- Email: [email protected]
Lack of correlation between cortical levels of choline acetyltransferase and learning scores in rats with globus pallidus lesions
402
Brain Research, 367 (1986) 402-404 Elsevier
BRE 21420
Lack of c o r r e l ~ ~ n learning ~
~ 1 levels of ~ m a~t~.raek,'foraea and in ~ s w ~ ~ s ~ ~ ~ons
ROBERT THOMPSON1, ROBERT B. GIBBS2, GINELL A. RISTIC2, CARL W. COTMAN2 and JEN YU l l Fairview State Hospital, Costa Mesa, and University of California lrvine Medical Center, Orange, and 2Universityof California, Irvine, CA (U.S.A.) (Accepted November 5th, 1985) Key words: globus pallidus - - nucleus basalis magnocellularis - - choline acetyltransferase - - learning
Previous work has shown that rats with lesions of the globus pallidus (GP) exhibit a generalized learning impairment. Data are presented suggesting that this impairment is not due to inadvertent damage to the nucleus basatis magaocell~aris. Rats with GP lesions evidenced a significant visual discrimination learning loss and a significant reduction in cortical choline aeetyltransferase (CHAT) activity. However, there was no significant correlation between the severity of the learning loss and the amount of reduction of cortical ChAT activity.
We have recently shown that young rats with neurotoxic lesions of the globus pallidus (GP) exhibit a global learning deficit characterized by a deficiency in learning visual and non-visual discrimination problems and a complex maze task 16. Intermingled among GP neurons (particularly at the ventromedial sector of the paliidum) are large cholinesterase-reactive cells that form a component of the nucleus basalis magnocellularis (NBM) and which provide cortical chotinergic input 3. In 1982, Whitehouse et al. 17,18 reported a loss of cells in the N B M as well as a reduction of cortical choline acetyltransferase (CHAT) activity in patients with Alzheimer's disease. Since then, numerous investigators have reported that neurotoxic lesions of the N B M in rats produce learning deficits on passive avoidance and spatial memory tasks 2.4.6-8.10.11.13. Therefore, the possibility must be considered that dysfunction of cortical cholinergic mechanisms is primarily responsible for the appearance of learning impairments in GP-lesioned rats. In the present study we assumed that, if the learning deficits we observed were due to partial destruction of the NBM, then the severity of the deficits should be correlated with the amount of cortical C h A T depletion produced by the lesion. Conse-
quently, C h A T activity was measured in various regions of cortex in each animal after producing bilateral lesions of the G P and subsequent behavioral testing. We report that there was no significant correlation between cortical C h A T activity and the severity of the learning deficit on a visual discrimination task, and conclude that the deficit is due to the loss of cells in the ventral GP and not to the depletion of cortical cholinergic innervation. A total of 31 weanling (21-23-day-old) male Sprag u e - D a w l e y albino rats (40-60 g) underwent surgery under deep chloral hydrate anesthesia (400 mg/kg). Lesions were produced by injecting 0,9/~1 ibotenic acid (IBO, 10 mg/ml) bilaterally into the GP of 12 animals (coordinates: ineissor 0,0, 7.0 mm anterior to lambda, 2.7 mm lateral to midtine, 5.9 mm ventral from the brain surface). Two additional groups (n = 10; n = 9) underwent the same surgical procedure as the IBO-lesioned animals, except that one group received no injection while the second received 0.9~1 of normal saline on each side. All operated animals were handled and weighed periodically, given food and water ad libitum and allowed to recover for 21 days prior to training on the visual discrimination task 14. Under the motive of es-
Correspondence: R. Thompson, Fairview State Hospital. Costa Mesa, CA 92626. U.S.A. 0006-8993/86/$03.50© 1986 Elsevier Science Publishers B.V. (Biomedical Division)
403 TABLE I Trial and error scores for all groups on the visual discrimination problem and body weight after a 3-week recovery period
Values represent mean + S.E.M. Statistical significance was determined by ANOVA and Newman-Keuls test. Group
