Neuropharmacology Vol. 31, No. 9, pp. 915-922, 1992 Printed in Great Britain. All rights reserved
0028-3908/92 $5.00 + 0.00 Copyright Q 1992 Pergamon Press Ltd
INVOLVEMENT OF CHOLINERGIC MECHANISMS IN IMPAIRMENT OF WORKING MEMORY IN RATS FOLLOWING BASOLATERAL AMYGDALOID LESIONS M.OHNO,*
T. YAMAMOTOand S. WATANABE
Department of Pharmacology, Faculty of Pharmaceutical Sciences, Kyushu University, Fukuoka 812, Japan (Accepted
3 March
1992)
Summary-In order to clarify the role of the amygdala in the working and reference memory of rats in the three-panel runway task, the effects of lesions of subnuclei of the amygdaloid complex on this behavior were studied. Rats that had been trained preoperatively, until they achieved the criterion of learning, were subjected to lesions of the amygdala. In the working memory task, lesions of the basolateral subdivision of the amygdala caused a significant increase in the number of errors (attempts to pass through two incorrect panels of the three panel-gates at four choice points), while lesions of the corticomedial amygdala had no effect on working memory errors. The increase in working errors, observed in basolateral amygdaloid-lesioned rats, declined gradually as retraining sessions were given once each day, reverting to control levels on and after the sixth session. In the reference memory task, the number of errors was not affected by lesions of the basolateral or corticomedial amygdala. The increase in working memory errors, induced by lesions of the basolateral amygdala was significantly reduced by intraperitoneal administration of the inhibitors of cholinesterase, tetrahydroaminoacridine (0.32-1.0 mg/kg) and physostigmine (0.032-O. 1 mg/kg), and the muscarinic receptor agonist, oxotremorine (0.1 mg/kg), before the runway test. These findings suggest that the basolateral amygdala is selectively involved in working memory but not in reference memory and that the lowering of central cholinergic function may account for the impairment of working memory, induced by lesions of the basolateral amygdala. Key words-amygdala,
brain lesions, working memory, reference memory, tetrahydroaminoacridine (THA), physostigmine, oxotremorine.
The amygdala is a subcortical structure, which occupies a strategic position in the circuitry of the limbic system and has been assumed to play an important role in the control, not only of emotional and motivational behavior, but also of memory functions. Patients with Alzheimer’s disease often show disturbances in limbic functions, such as emotional
deficits in addition to a profound decline in memory and intellectual acuity. Recently, it has been demonstrated that the amygdala undergoes severe neuropathological alterations in Alzheimer’s disease. A large number of senile plaques and neurofibrillary tangles, known as specific pathological hallmarks of Alzheimer’s disease, occur to varying degrees in all subnuclei of the amygdaloid complex in patients with Alzheimer’s disease (Brady and Mufson, 1990; Brashear, Godec and Carlsen, 1988; Unger, McNeill, Lapham and Hamill, 1988). A decrease in the amygdaloid area, as assessed by cresyl violet staining, is found in the brains of patients with Alzheimer’s disease (Herzog and Kemper, 1980). Consistently, it has also been reported that the activity of cholinergic markers, such as choline acetyltransferase (ChAT) and acetylcholinesterase (AChE), is markedly re-
*To whom correspondence
should be addressed. 915
duced in the amygdala from patients with Alzheimer’s disease (Brady and Mufson, 1990; Brashear et al., 1988; Davies and Maloney, 1976; Rossor, Iversen, Reynolds, Mountjoy and Roth, 1984). Thus, these neuropathological and biochemical changes in the amygdala may account for the marked memory loss and emotional disorders, seen in patients with Alzheimer’s disease. Although the role of the amygdala in learning and memory has been extensively studied in rats, monkeys and cats, few studies have examined memory impairments following lesions of the amygdala, confined to specific subnuclei (for review, see Sarter and Markowitsch, 1985). Lesions which simultaneously destroy a number of subnuclei, within the amygdaloid complex, preclude understanding of the relationship of specific nuclei to learning and memory. Since the amygdaloid complex has been divided into a basolateral and a corticomedial part, this study was undertaken to elucidate the differential roles of these two portions of the amygdala in the learning and memory of rats. It has been proposed that there are two different types of memory function in experimental animals, i.e. working memory and reference memory (Honig, 1978; Olton, Becker and Handelmann, 1979). Working memory allows animals to remember information that is useful for a single
M. Omo
916
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. It was previously reported that a three-panel runway task served well as a method for the study of learning and memory in the rat (Furuya, Yamamoto, Yatsugi and Ueki, 1988; Yamamoto, Yatsugi, Ohno, Furuya, Kitajima and Ueki, 1990b). With this experimental paradigm, memory can be divided into working and reference memory (Ohno, Yamamoto and Ueki, 1991). In the present study, the differential effects of lesions of the basolateral and corticomedial amygdala were investigated on working and reference memory, which was assessed in rats by the three-panel runway task. Furthermore, the effects of choline&c activating drugs, such as tetrahydroaminoa~dine (THA), physostigmine and oxotremorine on memory impairment in amygdaloid-lesioned rats were examined, as the amygdala receives a cholinergic innervation from the basal forebrain (Mesulam, Mufson, Wainer and Levey, 1983). METHODS
Subjects Eight- to lo-week-old male rats of the Wistar strain were obtained from Nippon SLC. The subjects 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 free-feeding level. The rats were housed in groups of 4 per cage under constant temperature (23 + 2°C) and a 12-hr light-dark cycle (light period: 07:00-19:00), with water freely available. Apparatus Working memory and reference memory were assessed with a three-panel runway apparatus (Fig. 1), that was introduced previously (Furuya et al., 1988;
et
al.
Ohno et al., 1991; Yamamoto et al., 1990b). In brief, this apparatus (175 x 36 x 25 cm) was composed of a start box, a goal box and four consecutive choice points inte~ening 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 2 food pellets (about 50mg each; Muromachi Kikai) as positive reinforcement. Acquisition training Initially, all the front stoppers were removed so that a rat could pass through any one of the three panel-gates 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 20sec. Once this time was reached, the rats were given 6 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 (Furuya et al., 1988; Yamamoto ef al., 199Ob). 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
A:
Start box
B: Goal box
Fig. 1. Schematic drawing of the three-panel runway apparatus. In the runway task, the rats have to pass four consecutive choice points to obtain food pellets placed in the goal box. Each choice point consisted of a gate with three panels. The rats could pass through only one of the three panel-gates (the correct panel-gate) at each choice point.
Amygdaloid complex and working memory
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 6 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 3 consecutive sessions. Surgery
The rats were subjected to lesions of the amygdala after they achieved the criterion of learning. Each rat was anesthetized with sodium pentobarbital (40 mg/kg, i.p.) and was fixed onto a stereotaxic instrument. Lesions of the amygdala were produced with a lesion generator (Muromachi Kikai; RFG4A), in which the temperature of the electrode tip could be monitored. The electrode (0.25 mm dia, 1.O-mm exposed tip) was inserted into the basolateral amygdala [anterior (A): -2.8 mm from bregma; lateral (L): k5.0 mm; horizontal (H): 8.8 mm] and corticomedial amygdala (A: - 2.8 mm, L: f 3.2 mm, H: 9.2 mm), according to the coordinates in the atlas of the rat brain of Paxinos and Watson (1982). The bilateral amygdaloid lesions were induced by an increase in the temperature of the tissue to 70°C within 15 set, this temperature being maintained for another 45 sec. Sham-operated rats were also fixed on the stereotaxic apparatus, under pentobarbital anesthesia and underwent small burr holes in the skull but they did not receive insertion of the electrode. The rats were allowed 5 days to recover from surgery before being used for testing. Thereafter, the runway test was given once each day for 7 consecutive
917
days. In different groups of amygdaloid-lesioned rats, the effects of the choline@ drugs on the impairment of memory were investigated 5 days after lesioning. Drugs
The drugs used in this study were tetrahydroaminoacridine hydrochloride (THA; Aldrich Chemical Co.), physostigmine salicylate (Sigma Chemical Co.) and oxotremorine sesquifumarate (Sigma Chemical Co.). All drugs were dissolved in distilled water and administered in a volume of 0.1 ml per 1OOg body weight. The doses expressed in terms of the
for the drugs
are
salt. Tetrahydroaminoacridine was administered intraperitoneally 20 min before the runway test, which was held 5 days after the amygdaloid lesions. Other drugs were given intraperitoneally 30 min before testing. Histology
After completion of the experiment, the rats given basolateral and corticomedial amygdaloid lesions were anesthetized with ether and the brain was perfused with 10% formalin solution, through the left cardiac ventricle. The sacrifice procedure conformed to the standards of animal procedures, established at Kyushu University. After the brain was removed, it was frozen and sliced to a thickness of 70pm. All sliced sections were stained with cresyl violet. The site and extent of each lesion of the amygdala were verified histologically. The extent of the minimum and maximum lesions of the basolateral and corticomedial amygdala, of all animals tested, is illustrated in Fig. 2.
