e m a j o r finding of the p r e s e n t study is t h a t cys-

d e p l e t i o n of b r a i n s o m a t o s t a t i n a g g r a v a t e d w o r k i n g m e m o r y deficits i n d u c e d by b l o c k a d e o f m u s c a r i n i c r e c e p t o r s may p r o v i d e evidence of s o m a t o s t a t i n e r g i c deficits c o n t r i b u t i n g in s o m e m a n n e r to the severity of m e m o r y d e c l i n e a s s o c i a t e d with cholinergic deficits in AD.

teamine-induced depletion of brain somatostatin imp a i r e d working m e m o r y w h e n a d m i n i s t e r e d t o g e t h e r

with 0.1 mg/kg scopolamine, which by itself did not affect working memory. This effect can not be nonspecific, e.g. m o t i v a t i o n a l o r a t t e n t i o n a l , since no effect on reference memory was found. Haroutunian et al. 11 r e p o r t e d , using passive a v o i d a n c e in rats, t h a t combi-

nation of forebrain cholinergic deficits and somatostatinergic deficits did not l e a d to any g r e a t e r i m p a i r m e n t o f m e m o r y function t h a n t h a t p r o d u c e d by

cholinergic deficits alone. This discrepancy may arise from differences in cholinergic manipulations; Haroutunian et al. it lesioned the nucleus basalis of Meynert to p r o d u c e f o r e b r a i n c h o l i n e r g i c deficits, which by itself had deleterious effects on memory processes, w h e r e a s we e m p l o y e d an ineffective d o s e o f scopol a m i n e to affect t h e p e r f o r m a n c e o n the t h r e e - p a n e l

runway task. At any rate, the present study clearly shows that depletion of brain somatostatin aggravates s c o p o l a m i n e - i n d u c e d deficit in w o r k i n g m e m o r y , i.e. acquisition o f new a n d v a r i a b l e i n f o r m a t i o n . Such in-

teractions between cholinergic and somatostatinergic functions a r e not f o u n d in r e f e r e n c e m e m o r y p e r f o r m a n c e , i.e. r e t e n t i o n or retrieval of previously a c q u i r e d

constant information. The present result does not, however, rule o u t a c h o l i n e r g i c / s o m a t o s t a t i n e r g i c int e r a c t i o n in t h e acquisition of r e f e r e n c e m e m o r y . W h i l e c o r r e l a t i o n s have b e e n f o u n d b e t w e e n t h e d e g r e e of c h o l i n e r g i c deficit a n d cognitive d e c l i n e in patients with Alzheimer's disease (AD) 2°, there occur d e g e n e r a t i o n s o f o t h e r n e u r o t r a n s m i t t e r systems such as peptidergic and monoaminergic systems in this

pathological state 1°. These non-cholinergic deficits may c o n t r i b u t e to t h e m e m o r y dysfunctions in A D p a t i e n t s ,

because clinical trials with cholinomimetic therapies have mostly failed to significantly alleviate cognitive deficits a s s o c i a t e d with this d i s e a s e 3'9't3. T h e s o m a t o -