n
Trials
Initial errors
Total errors
Body weight
No injection Saline IBO
10 9 12
31.8 _+2.7 34.0 _+3.0 69.8 -+ 7.0*
15.5 + 1.0 16.3 _+ 1.6 33.8 +_3.5*
2t.4 + 1.2 19.4 4- 2.2 43.7 + 4.2*
158,2 + 2.0 156.2 + 5.0 132.1 +_8.2*
* P < 0.01, statisticallysignificant difference from no injection and saline groups. c a p e - a v o i d a n c e of mild foot-shock ( 1 . 0 - 1 . 5 m A ) , each animal was required to a p p r o a c h the unlocked white (positive) card and to avoid the adjacent locked black (negative) card in o r d e r to gain entrance into the goal box. A n error was defined as an approach response to the black card which brought the animal's forefeet into contact with the electrified grid section which e x t e n d e d 8.0 cm in front of the negative card. Initial errors as well as multiple errors (intratrial repetitive approaches to the negative card) were recorded. The position of the positive and negative cards was switched from the right to the left side in a p r e a r r a n g e d ' r a n d o m ' order. Eight to 12 trials were given daily with an intertrial interval of 50-75 s. The criterion of learning consisted of the first appearance of either a 'perfect' or 'near-perfect' run of correct responses, having a probability of occurrence of 0.05 (ref. 12), followed by at least 75% correct responses in the subsequent block of 8 trials given on the next day. Following learning, all animals were sacrificed on the same day by decapitation and their brains dissected in a cold r o o m into frontal cortex, parietal cortex, occipital cortex, dorsal hippocampus a n d ventral hippocampus. C h A T activity was m e a s u r e d in each region by the m e t h o d of F o n n u m 5 and protein was measured according to the m e t h o d of Lowry et al. 9.
The remaining block of brain tissue was fixed with 30% sucrose/10% formal saline, sectioned at 40/~m, and stained with cresyl violet to d e t e r m i n e the locus and magnitude of the G P lesions. Histological analysis revealed that the locus, shape and size of the I B O lesions were similar to those described in our previous report. A s shown in Table I, the I B O - l e s i o n e d animals were' significantly impaired in learning the w h i t e - b l a c k discrimination in comparison to the controls. The level of C h A T activity in the various brain regions for each group is summarized in Table II. C h A T activity was decreased by 26% in the frontal cortex and 20% in the parietal cortex of the IBO-lesioned group relative to the controls. T h e r e was no significant effect on C h A T activity in the occipital cortex or the h i p p o c a m p u s and there was no significant effect of injecting saline alone. To d e t e r m i n e if the learning deficit was related to the reduction in cortical C h A T activity, r a n k - o r d e r correlations were c o m p u t e d b e t w e e n the individual learning scores of the 12 I B O - l e s i o n e d rats and their respective C h A T activity measures in the frontal cortex, parietal cortex and c o m b i n e d frontal-parietal cortex (see Table III). All correlations were in the same direction (lower C h A T levels m a t c h e d with greater trials and errors); however, none a p p r o a c h e d
TABLE II ChAT levels (nmol/mg protein~h) in 5 brain regions for all groups
Values represent mean + S.E.M. (no. of cases). Statistical significance was determined by ANOVA and Newman-Keuls test. Brain region
No injection
Saline
Frontal cortex Parietal cortex Occipital cortex Dorsal hippocampus Ventral hippocampus
72.1 _+2.2(10) 72.2 + 1.8 (9) 61.4 + 2.2 (9) 73.5 + 2.8 (9) 89.2 + 3.4 (9)
73.2 + 65.1 + 64.4 + 73.9 + 87.7 +
* P < 0.01, statisticallysignificant difference from no injection and saline groups.
IBO
1.7(9) 2.8 (9) 1.7 (9) 3.7 (9) 5.0 (9)
53.0 + 57.3 + 58.3 + 68.9 + 88.5 +
3.2(12)* 2.5 (11)* 1.8 (11) 1.6 (9) 3.2 (10)
404 TABLE II1
nificant correlation o f - - 0 . 5 6 between body weight at
Rank-order correlations between ChAT activity in the frontal and parietal cortex and trial and error scores on the visual discrimination task*
the outset of visual discrimination training and total errors to criterion.
