- Email: [email protected]
Behavioral impairments after lesions of the nucleus basalis by ibotenic acid and quisqualic acid
84
Brain Research, 555 (1991) 84-90 © 1991 Elsevier Science Publishers B.V. All rights reserved. 0006-8993/91/$03.50 ADONIS 00068993911682~Z
BRES 16828
Behavioral impairments after lesions of the nucleus basalis by ibotenic acid and quisqualic acid Donald J. Connor 1'3, Philip J. Langlais 1'3'4 and Leon J. Thal l'e 1Department of Neurosciences, University of California, San Diego, La Jolla, CA 92093 (U.S.A.), 2Neurology and 3Research Services, Veterans Administration Medical Center, 3350 La Jolla Village Dr., San Diego CA 92161 (U.S.A.) and 4Department of Psychology, San Diego State University, San Diego, CA 92182 (U.S.A.) (Accepted 26 February 1991)
Key words: Nucleus basalis; Ibotenic acid; Quisqualic acid; Water maze; Amino acid; Catecholamine; Indolamine
Ibotenic acid (IBO) or quisqualic acid (QUIS) was infused into the region of the nucleus basalis magnocellularis (NBm) in F344 rats in order to behaviorally and biochemically characterize the effects of these two neurotoxins. QUIS infusion resulted in a slightly higher depletion of choline acetyltransferase (CHAT) activity in both anterior and posterior regions of cortex than did lesions caused by infusion of IBO. Both QUIS- and IBO-treated rats demonstrated significantly longer latencies than controls to find a hidden platform in a Morris water maze task. In addition, QUIS-treated rats performed significantly better than IBO-treated rats in the water maze. Analysis of swim speed and open field behavior did not show significant differences in general motor activity. Passive avoidance retention was unaffected by either neurotoxin. Cortical amino acid levels, [3H]neurotensin binding, dopamine, norepinephrine, and serotonin levels were unaffected by either neurotoxin. The levels of HVA and 5-HIAA in the IBO and QUIS groups were significantly reduced compared to controls, but were not significantly different from each other. Histological examination showed greater damage to non-NBm structures with IBO than with QUIS, including the basolateral nucleus of the amygdala and the reticular formation of the thalamus. The greater behavioral deficit seen after IBO lesions may be due to damage to other areas rather than differences in the extent of depletion of cortical CHAT, amino acids, catecholamines or indolamines.
INTRODUCTION In the rat, the majority of cortical cholinergic innervation originates in a group of large neurons located v e n t r o m e d i a l to the globus pallidus and t e r m e d the nucleus basalis magnocellularis (NBm) 14. E x p e r i m e n t a l studies using pharmacological manipulations and lesion techniques to r e p r o d u c e the cholinergic deficit in animals have d e m o n s t r a t e d behavioral impairments on a variety of tasks including passive avoidance 5"1°, active avoidance 5'17, T-maze 1°'23'27, radial arm maze m'2a, and the water maze 2°,34. Excitatory amino acid analogues such as kainic acid and ibotenic acid ( I B O ) have been used to destroy N B m neurons since they have been shown to eliminate cell bodies in this area and leave fibers of passage intact m4. W o r k done in our l a b o r a t o r y has shown a correlation between the extent of cortical cholinergic denervation after I B O - i n d u c e d lesions of N B m and performance on a spatial m e m o r y task 19. The significance of this correlation has been brought into d o u b t by recent studies demonstrating that a n o t h e r neurotoxin, quisqualic acid ( Q U I S ) , destroys the N B m and produces depletions of cortical
C h A T activity equivalent to I B O but results in less behavioral i m p a i r m e n t in the w a t e r maze 2, split stem T-maze 33 and visual discrimination tasks 26. O n e study 2, however, did find an equivalent deficit on passive avoidance after lesions with both of these toxins. In the previous comparison of I B O and Q U I S groups on the water maze task 2, animals were trained on 4 trials per day over a 10-day period. Previous w o r k from o u r lab has shown that using 2 trials p e r day increases the sensitivity of the task to detect behavioral impairments 2°. Therefore, although a small deficit on the water maze task was seen with Q U I S 2, the degree of deficit may have been underestimated. M o r e o v e r , studies of the effect of I B O vs Q U I S on non-cholinergic cortical neurotransmitter or n e u r o m o d u l a t o r y systems have been limited to levels of cortical neurotensin and ketanserin r e c e p t o r binding 33. In the present study, groups of rats with I B O or Q U I S - i n d u c e d lesions of N B M were tested on a modified water maze training procedure. Cortical C h A T activity, catecholamine, indolamine and amino acid levels as well as neurotensin binding were subsequently measured to characterize the neurochemical effects of each of these toxins.
Correspondence: L.J. Thai, Neurology Service (127), VA Medical Center, 3350 La Jolla Village Drive, San Diego, CA 92161, U.S.A.
85 MATERIALS AND METHODS
Surgery Male Fisher-344 rats (280-300 g; Harlan Laboratory) received bilateral injections of either IBO (0.03 M) or QUIS (0.06 M) into the region of the NBM. Both toxins were dissolved in sterile 0.9% saline and brought to a pH = 7.4-7.6. Animals were placed on a small animal stereotaxic device (Kopf Instruments) in the flat skull position. Injections at two sites on one side (AP +8.1; Lat. 2.7, DV +2.2 and AP +7.2, Lat. 2.9, DV +3.4) 24 were made at a rate of 0.1 /A per min (0.5/zl anterior, 0.6/zl posterior for IBO and 0.5/A for QUIS at both sites). A 5-min diffusion period followed each injection. One week later the contralateral side was treated in the same manner. Control animals were prepared by lowering the injection needle into the cortex but no toxin was infused.
