Spatial learning deficits in the aged rat: neuroanatomical and neurochemical correlates

Brain Research Bulletin, Voi. 33, No. 5, pp. 489-500, 1994 Copyright 6 1994 Elsevier Science Ltd Printed in the USA. All rights reserved 0361-9230/94 $6.00 + .OO

Pergamon

Spatial Learning Deficits in the Aged Rat: Neur~anatomi~al and Neurochemical Correlates J. M. LEE,*$l E. R. ROSS,?’ A. GOWER,$ J. M. PARIS,* R. MARTENSSONg

AND S. A. LORENS*

Departments of *Pharmacology and fPathology, %-itch School ofMedicine, Loyola University, Maywood, IL 60153 $U.C.B. sa., Braine-L’Alleud, Belgium #Department of Dermatology, University of Lund, Lund, Sweden Received 29 March 1993; Accepted 4 October 1993 LEE, J. M., E. R. ROSS, A. GOWER, J. M. PARIS, R. MARTENSSON AND S. A. LORENS. Spatial learning de&its in the aged rat: ~earoana:om~cal and neuroc~emical correlates. BRAIN RES BULL 33(5) 489-500,1994.-To assess nemochemical

and neuroanatomjcal correlates of age and spatial learning, aged Sprague-Dawley male rats (20-22 mo) were divided into two groups based on their ability to locate a hidden platform in a Morris water maze. An “old good” group of rats acquired the task as rapidly as young (3-6 mo) animals, whereas an “old poor” group of rats failed to show improvement on subsequent testing days. Age-related changes included (a) a significant decrease in the number of choline acetyltransferase (CHAT) immunoreactive cells in the ventral cell group of the septal complex (28%); (b) a decrease in caudate dopamine levels (- 11%); and (c) an increase in 5-HLAA levels in the n. accumbens (+25%) and hippocampus (+lS%). Spatial learning related changes in aged rats included: (a) an increase in medial frontal cortex 5HIAA levels (52%) in the old good learners but not old poor learners with (b) a decrease in media! frontal cortex dopamine levels (-24%) only in the old poor learners group and (c) a decrease in n. accumbens DOPAC (-22%) and HVA (-23%) in the old good learners group only. The present study demonstrates age-related but not spatial learning related decrease in CHAT immunoreactive cells in the ventral cell group of the septal complex. Therefore, either the cholinergic cell loss in the septum is unrelated to the acquisition of spatial learning measured by the Morris water maze, or it is a permissive effect along with specific alterations in forebrain dopaminergic and serotonergic systems, particularly in the medial frontal cortex and n. accumbens. The above findings are consistent with findings seen in Alzheimer’s disease where both basal forebrain cholinergic nuclei and cortical projecting brainstem monoamine systems are affected. Choline acetyltransferase

Dopamine

Serotonin

Norepinephrine

Spatial learning

Aging

itive cell loss (25-50%) in the septal region but not in the nucleus basalis of aged rats (3,28). Discrepancies in the animal studies may be explained by any one or combination of factors: species, strain, and relative age differences at the time of study. More importantly, in a great majority of reports, there is no measurement of cognitive or behavioral capability. In addition to the cholinergic changes, central monoamine levels also are altered with age in both human and other mammals (for review see 44). Moreover, brainstem manoamine nuclei that project to the forebrain are affected in AD, Parkinson’s Disease and other neurodegenerative disease processes (12,13,44,85). The ascending monoamine systems have been hypothesized to be involved in learning. Although specific lesions of the serotonergic system with 5,7_dihydroxytryptamine do not alter spatial learning tasks, these lesions potentiate the learning deficits caused by basal forebrain cholinergic lesions (54). The medial frontal cortex (MFC), a component of the mesoiimboco~icai DA system which receives afferents from the ventral tegmental area, is involved with the processing of environmental spatial and temporal cues needed for memory acquisition (7). In addition, there is anatomical evidence that the basal forebrain cholinergic nuclei

BASED upon reports of basal forebrain cholinergic cell loss in Alzheimer’s disease (AD) (50,51,58,66,77,78) along with the experimental evidence that anti-cholinergic agents affect memory processes and cognition (33,36,76), an explosion has taken place in the experimental use of cholinergic drugs for the treatment of AD (4,6,8,17,69). This approach however, has had only limited if any success. It has been hypothesized, therefore, that other neurotransmitter systems or other factors are involved in the disease process (5,8,19,20,29,32,52,53,71-73,81,84). Any hypothesis regarding the etiology of AD must take into account degeneration of neurons in the chofinergic rich basal forebrain, deposition of amyloid plaques CAP), fo~ation of neurofibrillary tangles (NIT), decreases in cortical somatostatin levels and alterations in central monoamine concentrations. All of these abnormalities have been found in the vast majority of patients with AD (for review see 16). A major problem in this field is the lack an animal model to investigate pathologic aging processes. Reports of age-related cholinergic cell loss in humans has been equivocal (50,51,56,58,62,77,78; for review see 14). On the other hand, some studies show no change in cholinergic cell number (34,43), while others report acetylcholinesterase (AchE) pos-

1To whom requests for reprints should be addressed. ‘ Dedication: This manuscript is dedicated to the memory of E. R. “Manny” Ross, our dear friend and colleague who recently passed away. 489

490

l.EE ET AL.

receive afferent connections from these mesencephalic and pontine monoam~nergic nuclei (3574). In the present study we sought neuropathological, neuroanatomical and neurochemical correlates of impaired spatial learning acquisition in aged male Sprague-Dawley rats. Lesion and pharmacological studies have shown that spatial learning tasks, such as the acquisition in a Morris water maze, are impaired when cholinergic neurons in the septal and nucleus basalis regions are damaged or following anti-cholinergic drug treatment (75). Therefore, at the beginning of the study the aged rats (19-22 mo) were separated into two groups based upon their performance in the Morris water maze. One group was designated “old poor learners” and the other “old good learners”. These groups were compared to each other and to a young control group (36 mo) to determine if there was an accelerated aging component based upon neuroanatomical variables. We determined the number, area, and morphology of CHAT immunoreactive neurons in the basal forebrain and brainstem. In addition, we investigated forebrain monoamine levels in a similar but separate group of animals. METHODS

A. Animals Male Sprague-Dawley rats (IFFA-CREDO, Belgium) were 3 (young) or 19 (old) mo of age at the start of testing. The animals at the beginning of the study were 380 + 82 and 650 + 88 grams (mean 2 SD) in the young and old groups, respectively. They were housed in pairs in wire cages located in a temperature (22 f l”C), humidity (50-55%) and illumination (12 h lightdark cycle; with lights on at 7.00 h) controlled room and allowed ad lib access to tap water and standard rat chow. B. Behavioral Testing Spatial learning ability was tested using a Morris-type swimming maze (26,49). The maze consisted of a white plastic circular pool (1.4 m diameter x 45 cm height), filled to a depth of 33 cm with water made opaque with powered milk. The pool contained a platform (11 X 11 cm), hidden 3.0 cm below the surface of the water at a fixed position half way between the center and the perimeter of the pool, along an axis separating two quadrants. Each rat was required to escape from the water onto the platform using prominent extramaze cues. Each rat received four trials per day on four consecutive days. Each trial was initiated by placing the rat in the maze at a starting point midway between each quadrant at the perimeter of the pool. Thus, four different starting points were used each day, and in a randomized order. The time to reach the platform, the distance swum, and the pathway followed were recorded automatically using an overhead monitaring camera linked to a BBC computer system. Each rat was allowed to remain on the platform for 60 s between each trial. If the rat failed to reach the platform within 120 s, it was placed onto the platform by the experimenter. Separate groups of animals were selected for anatomicai and neurochemical studies 57 days following behavioral testing.