-2.3 mm
-2.8 mm
-3.3 mm
Fig. 2. Schematic drawing of the minimum (black area) and the maximum (black + stippled area) extent of lesions of the basolateral (A) and corticomedial (B) amygdala. The number on the right of each drawing shows the posterior distance of the plate from bregma. Serial plates were adapted from the atlas of the brain of the rat of Paxinos and Watson (1982).
918
M. OHNOet ai.
Anal.vsis of data 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 evaluation of 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 6 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 one-way analysis of variance (ANOVA), that was followed by Dunnett’s test when F ratios reached significance (P < 0.05). RESULTS
In the three-panel runway task, the random level of performance was 4 errors per trial or 24 errors per session. In the working memory task, the number of errors made from the second to the sixth trial (working memory errors) markedly decreased with repeated training, whereas the errors in the first trial remained constant at approximately 4 errors. About 15-20 training sessions were required for the rats to reach the criterion of less than 12 errors, summed across the 6 trials of a session. The latency was also reduced during repeated sessions and was stable from the tenth session on. In the reference memory task, the number of errors and latency in all 6 trials of a session decreased with the repetition of training. The rats could run the task within the 6-error criterion, summed across 6 trials, after they had had 5-10 training sessions. Lesions of the basolateral amygdala produced a significant increase in the number of errors, assessed by the working memory procedure, done in the three-panel runway task [F(l,lO) = 43.66, P < O.Ol] but did not affect the number of errors made in the first trial (Fig. 3 and Table 1). Lesions of the basolatera! amygdala prolonged the latency in the first trial [F(l,lO) = 6.95, P < 0.051, as well as the latency summed from the second to the sixth trial of a session [F(l,lO) = 13.55, P e 0.011. In contrast, rats with lesions of the corticomedial amygdala exhibited no increase in errors of working memory (mean + SE: 6.2 f 1.7) or errors in the first trial (5.3 + OS), as compared with those of sham-lesioned rats. Corticomedial amygdaloid lesions significantly prolonged the latency, recorded in the first trial [31.5 + 4.3 set, F(l,lO) = 6.25, P < 0.051, without affecting that recorded in trials second to sixth (68.8 f 12.9 WC). In the case of the reference memory task, neither lesions of the basolateral nor corticomedial amygdala produced increases in the number of errors or in the latency, across all 6 trials of a session (Fig. 3). The increase in working memory errors, induced by lesions of the basolateral amygdala, declined
gradually as subsequent sessions were given once each day, reverting to control levels on and after the sixth re-training session (Fig. 4). As shown in Fig. 5 and Table 1, administration of 0.32 and 1.0 mg/kg of THA, before the test session, significantly reduced the increase in working memory errors that would be expected to occur after lesions of the basolateral amygdala [F(2,15) = 13.06, P < O.Ol]. Treatment with THA also decreased the expected increase in latency from the second to the sixth trial [F(2,15) = 4.84, P < 0.051, an effect that reached significance only for the 1.0 mg/kg dose (P i 0.05). Basolateral amygdaloid-lesioned rats failed to run the task when they were given 3.2 mg/kg of THA (n = 2, data not shown). Physostigmine, at doses of 0.032 and 0.1 mg/kg, significantly reduced the increase in working errors induced by lesions of the basolateral amygdala [F(2,15) = 9.87, P < 0.011. A significant reduction in the increase of working errors was also seen when basolateral amygdaloidlesioned rats were treated with 0.1 mg/kg of oxotremorine before the runway test [F(l ,lO) = 26.96, P < 0.011. However, neither physosti~ine nor oxotremorine had an effect on the prolonged latency in rats with lesions of the basolateral amygdala. DISCUSSION
In this study, working memory, assessed in rats by a three-panel runway task, was signi~~ntly impaired by lesions of the basolateral amygdala but was not affected by lesions of the corticomedial amygdala. In contrast, lesions of neither the basolateral nor corticomedial amygdala had a significant effect on reference memory performance. These results suggest that the basolateral amygdala is selectively involved in working memory performance, in which animals are required to remember new information that is useful for only one session. Conceivably, the amyg daloid complex does not play a role in reference memory performance, in which animals have to retain constant info~ation, that is of continued value throughout all sessions. Recently, it has been reported that lesions of the basolateral amygdaloid nucleus induced with ibotenic acid significantly impair acquisition of passive avoidance, while basolatera! amygdaloid-lesioned rats exhibit retention of passive avoidance fairly well, as do sham-operated rats (Lorenzini, Bucherelli, Giachetti, Mugnai and Tassoni, 1991). Reversal learning of the deermouse, in a T-shaped water maze, which is related to the ability to respond to changing environmental demands, also deteriorated as a result of lesioning of the basolateral amygdala but not that of the medial amygdala (Elefthe~ou, Ehas and Norman, 1972). Clinically, the most obvious amygdaloid atrophy, observed in patients with Alzheimer’s disease, concerns the lateral and basolateral nuclei (Brady and Mufson, 1990). Thus, dysfunction of the basolatera1 portion of the amygdala may, at least in part,
Amygdaloid complex and working memory
919
Working Memory x
i ’ 123456
’
’
’
I
Ttw8 Rctorcnce Memory 5
4
123455
123456 TtWB
TM8
Fig. 3. Effects of lesions of the amygdala on working and reference memory errors and latency, in the three-panel runway task. Rats were trained preoperatively until they achieved the criterion and were then subjected to lesions of the amygdala [(O) sham, (0) basolateral amygdala, (A) corticomedial amygdala]. The runway test was given 5 days after surgery. Each point represents the mean f SE of errors and latencies for 6 animals, recorded in each trial of a session. contribute to severe deficiencies of the memory seen in patients with Alzheimer’s disease. There is much evidence that central cholinergic transmission plays a pivotal role in learning and Table
I.
memory. Drugs that block choline&z activity exert amnesic effects (Frumier, Herckat and Jarvik, 1976; Hagan, Jansen and Broekkamp, 1987; Sitaram, Weingartner and Gillin, 1978; Watts, Stevens and
tetrahydroaminoacridine (THA), physostigmine and oxotnmorineon increasesin workingmemoryerrors and latencv.inducedbv lesions of the basolateral am&alit. in the three-Dane1runwav task
Effects of
Numberof LeSiOIlS
Drug
Sham Basolateral amygdala Basolateral amygdala
THA
Basolateral amygdala
Physostigmine
BasoIateraI amygdala
OX0tTCIil0rilIC
mg/kg (i.p.) 0.32 1.0 0.032 0.1 0.1
N
Trial
6 6 6 6 6 6 6
4.3 f 4.3 f 4.5 f 3.8 4 4.2 & 4.3 f 4.0 f
I 0.6 0.7 0.6 0.5 0.6 0.6 0.5
Latency (see)
errors Trial 2-6 4.3 & 0.8 18.3 f 1.9$ 10.0 f 1.3” 7.7 f I .4** 9.7 f 1.9*+ 8.0 f I.4’* 7.3 f 0.8.”
Trial 18.7 f 37.0 k 29.0 * 39.5 f 36.5 f 33.7 + 29.5 *
1 2.8 6.4t 9.9 9.9 16.3 7.8 7.4
Trial 2-6 39.7 f 105.8 & SO.2f 52.3 f 63.8 i 62.3 f go.2 f
4.1 17.5t 9.6 6.7’ 16.4 12.6 13.4
Rats that had been trained preoperatively until they achieved the criterion, were subjected to lesionsof the basolateral amygdala. Tetrahydroaminoacridine was administered intraperitoneally 2Omin before the runway test, which was held 5 days after the lesions. Physostigminc and oxotremorine were given intraperitoneally 30min before the test session. Each value represents the mean f SE. The significance of the differences from sham-lesioned rats (tP < 0.05, $P < 0.01) and from basolateral amygdaloid-Iesioned rats (*P < 0.05, l*f < 0.01) was determined by a one-way ANOVA, followed by Dunneu’s test.