Biochemistry Following behavioral testing, the animals were sacrificed, the brain removed and rapidly cooled in ice-cold saline. The anterior and posterior cortices on both sides were then dissected and placed in cold, 50 mM phosphate buffer. The remainder of the brain was put into a 10% neutral buffered formalin solution for histology. The cortical tissue was sonicated and aliquots were taken for CHAT, catecholamine, indolamine, and amino acid analyses as described below. The remaining homogenate was frozen for neurotensin receptor binding analysis.
Choline acetyltransferase (CHAT) Aliquots of tissue were diluted in buffer containing 0.87 mM EDTA and 0.1% Triton X-100 (pH = 7.0). ChAT activity was measured by the conversion of [14C]acetyI-Co-A to [14C]acetylcholine as described by Fonnum 6.
Behavior Neurochemical analysis (NCA) Water maze Two weeks after the final surgery animals were tested on acquisition of a modified Morris water maze task as previously described 2°. Briefly, the maze consisted of a black circular pool, 152 cm in diameter, filled to a depth of 32 cm with clear water at a temperature of 25-26 °C. The platform was located in the SW quadrant. For each trial, animals were placed into the maze at one of four start points (designated N, S, E or W). The rat was given 90 s to find a platform hidden 3 cm below water level. If the rat was unable to find the platform by the end of the 90-s period, he was guided to it by the experimenter. The rat was allowed to remain on the platform for 10 s before removal. Testing consisted of one block of two consecutive trials, in a semi-random order with the first trial beginning from one of the long start points (N or E) and the second trial beginning from one of the closer start points (S or W). Acquisition was measured over 10 blocks (at 1 block per day). After the 10-block acquisition period, animals were removed from the water maze for 10 days and then tested for retention of the task on 4 consecutive blocks (2 trials/block/day). One day following the last block of retention testing, the platform was removed and animals were tested on a 90-s spatial probe trial. For the purposes of the spatial probe, the maze was divided into 4 equal quadrants with the platform in the third (SW) quadrant and 3 annuli determined by concentric circles drawn from the outer and inner edges of the platform. Behavioral recording was done utilizing a video tracking system (San Diego Instruments) that allowed quantification of latency, total swim distance and percent of distance swum in each quadrant and annulus. Data were analyzed by repeated measures ANOVA for acquisition and by t-test for retention.
Passive avoidance Following testing in the water maze, rats were trained on a single trial passive avoidance task in a Jarvik box 13. On the training day, animals were placed in the lighted chamber for a 10-s habituation period. The guillotine door was then opened and the latency to enter the dark chamber was recorded. The door was immediately shut and a shock was administered (0.5 mA DC, 0.5 s). Ten seconds later, the animal was removed and returned to the home cage. Retention testing was performed 72 h after the training trial and under the same conditions except that no shock was administered. Data analysis was performed using one-way ANOVA.
Open field activity Four to five days following the passive avoidance retention trial, general motor activity was assessed in a n6vel open field apparatus. The open field consisted of a 32-inch x 32-inch enclosure divided into sixteen 8-inch x 8-inch blocks. Horizontal activity was measured by the number of block crossings and ve~ical activity by the number of rearings. Activity was measured over a 5-min period with sub-analysis done on the first 90 s (to match the duration of water maze testing). Data were analyzed by one-way ANOVA.
Monoamines (1) Measurement of the monoamines norepinephrine (NE), dopamine (DA), serotonin (5-HT) and the free, uneonjugated metabolites 3-methoxy-4-hydroxyphenylglycol (MHPG), homovanillic acid (HVA) and 5-hydroxyindoleacetic acid (5-HIAA) was performed on 50-~1 aliquots of tissue extract. Clear supernatant extract was obtained by adding 10/~1 of 2.0 M PCA to 200/~1 of a homogenate and centrifuging (12,000 g, 4 °C) for 10 rain. A gradient high performance liquid chromatographic (HPLC) analyzer, equipped with 16 electrode eoulometric detectors and a refrigerated autosampler was used for the analysis (CEAS Model 55-0650, ESA, Inc. Bedford, MA) 21. This novel HPLC system provides measurement of both retention time (RT) and ratio of peak response at the dominant sensor to the preceding and following sensors. Sample peaks were compared to standard peaks for both RT and sensor ratio measures for compound identification. Quantification was achieved by comparison of the height of peaks from the sample to those of the standards on the dominant sensor. (2) Mobile phases were prepared from the following Stock solutions: A: 4 liters of 0.1 M phosphate buffer (monobasic), containing 55 mg SDS, 100 /~1 tetrahydrofuran, adjusted with ortho-phosphorie acid to pH 3.2. B: 4 liters of 0.1 M phosphate buffer (monobasic), containing 220 mg SDS, 2000 ml absolute methanol (Optima, Fisher), and adjusted as above to pH 3.45. Working mobile phase A contained 950 ml of Stock A and 50 ml of Stock B. Working mobile phase B was identical to Stock B. The first linear gradient began with 100% A and ended with 55% A:45% B at 47 min. These conditions were held for 1 min before beginning the second gradient which ended with 100% B at 57 min. Following an additional 3 min at 100% B, the analysis was terminated and abruptly switched to the initial conditions of 100% A. Samples were injected at 12.2 rain after initiation of the first gradient. A single 15-cm, 3-/~ Cls reverse-phase column (Pt#HR-150T, ESA, Inc.) was used for separation. All chromatographic runs were made at a constant temperature of 35 °C and a 1.0 ml/min flow rate. Electrochemical readings were made at 60 mV increments, with sensor 1 equal to 0 mV and sensor 16 equal to 900 mV.