buffered fixative with 20% sucrose for 1 h. The brain then was transferred to a storage solution containing 20% sucrose, 0.1% sodium azide, and 0.01% bacitracin in O.lM PB for 1-S days at 4°C prior to sectioning. From each animal CHAT, paired helical filamnent (PHF), cresyl violet (CV) and von Braunmuhls staining was performed on adjacent sections. Coronal sections (40 pm) were taken rostrocaudally at 0.2 mm intervals from the crossing of the corpus callosum through the hipp~ampus. In addition, 10 pm sections were taken from the frontal, temporal and occipital cortices as well as the hippocampus for staining by the hematoxylin and eosin (H and E), Bodian and Thioflavin S methods. For all immunocytochemical staining, the 40 km sections were initially placed in O.lM PB containing 0.1% sodium azide and 0.01% bacitracin for 24 h. D. CHAT Staining Visualization of CHAT immunoreactive neurons was achieved using a modification of the method of Sternberger et al. (68). A rat anti-CHAT monoclonal antibody was obtained from Boehringer-Mannheim Biochemicals (Indianapolis, IN). A working solution was made by dilution of 40 pg of stock with 4.0 ml of double distilled water with 0.1% sodium azide and 0.01% bacitracin. Sections were incubated in a final primary antibody concentration of 2.0 pg/ml in a diluent of 2.0% bovine serum albumin, 0.2% Triton X-100,0.1% sodium azide, 0.01% bacitratin and 10% normal goat serum. Sections were incubated for 24 h at 22°C in humidified chambers with constant agitation. The secondary antibody (goat anti-rat IgG) (Jackson Immunoresearch Lab Inc., Avondale, PA) was diluted 1:lOO in the diluent described above with the sodium azide omitted. Sections were incubated for 1 h at 22°C in h~idified chambers with agitation. This was followed by an incubation with rat peroxidase-antiperoxidase (PAP 1:lOO) (Sternberger-Meyer, Inc., Jarretsville, MD) for 1 h at 22°C. The free floating sections were then placed in 0.1% 3, 3-diaminobenzidine (DAB) and 0.033% hydrogen peroxide in O.lM PB for 12-15 min at 22°C. The reaction was stopped by placement in O.lM PB. Sections were mounted and dehydrated followed by rehydration and intensified by 0.1% osmium tetroxide for 30 s. CHAT immunoreactive cells were counted in each area by using a Baush and Lomb microprojection system. The cells counts were done by J.L. and S.L. who were blind to the animal subgroup designations at the time of cell counting. Cell counts were determined on multiple sections (rostra1 to caudal) at 0.2 mm intervals throughout the cell group designations (see below). Therefore, the cell number reported for each area represents the total number counted throughout the nucleus in an attempt to simulate the in vivo, three-dimensional structure of the nuclei. The cell number in each anatomical area was analyzed by ANOVA followed by a Duncan’s multiple range test. The cell areas and diameters were calculated using the BIOQUANT systern (Biometric, Inc., Nashville, TN). The statistical data analysis (STAT) and distribution (DIST) programs were used. Each immunoreactive cell was outlined manually by a cursor at 40X magnification, tabulated and stored by the BI~U~ programs described above. Thirty-three cells from each area were analyzed.

C. Perfusion, Blocking and Sectioning The animals were anesthetized with sodium pentobarbital(90 mg/kg, IP) and perfused transaortically with 200-250 ml of 0.1M ~i~~~siurn phosphate buffer (O.lM PB) at a rate of 500 ml/h at room temperature. mediately after beginning the perfusion, 300U of heparin was administered IV. This was followed by 500 ml of 4.0% paraformaldehyde and 0.1% glutaraldehyde in O.lM PB. The brain was removed and placed in the

E. Paired Helical Filament Staining Sections were floated in O.lM PB with 0.1% sodium azide and 0.1% bacitracin for 24 h, then transferred to 4% buffered formalin for 1wk, then transferred back to O.lM PB with sodium azide and bacitracin for 48 h before incubation in the primary antibody. A rabbit anti-human PHF antibody was provided by Dr. Dennis Selkoe (Brigham and Women’s Hospital, Harvard

491

SPATIAL LEARNING CORRELATES

Medical School, Boston, MA) and was used at 1:750 and incubated for 48 h at 0°C. The sections were developed using rabbit PAP (1:lOO) for 1 h followed by 0.05% DAB and H202. Each section was studied for the presence of immunoreactive PHF in the forebrain areas. Sections were taken rostrocaudally from the crossing of the corpus callosum through the hippocampus at 0.2 mm intervals. Frozen sections from the temporal lobe of a patient with known AD was used as a positive control. F. Special Stains

Routine hematoxylin and eosin, Bodian and von Braumuhl silver impregnation stains, as well as Thioflavin S staining for neuritic plaques and neurofibrillary tangles were done by a modification of the methods described in the Armed Forces Institute of Pathology (AFIP) manual (45). G. HPLC Determination

of Monoamines

and Their Metabolites

The amygdala, hippocampus, hypothalamus, medial frontal cortex, nucleus accumbens and neostriatum were dissected on a cold plate and the tissue was immediately frozen on dry ice and stored at -80°C. Tissue levels of monoamines and metabolites were determined within 3-4 wk by high performance liquid chromatography (HPLC) with electrochemical detection by a method previosuly described (41). H. Anatomical Terminology of the Basal Forebrain Nuclei (BFN) and Brainstem Cholinergic Cell Groups The septal complex (Fig. 1). CHAT immunoreactive perikarya in the septal complex were subdivided into four subgroups:

the ventral, dorsal, intermediate, and midline cell groups. This anatomical classification of septal CHAT neurons is based on the classification of Amaral and Kurz (1) and has been previosuly described (40). The neurons of the dorsal cell group are generally small and have few dendritic processes compared to other CHAT neurons in the BFN. The cell diameter for these cells is approximately 16 pm. At more caudal levels the cells lie dorsal to the anterior commissure. This cell group projects mainly to the cauda1 and temporal regions of the hippocampus (1,47). The ventral cell group is in an area that corresponds to what has been described as the vertical limb of the diagonal band of Broca. These cells first appear rostrally at the level of the crossing of the corpus callosum and are separated from the dorsal group by a band of noncholinergic cells. These cholinergic cells are generally larger, fusiform, multipolar, aligned parallel to one another and intensely stained. This cell group projects to the more rostra1 and dorsal aspects of the hippocampus (1). The intermediate cell group was initially defined by its projection sites. These cells were found to project to cingulate cortex (30%) and to other cortical areas rather than to the hippocampus (1). This group appears as a string of neurons that run from the dorsal group to the ventral group through areas where noncholinergic cells are abundant. Amaral and Kurz (1) suggest that because of their cortical projections these cells may form the rostra1 extension of the nucleus basalis of Meynert. The cells are generally more rounded with dendritic projections that lie perpendicular to the ventral cell group neurons. The midline group is found directly on the midline bounded bilaterally by the dorsal cell group. However, these neurons are

FIG. 1. Photomicrographs (X10) of eight coronal sections (40 pm), each separated by 0.2 mm, through the rostrocaudal (1-8) extent of the septal area showing the distribution of choline acetyltransferase immunoreactive perikarya (PAP method) and their subgroups. Abbreviations: cc, corpus callosum; d, dorsal subgroup; i, intermediate subgroup; m, midline subgroup; v, ventral subgroup.