M.
920
OHNO
et al.
Worklng Memory
berg, Pfefferbaum, Hollister and Kopell, 1978; Sitaram et al., 1978). The activity of ChAT has been shown to decrease markedly in the amygdala as well as in the hippocampus and cortex, in patients with Alzheimer’s disease (Davies and Maloney, 1976; Henke and Lang, 1983; Rossor et al., 1984). The importance of the cholinergic deficit for memory deficits is emphasized by a clinical study, suggesting that the degree of loss of choline@ function correlates with the severity of dementia in patients with Alzheimer’s disease (Perry, Tomlinson, Blessed, Bergmann, Gibson and Perry, 1978). In the present study, the effects of some cholinergic activating 1 drugs, including THA, which has been reported to be RI1 2 3 4 5 6 7 useful in the treatment of patients with Alzheimer’s SOuiOltS disease (Summers, Majovski, Marsh, Tachiki and Fig. 4. Changes of errors of working memory, assessed in Kling, 1986; Summers, Viesselman, Marsh and amygdaloid-lesioned rats by the three-panel runway task, Candelora, 1981) were investigated. The inhibitors when retraining sessions were repeated. Rats that had been trained preoperatively until they achieved the criterion (Pre) of AChE, THA and physostigmine, ameliorated were subjected to lesions of the amygdala [(C) sham, (0) significantly the impairment of working memory in basolateral amygdala, (A) corticomedial amygdala]. The rats with lesions of the basolateral amygdala, when rats had been retrained once each day for 7 consecutive they were administered before the test session. The days, starting from the fifth day (the first session) after effective doses of THA and physostigmine, in this surgery. Each point represents the mean f SE of errors for 6 animals, summed from the second to sixth trial of a study, have been shown to produce significant session. The significance of the differences from shaminhibition of the activity of AChE in the brain of the lesioned rats (**Pc 0.01) was determined by a one-way rat at the time of testing (Kwo-On-Yuen, Mandel, ANOVA, followed by Dunnett’s test. Chen and Thal, 1990; Mandel and Thal, 1988). Furthermore, the muscarinic receptor agonist, oxotremorine, also alleviated the working memory Robinson, 1981), whereas cholinergic activating deficit after lesions of the basolateral amygdala. drugs, such as muscarinic agonists and inhibitors These findings suggest that impairment of working of AChE, facilitate learning and memory both in memory, induced by lesions of the basolateral amyghumans and in animals (Baratti, Huygens, Miiio, dala, results from dysfunction of the central cholinMerlo and Gardella, 1979; Davis, Mohs, Tinklenergic system. ”
1
Worklng Yomory 50
40
3 30 E - 20 : 10
1
’
’
’
12345s
’
4
0
,
1
’
1
’
,
123456 Trlalr
Trlalr
Fig. 5. Effect of tetrahydroaminoacridine (THA) on increases in errors of working memory and latency, induced by lesions of the basolateral amygdala in the three-panel runway task. Rats that had been trained preoperatively until they achieved the criterion were subjected to lesions of the amygdala [(C) sham, (0) basolateral amygdala). Tetrahydroaminoacridine was administered intraperitoneally 20 min before the runway test, which was held 5 days after lesions of the basolateral amygdala [(A) 0.32 mg/kg, (A) 1.Omg/kg]. Each point represents the mean f SE of errors and latencies for 6 animals, recorded in each trial of a session.
Amygdaloid complex and working memory
The amygdaloid complex has been found to receive afferent cholinergic projections (Mesulam ef al., 1983), which are not distributed homogeneously throughout all amygdaloid subnuclei. The basolateral subdivision contains the greatest activity of both AChE and ChAT in the amygdaloid complex of the brain of the rat, while the corticomedial amygdaloid group is low in both enzymatic markers (Ben-Ari, Zigmond, Shute and Lewis, 1977). This finding is consistent with the present results, that working memory performance of rats was significantly impaired by lesions of the basolateral amygdala but not by those of the corticomedial amygdala. Immunohistochemical studies of ChAT and AChE, in combination with retrograde fluorescent tracing, have demonstrated that the major choline@ input to the basolateral amygdala derives from the basal forebrain (Carlsen, Zaborsxky and Heimer, 1985; Woolf and Butcher, 1982). All of the three cholinergic drugs which were used in this study also produce a marked improvement of learning and memory impairment in rats with lesions of the basal forebrain, as assessed by the radial maze, Morris water maze and passive avoidance tasks (Hodges, Ribeiro, Gray and Marchbanks, 1990; Miyamoto, Narumi, Ngaoka and Coyle, 1989; Ueki and Miyoshi, 1989). These findings indicate that loss of the choline@ innervation, from the basal forebrain into the basolateral amygdala, leads to a significant impairment in the working memory performance of rats. On the other hand, the basolatera1 amygdala is also interconnected with the mediodorsal thalamic nucleus (Krettek and Price, 1977; Ottersen and Ben-At-i, 1979), lesions of which result in impairment of the working memory performance of rats in the radial maze task (Stokes and Best, 1988, 1990). Therefore, the possible involvement of basolateral amygdaloid dysconnections from structures, other than the basal forebrain in the working memory deficiency, cannot be excluded. Rats normally perform a working memory task through mediation of the basolateral amygdala, as lesions of the basolateral amygdala caused performance to fall almost to chance levels during initial postoperative testing. However, the functional basolateral amygdala is not essential for working memory performance, since the impairment of working memory in basolateral amygdaloid-lesioned rats recovered to control levels after postoperative re-training sessions, given once each day. This finding is in contrast to results obtained with hippocampal-lesioned rats. Lesions of the dorsal hippocampus produced a significant impairment in working memory, assessed by the three-panel runway task, which was also alleviated by cholinergic drugs, such as physostigmine, as was that of basolateral amygdaloid-lesioned rats (Kitajima, Yamamoto, Ohno and Ueki, 1992; Yamamoto, Ohno, Kitajima and Ueki, 1990a). This impairment of working memory never recovered, even if 10 re-training sessions were given postoperatively (Kitajima et al., 1992). Similarly, rats with
921
destruction of the extrinsic connections of the hippocampus (the septum or fimbria-fomix anterior to the hippocampus) exhibit a severe deficit in working memory performance in the radial maze and show no signs of recovery of function when they are given 50 additional re-training sessions (Becker, Walker and Olton, 1980; Olton and Feustle, 1981; Olton, Walker and Gage, 1978). Thus, the functional hippocampus is required for the normal functioning of working memory. In fact, the intact septo-hippocampal cholinergic system will permit the rats to perform working memory tasks, without the mediation of the basolateral amygdala, as postoperative retraining sessions are repeated. However, it has been found that combined lesions of the amygdala and hippocampus in monkeys produce more severe and prolonged impairment on delayed nonmatching-tosample tasks, than either lesions of the amygdala or hippocampus alone (Bachevalier, Parkinson and Mishkin, 1985; Murray and Mishkin, 1983, 1984). In conclusion, the present study provides evidence that lesions of the basolateral amygdala lead to the selective impairment of working memory, without affecting reference memory, possibly through the lowering of the cholinergic function. Acknowledgemenrs-This research was supported in part by the Mochida Memorial Foundation for Medical and Pharmaceutical Research and by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture, Japan. REFERENCES Bachevalier J., Parkinson J. K. and Mishkin M. (1985) Visual recognition in monkeys: effects of separate vs combined transection of fornix and amygdalofugal pathways. Expl Brain Res. 57: 554-561. Baratti C. M., Huygens P., Mitio J., Merlo A. and Gardella J. (1979) Memory facilitation with posttrial injection of oxotremorine and physostigmine in mice. Psychopharmacology 64: 85-88.