Brain region
Frontal cortex Parietal cortex Frontal + parietal cortex
The finding that cortical C h A T activity scores
Initial
Total
were not significantly related to visual discrimination
errors
errors
-0.19 -0.15
-0.23 -0.15
-0.23 -0.20
learning scores suggests that learning deficits in our GP-lesioned animals resulted from damage to GP
-0.32
-0.39
-0.42
Trials
cells and not from inadvertent damage to NBM cells. Similarly, the possibility should be considered that
t A value of-0.50 or greater is needed for significance at the 0.05 level.
learning deficits reportedly due to destruction of the NBM are, in fact, the result of damage to the ventral
statistical significance (the correlation would have to
aspect of the GP. In a recent review article on brain systems and memory1 s, 10 separate studies reporting
reach 0.50 to be significant at the 0.05 level). The fact that all of the correlations were in the same direction may reflect slight differences in the severity of the
amnesic effects of lesions or electrical stimulation of the GP in rats are discussed. These findings, together
deficit due t o variation in lesion size. Consequently,
'subcortical dementia'a, provide clear evidence that further investigation of the pallidum holds promise in
no persuasive evidence was found to link the severity of the learning deficit to the degree of reduction in cortical C h A T activity. Interestingly, there was a sig-
1 Cummings, J.L. and Benson, D.F., Subcortical dementia, Arch. Neurol., 41 (1984) 24-28. 2 Dubois, B., Mayo, W., Agid, Y., LeMoal, M. and Simon, H., Profound disturbances of spontaneous and learned behaviors following lesions of the nucleus basalis magnocellularis in the rat, Brain Research, 338 (1985) 249-258. 3 Fibiger, H.C., The organization and some projections of chotinergic neurons of the mammalian forebrain, Brain Res. Rev., 4 (1982) 327-388. 4 Flicker, C., Dean, R.L., Watkins, D.L., Fisher, S.K. and Bartus, R.T., Behavioral and neurochemical effects following neurotoxic lesions of a major cholinergic input to the cerebral cortex in the rat, Pharmacol. Biochem. Behav., 18 (1983) 973-981. 5 Fonnum, F.A., A rapid radiochemical method for the determination of choline acetyltransferase, J. Neurochem., 24 (1975) 407-409. 6 Friedman, E., Lerer, B. and Kusier, J., Loss of cholinergic neurons in the rat neocortex produces deficits in passive avoidance learning, Pharmacol. Biochem. Behav.. 19 (1983) 309-312. 7 Helper, D.J., Olton, D.S., Wenk, G.L. and Coyle, J.T., Lesions in the nucleus basalis magnocellularis and medial septal area of rats produce qualitatively similar memory impairments, J. Neurosci., 5 (1985) 866-873. 8 Lo Conte, G., Bartolini, L., Casamenti, F., Marconcini-Pepeu, I. and Pepeu, G., Lesions of ~olinergic forebrain nuclei: changes in avoidance behavior and scopolamine actions, Pharmacol. Biochem. Behav., 17 (1982) 933-937. 9 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J., Protein measurement with the Folin phenol reagent, J. Biol. Chem.. 193 (1951)265-275.
with the growing literature implicating the GP in
deepening our knowledge about the neurobiological mechanisms of learning and memory.
10 Miyamoto. M.. Shintani. M.. Nagaoka. A and Nagawa. Y.. Lesioning of the rat basal forebrain leads to memory impairments in passwe and active avoidance tasks. Brain Research. 328 (1985) 97-104. 11 Murray, C.L. and FiNger, H.C.. Learning and memory deficits after lesions of the nucleus basalis magnocellularis: reversal by physostigmine. Neuroscience. 15 ~1985) 1025-1032. 12 Runnels. L.K.. Thompson. R and Runnels. P.. Near-perfect runs as a learning criterion. J. Math. Psychol.. 5 11968) 362-368. 13 Salamone. J.D.. Beart. P.M.. Alpert. J.E. and Iversen S.D.. Impmrment in T-maze reinforced alternation performance following nucleus basalis magnocellularis lesions in rats. Behav. Brain Res. 1311984)63-70 14 Thompson. R.. A Behavioral Atlas of the Rat Brain Oxford University Press. New York. 1978. 15 Thompson. R.. Brain systems and long-term memory, Behav. Neurol. Biol., 37 (1983) 1-45. 16 Thompson. R.. Gibbs, R.B.. Ristic, G.A.. Cotman, C.W. and Yu, J., Learning deficits in rats with early neurotoxic lesions to the globus pallidus, substantia nigra, median raphe, or pontine reticular formation, submitted. 17 Whitehouse, P.J., Price, D.L., Clark. A.W., Coyle. J.T. and DeLong, MR.. Alzheimer's disease: evidence for selective loss of cholinergic neurons in the nucleus basalis, Ann. Neurol., 10 (1981) 122-126. 18 Whitehouse, P.J., Price, D.L., Struble, R.G., Clark, A.W., Coyle, J.T. and DeLong, M.R., Alzheimer's disease and senile dementia: loss of neurons in the basal forebrain, Science, 215 (1982) 1237-1239.