492

separated from the dorsal cells by an area of noncholinergic cells. These cells are more intensely stained than in the dorsal group. However, like the dorsal group they have few dendritic processes. They have similar projection sites as the dorsal cell group K47). The magnocellularpreoptic ceil (MaPO) group (Fig. 2). This group of cells is found lateral to the more caudal levels of what has been described as the horizontal limb of the diagonal band of Broca (hlDB). This cell group also has been described as the substantia innomatia. It begins rostrally from an area lateral to the hlDB and optic tracts and ventral to the anterior ~ommissure and extends caudally to the beginning of the more ventromedial cells of the nucleus basalis. The cells are generally rounded (diameter = 18 pm), multipolar and randomly dispersed throughout the cell group. Most of the cells in this area project to cortical areas and not to hippocampal sites (47,61,64). However, the more ventral cells in this area do send afferents to the olfactory bulb. The nucleus basalis magnocellularis (NBM) and ventral globus pallidus cell groups (Fig, 2). This cell group begins as a caudal extension of the MaPO cell group that swings dorsally and medially into the internal capsule. The NBM cells are aligned parallel to one another. The cells are fusiform, multipolar, intensely stained and have an approximate diameter of 20 grn (59,61,64). The brainstem peripeduncular tegmental (PPT) and lateral dorsal tegmental (LDTn) cell groups (Fig. 3). The PPTn begins rostally just caudal and lateral to the substantia nigra in the mes-

LEE ET AL.

encephalon and extends as a diffuse cluster of cells in the lateral pontine regions extending into the central gray. This nucleus then swings dorsally and medially to merge with the LDTn that lies adjacent to the fourth ventricle lateral to the NTD. The LDTn extends caudally approximately 0.2mm and ends at the level of the exit of the fifth cranial nerve (37,60). I. Statistics Behavioral data was analyzed by a two-way analysis of variance (ANOVA) with repeated measures. Anatomical and neurochemical data were analyzed by an ANOVA. Posthoc comparisons of group means were obtained by a Duncan’s multiple range test for all parameters. All significant differences reported are at the p < 0.05 level or less. RESULTS

Selection of Rats for Anatomical and ~eurochemica~ Studies A. Anatomical studies. The young rats (3-6 mo) were selected at random from a large group of animals which were tested in the water maze. The performance of the young rats (N = 6) did not differ from the remaining pool of young animals (data not shown). Old rats were allocated to the “good learners” group on the basis of learning curves which were similar to those obtained in young rats. Their latencies and distances swum decreased progressively each day, and they showed an obvious awareness of

2. Photomicrographs (X10) of six coronal sections (40 pm), each separated at 0.2 mm, through the rostrocaudal (l-6) extent of the ver ltral dum showing the distribution of choline acetyltransferase imnmnoreactive perikarya (PAP method) of the two subgroups. Abbrebfiations: m, =agnocellular preoptic cell group; b, nucleus basalis magnocellularis cell group.

493

SPATIAL LEARNING CORRELATES

8 FIG. 3. Photomicro~ap~s (X10) of 8 hor~ontal sections (40 pm), each separated by approximately 0.2 mm, through the rostr~audal extent of the lower mesencephalon and the pons showing the distribution of choline acetyltransferase immunoreactive perikarya (PAP method) and the two subgroups. Abbreviations: p, peduncular tegmental cell group; 1, lateral dorsal tegmental cell group.

the location of the platform as evidenced from their swim pathways on Day 4. Old rats failing to show such progressive learning curves and evidence of orientation towards the platform were designated as “poor learners.” The latency and distances swum for the young rats and the subgroups of good and poor old rats are illustrated in Figs. 4A and 4B. The impaired ability of the

old poor rats is evidenced by their prolonged latencies and increased distances swum. Si~ificant group differences for latency [F(2,12) = 17.34; p < 0.011 as well as for distance [F(2,12) = 19.02; p < 0.011 were observed. The old poor learners were significantly different from both old good and young animals on Day 4 for latency and distance (Duncan’s multiple range test). On Day 3, the old poor learners differed only from the young animals. Both the young and old good learners showed a significant improvement in both latency and distance across days [F(3,36) = 5.72; p < 0.05 and F(3,36) = 9.13; p < 0.05, rea. LATENCY

spectively]. This was not so for the old poor learners which showed no change from baseline levels on Day 1. There was no effect on swimming speed between any of the three groups

[F(2,12) = 3.37; p = O.OS].Therefore, the change in latency in the old poor learners is not a result of performance variables

between the old good learners and old poor learners. B. Neurochemical studies In the group of animals used for neur~hemic~ parameters, there was a significant effect across groups for latency [F(2,18) = 29.4; p < O.OOl] as well as for distance [F(2,18) = 14.61;~ < O.Ol]. Using a Duncan’s multiple range test, the old poor learner group was significantly different from the old good learners and young controls on Day 4 with respect to both the latency and distance parameters. Both the young group and the old good learners showed a significant improvement in latency and distance across days [F(3,54) = 8.6; p < 0.05 and F = (3,54) = 8.3; p < 0.05, respectively]. The old poor learners failed to show improvement from baseline levels on Day 1.

b. DISTANCE HISTOLOGICAL.

-::?:

FIG. 4. A) Data represents the mean latency (s) + SEM for each group of rats to reach the hidden platform on successive testing days. B) Data represents the mean distance swum (m) + SEM for each group of rats before reaching the hidden platform. A----A young group; W---m old good learners group; O----O old poor learners group.

STUDIES

Multiple sections (10 pm) were taken from frontal, temporal and occipital cortices as well as from the hippocampus. There was no evidence of infarction, hypoxia, gliosis or alterations in cortical or hippocampal lamination in any of the old rats in comparison to the young animals. There was no perivascular lymphocytic cuffing or myelin degeneration noted in any of the sections. The brains of the old and yound animals were virtually indistinguishable. With Bodian and von Braunmuhll silver stains, there was no evidence of “flame-like” neurons suggestive of NET in any of the three groups. In addition, there was no evidence of amyloid plaques in any of the aged animals. Both plaques and tangles were evident in the stained tissue from an Alzheimer’s case using both these stains. Thioflavin S staining also failed to reveal evidence of amyloid deposits in either the old or young animals. The fact that the Bodian stain did not demonstrate any NFT’s and plaques was confirmed by the absence of staining with anti-

494

LEE El‘ AL.

TABLE

I

CHAT IMMUNOREACTIVE CELL NUMBER Septal Area

GIOUP

Young Old Good Old Poor

Dorsal

Midline

Intermediate

Ventral

Total

810 t 73 756 i 53 803 +- 59

82 + 4 85 _f 8 81 k 7

699 -+ 69 744 ? 106 766 + 50

1105 + 80 815 + 65* 774 2 48*

2625 -+ 138 2431 2 208 2424 k 149

Caudal Basal Forebrain and Brainstem Regions

Young Old Good Old Poor

MaPO

NBM

PPTn

LDT

1206 2 51 1013 f 75 1132 -t 67

1357 -+ 72 1402 k 77 1397 -t 40

719 + 29 721 t 55 764 ? 47

374 +- 19 364 i 46 311 t 11

Data presented as group means + SEM (N = 4-6). * Significantly different than young group; p < 0.05. MaPO = magnocellular preoptic area; NBM = nucleus basalis magnocellularis;

PPTn = pedun-

culopontine tegmental nucleus; LDTn = lateral dorsal tegmental nucleus.

Therefore, a large part of the error in any measurement is reflected in the intraassay variability. 2. Cell area. There were no significant changes in the cell areas in the dorsal and midline group nor in the ventral group (Table 2). For area measurements, cells from both the dorsal and midline groups were used and designated “dorsal.” In addition, neither the equivalent diameters, the median cell area, nor the cumulative frequency distribution (data not shown) of cell areas were different in any of the three groups.