Becker J. T., Walker J. A. and Olton D. S. (1980) Neuroanatomical bases of spatial memory. Brain Rex 200: 307-320. Ben-Ari Y., Zigmond R. E., Shute C. C. D. and Lewis P. R. (1977) Regional distribution of choline acetyltransferase and acetylcholinesterase within the amygdaloid complex and stria terminalis system. Brain Res. 1% 435-445. Brady D. R. and Mufson E. J. (1990) Amygdaloid pathology in Alzheimer’s disease: qualitative and quantitative analysis. Dementia 1: 5-17. Brashear H. R., Godec M. S. and Carlsen J. (1988) The distribution of neuritic plaques and acetylcholinesterase staining in the amygdala in Alzheimer’s disease. Neurology 38: 169&1699. Carlsen J., aborszky L. and Heimer L. (1985) Cholinergic projections from the basal forebrain to the basolateral amygdaloid complex: a combined retrograde fluorescent and immunohistochemical study. J. camp. Neural. 234: 155-167. Davies P. and Maloney A. J. F. (1976) Selective loss of central choline&c neurons in Alzheimer’s disease. Lancer 2: 1403. Davis K. L., Mohs R. C., Tinklenberg J. R., Pfefferbaum A., Holhster L. E. and Kopell B. S. (1978) Physostigmine:
922
M. OHNO et al.
improvement of long-term memory processes in normal humans, Science 20& 272-274. __ Eleftheriou B. E.. Elias M. F. and Norman R. L. (1972) Effects of amygdaloid lesions on reversal learning in the deermouse. Physiol. Behav. 9: 69-73. Frumier M. J., Herckar V. R. and Jarvik M. E. (1976) Amnesic actions of diazepam and scopolamine in man. Anesthesiology 45: 406-410.
Furuya Y., Yamamoto T., Yatsugi S. and Ueki S. (1988) A new method for studying working memory by using the three-aanel runway avvaratus in rats. Jan. J. Phurmac. 46: 183-188. - -Hagan J. J., Jansen J. H. M. and Broekkamp C. L. E. (1987) Blockade of suatial learning bv the M, muscarinic antagonist piren2e~ine. Psy~hop~ur~u~olog~ 93: 470-476. Henke H. and Lang W. (1983) Cholinergic enzymes in neocortex, hippccampus and basal forebrain of nonneurological and senile dementia of Alzheimer-type patients. Brain Res. 267: 281-291. Herzog A. G. and Kemper T. L. (1980) Amygdaloid changes in aging and dementia. Archs Neurol. 37: 625-629. Hodges )I., Ribeiro A. M., Gray J. A. and Marchbanks R. M. (1990) Low dose tetrahydroaminoacridine (THA) improves cognitive function but does not affect brain acetvlcholine in rats. Pharmac. B&hem. Behuv. 36: 2911298. Honig W. K. (1978) Studies of working memory in the pigeon. In: Cognitive Processes in Animal Behavior (Hulse S.-H., Fowler‘H. and Honig W. K., Eds), pp. 211-248. Erlbaum, Hillsdale, New Jersey. Kitajima I., Yamamoto T., Ohno M. and Ueki S. (1992) Working and reference memory in rats in the three-panel runway task following dorsal hippocampal lesions. Jup. J. Phurmuc. 58: 175-183.
Krettek J. E. and Price J. L. (1977) Projections from the amygdaloid complex to the cerebral cortex and thalamus in the rat and cat. J. romp. Neural. 172: 687-722. Kw~On-Yuen P. F., Mandel R., Chen A. D. and Thai L. J. (1990) Tetrahydroaminoacridine improves the spatial acquisition deficit produced by nucleus basalis lesions in rats. Expl Nevroi. 10s: 221-228. Lorenzini C. A., Bucherelli C., Giachetti A.. Mugnai L. and Tassoni G. (1991) Effects of nucleus basolatecdlis amygdalae neurotoxic lesions on aversive conditioning in the rat. Physiol. Behuv. 49: 765-770. Mandel R. J. and Thai L. J. (1988) Physosti~ine improves water maze performance following nucleus basalis magnocellularis lesions in rats. Psychophurmucology _. 96: _ 421-425. Mesulam M.-M., Mufson E. J., Wainer B. H. and Levey A. I. (1983) Central cholinergic pathways in the rat: an overview based on an alternative nomenclature (ChlCh6). Neuroscience 10: 1185-1201. Miyamoto M., Narumi S., Nagaoka A. and Coyle J. T. (1989) Effects of continuous infusion of cholinergic drugs on memory impairment in rats with basal forebrain lesions. J. Pharmae. exp. Ther. 24& 825-835. Murray E. A. and Mishkin M. (1983) Severe tactual memory deficits in monkeys after combined removal of the amygdala and hippocampus. Brain Res. 270: 34&344. Murray E. A. and Mishkin M. (1984) Severe tactual as well as visual memory deficits follow combined removal of the amygdala and hippocampus in monkeys. J. Neurosci. 4: 2565-2.580. Ohno M., Yamamoto T. and Ueki S. (1991) Effect of the h--receptor agonist, U-50,488H, on cerebral ischcmiainduced impairment of working memory assessed in rats by a three-panel runway task. Eur. J. Phurmuc. 193: 357-361.
Olton D. S. and Feustle W. A. (1981) Hippocampal function required for nonspatial working memory. Exi/ Bruin Res. 41: 380-389. Olton D. S., Becker J. T. and Handelmann G. E. (1979) Hippocampus, space, and memory. Behav. Bruin Sci. 2: 313-365. Olton D. S., Walker J. A. and Gage F. H. (1978) Hippocampal connections and spatial discrimination. Bruin Res. 139: 295-308. Ottersen 0. P. and Ben-Ari Y. (1979) Afferent connections to the amygdaloid complex of the rat and cat. J. camp. Neural. 187: 401-424. Paxinos G. and Watson C. (1982) The Rut Brain in Stereotuxic Coordinutes. Academic Press, New York. 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. Rossor M. N., lversen L, L., Reynolds G. P., Mountjoy C. Q. and Roth M. (1984) Neurochemical characteristics of early and late onset types of Alzheimer’s disease. Br. Med. J. 288: 961-964. Sarter M. and Markowitsch H. J. (1985) Involvement of the amygdala in learning and memory: a critical review, with emphasis on anatomical relations. Behuv. Neurosci. 99: 342-380. Sitaram N., Weingartner H. and Gillin J. C. (1978) Human serial learning: enhancement with arecholine and choline and impairment with scopolamine. Science 201: 274-276. Stokes K. A. and Best P. J. (1988) Mediodorsal thalamic lesions impair radial maze performance in the rat. Behuv. Neurosci. 102: 294-300.
Stokes K. A. and Best P. J. (1990) Mediodorsal thalamic lesions impair “reference” and “working” memory in rats. Physiol. Be&v. 47: 471-476. Summers W. K., Viesselman J. O., Marsh G. M. and Candelora K. (1981) Use of THA in treatment of Alzheimer-like dementia: pilot study in twelve patients. Biol. Psychiut. 16: 145-153. Summers W. K., Majovski L. V., Marsh G. M., Tachiki K. and Kling A. (1986) Oral tetrahydroaminoacridine in long-term treatment of senile dementia. Alzheimer tvpe. -. New Engl. J. Med. 315: 1241-1245. Ueki A. and Miyoshi K. (1989) Effects of cholinergic drugs on learning impairment in ventral globus palliduslesioned rats. J. Neural. Set. 90: l-21. Unger J. W., McNeil1 T. H., Lapham L. L. and Hamill R. W. (1988) Neuropeptides and neuropathology in the amygdala in’ Alzheimer’s disease: rclationship%etween somatostatin, neuropeptide Y and subregional distribution of neuritic plaques. Bruin Res. 452: 293-302. Watts J., Stevens R. and Robinson C. (1981) Effects of scopolamine on radial maze performance in rats. Physiol. Behuv. f6: 845-851.
Woolf N. J. and Butcher L. L. (1982) Cholinergic projections to the basolateral amygdala: a combined Evans Blue and a~tylchoiines~rase analysis. Bruin Res. Ml. 8: 751-763. Yamamoto T., Ohno M., Kitajima 1. and Ueki S. (1990a) Effects of nootropic drugs -on experimentally induced amnesia of rats in the three-panel runway task. In: B&c, Clinical, and Therapeutic Aspects of Alrh~imer’s and Parkinson’s Diseases (Nagatsu T., Fisher A. and Yoshida M.,
Eds), Vol. 2, pp. 439-443. Plenum Press, New York. Yamamoto T., Yatsugi S., Ohno M., Furuya Y., Kitajima I. and Ueki S. (1990b) Minaprine improves impairment of working memory induced by scopolamine and cerebral ischemia in rats. Psychophurmucology 10th 316-322.