This research was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture, Japan. 1 Aguilar-Roblero, R., Shibata, S., Speh, J.C., Drucker-Colln, R. and Moore, R.Y. (1992) Morphological and functional development of the suprachiasmatic nucleus in transplanted fetal hypothalamus, Brain Res., 580, 288-296. 2 Beal, M.F., Mazurek, M.F., Yran, V.T., Chattha, G., Bird, E.D. and Martin, J.B. (1985) Reduced numbers of somatostatin receptors in the cerebral cortex in Alzheimer's disease, Science, 229, 289-291. 3 Bruno, G., Mohr, E., Gillespie, M., Fedio, P. and Chase, T.N. (1986) Muscarinic agonist therapy of Alzheimer's Disease: a clinical trial of RS-86, Arch. NeuroL, 43, 659-661. 4 Davies, P., Katzman, R. and Terry, R.D. (1980) Reduced somatostatin-like immunoreactivity in cerebral cortex from cases of Alzheimer disease and Alzheimer senile dementa, Nature, 288, 279-280. 5 DeNoble, V.J., Hepler, D.J. and Barto, R.A. (1989) Cysteamineinduced depletion of somatostatin produces differential cognitive deficits in rats, Brain Res., 482, 42-48. 6 DiLiberato, E.J., DiStefano, V. and Smith, J.C. (1973) Mechanism and kinetics of the inhibition of dopamine-/3-hydroxylase by 2-mercaptoethylguanidine, Biochem. Pharmacol., 22, 2961-2972. 7 Fitzgerald, L.W. and Dokla, C.P.J. (1989) Morris water task impairment and hypoactivity following cysteamine-induced reductions of somatostatin-like immunoreactivity, Brain Res., 505, 24625o. 8 Furuya, Y., Yamamoto, T., Yatsugi, S. and Ueki, S. (1988) A new method for studying working memory by using the three-panel runway apparatus in rats, Jpn. J. Pharmacol., 46, 183-188. 9 Gottfries, C.G. (1990) Pharmacological treatment strategies in Alzheimer type dementia, Eur. NeuropsychopharmacoL, 1, 1-5. 10 Hardy, J., Adolfsson, R., Alafuzoff, I., Bueht, G., Marcusson, J., Nyberg, P., Perdahl, E., Wester, P. and Winblad, B. (1985) Transmitter deficits in Alzheimer's disease, Neurochem. Int., 7, 545-563. 11 Haroutunian, V., Kauof, P.D. and Davis, K.L. (1989) Interactions of forebrain cholinergic and somatostatinergic systems in the rat, Brain Res., 496, 98-104. 12 Haroutunian, V., Martin, R., Campbell, G.A., Tsuboyama, G.K. and Davis, K.L. (1987) Cysteamine-induced depletion of central somatostatin-like immunoreaetivity: effects on behavior, learning, memory and brain neurochemistry, Brain Res., 403, 234-242. 13 Hollander, E., Mohs, R.C. and Davis, K.L. (1986) Cholinergic approaches to the treatment of Alzheimer's disease, Br. Med. Bull., 42, 97-100. 14 Honig, W.K., Studies of working memory in the pigeon. In S.H. Hulse, H. Fowler and W.K. Honig (Eds.), Cognitive Processes in Animal Behavior, Lawrence Erlbaum, Hillsdale, 1978, pp. 211248. 15 Hsu, S.-M., Raine, L. and Fanger, H. (1981) Use of avidinbiotin-peroxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabeled antibody (PAP) procedures, J. Histochem. Cytochem., 29, 577-580.

353 16 Ohno, M., Yamamoto, T. and Ueki, S. (1991) Effect of the K-receptor agonist, U-50,488H, on cerebral ischemia-induced impairment of working memory assessed in rats by a three-panel runway task, Eur. J. Pharmacol., 193, 357-361. 17 Ohno, M., Yamamoto, T. and Watanabe, S. (1992) Effects of intra-hippocampal injections of N-methyl-o-aspartate receptor antagonists and scopolamine on working and reference memory assessed in rats by a three-panel runway task, J. Pharmacol. Exp. Then., 263, 943-950. 18 Olton, D.S., Becker, J.T. and Handelmann, G.E. (1979) Hippocampus, space, and memory, Behav. Brain Sci., 2, 313-365. 19 Palkovits, M., Brownstein, M.J., Eiden, L.E., Beinfeld, M.C., Russell, J., Arimura, A. and Szabo, S. (1982) Selective depletion of somatostatin in rat brain by cysteamine, Brain Res., 240, 178-180. 20 Perry, E.K., Tomlinson, B.E., Blessed, G., Bergmann, K., Gibson, P.H. and Perry, R.H. (1978) Correlation of cholinergic abnormalities with senile plaques and mental test scores in senile dementia, Br. Med. J., 2, 1457-1459.

21 Reinikainen, K.J., Riekkinen, P.J., Jolkkonen, J., Kosma, V.-M. and Soininen, H. (1987) Decreased somatostatin-like immunoreactivity in cerebral cortex and cerebrospinal fluid in Alzheimer's disease, Brain Res., 402, 103-108. 22 Rossor, M.N., Emson, P.C., Mountjoy, C.Q., Roth, M. and Iversen, L.L. (1980) Reduced amounts of immunoreactive somatostatin in the temporal cortex in senile dementia of Alzheimer type, Neurosci. Lett., 20, 373-377. 23 Sagar, S.M., Landry, D., Millard, W.J., Badger, T.M., Arnold, M.A. and Martin, J.B. (1982) Depletion of somatostatin-like immunoreactivity in the rat central nervous system by cysteamine, J. Neurosci., 2, 225-231. 24 Yamamoto, T., Yatsugi, S., Ohno, M., Furuya, Y., Kitajima, I. and Ueki, S. (1990) Minaprine improves impairment of working memory induced by scopolamine and cerebral ischemia in rats, Psychopharmacology, 100, 316-322.