PHF antibody. There was no evidence of tangles or positive immunoreactive cells in the frontal or entorhinal cortex nor in the cholinergic rich basal forebrain nuclei. Summary There were no neuropathological changes noted in either of the two older groups of animals when compared to each other or to the younger group. CHAT IMMUNOREAC’I’IVE NEURONS IN THE BASAL FOREBRAIN AND PONTINE NUCLEI

B. The Magnocellular Cell Groups

A. The Medial Septal Cell Groups

1. Cell number. There was no significant change in the number of CHAT immunoreactive cells in either cell group. Although there is a trend toward a decrease in cell number in the old groups in the MaPO area, the ANOVA did not reach significance [F(2,12) = 1.842,~ = 0.201 (Table 1). 2. Cell area. There was no significant change in cell area in the MaPO cell group. In addition, there was no difference in the equivalent diameter, median area, nor cumulative frequency distribution (data not shown) between young animals and either old animal groups. In the NBM, an ANOVA of area but not diameter revealed a significant effect [F(2,11) = 4.42; p < 0.051 when

1. Cell number. There was a significant decrease in the number of CHAT immunoreactive cells in the ventral cell group both in the old good and old poor learners when compared to controls [F(2,12) = 6.457; p < 0.011 (Table 1). However, there was no difference between the old good and old poor groups. The differences were 26% and 30%, respectively. There was no change in the number of cells in the dorsal, midline or intermediate cell groups between the three groups. The interassay variability was approximately 16%. The intraassay variability between the two cell counters (S.L. and J.L.) was 10% and ll%, respectively.

TABLE

2

CHAT IMMUNOREACIWE Group Area (pm’) Young Old Good Old Poor Diameter (pm) Young Old Good Old Poor

142 137 144

+ 7 f 3 t5

15.8 -e 0.5 15.6 + 0.2 15.9 4 0.3

CELL SIZE

170 170 179

i- 8 2 3 -f 10

17.3 t17.3 2 17.7 2

* p < 0.05 compared to young group. Data represents group means + SEM (N = 4-6).

NBM

MaPO

Ventral

Dorsal

Preoptic and Nucleus Basalis of Meynert

0.5 0.1 0.5

157 163 167

k 12 2 4 t- 7

16.3 2 17.0 t 17.2 2

0.7 0.2 0.4

201 171 185

f9 ? 3* t4

18.8 i- 0.5 17.6 k 0.2 18.1 ? 0.2

SPATIAL

LEARNING

followed by a Duncan’s multiple range test (p < 0.05) between the young and old good learners (Table 2). If the old animals were grouped together, a t-test revealed that the there is a significant effect of age alone between the old and young animals (t = 2.352; p < 0.05). The data suggest that either a subpopulation of large cells is dropping out in the two old groups when compared to the young group or a subgroup of large cells shrink with age. If a subpopulation of cells was dropping out, then the distribution of cell areas would change from a bimodal to a normal distribution. To test this hypothesis, a Kolmogorov-Smirnov Test for normal distribution was done (39). The young animal cell areas in the NBM were not normally distributed @ < 0.001, N = 11, Max Diff = 0.660). This means that there is at least a bimodal distribution suggesting two or more subpopulations of neurons in the area. However, the cell areas in both the old groups also were not normally distributed @ < 0.05, N = 11, Max Diff = 0.409). Therefore, it may be that a subpopulation of large cells may be shrinking in the old animals. This is supported by the finding that there is no cell loss in the NBM, arguing against a large cell drop out with age. C. Brainstem

495

CORRELATES

Cell Groups

There was no significant difference in the number of cells in either the PPTn or in the LDTn among the three groups (Table 1). However, closer examination of the data suggests that the LDTn in the old poor learners have a decreased number of cells when compared to the young controls and that the ANOVA analysis may be insignificant because of the variability in the old good learners.

TABLE

3

REGIONAL INDOLAMINE LEVELS Area

Serotonin

Group

5HIAA

(rig/g tissue) Caudate

N.ACC.

MFC

AMG

HYPO

HIPP

658 651 634 1290 1291 1355 928 887 871 1256 1149 1094 984 1009 807 683 628 656

Young (I) Old Good (II) Old Poor (III) I II III I II III I II III I II III I II III

2 z + k 2 2 2 2 -e 2 2 2 2 2 f 2 + 2

56 100 74 188 173 159 170 224 224 120 162 99 138 169 147 88 63 92

321 360 355 341 417 439 203 310 241 472 486 459 567 633 621 289 345 338

+ 30 2 49 ? 45 r+_42 2 79* k 54* k 52 k 68$ 2 38 2 58 k 92 k 56 2 97 c 37 2 72 2 24 k 43* 2 47*

Mean 2 SD; N = 7 per group. N. ACC. = n. accumbens, HYPO = hypothalamus, AMG = amygdala, HIPP = hippocampus, MFC = medial frontal cortex. * p < 0.05 compared to young group (I). t p < 0.05 compared to both young (I) and old poor (III) groups.

Summary of CHAT Immunocytochemistry There was a significant aging effect on the number of immunoreactive cells in the ventral cell group of the medial septal complex. There was no evidence of an accelerative aging process in the old poor learning group since there was no difference between the old good and old poor animals. In the NBM, there was a significant effect with age on cell area when the old good learners were compared to young controls. However, since the NBM cell areas in the old good and old poor learners were not different, there was no evidence of an accelerated aging process. In the LDTn, the data suggests a trend toward a decrease in the cell number in the old poor learners. NEUROCHEMICAL

PARAMETERS

A. Effects on Serotonin and 5-HIAA Levels In the brain areas evaluated no significant alteration in serotonin levels was found (Table 3). On the other hand, in the n. accumbens and hippocampus, there were significant increases in 5-HIAA levels in both the old good and old poor group when compared to the young group. In the nucleus accumbens the increases were 22% and 29% in the old good and old poor, respectively [F(2,18) = 5.55; p < 0.051. In the hippocampus, the increases were 19% and 17%, respectively [F(2,18) = 4.25; p < 0.05). In the medial frontal cortex, there was a significant 52% increase in 5-HIAA levels in the old good group compared to the young group, which was not seen in the old poor group [F(2,18) = 6.86; p < 0.011. The old poor and old good groups were also statistically different from one another (p < 0.05). As an indirect measure of turnover, the 5-HIIAA/SHT ratios were calculated. There were significant elevations only in the same groups that had increased 5-HIAA levels alone (data not shown).

B. Effects on Dopamine, DOPAC and WA

Levels

There was a significant effect of age alone on caudate DA levels as evidenced by a 6% and 11% decrease in the old good and old poor learners, respectively [F(2,18) = 6.98; p < 0.011. There were no differences in DA levels between the two old groups. No significant changes in DA levels were found in the amygdala, n. accumbens, hippocampus or in the hypothalamus. On the other hand, there was a significant 24% decrease in medial frontal cortical DA levels in the old poor group when compared to the young controls [F = (2,18) = 4.50;~ < 0.051. This decrease was not seen in the old good group. DOPAC levels were significantly lower only in the n. accumbens, where a 22% decrease in the old good learners was found. This decrease in DOPAC was not seen in the old poor learners (8%). The old poor and the old good groups were statistically different from one another [F(2,18) = 6.12;~ < O.Ol]. HVA was measured only in the caudate and n. accumbens. There was no significant changes in the caudate. However, like the DOPAC levels in the n. accumbens, HVA levels were decreased 23% in the old good learners but not in the old poor learners [F(2,18) = 4.62; p < 0.051. Again, this effect in the old good learners was statistically different from both the young and the old poor groups. As an indirect measure of turnover, the DOPAC/DA and HVA/DA rations were calculated. There were no differences found between any of the three groups in any brain area investigated (data not shown). C. Effects on Norepinephrine

Levels

No significant differences in NE levels were found in any of the forebrain areas investigated (Table 4).

496

LEE ET Ai..