0028-3908/92 $5.00 + 0.00 Copyright Q 1992 Pergamon Press Ltd
INVOLVEMENT OF CHOLINERGIC MECHANISMS IN IMPAIRMENT OF WORKING MEMORY IN RATS FOLLOWING BASOLATERAL AMYGDALOID LESIONS M.OHNO,*
T. YAMAMOTOand S. WATANABE
Department of Pharmacology, Faculty of Pharmaceutical Sciences, Kyushu University, Fukuoka 812, Japan (Accepted
3 March
1992)
Summary-In order to clarify the role of the amygdala in the working and reference memory of rats in the three-panel runway task, the effects of lesions of subnuclei of the amygdaloid complex on this behavior were studied. Rats that had been trained preoperatively, until they achieved the criterion of learning, were subjected to lesions of the amygdala. In the working memory task, lesions of the basolateral subdivision of the amygdala caused a significant increase in the number of errors (attempts to pass through two incorrect panels of the three panel-gates at four choice points), while lesions of the corticomedial amygdala had no effect on working memory errors. The increase in working errors, observed in basolateral amygdaloid-lesioned rats, declined gradually as retraining sessions were given once each day, reverting to control levels on and after the sixth session. In the reference memory task, the number of errors was not affected by lesions of the basolateral or corticomedial amygdala. The increase in working memory errors, induced by lesions of the basolateral amygdala was significantly reduced by intraperitoneal administration of the inhibitors of cholinesterase, tetrahydroaminoacridine (0.32-1.0 mg/kg) and physostigmine (0.032-O. 1 mg/kg), and the muscarinic receptor agonist, oxotremorine (0.1 mg/kg), before the runway test. These findings suggest that the basolateral amygdala is selectively involved in working memory but not in reference memory and that the lowering of central cholinergic function may account for the impairment of working memory, induced by lesions of the basolateral amygdala. Key words-amygdala,
brain lesions, working memory, reference memory, tetrahydroaminoacridine (THA), physostigmine, oxotremorine.
The amygdala is a subcortical structure, which occupies a strategic position in the circuitry of the limbic system and has been assumed to play an important role in the control, not only of emotional and motivational behavior, but also of memory functions. Patients with Alzheimer’s disease often show disturbances in limbic functions, such as emotional
deficits in addition to a profound decline in memory and intellectual acuity. Recently, it has been demonstrated that the amygdala undergoes severe neuropathological alterations in Alzheimer’s disease. A large number of senile plaques and neurofibrillary tangles, known as specific pathological hallmarks of Alzheimer’s disease, occur to varying degrees in all subnuclei of the amygdaloid complex in patients with Alzheimer’s disease (Brady and Mufson, 1990; Brashear, Godec and Carlsen, 1988; Unger, McNeill, Lapham and Hamill, 1988). A decrease in the amygdaloid area, as assessed by cresyl violet staining, is found in the brains of patients with Alzheimer’s disease (Herzog and Kemper, 1980). Consistently, it has also been reported that the activity of cholinergic markers, such as choline acetyltransferase (ChAT) and acetylcholinesterase (AChE), is markedly re-
*To whom correspondence
should be addressed. 915
duced in the amygdala from patients with Alzheimer’s disease (Brady and Mufson, 1990; Brashear et al., 1988; Davies and Maloney, 1976; Rossor, Iversen, Reynolds, Mountjoy and Roth, 1984). Thus, these neuropathological and biochemical changes in the amygdala may account for the marked memory loss and emotional disorders, seen in patients with Alzheimer’s disease. Although the role of the amygdala in learning and memory has been extensively studied in rats, monkeys and cats, few studies have examined memory impairments following lesions of the amygdala, confined to specific subnuclei (for review, see Sarter and Markowitsch, 1985). Lesions which simultaneously destroy a number of subnuclei, within the amygdaloid complex, preclude understanding of the relationship of specific nuclei to learning and memory. Since the amygdaloid complex has been divided into a basolateral and a corticomedial part, this study was undertaken to elucidate the differential roles of these two portions of the amygdala in the learning and memory of rats. It has been proposed that there are two different types of memory function in experimental animals, i.e. working memory and reference memory (Honig, 1978; Olton, Becker and Handelmann, 1979). Working memory allows animals to remember information that is useful for a single
M. Omo
916
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. It was previously reported that a three-panel runway task served well as a method for the study of learning and memory in the rat (Furuya, Yamamoto, Yatsugi and Ueki, 1988; Yamamoto, Yatsugi, Ohno, Furuya, Kitajima and Ueki, 1990b). With this experimental paradigm, memory can be divided into working and reference memory (Ohno, Yamamoto and Ueki, 1991). In the present study, the differential effects of lesions of the basolateral and corticomedial amygdala were investigated on working and reference memory, which was assessed in rats by the three-panel runway task. Furthermore, the effects of choline&c activating drugs, such as tetrahydroaminoa~dine (THA), physostigmine and oxotremorine on memory impairment in amygdaloid-lesioned rats were examined, as the amygdala receives a cholinergic innervation from the basal forebrain (Mesulam, Mufson, Wainer and Levey, 1983). METHODS
Subjects Eight- to lo-week-old male rats of the Wistar strain were obtained from Nippon SLC. The subjects 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 free-feeding level. The rats were housed in groups of 4 per cage under constant temperature (23 + 2°C) and a 12-hr light-dark cycle (light period: 07:00-19:00), with water freely available. Apparatus Working memory and reference memory were assessed with a three-panel runway apparatus (Fig. 1), that was introduced previously (Furuya et al., 1988;
et
al.
Ohno et al., 1991; Yamamoto et al., 1990b). In brief, this apparatus (175 x 36 x 25 cm) was composed of a start box, a goal box and four consecutive choice points inte~ening 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 2 food pellets (about 50mg each; Muromachi Kikai) as positive reinforcement. Acquisition training Initially, all the front stoppers were removed so that a rat could pass through any one of the three panel-gates 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 20sec. Once this time was reached, the rats were given 6 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 (Furuya et al., 1988; Yamamoto ef al., 199Ob). 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
A:
Start box
B: Goal box
Fig. 1. Schematic drawing of the three-panel runway apparatus. In the runway task, the rats have to pass four consecutive choice points to obtain food pellets placed in the goal box. Each choice point consisted of a gate with three panels. The rats could pass through only one of the three panel-gates (the correct panel-gate) at each choice point.
Amygdaloid complex and working memory
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 6 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 3 consecutive sessions. Surgery
The rats were subjected to lesions of the amygdala after they achieved the criterion of learning. Each rat was anesthetized with sodium pentobarbital (40 mg/kg, i.p.) and was fixed onto a stereotaxic instrument. Lesions of the amygdala were produced with a lesion generator (Muromachi Kikai; RFG4A), in which the temperature of the electrode tip could be monitored. The electrode (0.25 mm dia, 1.O-mm exposed tip) was inserted into the basolateral amygdala [anterior (A): -2.8 mm from bregma; lateral (L): k5.0 mm; horizontal (H): 8.8 mm] and corticomedial amygdala (A: - 2.8 mm, L: f 3.2 mm, H: 9.2 mm), according to the coordinates in the atlas of the rat brain of Paxinos and Watson (1982). The bilateral amygdaloid lesions were induced by an increase in the temperature of the tissue to 70°C within 15 set, this temperature being maintained for another 45 sec. Sham-operated rats were also fixed on the stereotaxic apparatus, under pentobarbital anesthesia and underwent small burr holes in the skull but they did not receive insertion of the electrode. The rats were allowed 5 days to recover from surgery before being used for testing. Thereafter, the runway test was given once each day for 7 consecutive
917
days. In different groups of amygdaloid-lesioned rats, the effects of the choline@ drugs on the impairment of memory were investigated 5 days after lesioning. Drugs
The drugs used in this study were tetrahydroaminoacridine hydrochloride (THA; Aldrich Chemical Co.), physostigmine salicylate (Sigma Chemical Co.) and oxotremorine sesquifumarate (Sigma Chemical Co.). All drugs were dissolved in distilled water and administered in a volume of 0.1 ml per 1OOg body weight. The doses expressed in terms of the
for the drugs
are
salt. Tetrahydroaminoacridine was administered intraperitoneally 20 min before the runway test, which was held 5 days after the amygdaloid lesions. Other drugs were given intraperitoneally 30 min before testing. Histology
After completion of the experiment, the rats given basolateral and corticomedial amygdaloid lesions were anesthetized with ether and the brain was perfused with 10% formalin solution, through the left cardiac ventricle. The sacrifice procedure conformed to the standards of animal procedures, established at Kyushu University. After the brain was removed, it was frozen and sliced to a thickness of 70pm. All sliced sections were stained with cresyl violet. The site and extent of each lesion of the amygdala were verified histologically. The extent of the minimum and maximum lesions of the basolateral and corticomedial amygdala, of all animals tested, is illustrated in Fig. 2.