Brain Research, 367 (1986) 402-404 Elsevier
BRE 21420
Lack of c o r r e l ~ ~ n learning ~
~ 1 levels of ~ m a~t~.raek,'foraea and in ~ s w ~ ~ s ~ ~ ~ons
ROBERT THOMPSON1, ROBERT B. GIBBS2, GINELL A. RISTIC2, CARL W. COTMAN2 and JEN YU l l Fairview State Hospital, Costa Mesa, and University of California lrvine Medical Center, Orange, and 2Universityof California, Irvine, CA (U.S.A.) (Accepted November 5th, 1985) Key words: globus pallidus - - nucleus basalis magnocellularis - - choline acetyltransferase - - learning
Previous work has shown that rats with lesions of the globus pallidus (GP) exhibit a generalized learning impairment. Data are presented suggesting that this impairment is not due to inadvertent damage to the nucleus basatis magaocell~aris. Rats with GP lesions evidenced a significant visual discrimination learning loss and a significant reduction in cortical choline aeetyltransferase (CHAT) activity. However, there was no significant correlation between the severity of the learning loss and the amount of reduction of cortical ChAT activity.
We have recently shown that young rats with neurotoxic lesions of the globus pallidus (GP) exhibit a global learning deficit characterized by a deficiency in learning visual and non-visual discrimination problems and a complex maze task 16. Intermingled among GP neurons (particularly at the ventromedial sector of the paliidum) are large cholinesterase-reactive cells that form a component of the nucleus basalis magnocellularis (NBM) and which provide cortical chotinergic input 3. In 1982, Whitehouse et al. 17,18 reported a loss of cells in the N B M as well as a reduction of cortical choline acetyltransferase (CHAT) activity in patients with Alzheimer's disease. Since then, numerous investigators have reported that neurotoxic lesions of the N B M in rats produce learning deficits on passive avoidance and spatial memory tasks 2.4.6-8.10.11.13. Therefore, the possibility must be considered that dysfunction of cortical cholinergic mechanisms is primarily responsible for the appearance of learning impairments in GP-lesioned rats. In the present study we assumed that, if the learning deficits we observed were due to partial destruction of the NBM, then the severity of the deficits should be correlated with the amount of cortical C h A T depletion produced by the lesion. Conse-
quently, C h A T activity was measured in various regions of cortex in each animal after producing bilateral lesions of the G P and subsequent behavioral testing. We report that there was no significant correlation between cortical C h A T activity and the severity of the learning deficit on a visual discrimination task, and conclude that the deficit is due to the loss of cells in the ventral GP and not to the depletion of cortical cholinergic innervation. A total of 31 weanling (21-23-day-old) male Sprag u e - D a w l e y albino rats (40-60 g) underwent surgery under deep chloral hydrate anesthesia (400 mg/kg). Lesions were produced by injecting 0,9/~1 ibotenic acid (IBO, 10 mg/ml) bilaterally into the GP of 12 animals (coordinates: ineissor 0,0, 7.0 mm anterior to lambda, 2.7 mm lateral to midtine, 5.9 mm ventral from the brain surface). Two additional groups (n = 10; n = 9) underwent the same surgical procedure as the IBO-lesioned animals, except that one group received no injection while the second received 0.9~1 of normal saline on each side. All operated animals were handled and weighed periodically, given food and water ad libitum and allowed to recover for 21 days prior to training on the visual discrimination task 14. Under the motive of es-
Correspondence: R. Thompson, Fairview State Hospital. Costa Mesa, CA 92626. U.S.A. 0006-8993/86/$03.50© 1986 Elsevier Science Publishers B.V. (Biomedical Division)
403 TABLE I Trial and error scores for all groups on the visual discrimination problem and body weight after a 3-week recovery period
Values represent mean + S.E.M. Statistical significance was determined by ANOVA and Newman-Keuls test. Group
n
Trials
Initial errors
Total errors
Body weight
No injection Saline IBO
10 9 12
31.8 _+2.7 34.0 _+3.0 69.8 -+ 7.0*
15.5 + 1.0 16.3 _+ 1.6 33.8 +_3.5*
2t.4 + 1.2 19.4 4- 2.2 43.7 + 4.2*
158,2 + 2.0 156.2 + 5.0 132.1 +_8.2*
* P < 0.01, statisticallysignificant difference from no injection and saline groups. c a p e - a v o i d a n c e of mild foot-shock ( 1 . 0 - 1 . 5 m A ) , each animal was required to a p p r o a c h the unlocked white (positive) card and to avoid the adjacent locked black (negative) card in o r d e r to gain entrance into the goal box. A n error was defined as an approach response to the black card which brought the animal's forefeet into contact with the electrified grid section which e x t e n d e d 8.0 cm in front of the negative card. Initial errors as well as multiple errors (intratrial repetitive approaches to the negative card) were recorded. The position of the positive and negative cards was switched from the right to the left side in a p r e a r r a n g e d ' r a n d o m ' order. Eight to 12 trials were given daily with an intertrial interval of 50-75 s. The criterion of learning consisted of the first appearance of either a 'perfect' or 'near-perfect' run of correct responses, having a probability of occurrence of 0.05 (ref. 12), followed by at least 75% correct responses in the subsequent block of 8 trials given on the next day. Following learning, all animals were sacrificed on the same day by decapitation and their brains dissected in a cold r o o m into frontal cortex, parietal cortex, occipital cortex, dorsal hippocampus a n d ventral hippocampus. C h A T activity was m e a s u r e d in each region by the m e t h o d of F o n n u m 5 and protein was measured according to the m e t h o d of Lowry et al. 9.