DJSCUSSION

In the present study, we investigated an animal model for normal and accelerated pathologic aging processes by separating animals of the same chronological age into “old good learners” and “old poor learners’” based on their ability to solve a spatial learning task (26,49). A similar pattern is seen clinically with AD patients who show significant deficits in their cognitive skills compared to nonaffected individuals of the same chronological age. The use of rats for aging studies may have some inherent difficulties since none of the neuropathological findings of AD was demonstrable (i.e., no evidence of seniIe plaques, Nff’s, gliosis or vascular deposition of amyIoid). The old animals in the present study were 22-24 mo oId at the time of sacrifice. The rn~irn~ life span of Sprague-Dawiey rats is approximately 36 mo. Therefore, the rats used are late middle to earIy aged with regard to their average life span. Nonetheless, the behavioral deficits seen in the present study either precede or are not reflected in overt CNS neuropathology, Other studies using aged animals have sought to document similar neuropathological changes to those found in AD. Indeed, higher primates and other aged mammals, such as dogs and polar bears, form amyloid plaques with increasing age (63,82). It has been reported that rats which received injections of the cholinotoxic agent AF64A in the nucleus basalis developed amyioid “plaque like” formations in the cerebral cortex 1 yr postinjection (2). On the other hand, there have been relatively few reports of NF’T formation in mammals other than primates (83). In the present study, we have demonstrated that the CHAT immunoreactive cell number is decreased with age in the ventral CHAT ceil group of the septal complex. According to previous studies (1,47,61,64), the ventral CHAT cell group is the origin

of a cholinergic projection to the dorsorostral hippocampal formation. In contrast, the dorsal and midline cell groups which project to more caudal and temporal regions of the hippocampal formation (1,47) were not affected with age. In animals studies. the effects of aging on cholinergic basal forebrain cell groups is equivocal. There have been reports of no changes (34,43) or decreases in AchE positive neurons in the septal region with age (3,28). In fact, a study by Fischer et al. (24) found that there are decreases in septal AchE size and number, which significantly correlated with the degree of behavioral impairment used in the Morris water maze in female Sprague-Dawlev rats. In contrast, it has been reported that there are no difference-in nucleus basalis AchE positive cells in 24 ma old rats (18). However, AchE is a nonspecific marker for cholinergic neurons, since many noncholinergic neurons can be positively stained by this method. The cholinergic cell loss in the ventral group suggests that the cells have dropped out as a primary process. This is in contrast to the cortical projecting CHAT neurons of the nucleus basalis where we found only cell shrinkage. The shrinkage of cells in the nucleus basalis may be due to secondary effects of postsynaptic cortical cell loss (48,65,67). Of interest is the recent finding that fimbria transsection decreases the number of septal CHAT immunoreactive neurons which can be reversed by treatment with nerve growth factor either given immediately or after a twoweek delay (38). This suggests that there is only an apparent loss of cells as measured by loss of CHAT immunoreactivity. It may be that following nerve growth factor administration CHAT enzyme levels are reexpressed or increased. Therefore, the cholinergic cells may not die but rather are in a quiescent state. fn addition, a recent study has shown that nerve growth factor can ameliorate both cholinergic cell atrophy measured by AchE stain-

TABLE 4 REGIONAL Area

Group

CATHECHOLAMINE

Dopamine

LEVELS DOPAC (ng!g

Caudate

N. ACC.

MFC

AMG

HYPO

HIPP

Young (I) Old Good (II) OId Poor (III) I II III I II III I II III I II III I II III

61.50 z?z383 5767 L 374* 5477 t 234” 5915 + 778 5325 L 576 5764 rt 305 1402 22 128 t: 23 106 -e 19* 563 L 112 509 _’ 55 4.58 t 80 369 rt 75 403 +: 82 411 r 98 56 ir 17 50 + 10 46rt 11

706 t 72 688 2 65 639 _t 50 890 2 97 695 2 113t 816 k 104 362 11 342 14 30% 6 36? 11 4.5, 10 482 8 1022 12 107 -+ 15 1192 29 N.D. ND. N.D.

HVA

NE

366 2 42 348 t 63 391 t- 29 332 + 49 256 t 607 313 k 33 N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D. N.D.

N.T. N.T. N.T. N.T. N.T. N.T. 458 L 27 518 rt 105 485 If 45 639 L 96 5.58 i- 97 598 I 51 N.T.

tissue)

Mean 2 SD N = 7 per group. N. ACC = n. aeeumbens, HYPO = hypothalamus, AMG = amypdala, HIPP = hippocampus, MFC = medial frontal cortex. * p < 0.05 compared to young group (I). t p < 0.05 compared to both young (I) and old poor (III) groups. N.D. = not detectable. N.T. = not tested.

N.T.

N.T. 365 L 63 347 L 44 355 t 21

SPATIAL LEARNING CORRELATES

ing, as well as an associated spatial learning deficit in aged rats (23). Using Nissl stains, negative correlation’s with age and cell number in the NBM of postmortem human brains have been reported (14,46). However, other studies have not been able to confirm this association (9,10,79). Interestingly, a 30% decrease in cell number has been reported in the vertical limb of the diagonal band of Broca (18) in AD. This area overlaps with the rat ventral cell group designated in the present study, where we found a significant decrease in CHAT cell number. In an attempt to develop animal models for AD, lesions of the acetylcholinerich septal and nucleus basalis regions have been analyzed. Neurochemical and electrolytic lesions of the medial septum cause impairment in memory acquisition (11,36,80). Similar results are found with lesions of the more caudal regions of the basal forebrain nuclei including the cortical projecting nucleus basalis nucleus (33,36,76). Specifically, lesions of the septal region as well as the nucleus basalis cause impairment in acquisition learning in the Morris water maze task (76). The evidence that deficits in the cholinergic system of the basal forebrain are involved in spatial learning deficits is based on numerous studies (15,23,70,75,76). In the present study, because both the old good and old poor groups have decreased cholinergic cell number in the ventral cell group, either this specific cholinergic cell loss is not related to the learning deficits or it is permissive for other neurotransmitter systems alterations to produce the deficit. It is also possible that functional alterations of the cholinergic system exist independent of cell number or size. Alterations in the cortical projecting brainstem monoamine systems are among the many neurotransmitter systems that have been implicated in both normal and pathologic aging processes. The most consistent monoamine alteration in both animal and human studies of normal aging are decreased caudate DA and hy~~alamic NE levels (21,22). In the present study, there were effects of age alone on DA levels in specific rat brain regions. Consistent with other studies, caudate DA levels are decreased with age (27,31). This presumably is due to substantia nigra neuronal cell loss with age since the DOPAC/DA and HVA/DA ratios, an indication of DA metabolism, were not altered. On the other hand, in the medial frontal cortex, there was a significant reduction in the DA content only in the old poor learners when compared to the young group. In the nucleus accumbens, there was a significant decrease in DOPAC and HVA in the old good learners but not in the old poor learners. There has been equivocal findings in the S-HT system. No changes or slight decreases in .5-HT ~ncentrations with generalized increases in .5-HIAA in aged animals have been reported (27,30,31). In all brain areas investigated in the present study, there was no effect of age on 5-HT levels. However, in the n. accumbens and hippocampus, there was a significant increase in 5-HIAA levels (Table 3) and 5-HIAA/5HT ratios in both old good and poor learners. In the hypothalamus and the caudate, there was a trend toward increases in both these parameters. This increase in 5-HIAA and as a consequence the 5-HIAA/S-HT ratio reflects either an increased turnover or increased metabolism of 5-HT. Since S-HT is a substrate for MAO A, which in contrast to MAO B is not increased with age, it is less likely that an increased intracellular metabolism of 5-HT is occurring. In addition, because the 5-HT levels are not altered, it is difficult to hypothesize raphe neuronal cell loss with age. In fact, in both humans and rats, the number of serotonergic cells of the raphe do not appear to be decreased with age (personal observations). Therefore, the present results indicate a compensatory increase