-2.3 mm
-2.8 mm
-3.3 mm
Fig. 2. Schematic drawing of the minimum (black area) and the maximum (black + stippled area) extent of lesions of the basolateral (A) and corticomedial (B) amygdala. The number on the right of each drawing shows the posterior distance of the plate from bregma. Serial plates were adapted from the atlas of the brain of the rat of Paxinos and Watson (1982).
918
M. OHNOet ai.
Anal.vsis of data 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 evaluation of 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 6 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 one-way analysis of variance (ANOVA), that was followed by Dunnett’s test when F ratios reached significance (P < 0.05). RESULTS
In the three-panel runway task, the random level of performance was 4 errors per trial or 24 errors per session. In the working memory task, the number of errors made from the second to the sixth trial (working memory errors) markedly decreased with repeated training, whereas the errors in the first trial remained constant at approximately 4 errors. About 15-20 training sessions were required for the rats to reach the criterion of less than 12 errors, summed across the 6 trials of a session. The latency was also reduced during repeated sessions and was stable from the tenth session on. In the reference memory task, the number of errors and latency in all 6 trials of a session decreased with the repetition of training. The rats could run the task within the 6-error criterion, summed across 6 trials, after they had had 5-10 training sessions. Lesions of the basolateral amygdala produced a significant increase in the number of errors, assessed by the working memory procedure, done in the three-panel runway task [F(l,lO) = 43.66, P < O.Ol] but did not affect the number of errors made in the first trial (Fig. 3 and Table 1). Lesions of the basolatera! amygdala prolonged the latency in the first trial [F(l,lO) = 6.95, P < 0.051, as well as the latency summed from the second to the sixth trial of a session [F(l,lO) = 13.55, P e 0.011. In contrast, rats with lesions of the corticomedial amygdala exhibited no increase in errors of working memory (mean + SE: 6.2 f 1.7) or errors in the first trial (5.3 + OS), as compared with those of sham-lesioned rats. Corticomedial amygdaloid lesions significantly prolonged the latency, recorded in the first trial [31.5 + 4.3 set, F(l,lO) = 6.25, P < 0.051, without affecting that recorded in trials second to sixth (68.8 f 12.9 WC). In the case of the reference memory task, neither lesions of the basolateral nor corticomedial amygdala produced increases in the number of errors or in the latency, across all 6 trials of a session (Fig. 3). The increase in working memory errors, induced by lesions of the basolateral amygdala, declined
gradually as subsequent sessions were given once each day, reverting to control levels on and after the sixth re-training session (Fig. 4). As shown in Fig. 5 and Table 1, administration of 0.32 and 1.0 mg/kg of THA, before the test session, significantly reduced the increase in working memory errors that would be expected to occur after lesions of the basolateral amygdala [F(2,15) = 13.06, P < O.Ol]. Treatment with THA also decreased the expected increase in latency from the second to the sixth trial [F(2,15) = 4.84, P < 0.051, an effect that reached significance only for the 1.0 mg/kg dose (P i 0.05). Basolateral amygdaloid-lesioned rats failed to run the task when they were given 3.2 mg/kg of THA (n = 2, data not shown). Physostigmine, at doses of 0.032 and 0.1 mg/kg, significantly reduced the increase in working errors induced by lesions of the basolateral amygdala [F(2,15) = 9.87, P < 0.011. A significant reduction in the increase of working errors was also seen when basolateral amygdaloidlesioned rats were treated with 0.1 mg/kg of oxotremorine before the runway test [F(l ,lO) = 26.96, P < 0.011. However, neither physosti~ine nor oxotremorine had an effect on the prolonged latency in rats with lesions of the basolateral amygdala. DISCUSSION
In this study, working memory, assessed in rats by a three-panel runway task, was signi~~ntly impaired by lesions of the basolateral amygdala but was not affected by lesions of the corticomedial amygdala. In contrast, lesions of neither the basolateral nor corticomedial amygdala had a significant effect on reference memory performance. These results suggest that the basolateral amygdala is selectively involved in working memory performance, in which animals are required to remember new information that is useful for only one session. Conceivably, the amyg daloid complex does not play a role in reference memory performance, in which animals have to retain constant info~ation, that is of continued value throughout all sessions. Recently, it has been reported that lesions of the basolateral amygdaloid nucleus induced with ibotenic acid significantly impair acquisition of passive avoidance, while basolatera! amygdaloid-lesioned rats exhibit retention of passive avoidance fairly well, as do sham-operated rats (Lorenzini, Bucherelli, Giachetti, Mugnai and Tassoni, 1991). Reversal learning of the deermouse, in a T-shaped water maze, which is related to the ability to respond to changing environmental demands, also deteriorated as a result of lesioning of the basolateral amygdala but not that of the medial amygdala (Elefthe~ou, Ehas and Norman, 1972). Clinically, the most obvious amygdaloid atrophy, observed in patients with Alzheimer’s disease, concerns the lateral and basolateral nuclei (Brady and Mufson, 1990). Thus, dysfunction of the basolatera1 portion of the amygdala may, at least in part,
Amygdaloid complex and working memory
919
Working Memory x
i ’ 123456
’
’
’
I
Ttw8 Rctorcnce Memory 5
4
123455
123456 TtWB
TM8
Fig. 3. Effects of lesions of the amygdala on working and reference memory errors and latency, in the three-panel runway task. Rats were trained preoperatively until they achieved the criterion and were then subjected to lesions of the amygdala [(O) sham, (0) basolateral amygdala, (A) corticomedial amygdala]. The runway test was given 5 days after surgery. Each point represents the mean f SE of errors and latencies for 6 animals, recorded in each trial of a session. contribute to severe deficiencies of the memory seen in patients with Alzheimer’s disease. There is much evidence that central cholinergic transmission plays a pivotal role in learning and Table
I.
memory. Drugs that block choline&z activity exert amnesic effects (Frumier, Herckat and Jarvik, 1976; Hagan, Jansen and Broekkamp, 1987; Sitaram, Weingartner and Gillin, 1978; Watts, Stevens and
tetrahydroaminoacridine (THA), physostigmine and oxotnmorineon increasesin workingmemoryerrors and latencv.inducedbv lesions of the basolateral am&alit. in the three-Dane1runwav task
Effects of
Numberof LeSiOIlS
Drug
Sham Basolateral amygdala Basolateral amygdala
THA
Basolateral amygdala
Physostigmine
BasoIateraI amygdala
OX0tTCIil0rilIC
mg/kg (i.p.) 0.32 1.0 0.032 0.1 0.1
N
Trial
6 6 6 6 6 6 6
4.3 f 4.3 f 4.5 f 3.8 4 4.2 & 4.3 f 4.0 f
I 0.6 0.7 0.6 0.5 0.6 0.6 0.5
Latency (see)
errors Trial 2-6 4.3 & 0.8 18.3 f 1.9$ 10.0 f 1.3” 7.7 f I .4** 9.7 f 1.9*+ 8.0 f I.4’* 7.3 f 0.8.”
Trial 18.7 f 37.0 k 29.0 * 39.5 f 36.5 f 33.7 + 29.5 *
1 2.8 6.4t 9.9 9.9 16.3 7.8 7.4
Trial 2-6 39.7 f 105.8 & SO.2f 52.3 f 63.8 i 62.3 f go.2 f
4.1 17.5t 9.6 6.7’ 16.4 12.6 13.4
Rats that had been trained preoperatively until they achieved the criterion, were subjected to lesionsof the basolateral amygdala. Tetrahydroaminoacridine was administered intraperitoneally 2Omin before the runway test, which was held 5 days after the lesions. Physostigminc and oxotremorine were given intraperitoneally 30min before the test session. Each value represents the mean f SE. The significance of the differences from sham-lesioned rats (tP < 0.05, $P < 0.01) and from basolateral amygdaloid-Iesioned rats (*P < 0.05, l*f < 0.01) was determined by a one-way ANOVA, followed by Dunneu’s test.