Amino acid analysis Amino acid levels were determined by HPLC with electrochemical detection as described previously 16. Aliquots of the perchloric acid extract described above were pretreated with orthophthaldialdehyde (OPA) and then injected onto an 8-g Bondpak Cls reverse-phase column (Waters Corp.). The mobile phase consisted of a 0.1-M sodium phosphate buffer, pH = 5.25 containing 34% methanol and 1.5% acetonitrile (v/v), and defivered at a flow rate of 1.8 ml/min. A coulometric electrochemical detector (Coulchem model 5100A, ESA Inc.) equipped with a conditioning cell (+80 mV) and a dual electrode analytical cell (Model 5011) set at +350 mV and +800 mV respectively was used to determine amino acid
86 100 90 80 ~-~
~
O
N
'
F
R
~ ~ O L
IBO OUlS
70 so
50
2o
0 Z UJ
~
t~ Z
8
90 80 70 60 50 40 30
[] []
EllOCK9*1:3 BLOCK1I+1~
20
10-
10 0
TRIAL BLOCKS
Fig. 1. Latency (in s) to find and climb onto the platform in the Morris water maze test during acquisition training. Each point represents the mean of 2 trials/block for needle drop control rats (CONTROL; n = 10), rats with ibotenic acid lesions of NBm (IBO; n = 10) and for rats with quisqualic acid lesions of NBm (QUIS; n = 10). All groups are significantly different from each other.
concentration in 10/~1 of extract. Final concentration of amino acid is expressed as nanomoles per mg of tissue protein 18. Neurotensin receptor binding The number of neurotensin receptors in anterior and posterior cortex was determined by the binding of [3H]neurotensin ([3H]NT) to tissue homogenates 3°. Tissue homogenate was thawed and washed 3x in 50 mM Tris buffer (pH = 7.4) then resuspended (approx. 25 mg/ml). The washed tissues (100 ml) were added to tubes containing 800/al of reaction buffer (50 mM "Iris, 0.1% BSA, 0.2 mM Bacitracin and 1 mM EDTA) and 100 ~1 of [3H]NT (124 nM, 80.7 Ci/mmol) and incubated at 25 °C for 45 min. The reaction solution was then applied to filters presoaked in polyethylenimine (0.1%) on a Brandel cell harvester. Non-specific binding was determined by binding of [3H]NT in the presence of 10 mM cold neurotensin. Histology Three days after being placed into the formalin solution, the subcortical block was transferred to sucrose (10%). Frozen sections, 40-/~m thick, were cut from approximately 6.5 to 9.5 mm anterior to the intraaural line24 on a sliding microtome. To assess the extent of denervation, acetylcholinesterase (ACHE) staining was done according to a modification9 of the Koelle method. Cresyl violet staining was used to assess non-specific neuronal damage. Statistics The water maze data were analyzed by repeated measures ANOVA and subsequent Newman-Keuis post-hoe analysis of specific group differences. All other behavioral data and biochemical assays were analyzed by one-way ANOVA and post-hoc comparisons between groups were accomplished with NewmanKeuls test. RESULTS Water maze The I B O animals had the longest latencies for finding the platform, the Q U I S were intermediate and the controls performed the best across the 10 water maze acquisition blocks (Fig. 1). There were significant overall group and block effects on latency to find the platform (repeated measures A N O V A , P < 0.001). The interaction effect (group x block) was not significant (P > 0.1).
CONTROL
QUIS
IBO
GROUP
Fig. 2. Latency (in s) to find and climb onto the platform in the Morris water maze test before and after a 10-day retention interval. The mean latency on blocks 9 + 10 (total 4 trials) prior to the retention period are compared to mean latency on blocks 11 + 12 (total 4 trials) obtained after the retention interval for needle drop control rats (CONTROL; n = 10), rats with ibotenic acid lesions of NBm (IBO: n = 10) and for rats with quisqualic acid lesions of NBm (QUIS; n = 10). There is no increase in latency to find the platform for any of the groups after the retention interval. Bars represent S.E.M.
All 3 groups (collapsed across blocks) were significantly different from each other (NK P < 0.025). This pattern of longest latencies in the I B O group, intermediate latencies in the Q U I S group and shortest time to reach the platform for the control animals was also observed on the retention trials. The 10-day retention interval produced no change in latency to find the platform for any of the 3 groups, t-Tests on the means of the last two acquisition blocks (9 + 10) vs the first two retention blocks (11 + 12) were non-significant for all groups ( P > 0.4) (Fig. 2). A more sensitive m e a s u r e m e n t of retention may be achieved by comparing the first trial on the last acquisition day (block 10) with the first trial on the first retention day (block 11). This comparison also failed to show any effect of the retention interval on latency or distance measures. There was no difference in swim speed between the groups during water maze testing. The spatial probe data over the 90-s interval showed a significant group effect ( A N O V A P < 0.001, Fig. 3A). Controls swam significantly greater distances in a n n u l u s 2 (the platform annulus), than the Q U I S group, which swam more than the I B O animals in this annulus (NK P < 0.025). There was a tendency for the controls to swim more distance in the platform quadrant than either of the lesion groups, but this difference was not statistically significant (Fig. 3B). Passive avoidance All groups demonstrated short but comparable latencies during training. All groups demonstrated comparably long latencies during retention testing 72 h later (Fig. 4).
87 TABLE I
(B)
(A) 30
50'
Choline acetyltransferase (CHAT) activity
T
Data are expressed as nanomoles of acetylcholine formed per hour per milligram protein (mean + S.E.M.) and as percent decrease in ChAT activity relative to controls. All groups were significantly different from each other in both areas (NK P < 0.01). CON, needle drop control rats; QUIS, rats with bilateral quisqualic acid lesions of NBm; IBO, rats with bilateral ibotenic lesions of NBm.
4O IOm
Z
30
,,,~_
O,',, n-
10 w Q.
0
Cortex 0
CONTROL QUIS IBO GROUP
""%"
CONTROL QUIS
IBO
GROUP
Fig. 3. Distribution of distance swum during the 90-s spatial probe trial of the water maze test conducted after the final retention trial. Data represent the mean distance traveled in annulus 2 (A) and quadrant 3 (B) as a percent of the total distance traveled. All groups are significantly different from each other in distance swum in annulus 2, the platform location (Newman-Keuls post hoe-test, P < 0.05).