497

in 5-HT turnover with age in old good learners which was not found in the MFC of the old poor learners. The monoamine alterations reported here are associated with learning deficits in the old poor learners. The lowered DA levels and the lack of compensatory increase in S-HIAA in the medial frontal cortex seen in the old good learners suggests that this area may be involved in spatial learning deficits. Therefore, the lowered dopamine levels in the old poor group could reflect an altered mesolimbic DA system specifically in the area that projects to the frontal cortices. Previously we have shown decreased MFC DOPAC levels with age in Fischer 344 rats (42). Why the old poor group did not show a similar dopamine deficit in the n. accumbens, also a projection site of the mesolimbic dopamine system, cannot be readily explained at this time. Lesion studies of the medial frontal cortex show that this area is involved with processing of spatial and environmental information (7). The Morris water maze task does involve spatial cues. Therefore, it is possible that alterations of the monoamine levels in this region may affect the ability of the animals to learn. In a similar study, Gallagher et al. (27) correlated CHAT enzyme activity and monoamine levels with the acquisition of spatial learning in the Morris water maze in Long-Evans rats. They found that CHAT enzyme activity in basal forebrain, striatum and frontal cortex was the best predictor of the behavioral deficits. No direct measurements of CHAT immunoreactive cell number in the forebrain cholinergic nuclei were done. They also found that decreases in basal forebrain DA and DOPAC levels were correlated with impaired spatial learning in aged rats. However, they did not find any differences in frontal cortex 5-HIAA or DA levels between the old impaired and old nonimpaired animals. To explain the differences between our study and theirs, it could be that our specific dissection was of the medial portion of the frontal cortex and not the entire frontal cortex. Again, some of these neurochemical findings are consistent with normal and pathologic aging in humans. There are increases in CSF 5-HIAA levels with age but are decreased relative to agematched controls in patients with AD (55). It has been shown that frontal cortical DA levels were significantly decreased in patients with AD (44,85). These data suggest that in normal aging there is a compensatory increase in 5-HT turnover which is not present or is decreased in pathologic aging. Although there appears to be a consistent age-related increase in the 5-HIAA/5HT ratios, this may be due to either a decrease in 5-HT levels or an increase in 5-HIAA levels (27,31). Neuroanatomical studies have shown that the cholinergic basal forebrain nuclei have afferent projections from the serotonergic raphe nuclei, the noradrenergic locus ceruleus and the dopaminergic ventral tegmental area (35,74). It has been demonstrated that the serotonergic and noradrenergic systems are specifically involved in AD based on the findings of cell loss and the presence of neurofibrillary tangles in the raphe and the locus ceruleus, as well as substantial decreases in temporal lobe 5-HT and 5-HIAA levels and S-HT reuptake in AD (12,13,55). In addition, recent work by Perry et al. (57) suggest that there is a specific interaction of the cholinergic and monoaminergic systems in diffuse Lewy Body disease (DLBD). In DLBD, intraneuronal Lewy bodies are found rather than classical neurofibrillary tangles. At present, there is debate in the literature as to whether DLBD represents a variant of AD or whether it is a distinct pathologic entity. In summary, either the cholinergic cell alterations in the basal forebrain nuclei found in the present study are unrelated to the acquisition of spatial learning or both the cholinergic changes along with alterations in the ascending dopaminergic and serotonergic systems are needed to produce a behavioral deficit as measured by the Morris water maze. The clinical lit-

498

LEE ET AI_

erature would support the have had limited success Subsequent to these early icits, there now have been the basal forebrain nuclei basal forebrain

cholinergic

latter hypothesis. Cholinergic drugs in patients with AD (4,6,8,17,69). findings of primary cholinergic defnumerous reports of interactions of and monoamine systems. Both the nuclei and the cortical

projecting

1. Amaral, D. G.; Kurz, J. An analysis of the origins of the cholinergic and noncholinergic septal projections to the hippocampal formation of the rat. J. Comp. Neuro. 240:37-59; 1985. 2. Arendash, G. W.; Millard, W. J.; Dunn, A. J.; Meyer, E. M. Longterm neuropathologicai and neur~hemical effects of nucleus basalis lesions in the rat. Science 238:952-9X; 1987. 3. Biegon, A.; Greenberger, V.; Segal, M. Quantitative histochemisty of brain acetylcholinestrase and learning rate in aged rat. Neurobiol. Aging 7~215-217; 1986. 4. Boyd, W. D.; Graham-White, J.; Blackwood, G.; Glen, 1.; McQueen, J. Clinical effects of choline in Alzheimer senile dementia. Lancet 2:711; 1977. 5. Breitner, J.; Folstein, M.; Murphy, E. Familial aggregation in Alzheimer dementia-I. A model for age-dependent expression of an autosomal dominant gene. J. Psychiat. Res. 20:31-43; 1986. 6. Brinkman, S. D.; Gershon, S. Measurement of cholinergic drug effects on memory in Aizheimer’s disease. Neurobiol. Aging 4:139145; 1983. 7. Brozoski, T. J.; Brown, R. M.; Rosvold, H. E.; Goldman, P. S. Cognitive deficit caused by regional depletion of dopamine in prefrontal cortex of Rhesus monkey. Science 205:929-932; 1979. 8. Christie, J. E.; Shering A.; Ferguson, J.; Glen, A. Physostigmine and arecoline: Effects of intravenous infusion in Alzheimer’s presenile dementia. Br. J. Psychiatr. 138:46-50; 1981. 9. Chui, H. C.; Bondareff, W.; Zarow, C.; Slager, U. Stability of neuronal number of the human nucleus basalis of Meynert with age. Neurobiol. Aging 5:83-88; 1984. IO. Clark, A. W.; Parhad, I. M.; Folstein, S. E.; Whitehouse, P. J.: Hedreen, J. C.; Price, D. L.; Chase, G. A. The nucleus basalis in Huntington’s disease. Neurology 33:1262-1267; 1983. 11. Cormier, S. M. A match-~smatch theory of limbic system function. Physiol. Psychol. 93-36; 1981. 12. Cross, A. J.; Crow, T. J.; Johnson, J. A.; Johnson, J.; Perry, E. K.; Perry, R. H.; Blessed, G.; Tomlinson, B. E. Sudies on neurotransmitter receptor systems in neocortex and hippocampus in senile dementia of the Azheimer type, J. Neurol. Sci. 64:109-111; 1984. 13. Cross, A.; Crow, T.; Ferrier, I.; Johnson, J.; Bloom, S.; Corsellis, J. Serotonin receptor changes in dementia of the Alzheimer type. J. Neurochem. 43:1574-1581; 1983. 14. Decker, M. W. The effects of aging on hippocampal and cortical projections of the forebrain cholinergic system. Brain Res. 434:423438; 1987. 1.5. Dunnett, S. B.; Toniolo, G.; Fine, A.; Ryan, C. N.; Bjorklund, A.; Iversen, S. D. T~nspl~~tion of embryonic ventral forebrain neurons to the neocortex of rats with lesions of nucleus basalis magnocellularis II. Sensorimotor and learning impatiments. Neurosci. 16:787-797; 1985. 16. Duyckaerts, C.; Brion, J. P.; Hauw, J. J.; Flament-Durand, J. Quantitative assessment of the density of neurofibrillary tangles and senile plaques in senile dementia of the Aizheimer type: Comparison of immunocyt~hem~~y with a specific antibody and Bodian method. Acta. Neuropathol. 73:167-170; 1987. 17. Etienne, P.; Gauthier, S.; Johnson, G.; Collier, B.; Mendis, T.; Dastoor, D.; Cole, M.; Muller, H. Clinical effects of choline in Alzheimer’s disease. Lancet 1:508-509; 1978. 18. Etienne, P.; Robitaille, Y.; Wood, P.; Gauthier, S.; Nair, N. P; Quirion, R. Nucleus basalis neuronal loss, neuritic plaques and choline acetyitransferase activity in advanced Alzheimer’s disease. Neuroscience 19:1279-1291; 1986. 19. Fillit, H.; Luine, V. N.; Reisberg, B.; Amador, R.; McEwen, B.; Zabriskie, J. B. Studies of the specificity of antibrain antibodies in

brainstem monoamine systems are affected in AD (25). Since the MFC has been implicated as an area involved with spatial learning tasks and receives projections from both the basal forebrain as well as the brainstem monoamine systems, this area may be of particular interest for future investigation of learning deficits.

20.

21. 22.

23.

24.

25.

26. 27.

28.

29.

30.

31.

32. 33.

34. 35. 36.