M.
920
OHNO
et al.
Worklng Memory
berg, Pfefferbaum, Hollister and Kopell, 1978; Sitaram et al., 1978). The activity of ChAT has been shown to decrease markedly in the amygdala as well as in the hippocampus and cortex, in patients with Alzheimer’s disease (Davies and Maloney, 1976; Henke and Lang, 1983; Rossor et al., 1984). The importance of the cholinergic deficit for memory deficits is emphasized by a clinical study, suggesting that the degree of loss of choline@ function correlates with the severity of dementia in patients with Alzheimer’s disease (Perry, Tomlinson, Blessed, Bergmann, Gibson and Perry, 1978). In the present study, the effects of some cholinergic activating 1 drugs, including THA, which has been reported to be RI1 2 3 4 5 6 7 useful in the treatment of patients with Alzheimer’s SOuiOltS disease (Summers, Majovski, Marsh, Tachiki and Fig. 4. Changes of errors of working memory, assessed in Kling, 1986; Summers, Viesselman, Marsh and amygdaloid-lesioned rats by the three-panel runway task, Candelora, 1981) were investigated. The inhibitors when retraining sessions were repeated. Rats that had been trained preoperatively until they achieved the criterion (Pre) of AChE, THA and physostigmine, ameliorated were subjected to lesions of the amygdala [(C) sham, (0) significantly the impairment of working memory in basolateral amygdala, (A) corticomedial amygdala]. The rats with lesions of the basolateral amygdala, when rats had been retrained once each day for 7 consecutive they were administered before the test session. The days, starting from the fifth day (the first session) after effective doses of THA and physostigmine, in this surgery. Each point represents the mean f SE of errors for 6 animals, summed from the second to sixth trial of a study, have been shown to produce significant session. The significance of the differences from shaminhibition of the activity of AChE in the brain of the lesioned rats (**Pc 0.01) was determined by a one-way rat at the time of testing (Kwo-On-Yuen, Mandel, ANOVA, followed by Dunnett’s test. Chen and Thal, 1990; Mandel and Thal, 1988). Furthermore, the muscarinic receptor agonist, oxotremorine, also alleviated the working memory Robinson, 1981), whereas cholinergic activating deficit after lesions of the basolateral amygdala. drugs, such as muscarinic agonists and inhibitors These findings suggest that impairment of working of AChE, facilitate learning and memory both in memory, induced by lesions of the basolateral amyghumans and in animals (Baratti, Huygens, Miiio, dala, results from dysfunction of the central cholinMerlo and Gardella, 1979; Davis, Mohs, Tinklenergic system. ”
1
Worklng Yomory 50
40
3 30 E - 20 : 10
1
’
’
’
12345s
’
4
0
,
1
’
1
’
,
123456 Trlalr
Trlalr
Fig. 5. Effect of tetrahydroaminoacridine (THA) on increases in errors of working memory and latency, induced by lesions of the basolateral amygdala in the three-panel runway task. Rats that had been trained preoperatively until they achieved the criterion were subjected to lesions of the amygdala [(C) sham, (0) basolateral amygdala). Tetrahydroaminoacridine was administered intraperitoneally 20 min before the runway test, which was held 5 days after lesions of the basolateral amygdala [(A) 0.32 mg/kg, (A) 1.Omg/kg]. Each point represents the mean f SE of errors and latencies for 6 animals, recorded in each trial of a session.
Amygdaloid complex and working memory
The amygdaloid complex has been found to receive afferent cholinergic projections (Mesulam ef al., 1983), which are not distributed homogeneously throughout all amygdaloid subnuclei. The basolateral subdivision contains the greatest activity of both AChE and ChAT in the amygdaloid complex of the brain of the rat, while the corticomedial amygdaloid group is low in both enzymatic markers (Ben-Ari, Zigmond, Shute and Lewis, 1977). This finding is consistent with the present results, that working memory performance of rats was significantly impaired by lesions of the basolateral amygdala but not by those of the corticomedial amygdala. Immunohistochemical studies of ChAT and AChE, in combination with retrograde fluorescent tracing, have demonstrated that the major choline@ input to the basolateral amygdala derives from the basal forebrain (Carlsen, Zaborsxky and Heimer, 1985; Woolf and Butcher, 1982). All of the three cholinergic drugs which were used in this study also produce a marked improvement of learning and memory impairment in rats with lesions of the basal forebrain, as assessed by the radial maze, Morris water maze and passive avoidance tasks (Hodges, Ribeiro, Gray and Marchbanks, 1990; Miyamoto, Narumi, Ngaoka and Coyle, 1989; Ueki and Miyoshi, 1989). These findings indicate that loss of the choline@ innervation, from the basal forebrain into the basolateral amygdala, leads to a significant impairment in the working memory performance of rats. On the other hand, the basolatera1 amygdala is also interconnected with the mediodorsal thalamic nucleus (Krettek and Price, 1977; Ottersen and Ben-At-i, 1979), lesions of which result in impairment of the working memory performance of rats in the radial maze task (Stokes and Best, 1988, 1990). Therefore, the possible involvement of basolateral amygdaloid dysconnections from structures, other than the basal forebrain in the working memory deficiency, cannot be excluded. Rats normally perform a working memory task through mediation of the basolateral amygdala, as lesions of the basolateral amygdala caused performance to fall almost to chance levels during initial postoperative testing. However, the functional basolateral amygdala is not essential for working memory performance, since the impairment of working memory in basolateral amygdaloid-lesioned rats recovered to control levels after postoperative re-training sessions, given once each day. This finding is in contrast to results obtained with hippocampal-lesioned rats. Lesions of the dorsal hippocampus produced a significant impairment in working memory, assessed by the three-panel runway task, which was also alleviated by cholinergic drugs, such as physostigmine, as was that of basolateral amygdaloid-lesioned rats (Kitajima, Yamamoto, Ohno and Ueki, 1992; Yamamoto, Ohno, Kitajima and Ueki, 1990a). This impairment of working memory never recovered, even if 10 re-training sessions were given postoperatively (Kitajima et al., 1992). Similarly, rats with
921
destruction of the extrinsic connections of the hippocampus (the septum or fimbria-fomix anterior to the hippocampus) exhibit a severe deficit in working memory performance in the radial maze and show no signs of recovery of function when they are given 50 additional re-training sessions (Becker, Walker and Olton, 1980; Olton and Feustle, 1981; Olton, Walker and Gage, 1978). Thus, the functional hippocampus is required for the normal functioning of working memory. In fact, the intact septo-hippocampal cholinergic system will permit the rats to perform working memory tasks, without the mediation of the basolateral amygdala, as postoperative retraining sessions are repeated. However, it has been found that combined lesions of the amygdala and hippocampus in monkeys produce more severe and prolonged impairment on delayed nonmatching-tosample tasks, than either lesions of the amygdala or hippocampus alone (Bachevalier, Parkinson and Mishkin, 1985; Murray and Mishkin, 1983, 1984). In conclusion, the present study provides evidence that lesions of the basolateral amygdala lead to the selective impairment of working memory, without affecting reference memory, possibly through the lowering of the cholinergic function. Acknowledgemenrs-This research was supported in part by the Mochida Memorial Foundation for Medical and Pharmaceutical Research and by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture, Japan. REFERENCES Bachevalier J., Parkinson J. K. and Mishkin M. (1985) Visual recognition in monkeys: effects of separate vs combined transection of fornix and amygdalofugal pathways. Expl Brain Res. 57: 554-561. Baratti C. M., Huygens P., Mitio J., Merlo A. and Gardella J. (1979) Memory facilitation with posttrial injection of oxotremorine and physostigmine in mice. Psychopharmacology 64: 85-88.