CON QUIS IBO
Anterior (n = 10)
Posterior (n = 10)
ChAT
ChAT
(% Depletion)
50.5 _ 0.88 18.5+ 0.41 (63%) 22.6_+0.69 (55%)
(% Depletion)
47.0 + 1.17 21.3 + 0.57 (54%) 3 0 . 2 _ + 0 . 7 9(36%)
Monoamines Open field activity No differences were found between groups on horizontal or vertical activity measures for either the first 90 s of the trial or the total time (5 min) (Fig. 5), although there was a non-significant tendency for the IBO group to have a greater n u m b e r of horizontal crossings over the 5-min interval (P < 0.08).
Choline acetyltransferase (CHAT) Both I B O and Q U I S caused significant depletions of C h A T activity in the anterior and posterior cortical regions ( A N O V A , P < 0.001), with Q U I S causing significantly more depletion than I B O in both regions (Table I; N K P < 0.01).
The concentrations of NE, D A and 5-HT in the anterior and posterior regions of cortex of the two lesion groups were not different from controls. However, small but significant differences were found for 3 of the metabolites in anterior cortex (Table II). In this region, a significant decrease (NK, P < 0.05) from controls was observed in H V A and 5 - H I A A levels ( > 3 0 % and 15% respectively) of the I B O and Q U I S groups. A significant decrease in free M H P G levels relative to controls (26%) was seen only in the I B O group. In the posterior contex, the only significant difference observed was a lower H V A level in the I B O group (23%; data not shown). No significant differences b e t w e e n I B O and Q U I S groups were found for any of the catecholamines, indolamines, or their metabolites in either anterior or posterior cortex.
(A)
[ ] CONTROL >0 Z IM I'<[
[ ] QUIS
500
[] mo
400
T ~
300
O CC ~ Z ~ I,-
200
I,U F-
100
(n)
7 0"
T
6 0-
|
350_z
40- ~ 30"
~/
:~ 20-
d.
100
0-90
~r~ CONTROL 1 4 OUlS [] IBO ._ 1 2
1
0-300
~
INTERVAL (seconds) TRAINING
RETENTION
Fig. 4. Latency (in s) to enter the dark chamber of a passive avoidance box. The mean latency on a single training trial and a retention trial conducted 72 h after the training trial are shown for needle drop control rats (CONTROL; n = 10), rats with ibotenic acid lesions of NBm (IBO; n = 10) and for rats with quisqualic acid lesions of NBm (QUIS; n = 10). No differences were found between the groups on either the training or the retention day. Bars represent S.E.M.
• CONTROL [] OUlS IBO
g,o i ~)
8 6 4 2 0
0-90
INTERVAL
0-300 (seconds)
Fig. 5. Distribution of the number of movements in an open field activity test. Data are the mean number of block crossings in the horizontal (A) and the number of rearings in vertical (B) planes during the first 90 s and the entire 0- to 300-s interval for needle drop control rats (CONTROL; n = 10), rats with ibotenic acid lesions of NBm (IBO; n = 10) and for rats with quisqualic acid lesions of NBm (QUIS; n = 10). There was no significant difference between groups for either the 0- to 90- or 0- to 300-s intervals for horizontal or vertical movements. Bars represent S.E.M.
88 'FABLE II
"FABLE IV
Catecholamine and indolamine concentration in anterior cortex
[~H] Neurotensin binding ( B ,,,,~) in cortex
Data are expressed as n a n o g r a m s of c o m p o u n d per milligram protein (mean + S.E.M.). *P < 0.05 from controls, ( N e w m a n - K e u l s post hoc test). C O N , needle drop control rats; Q U I S , rats with bilateral quisqualic acid lesions of N B m ; IBO, rats with bilateral ibotenic lesions of NBm.
Data are expressed as femtomoles per milligram protein ( m e a n +_ S.E.M.). C O N , needle drop control rats; Q U I S , rats with bilateral quisqualic acid lesions of N B m ; IBO, rats with bilateral ibotenic lesions of N B m Anterior (n = 10)
Posterior (n = 10)
81.31 + 5.84 82.50 + 4.45 82.22 _+ 6.73
103.05 + 10.83 101.93 _ 7.26 103.28 -+ 6.67
Anterior cortex
NE MHPG DA HVA 5-HT 5-HIAA
C O N (n = 9)
QUIS (n = 10)
I B O (n = 10)
0.991 0.046 0.078 0.054 1.993 0.947
0.945 0.040 0.077 0.037 1.921 0.797
0.921 0.034 0.075 0.032 1.900 0.771
+ + + + + +
0.044 0.003 0.010 0.004 0.097 0.055
+ 0.020 __+0.003 + 0.006 + 0.002* + 0.064 + 0.033*
+ + + + + +
0.060 0.002* 0.010 0.002* 0.097 0.033*
Amino acids There were no significant differences between groups for any of the amino acids measured in either anterior or posterior cortex (Table III, anterior cortical data; posterior cortical data not shown).
CON QUIS IBO
Both neurotoxins resulted in damage to the medial aspect of the reticular nucleus of the thalamus without apparent damage to adjacent thalamic nuclei. The basolateral nucleus of the amygdala was also affected by both toxins, but to a much greater extent by IBO than by QUIS. The hippocampus, striatum, nucleus of the lateral optic tract and the cortical-amygdala transition zone were not affected by either neurotoxin. DISCUSSION
Neurotensin binding No difference was found between groups for the Bmax of [3H]NT binding in the anterior or posterior cortex (Table IV).
Histology Histological evaluation of the tissue block containing the NBm by Cresyl violet and AChE staining showed slight cell loss and macrophage accumulation in the dorsal globus pallidus in the IBO group but almost no cell loss in the QUIS group. The magnocellular cholinergic neurons in the substantia innominata and ventral globus pallidus were affected by both toxins, but the degree of cell loss and gliosis was much less in the QUIS group.