37.

alzheimer’s disease. In: Hutton, J. T.; Kenny, A. D., eds. Senile dementia of Alzheimer type. New York: Alan R. Liss, Inc.; 1985:307-318. Fillit, H. M.; Kemeny, E.; Luine, V.; Weksler, M. E.; Zabriskie, J. B. Antivascular antibodies in the sera of patients with senile dementia of the Alzheimer’s type. J. Gerontol. 42:180-1&t; 1987. Finch, C. E. Catecholamine metabolism in brains of aging male mice. Brain Res. 52:261-276;1973. Finch, C. E. Age-related changes of catecholamines: A synopsis of findings in C57/6J mice and other rodent models. Adv. Exp. Med. Biol. 113:15-39; 1978. Fine, A.; Dunnett, S. B.; Bjor~und, A.; Iversen, S. D. Cholinergic ventral forebrain grafts into the neocortex improve passive avoidance memory in a rat model of Alzheimer’s disease. Proc. Natl. Acad. Sci. 82~5227-5230; 1985. Fischer, W., Gage, F. H.; Bjorklund, A. Degenerative changes in forebrain cholinergic nuclei correlate with cognitive impairments in aged rats. Eur. J. Neurosci. 1:34-45; 1989. Francis, P.; Palmer, A.; Sims, N.; Bowen, D.; Davison, A.; Esiri, M.; Neary, D.; Snowden, J.; Wilcock, G. Neurochemical studies of early onset Alzheimer’s disease: Possible influence on treatment. New Engl. J. Med. 313:7-11; 1985. Gage, F. H.; Dunnett, S. B.; Bjorklund, A. Spatial learning and motor deficits in aged rats. Neurobiol. Aging 5:43-48; 1984. Gallagher, M.; Burwell, R, Kodsi, M.; McKinney, M.; Southerland, S.; Vella-Rountree, L.; Lewis, M. Markers for biogenic amines in the aged rat brain: Relationship to decline in spatial learning ability. Neuribio. Aging 11507-514; 1990. Gilad. G. M.: Rabev. J. M.: Tizabi, Y.: Gilad, V. H. Age-dependent loss and co&pens&&y changes of septohippocampal choiinergic neurons in two rat strains differing in longevity and response to stress. Brain Res. 436:311-322; 1987. Glenner, G.; Wong, C. Aizheimer’s disease and Down’s syndrome sharing a unique cerebrovascular amyloid fibril protein. Biochem. Biophys. Res. Commun. 122:1131-1135; 1984. Goldman-Rakic, P. S.; Brown, R. M. Regional changes of monoamiues in cerebral cortex and subcortical structures of aging Rhesus monkeys. Neurosci. 6:177-187; 1981. Gozlai, H.; Daval, G.; Verge, D.; Spampinato, U.; Fattaccini, C. M.; Gallissot. M. C.: Mestikawv, S.; Hamon, M. Aging - - associated changes in serotoninergic aid dopaminergic pre and postsynaptic neurochemical markers in the rat brain. Neurobio. Aging It: 437449; 1990. Ha&in&i, V.: Iliff, M.; Zilhka, E.; DuBoulay, G.; McAllister, V.; Marshal, J.; Russell, R.; Symon, L. Cerebral blood flow in dementia. Arch. Neural. 32:632-637; 1975. Hepler, D. J.; Olton, D. S.; Wenk, G. L.; Coyle, J. T. Lesions in nucleus basalis magnocellularis and medial septal area of rats produce qualitatively similar memory impairments. J. Neurosci S:866873; 1985. Hornberger, J. C.; Buell, S. J.; Flood, D. G.; McNeilI, T. H.; Coteman, P. D. Stability of numbers but not size of mouse forebrain choline@ neurons to 53 mo. Neurobiol. Aging 6:269-275; 1985. Jones, B.; Yang, T. The efferent projections from the reticular formation and the locus coeruleus studied by anterograde and retograde axonal transport in the rat. J. Comp. Nehrol. 24256-92; 1985. Kesner. R. P.: Crutcher, K. A.; Measom, M. 0. Medial septal and nucleus basalis ma~~~l~ularis lesions produce order mem&y deficits in rats which mimic symptomatology of Alzhemier’s disease. Neurobioi. Aging 71287-295; 1986. Kimura, H.; McGeer, P. L.; Peng, J. H. Choline acetyltransferasecontaining neurons in the rat brain. In: Bjorklund, A.; Hokfelt, T.;

SPATIAL

LEARNING

499

CORRELATES

Kuhur, M. J., eds. Handbood of chemical neuroanatomy. Amsterdam: Elsevier Science Publishers; 1984: 51-67. 38. Koliatosos, V. E.; Applegate, M. D.; Kmrsel, B.; Junard, E. 0.; Burton, L. E.; Mobley, W. C.; Hefti, F.; Price, D. L. Recombinant human nerve growth factor prevents retrograde degenertion of axotomized basal forebrain cholinergic neurons in the rat. Exp. Nemo]. 112:161-173; 1991. 39. Lilifors, H. W. On the Kolmogorov-Smirnov test for normality with mean and variance unknown. J. Am. Statistical Assoc. 64:399-402; 1967. 40. Lorens, S. A.; Kindel, G.; Dong, X. W.; Lee, J. M.; Hanin, I. Septal choline acetyltransferase immunoreactive neurons: Dose-dependent effects of AF64A. Brain. Res. Bull. 26:965-971; 1991. 41. Lorens, S.; Hata, N.; Handa, R.; Van De Kar, L.; Guschwan, M.; Goral, J.; Lee, J.; Hamilton, M.; Bethea, C., Clancy, J. Neuroendocrine, endocrine and immunological responses to stress in young and old Fischer 344 male rats. Neurobio. Aging 11:139-150; 1990. 42. Lorens, S.; Guschwan, M.; Hata, N.; Van De Kar, L.; Walenga, J.; Fareed, J. Behavioral, endocrine and neurochemical effects of sulfomucopolysaccharide treatment in aged Fischer 344 male rat. Sem. Thromb. Hemost. 17 (supp):164-173; 1991. 43. Luine, V. N.; Renner, K. J.; Heady, S.; Jones, K. J. Age and sexdependent decreases in CHAT in basal forebrain nuclei. Neurobiol. Aging 7:193-198; 1986. 44. Mann, D.; Lincoln, J.; Yates, P.; Stamp, J.; Toper, J. Changes in monoamine containing neurons of the human CNS in senile dementia. Br. J. Psychiat. 136533-541; 1980. 45. Manual of histologic and special staining techniques: AFIP. New York: McGraw-Hill Co., Inc.; 1960. 46. McGee& P. L.; McGeer, E. G.; Suzuki, J.; Dolman, C. E.; Nagai. T. Aging, Alzheimer’s disease and the cholinergic system of the basal forebrain. Neurology 34:741-745; 1984. 47. McKinney, M.; Coyle, J. T.; Hedreen, J. C. Topographic analysis of the innervation of the rat neocortex and hippocampus by the basal forebrain svstem. J. Coma. Neurol. 217:103-121: 1983. 48. Mesulam, M. M.; Mufson, E. J.; Rogers, J. Ageirelated shrinkage of cortically projecting cholinergic neurons: A selective effect. Ann. Neural. 22:31-36; 1987. 49. Morris, R. Development of a water maze procedure for studying spatial learning in the rat. J. Neurosci. Meth. 11:47-60; 1984. 50. Nagai, T.; McGeer, P.; Peng, J.; McGeer, E.; Dolman, C. Choline acetyitransfer~e imm~nohist~hemistry in brains of Alzheimer’s disease. Neurosci. Letters 36:195-199; 1983. 51. Nakano, I.; Hirano, A. Loss of large neurons of the medial septal nucleus in an autopsy case of Alzheimer’s disease. J. Neuropathol. Exp. Neurol. 41:341; 1982. 52. Nandy, K. In Katzman, R.; Terry, R.; Bick, K., eds. Alzheimer’s disease and related disorders. New York: Raven Press; 1978. 53. Newton, R. D. The idenity of Alzheimer’s disease and senile dementia and their relationship to senility. J. Met. Sci. 94225-249; 1948. 54. Nisson, 0. G.; Strcker, R. E.; Daszuta, A.; Bjorklund, A. Combined cholinergic and serotonergic denervation of the forebrain produce severe deficits in a spatial learning task in the rat. Brain Res. 453:235-246: 1988. 55. Palmer, A.; Francis, P.; Benton, J.; Sims, N.; Mann, D.; Neary, D.; Snowden, J.; Bowen, D. J. Presvnantic serotonernic dvsfunction in patients with Alzheimer’s d&e&e. J. Neurosci. 4&S-15; 1987. 56. Pearson, R. C. A.; Sofroniew, M. V.; Cuello, A. C.; Powell, T. P. S.; Eckenstein, F.; Esiri, M. M.; Wilcock, G. K. Persistence of cholinergic neurons in the basal nucleus in a brain with senile dementia of the Alzheimer’s type demonstrated by immunohistochemical staining for choline acetyltransferase. Brain Res. 289:375-379; 1983. 57. Perry, E.; Marshall, E.; Perry, R. Cholinergic and dopaminergic activities in senile dementia of the Lewv bodv tvoe. Dis. Assoc. Disord. 487-95; 1990. 58. Rogers, B.; Brogan, D.; Mirra, S. The Nucleus basalis of Meynert in neurological disease: A quantitative morphological study. Ann. Neural. 17:163-170; 1985. I