The remaining block of brain tissue was fixed with 30% sucrose/10% formal saline, sectioned at 40/~m, and stained with cresyl violet to d e t e r m i n e the locus and magnitude of the G P lesions. Histological analysis revealed that the locus, shape and size of the I B O lesions were similar to those described in our previous report. A s shown in Table I, the I B O - l e s i o n e d animals were' significantly impaired in learning the w h i t e - b l a c k discrimination in comparison to the controls. The level of C h A T activity in the various brain regions for each group is summarized in Table II. C h A T activity was decreased by 26% in the frontal cortex and 20% in the parietal cortex of the IBO-lesioned group relative to the controls. T h e r e was no significant effect on C h A T activity in the occipital cortex or the h i p p o c a m p u s and there was no significant effect of injecting saline alone. To d e t e r m i n e if the learning deficit was related to the reduction in cortical C h A T activity, r a n k - o r d e r correlations were c o m p u t e d b e t w e e n the individual learning scores of the 12 I B O - l e s i o n e d rats and their respective C h A T activity measures in the frontal cortex, parietal cortex and c o m b i n e d frontal-parietal cortex (see Table III). All correlations were in the same direction (lower C h A T levels m a t c h e d with greater trials and errors); however, none a p p r o a c h e d
TABLE II ChAT levels (nmol/mg protein~h) in 5 brain regions for all groups
Values represent mean + S.E.M. (no. of cases). Statistical significance was determined by ANOVA and Newman-Keuls test. Brain region
No injection
Saline
Frontal cortex Parietal cortex Occipital cortex Dorsal hippocampus Ventral hippocampus
72.1 _+2.2(10) 72.2 + 1.8 (9) 61.4 + 2.2 (9) 73.5 + 2.8 (9) 89.2 + 3.4 (9)
73.2 + 65.1 + 64.4 + 73.9 + 87.7 +
* P < 0.01, statisticallysignificant difference from no injection and saline groups.
IBO
1.7(9) 2.8 (9) 1.7 (9) 3.7 (9) 5.0 (9)
53.0 + 57.3 + 58.3 + 68.9 + 88.5 +
3.2(12)* 2.5 (11)* 1.8 (11) 1.6 (9) 3.2 (10)
404 TABLE II1
nificant correlation o f - - 0 . 5 6 between body weight at
Rank-order correlations between ChAT activity in the frontal and parietal cortex and trial and error scores on the visual discrimination task*
the outset of visual discrimination training and total errors to criterion.
Brain region
Frontal cortex Parietal cortex Frontal + parietal cortex
The finding that cortical C h A T activity scores
Initial
Total
were not significantly related to visual discrimination
errors
errors
-0.19 -0.15
-0.23 -0.15
-0.23 -0.20
learning scores suggests that learning deficits in our GP-lesioned animals resulted from damage to GP
-0.32
-0.39
-0.42
Trials
cells and not from inadvertent damage to NBM cells. Similarly, the possibility should be considered that
t A value of-0.50 or greater is needed for significance at the 0.05 level.