T A B L E III A m i n o acid concentration in anterior cortex Data are expressed as n a n o m o l e s of amino acid per milligram protein
(mean + S.E.M.). CON, needle drop control rats; QUIS, rats with bilateral quisqualic acid lesions of NBm; IBO, rats with bilateral ibotenic lesions of NBm.
Aspartate Serine Glutamine Glutamate Glycine Taurine Alanine Ethanolamine GABA
C O N (n = 8)
QUIS (n = 10)
I B O (n = 9)
28.82 7.92 34.69 109.63 6.41 48.44 5.56 2.84 13.07
38.71 8.77 37.88 117.39 6.79 52.81 5.86 3.40 13.96
33.62 +_ 2.23 8.81 + 0.58 36.55 + 3.09 113.66 _+ 8.02 7.40 + 0.68 54.94 + 3.66 5.88 -+ 0.43 2.90 + 0.14 15.44 + 1.89
+ 2.05 -+ 0.46 -+ 1.72 + 7.08 +- 0.41 _+ 2.59 + 0.36 + 0.16 + 0.93
+- 4.90 + 0.543 + 2.52 -+ 7.72 -+ 0.35 + 3.30 + 0.20 +- 0.25 + 0.95
In agreement with our previous studies t9'2°, this study demonstrated that bilateral I130 lesions of the NBm produced behavioral impairment on the acquisition of a spatial memory task without causing a deficit in the retention of this task. The acquisition deficit was significantly less severe in the animals treated with QUIS. The behavioral difference cannot be attributed to motor disturbances since general motor activity, water maze speed and passive avoidance retention were not significantly affected by NBm lesion with either neurotoxin. Despite the greater behavioral impairment of the IBO group on the water maze task, ChAT depletion was greater in both anterior and posterior regions of cortex of the QUIS group. These data, in conjunction with the behavioral data, agree with previous s t u d i e s 2'26"33 and demonstrate that the extent of cortical ChAT depletion does not necessarily correlate with the degree of behavioral impairment. In the water maze task, the annulus measure of the spatial probe showed a good correlation with the latency to find the platform. The distance swum in quadrant 3 (platform quadrant; Fig. 3B) showed a tendency for both lesion groups to swim less in this quadrant than controls. Observation of the animals during the task revealed that controls had a greater tendency to circle in the area of the platform while QUIS and IBO animals tended to push off from one wall to another for pool crossings. This was supported by the annulus data which showed that the group that swam the most in annulus 2 (platform annulus) also had the shortest latency and the group that
89 swam the least in annulus 2 had the longest latency (Fig. 1 vs Fig. 3A). The possibility that IBO and QUIS could differ with respect to the cholinergic neuronal subpopulation affected in the NBm is supported by evidence that IBO and QUIS bind to different subclasses of glutamate receptors with IBO preferring the N M D A subtype and QUIS utilizing the quisqualate subtype 3'4. However, another study measuring second messenger effects indicates that the two toxins may act at the same receptor with different affinities 29. If the behavioral impairment is based purely on a NBm cholinergic lesion, then by subtracting the depletion caused by QUIS in this study (54--63%) from the total cortical innervation by NBm neurons (70%) 14, it can be concluded that less than 15% of the NBMcortical cholinergic input is left to explain the difference between IBO and QUIS. That is, less than 15% of the cortical projecting cholinergic neurons in the NBm must be responsible for the behavioral differences between the two neurotoxins if this hypothesis is true. Although interactions between the noradrenergic, dopaminergic, serotonergic and cholinergic systems have been previously demonstrated in a variety of models 7' 11,12.25.28,32, changes in cortical levels of these substances do not explain the more severe behavioral impairments produced by IBO. Levels of each of the neurotransmitters were unchanged between all groups. Significant differences from control levels were noted for some of the monoamine metabolites, but these metabolites were not different between the two lesion groups. Similarly, cortical amino acid levels were not different in the two populations. However, it should be noted that the amino acid levels in extracts of cortex reflect the metabolic as well as the neurotransmitter pools of these compounds. Measures of in vivo release of the monoamines, amino acids and acetylcholine using microdialysis techniques are needed to more accurately assess the utilization and turnover of these transmitters following IBO and QUIS lesions. It is also possible that IBO may have a greater tendency to diffuse to other structures such as the globus pallidus and amygdala. This difference was suggested by studies in which both neurotoxins were injected into the lateral hypothalamus and different diffusion patterns were seen 8. Other s t u d i e s 2'26 have found that in some cases after IBO infusion to the NBm, surrounding structures showed cell loss and gliosis. Our results showed slight damage to dorsal globus pallidus with more REFERENCES 1 Coyle, J.T., Molliver, M.E. and Kuhar M.J., In situ injection of kainic acid: a new method for selectively lesioning neuronal cell bodies while sparing axons of passage, J. Comp. Neurol., 180 (1978) 301-324.
extensive damage to the thalamic reticular nucleus and the basolateral nucleus of the amygdala with IBO as compared to QUIS. Contrary to reports from other labs 2"5A°, this study, as well as our previous studies 15'19, failed to detect any differences between IBO and controls or between QUIS and control animals in passive avoidance training or in retention at 72 h after shock. A ceiling effect is possible but our titration studies indicate that controls are just at optimal retention (unpublished observations) and therefore if IBO or QUIS produces an impairment of passive avoidance it would be very subtle. It is possible that studies demonstrating a passive avoidance deficit involve a more extensive lesion and damage to structures other than the NBm as seen in previous work from our laboratory 31. Other groups have also demonstrated a passive avoidance impairment with lesions of the dorsolateral globus pallidus 5. A final possibility is that since several groups chose a 24-h retention interval, they may be testing their groups before consolidation has been completely established. In conclusion, although a wide variety of cortical neurochemical systems were studied, none were able to explain the behavioral difference between the IBO and QUIS groups on water maze acquisition. Cortical ChAT levels showed a large decrease after both IBO and QUIS lesion of the NBm (Table I), as did HVA (Table II), but neither corresponded with the order of behavioral impairment caused by the two toxins. Further, no change was seen in cortical levels of DA, NE, 5-HT, amino acids or neurotensin binding. At present, the reason for the milder behavioral deficit when QUIS is used as the neurotoxin is unknown. It is possible that IBO destroyed other unknown projections to the cortex that were not examined. Alternately, important NBm pathways projecting to non-cortical structures may have been damaged by the IBO but not the QUIS lesion. Observation of damage to adjacent areas (amygdala, reticular nucleus of the thalamus) after IBO infusion may also have contributed to the behavioral difference.