..,‘

59. Rye, D. B.; Wainer, B. H.; Mesulam, M. M.; Mufson, E. J.; Saper, C. B. Cortical projections arising from the basal forebrain: A study of cholinergic and noncholinergic components employing combined retrograde traces and imm~ohist~hemical localization of choline acetyltransferase. Neurosci. 13:627-643; 1984. 60. Rye, D. B.; Saper, C. B.; Lee, H. J.; Wainer, B. H. Pedunculopontine tegmental nucleus of the rat: Cytoarchitecture, cytochemistry, and some extrapyramidal connections of the mesopontine tegmentum. J. Comp. Neurol. 259:483-528; 1987. 61. Saper, C. B. Organization of cerebral cortical afferent systems in the rat. II. Ma~ocellular basal nucleus. J. Comp. Neurol. 222:313-342; 1984. 62. Saper, C. B.; German, D. C.; White, C. L. Neuronal pathology in the nucleus basalis and associated cell groups in senile dementia of the Alzheimer’s type: Possible role in cell lo&. Neurology 35:10891095; 1985. 63. Selkoe, D. J.; Bell, D. S.; Podlisny, M. Et.; Price, D. L.; Cork, L. C. Conservation of brain amyloid proteins in aged mammals and humans with Alzheimer’s disease. Science 235:873-877; 1987. 64. Semba, K.; Reiner, P. B.; McGeer, E. G. Fibiger, H. C. Brainstem afferents to the magnocellular basal forebrain studied by axonal transport, immunohistochemistry, and electrophysiology the rat. J. Comp. Neural. 267:433-453; 198%. 65. Sofroniew, M. V.; Pearson, R. C.; Powell, T. P. The cholinergic nuclei of the basal forebrain of the rat: Normal structure, development and experimentally induced degeneration. Brain Res. 411:310331; 1987. 66. Spencer, D. G.; Hovath, E.; Luiten, P.; Schuurman, T.; Traber, J. Novel approaches in the study of brain acetylcholine function: Neuropha~acology, neuroanatomy and behavior. In: Traber, J.; Gispen, W. H., eds. Senile dementia of the Alzheimer type. Berlin: SpringerVerlag; 1985:325-342. 67. Stephens, P. H.; Tagari, P.; Cuello, A. C. Ethylcholine mustard aziridinium ion lesions of the rat cortex result in retrograde degeneration of basal forebrain cholinergic neurons: Implications for animal models of neurodegenerative disease. Neurochem. Res. 12613-618; 1987. 68. Sternberger, L. A.; Hardy, P. H.; Cuculis, J. J.; Meyer, H. G. The unlabled antibody-enzyme method of immunohistochemistry. Preparation and properties of soluble antigen-antibody compless (PAP) and its use in identification of spirochetes. J. Histochem. Cytochem. 1%:315-333; 1970. 69. Summers., W.:, Maiovski. , I L.:/ Marsh. -----, G.; Tachiki, K.; Kling, A. Oral tetrahydroaminoacridine in long-term treatment of senile dementia, Alzheimer type. New Engl. J. Med. 31.51241-1245; 1986. 70 Sutherland, R. D.; Whishaw, I. Q.; Regehr, J. C. Cholinergic receptor blockage impairs spatial localization by use of distal cues in the rat. J. Comp. Physiol. Psychol. 96:563-573; 1982. 71 Tomlinson, B. E.; Blessed, G.; Roth, M. Obse~ations of the brains of nondemented old people. 3. Neurol. Sci. 7:331-34% 1968. 72. Tomlinson, B. E.; Blessed, G.; Roth, M. Observations on the brain of demented old people. J. Neurol. Sci. 11:205-242; 1970. 73. Troncoso, J.; Price, D. L.; Griffin, J. W. Neurofibrillary axonal pathology in aluminum intoxication. Ann. Neurol. 12:278-283; 1982. 74. Vertes, R. P. Brainstem afferents to the basal forebrain in the rat. Neurosci. 24:907-935; 1988. 7.5. Watanabe, H.; Shimizu, H. Effect of anticholinergic drugs on striatal acetylcholine release and motor activity in freely moving rats studied by brain microdialysis. Jap. J. Pharmacol 51:75-82; 1989. 76. Whishaw, 1. Q.; O’Connor, W. T.; Dunnett, S. B. Disruption of central cholinergic systems in the rat by basal forebrain lesions or atropine: Effects on feeding, sensorimotor behavior, locomotor activity and spatial navigation. Behav. Brain Res. 17:103-115; 1985. 77. Whitehouse, P.; Price, D.; Clark, A.; Coyle, J.; DeLong, M. Alzheimer’s disease and senile dementia: Loss of neurons in basal forebrain. Science 215:1237-1239; 1982. 78. Whitehouse, P.; Price, D.; Clark, A.; Coyle, J.; DeLong, M. Alzheimer’s disease: Evidence for selective loss of cholinergic neurons in the nucleus basalis. Ann Neurol. 10:122-126; 1981.

500

79. Whitehouse, P.; Parhad, I.; Herdeen, J.; Clark, A.; White, C.; Struble, R.; Price, D. Integrity of the nucleus basalis of Meynert in normal aging. Neural. 33:159 (Abst); 1983. 80. Winocur, G., The Hippocampus and cue utilization. Physiol. Psycho]. 8280-288; 1980. 81. Wisniewski, H.; KozIowski, P. Evidence for blood-brain barrier injury in AIzheimer’s disease. Ann. N.Y. Acad. Sci. 3%:119-127; 1982. 82. Wisniewski, H. M.; Johnson, A. B.; Raine, C. S.; Kay, W.; Terry, R. D. Senile plaques and cerebral amyloidosis in aged dogs. Lab. Invest. 23:287-296; 1970.

LEE ET AL.

X3. Wisniewski, H. M.; Sturman, J. A.; Shek, J. W. Aluminum chloride induced neurofibrillary tangles in the developing rabbit: A chronic animal model. Ann. Neural. 8:479-490; 1980. 84. Yano, I.; Uebayashi, Y.; Yoshida, S. Experimental study of degenerative CNS disease in monkeys. Magnesium 4:477-X3: 1985. 85. Zweig, R.; Ross, C.; Hedreen, J.; Steele, C.; Cardillo, J.; Whitehouse, P.; Folstein, M.; Price, D. The neuropathology of aminergic nuclei in Alzheimer’s disease Ann. Neurol 24:233-242; 1988.