learning deficits reportedly due to destruction of the NBM are, in fact, the result of damage to the ventral
statistical significance (the correlation would have to
aspect of the GP. In a recent review article on brain systems and memory1 s, 10 separate studies reporting
reach 0.50 to be significant at the 0.05 level). The fact that all of the correlations were in the same direction may reflect slight differences in the severity of the
amnesic effects of lesions or electrical stimulation of the GP in rats are discussed. These findings, together
deficit due t o variation in lesion size. Consequently,
'subcortical dementia'a, provide clear evidence that further investigation of the pallidum holds promise in
no persuasive evidence was found to link the severity of the learning deficit to the degree of reduction in cortical C h A T activity. Interestingly, there was a sig-
1 Cummings, J.L. and Benson, D.F., Subcortical dementia, Arch. Neurol., 41 (1984) 24-28. 2 Dubois, B., Mayo, W., Agid, Y., LeMoal, M. and Simon, H., Profound disturbances of spontaneous and learned behaviors following lesions of the nucleus basalis magnocellularis in the rat, Brain Research, 338 (1985) 249-258. 3 Fibiger, H.C., The organization and some projections of chotinergic neurons of the mammalian forebrain, Brain Res. Rev., 4 (1982) 327-388. 4 Flicker, C., Dean, R.L., Watkins, D.L., Fisher, S.K. and Bartus, R.T., Behavioral and neurochemical effects following neurotoxic lesions of a major cholinergic input to the cerebral cortex in the rat, Pharmacol. Biochem. Behav., 18 (1983) 973-981. 5 Fonnum, F.A., A rapid radiochemical method for the determination of choline acetyltransferase, J. Neurochem., 24 (1975) 407-409. 6 Friedman, E., Lerer, B. and Kusier, J., Loss of cholinergic neurons in the rat neocortex produces deficits in passive avoidance learning, Pharmacol. Biochem. Behav.. 19 (1983) 309-312. 7 Helper, D.J., Olton, D.S., Wenk, G.L. and Coyle, J.T., Lesions in the nucleus basalis magnocellularis and medial septal area of rats produce qualitatively similar memory impairments, J. Neurosci., 5 (1985) 866-873. 8 Lo Conte, G., Bartolini, L., Casamenti, F., Marconcini-Pepeu, I. and Pepeu, G., Lesions of ~olinergic forebrain nuclei: changes in avoidance behavior and scopolamine actions, Pharmacol. Biochem. Behav., 17 (1982) 933-937. 9 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J., Protein measurement with the Folin phenol reagent, J. Biol. Chem.. 193 (1951)265-275.
with the growing literature implicating the GP in
deepening our knowledge about the neurobiological mechanisms of learning and memory.
10 Miyamoto. M.. Shintani. M.. Nagaoka. A and Nagawa. Y.. Lesioning of the rat basal forebrain leads to memory impairments in passwe and active avoidance tasks. Brain Research. 328 (1985) 97-104. 11 Murray, C.L. and FiNger, H.C.. Learning and memory deficits after lesions of the nucleus basalis magnocellularis: reversal by physostigmine. Neuroscience. 15 ~1985) 1025-1032. 12 Runnels. L.K.. Thompson. R and Runnels. P.. Near-perfect runs as a learning criterion. J. Math. Psychol.. 5 11968) 362-368. 13 Salamone. J.D.. Beart. P.M.. Alpert. J.E. and Iversen S.D.. Impmrment in T-maze reinforced alternation performance following nucleus basalis magnocellularis lesions in rats. Behav. Brain Res. 1311984)63-70 14 Thompson. R.. A Behavioral Atlas of the Rat Brain Oxford University Press. New York. 1978. 15 Thompson. R.. Brain systems and long-term memory, Behav. Neurol. Biol., 37 (1983) 1-45. 16 Thompson. R.. Gibbs, R.B.. Ristic, G.A.. Cotman, C.W. and Yu, J., Learning deficits in rats with early neurotoxic lesions to the globus pallidus, substantia nigra, median raphe, or pontine reticular formation, submitted. 17 Whitehouse, P.J., Price, D.L., Clark. A.W., Coyle. J.T. and DeLong, MR.. Alzheimer's disease: evidence for selective loss of cholinergic neurons in the nucleus basalis, Ann. Neurol., 10 (1981) 122-126. 18 Whitehouse, P.J., Price, D.L., Struble, R.G., Clark, A.W., Coyle, J.T. and DeLong, M.R., Alzheimer's disease and senile dementia: loss of neurons in the basal forebrain, Science, 215 (1982) 1237-1239.