Acknowledgements. The authors gratefully acknowledge the assistance of Andrew Chen and Mark Wardlow in the neurochemical assays and Barbara Reader in the preparation of this manuscript. This work was supported by funds from the VA Merit Program (P.J.L. and L.J.T.), an NSF grant to L.J.T. and NIH/DRR ($10 RRO 4754-01 to P.J.L.).
2 Dunnett, S.B., Whishaw, I.Q., Jones, G.H. and Bunch, S.T., Behavioral, biochemical and histochemical effects of different neurotoxic amino acids injected into nucleus basalis magnocellularis of rats, Neuroscience, 20 (1987) 653-669. 3 Fagg, G.E. and Matus, A., Selective association of N-methyl aspartate and quisqualate types of L-glutamate receptor with
90
4 5
6 7 8
9 10 11
12
13 14 15
16
17
18 19
brain postsynaptic densities, Proc. Natl. Acad. Sci. U.S.A., 81 (1984) 6876-6880. Fagg, G.E., Foster, A.C. and Ganong, A.H., Excitatory amino acid synaptic mechanisms and neurological function, Trends Pharmacol. Sci., 7 (1986) 357-363. Flicker, C., Dean, R.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. Fonnum, E, Radiochemical micro assays for the determination of choline acetyltransferase and acetylcholinesterase activities, Biochem. J., 115 (1969) 465-472. Harrell, L.E. and Parsons, D.S., The role of cortical sympathetic ingrowth in the behavioral effects of nucleus basalis magnocellularis lesions, Brain Research, 474 (1988) 353-358. Hastings, M.N., Winn, P. and Dunnett, S.B., Neurotoxic amino acid lesions of the lateral hypothalamus: a parametric comparison of the effects of ibotenate, N-methyi-o.L-aspartate and quisqualate in the rat, Brain Research, 360 (1985) 248-256. Hedreen, J.C., Bacon, J.C. and Price, D.L., A modified method to visualize acetylcholinesterase-containingaxons, J. Histochem, Cytochem., 33 (1985) 134-140. Hepler, D.J., Wenk, G.L., Cribbs, B.L., Olton, D.S. and Coyle, J.T., Memory impairments following basal forebrain lesions, Brain Research, 346 (1985) 8-14. Hortnagl, H., Potter, P.E. and Hanin, I., Effect of cholinergic deficit induced by ethylcholine aziridinium (AF64A) on noradrenergic and dopaminergic parameters in rat brain, Brain Research, 421 (1987) 75-84. Hortnagl, H., Potter, P.E., Kindel, G. and Hanin, I., Noradrenaline depletion protects cholinergic neurons in rat hippocampus against AF64A-induced damage, J. Neurosci. Methods, 27 (1989) 103-108. Jarvik, M.E. and Kopp, R., An improved one-trial passive avoidance learning situation, Psychol. Rep., 21 (1967) 221-224. Johnston, M.V., McKinney, M. and Coyle, J.T., Neocortical cholinergic innervation: a description of extrinsic and intrinsic components in the rat, Exp. Brain Res., 43 (1981) 159-172. Kwo-On-Yuen, P.E, Mandel, R., Chen, A.D. and Thai, L.J., Tetrahydroaminoacridine improves the spatial acquisition deficit produced by nucleus basalis lesions in rats, Exp. Neurol., 108 (1990) 221-228. Langlais, P.J., Mair, R.G., Anderson, C.D. and McEntee, W.J., Long-lasting changes in regional brain amino acids and monoamines in recovered pyrithiamine treated rats, Neurochem. Res., 13 (1988) 1199-1206. Lo Conte, G., Bartolini, L., Casamenti, E, Marconcini-Pepeu, I. and Pepeu, G., Lesions of cholinergic forebrain nuclei: changes in avoidance behavior and scopolamine actions, Pharmacol. Biochem. Behav., 17 (1982) 933-937. Lowry, O.H., Rosenberg, N.J., Farr, A.L. and Randall, R.J., Protein measurement with the Folin phenol reagent, J. Biol. Chem., 193 (1951) 265-275. Mandel, R.J., Gage, EH. and Thai, L.J., Spatial learning in rats: correlation with cortical choline acetyltransferase and improvement with NGF following NBM damage, Exp. NeuroL, 104 (1989) 208-217.