Becker J. T., Walker J. A. and Olton D. S. (1980) Neuroanatomical bases of spatial memory. Brain Rex 200: 307-320. Ben-Ari Y., Zigmond R. E., Shute C. C. D. and Lewis P. R. (1977) Regional distribution of choline acetyltransferase and acetylcholinesterase within the amygdaloid complex and stria terminalis system. Brain Res. 1% 435-445. Brady D. R. and Mufson E. J. (1990) Amygdaloid pathology in Alzheimer’s disease: qualitative and quantitative analysis. Dementia 1: 5-17. Brashear H. R., Godec M. S. and Carlsen J. (1988) The distribution of neuritic plaques and acetylcholinesterase staining in the amygdala in Alzheimer’s disease. Neurology 38: 169&1699. Carlsen J., aborszky L. and Heimer L. (1985) Cholinergic projections from the basal forebrain to the basolateral amygdaloid complex: a combined retrograde fluorescent and immunohistochemical study. J. camp. Neural. 234: 155-167. Davies P. and Maloney A. J. F. (1976) Selective loss of central choline&c neurons in Alzheimer’s disease. Lancer 2: 1403. Davis K. L., Mohs R. C., Tinklenberg J. R., Pfefferbaum A., Holhster L. E. and Kopell B. S. (1978) Physostigmine:
922
M. OHNO et al.
improvement of long-term memory processes in normal humans, Science 20& 272-274. __ Eleftheriou B. E.. Elias M. F. and Norman R. L. (1972) Effects of amygdaloid lesions on reversal learning in the deermouse. Physiol. Behav. 9: 69-73. Frumier M. J., Herckar V. R. and Jarvik M. E. (1976) Amnesic actions of diazepam and scopolamine in man. Anesthesiology 45: 406-410.
Furuya Y., Yamamoto T., Yatsugi S. and Ueki S. (1988) A new method for studying working memory by using the three-aanel runway avvaratus in rats. Jan. J. Phurmac. 46: 183-188. - -Hagan J. J., Jansen J. H. M. and Broekkamp C. L. E. (1987) Blockade of suatial learning bv the M, muscarinic antagonist piren2e~ine. Psy~hop~ur~u~olog~ 93: 470-476. Henke H. and Lang W. (1983) Cholinergic enzymes in neocortex, hippccampus and basal forebrain of nonneurological and senile dementia of Alzheimer-type patients. Brain Res. 267: 281-291. Herzog A. G. and Kemper T. L. (1980) Amygdaloid changes in aging and dementia. Archs Neurol. 37: 625-629. Hodges )I., Ribeiro A. M., Gray J. A. and Marchbanks R. M. (1990) Low dose tetrahydroaminoacridine (THA) improves cognitive function but does not affect brain acetvlcholine in rats. Pharmac. B&hem. Behuv. 36: 2911298. Honig W. K. (1978) Studies of working memory in the pigeon. In: Cognitive Processes in Animal Behavior (Hulse S.-H., Fowler‘H. and Honig W. K., Eds), pp. 211-248. Erlbaum, Hillsdale, New Jersey. Kitajima I., Yamamoto T., Ohno M. and Ueki S. (1992) Working and reference memory in rats in the three-panel runway task following dorsal hippocampal lesions. Jup. J. Phurmuc. 58: 175-183.
Krettek J. E. and Price J. L. (1977) Projections from the amygdaloid complex to the cerebral cortex and thalamus in the rat and cat. J. romp. Neural. 172: 687-722. Kw~On-Yuen P. F., Mandel R., Chen A. D. and Thai L. J. (1990) Tetrahydroaminoacridine improves the spatial acquisition deficit produced by nucleus basalis lesions in rats. Expl Nevroi. 10s: 221-228. Lorenzini C. A., Bucherelli C., Giachetti A.. Mugnai L. and Tassoni G. (1991) Effects of nucleus basolatecdlis amygdalae neurotoxic lesions on aversive conditioning in the rat. Physiol. Behuv. 49: 765-770. Mandel R. J. and Thai L. J. (1988) Physosti~ine improves water maze performance following nucleus basalis magnocellularis lesions in rats. Psychophurmucology _. 96: _ 421-425. Mesulam M.-M., Mufson E. J., Wainer B. H. and Levey A. I. (1983) Central cholinergic pathways in the rat: an overview based on an alternative nomenclature (ChlCh6). Neuroscience 10: 1185-1201. Miyamoto M., Narumi S., Nagaoka A. and Coyle J. T. (1989) Effects of continuous infusion of cholinergic drugs on memory impairment in rats with basal forebrain lesions. J. Pharmae. exp. Ther. 24& 825-835. Murray E. A. and Mishkin M. (1983) Severe tactual memory deficits in monkeys after combined removal of the amygdala and hippocampus. Brain Res. 270: 34&344. Murray E. A. and Mishkin M. (1984) Severe tactual as well as visual memory deficits follow combined removal of the amygdala and hippocampus in monkeys. J. Neurosci. 4: 2565-2.580. Ohno M., Yamamoto T. and Ueki S. (1991) Effect of the h--receptor agonist, U-50,488H, on cerebral ischcmiainduced impairment of working memory assessed in rats by a three-panel runway task. Eur. J. Phurmuc. 193: 357-361.
Olton D. S. and Feustle W. A. (1981) Hippocampal function required for nonspatial working memory. Exi/ Bruin Res. 41: 380-389. Olton D. S., Becker J. T. and Handelmann G. E. (1979) Hippocampus, space, and memory. Behav. Bruin Sci. 2: 313-365. Olton D. S., Walker J. A. and Gage F. H. (1978) Hippocampal connections and spatial discrimination. Bruin Res. 139: 295-308. Ottersen 0. P. and Ben-Ari Y. (1979) Afferent connections to the amygdaloid complex of the rat and cat. J. camp. Neural. 187: 401-424. Paxinos G. and Watson C. (1982) The Rut Brain in Stereotuxic Coordinutes. Academic Press, New York. 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. Rossor M. N., lversen L, L., Reynolds G. P., Mountjoy C. Q. and Roth M. (1984) Neurochemical characteristics of early and late onset types of Alzheimer’s disease. Br. Med. J. 288: 961-964. Sarter M. and Markowitsch H. J. (1985) Involvement of the amygdala in learning and memory: a critical review, with emphasis on anatomical relations. Behuv. Neurosci. 99: 342-380. Sitaram N., Weingartner H. and Gillin J. C. (1978) Human serial learning: enhancement with arecholine and choline and impairment with scopolamine. Science 201: 274-276. Stokes K. A. and Best P. J. (1988) Mediodorsal thalamic lesions impair radial maze performance in the rat. Behuv. Neurosci. 102: 294-300.
Stokes K. A. and Best P. J. (1990) Mediodorsal thalamic lesions impair “reference” and “working” memory in rats. Physiol. Be&v. 47: 471-476. Summers W. K., Viesselman J. O., Marsh G. M. and Candelora K. (1981) Use of THA in treatment of Alzheimer-like dementia: pilot study in twelve patients. Biol. Psychiut. 16: 145-153. Summers W. K., Majovski L. V., Marsh G. M., Tachiki K. and Kling A. (1986) Oral tetrahydroaminoacridine in long-term treatment of senile dementia. Alzheimer tvpe. -. New Engl. J. Med. 315: 1241-1245. Ueki A. and Miyoshi K. (1989) Effects of cholinergic drugs on learning impairment in ventral globus palliduslesioned rats. J. Neural. Set. 90: l-21. Unger J. W., McNeil1 T. H., Lapham L. L. and Hamill R. W. (1988) Neuropeptides and neuropathology in the amygdala in’ Alzheimer’s disease: rclationship%etween somatostatin, neuropeptide Y and subregional distribution of neuritic plaques. Bruin Res. 452: 293-302. Watts J., Stevens R. and Robinson C. (1981) Effects of scopolamine on radial maze performance in rats. Physiol. Behuv. f6: 845-851.
Woolf N. J. and Butcher L. L. (1982) Cholinergic projections to the basolateral amygdala: a combined Evans Blue and a~tylchoiines~rase analysis. Bruin Res. Ml. 8: 751-763. Yamamoto T., Ohno M., Kitajima 1. and Ueki S. (1990a) Effects of nootropic drugs -on experimentally induced amnesia of rats in the three-panel runway task. In: B&c, Clinical, and Therapeutic Aspects of Alrh~imer’s and Parkinson’s Diseases (Nagatsu T., Fisher A. and Yoshida M.,
Eds), Vol. 2, pp. 439-443. Plenum Press, New York. Yamamoto T., Yatsugi S., Ohno M., Furuya Y., Kitajima I. and Ueki S. (1990b) Minaprine improves impairment of working memory induced by scopolamine and cerebral ischemia in rats. Psychophurmucology 10th 316-322.