20 Mandel, R.J., Gage, EH. and Thai, L.J., Enhanced detection of nucleus basalis magnocellularis lesion-induced spatial learning deficit in rats by modification of training regimen, Behav. Brain Res., 31 (1989) 221-229. 21 Matson, W.R., Gamache, P.G., Beal, M.E and Bird, E.D., EC array sensor concepts and data, Life Sci., 41 (1987) 905-908. 22 Murray, C.L. and Fibiger, H.C., Learning and memory deficits after lesions of the nucleus basalis magnocellularis: reversal by physostigmine, Neuroscience, 14 (1985) 1025-1032. 23 Nakamura, S., Nakagawa, Y., Kawai, M., Tohyama, M. and Ishihara, T., AF64A (ethylcholine aziridinum ion)-induced basal forebrain lesion impairs maze performance, Behav. Brain Res., 29 (1988) 119-126. 24 Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic Press, New York, 1986. 25 Robbins, T.W. and Everitt, B.J., Comparative functions of the central noradrenergic, dopaminergic and cholinergic systems, Neuropharmacology, 26 (1987) 893-901. 26 Robbins, T.W., Everitt, B.J., Ryan, C.N., Marston, H.M., Jones, G.H. and Page, K.J., Comparative effects of quisqualic and ibotenic acid-induced lesions of the substantia innominata and globus pallidus on the acquisition of a conditional visual discrimination: differential effects on cholinergic mechanisms, Neuroscience, 28 (1989) 337-352. 27 Salamone, J.D., Beart, P.M., Aipert, J.E. and Iversen, S.D., Impairment in T-maze reinforced alternation performance following nucleus basalis magnocellularis lesions in rats, Behav. Brain Res., 13 (1984) 63-70. 28 Sara, S.J., Noradrenergic-cholinergic interaction: its possible role in memory dysfunction associated with senile dementia, Arch. Gerontol. Geriatr., Suppl. 1 (1989)99-108. 29 Schoepp, D.D. and Johnson, B.G., Excitatory amino acid agonist-antagonist interactions at 2-amino-4-phosphonobutyric acid-sensitive quisqualate receptors coupled to phosphoinositide hydrolysis in slices of rat hippocampus, J. Neurochem., 50 (1988) 1605-1613. 30 Schotte, A. and Laduron, P.M., Different postnatal ontogeny of two [3H]neurotensin binding sites in rat brain, Brain Research, 408 (1987) 326-328. 31 Thai, L.J., Dokla, C.P.J. and Armstrong, D.M., Nucleus basalis magnocellularis lesions: lack of biochemical and immunocytochemical recovery and effect of cholinesterase inhibitors on passive avoidance, Behav. Neurosci., 102 (1988) 852-860. 32 Wenk, G., Hughey, D., Boundy, V., Kim, A., Walker, L. and Olton, D., Neurotransmitters and memory: role of cholinergic, serotonergic and noradrenergic systems, Behav. Neurosci., 101 (1987) 325-332. 33 Wenk, G.L., Markowska, A.L. and Olton, D.S., Basal forebrain lesions and memory: alterations in neurotensin, not acetylcholine, may cause amnesia, Behav. Neurosci., 103 (1989) 765-769. 34 Whishaw, I.Q., O'Connor, W.T. and 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 (1985) 103-115.
Brain Research, 555 (1991) 84-90 © 1991 Elsevier Science Publishers B.V. All rights reserved. 0006-8993/91/$03.50 ADONIS 00068993911682~Z
BRES 16828
Behavioral impairments after lesions of the nucleus basalis by ibotenic acid and quisqualic acid Donald J. Connor 1'3, Philip J. Langlais 1'3'4 and Leon J. Thal l'e 1Department of Neurosciences, University of California, San Diego, La Jolla, CA 92093 (U.S.A.), 2Neurology and 3Research Services, Veterans Administration Medical Center, 3350 La Jolla Village Dr., San Diego CA 92161 (U.S.A.) and 4Department of Psychology, San Diego State University, San Diego, CA 92182 (U.S.A.) (Accepted 26 February 1991)
Key words: Nucleus basalis; Ibotenic acid; Quisqualic acid; Water maze; Amino acid; Catecholamine; Indolamine
Ibotenic acid (IBO) or quisqualic acid (QUIS) was infused into the region of the nucleus basalis magnocellularis (NBm) in F344 rats in order to behaviorally and biochemically characterize the effects of these two neurotoxins. QUIS infusion resulted in a slightly higher depletion of choline acetyltransferase (CHAT) activity in both anterior and posterior regions of cortex than did lesions caused by infusion of IBO. Both QUIS- and IBO-treated rats demonstrated significantly longer latencies than controls to find a hidden platform in a Morris water maze task. In addition, QUIS-treated rats performed significantly better than IBO-treated rats in the water maze. Analysis of swim speed and open field behavior did not show significant differences in general motor activity. Passive avoidance retention was unaffected by either neurotoxin. Cortical amino acid levels, [3H]neurotensin binding, dopamine, norepinephrine, and serotonin levels were unaffected by either neurotoxin. The levels of HVA and 5-HIAA in the IBO and QUIS groups were significantly reduced compared to controls, but were not significantly different from each other. Histological examination showed greater damage to non-NBm structures with IBO than with QUIS, including the basolateral nucleus of the amygdala and the reticular formation of the thalamus. The greater behavioral deficit seen after IBO lesions may be due to damage to other areas rather than differences in the extent of depletion of cortical CHAT, amino acids, catecholamines or indolamines.
INTRODUCTION In the rat, the majority of cortical cholinergic innervation originates in a group of large neurons located v e n t r o m e d i a l to the globus pallidus and t e r m e d the nucleus basalis magnocellularis (NBm) 14. E x p e r i m e n t a l studies using pharmacological manipulations and lesion techniques to r e p r o d u c e the cholinergic deficit in animals have d e m o n s t r a t e d behavioral impairments on a variety of tasks including passive avoidance 5"1°, active avoidance 5'17, T-maze 1°'23'27, radial arm maze m'2a, and the water maze 2°,34. Excitatory amino acid analogues such as kainic acid and ibotenic acid ( I B O ) have been used to destroy N B m neurons since they have been shown to eliminate cell bodies in this area and leave fibers of passage intact m4. W o r k done in our l a b o r a t o r y has shown a correlation between the extent of cortical cholinergic denervation after I B O - i n d u c e d lesions of N B m and performance on a spatial m e m o r y task 19. The significance of this correlation has been brought into d o u b t by recent studies demonstrating that a n o t h e r neurotoxin, quisqualic acid ( Q U I S ) , destroys the N B m and produces depletions of cortical
C h A T activity equivalent to I B O but results in less behavioral i m p a i r m e n t in the w a t e r maze 2, split stem T-maze 33 and visual discrimination tasks 26. O n e study 2, however, did find an equivalent deficit on passive avoidance after lesions with both of these toxins. In the previous comparison of I B O and Q U I S groups on the water maze task 2, animals were trained on 4 trials per day over a 10-day period. Previous w o r k from o u r lab has shown that using 2 trials p e r day increases the sensitivity of the task to detect behavioral impairments 2°. Therefore, although a small deficit on the water maze task was seen with Q U I S 2, the degree of deficit may have been underestimated. M o r e o v e r , studies of the effect of I B O vs Q U I S on non-cholinergic cortical neurotransmitter or n e u r o m o d u l a t o r y systems have been limited to levels of cortical neurotensin and ketanserin r e c e p t o r binding 33. In the present study, groups of rats with I B O or Q U I S - i n d u c e d lesions of N B M were tested on a modified water maze training procedure. Cortical C h A T activity, catecholamine, indolamine and amino acid levels as well as neurotensin binding were subsequently measured to characterize the neurochemical effects of each of these toxins.
Correspondence: L.J. Thai, Neurology Service (127), VA Medical Center, 3350 La Jolla Village Drive, San Diego, CA 92161, U.S.A.
85 MATERIALS AND METHODS
Surgery Male Fisher-344 rats (280-300 g; Harlan Laboratory) received bilateral injections of either IBO (0.03 M) or QUIS (0.06 M) into the region of the NBM. Both toxins were dissolved in sterile 0.9% saline and brought to a pH = 7.4-7.6. Animals were placed on a small animal stereotaxic device (Kopf Instruments) in the flat skull position. Injections at two sites on one side (AP +8.1; Lat. 2.7, DV +2.2 and AP +7.2, Lat. 2.9, DV +3.4) 24 were made at a rate of 0.1 /A per min (0.5/zl anterior, 0.6/zl posterior for IBO and 0.5/A for QUIS at both sites). A 5-min diffusion period followed each injection. One week later the contralateral side was treated in the same manner. Control animals were prepared by lowering the injection needle into the cortex but no toxin was infused.
Biochemistry Following behavioral testing, the animals were sacrificed, the brain removed and rapidly cooled in ice-cold saline. The anterior and posterior cortices on both sides were then dissected and placed in cold, 50 mM phosphate buffer. The remainder of the brain was put into a 10% neutral buffered formalin solution for histology. The cortical tissue was sonicated and aliquots were taken for CHAT, catecholamine, indolamine, and amino acid analyses as described below. The remaining homogenate was frozen for neurotensin receptor binding analysis.
Choline acetyltransferase (CHAT) Aliquots of tissue were diluted in buffer containing 0.87 mM EDTA and 0.1% Triton X-100 (pH = 7.0). ChAT activity was measured by the conversion of [14C]acetyI-Co-A to [14C]acetylcholine as described by Fonnum 6.
Behavior Neurochemical analysis (NCA) Water maze Two weeks after the final surgery animals were tested on acquisition of a modified Morris water maze task as previously described 2°. Briefly, the maze consisted of a black circular pool, 152 cm in diameter, filled to a depth of 32 cm with clear water at a temperature of 25-26 °C. The platform was located in the SW quadrant. For each trial, animals were placed into the maze at one of four start points (designated N, S, E or W). The rat was given 90 s to find a platform hidden 3 cm below water level. If the rat was unable to find the platform by the end of the 90-s period, he was guided to it by the experimenter. The rat was allowed to remain on the platform for 10 s before removal. Testing consisted of one block of two consecutive trials, in a semi-random order with the first trial beginning from one of the long start points (N or E) and the second trial beginning from one of the closer start points (S or W). Acquisition was measured over 10 blocks (at 1 block per day). After the 10-block acquisition period, animals were removed from the water maze for 10 days and then tested for retention of the task on 4 consecutive blocks (2 trials/block/day). One day following the last block of retention testing, the platform was removed and animals were tested on a 90-s spatial probe trial. For the purposes of the spatial probe, the maze was divided into 4 equal quadrants with the platform in the third (SW) quadrant and 3 annuli determined by concentric circles drawn from the outer and inner edges of the platform. Behavioral recording was done utilizing a video tracking system (San Diego Instruments) that allowed quantification of latency, total swim distance and percent of distance swum in each quadrant and annulus. Data were analyzed by repeated measures ANOVA for acquisition and by t-test for retention.
Passive avoidance Following testing in the water maze, rats were trained on a single trial passive avoidance task in a Jarvik box 13. On the training day, animals were placed in the lighted chamber for a 10-s habituation period. The guillotine door was then opened and the latency to enter the dark chamber was recorded. The door was immediately shut and a shock was administered (0.5 mA DC, 0.5 s). Ten seconds later, the animal was removed and returned to the home cage. Retention testing was performed 72 h after the training trial and under the same conditions except that no shock was administered. Data analysis was performed using one-way ANOVA.
Open field activity Four to five days following the passive avoidance retention trial, general motor activity was assessed in a n6vel open field apparatus. The open field consisted of a 32-inch x 32-inch enclosure divided into sixteen 8-inch x 8-inch blocks. Horizontal activity was measured by the number of block crossings and ve~ical activity by the number of rearings. Activity was measured over a 5-min period with sub-analysis done on the first 90 s (to match the duration of water maze testing). Data were analyzed by one-way ANOVA.
Monoamines (1) Measurement of the monoamines norepinephrine (NE), dopamine (DA), serotonin (5-HT) and the free, uneonjugated metabolites 3-methoxy-4-hydroxyphenylglycol (MHPG), homovanillic acid (HVA) and 5-hydroxyindoleacetic acid (5-HIAA) was performed on 50-~1 aliquots of tissue extract. Clear supernatant extract was obtained by adding 10/~1 of 2.0 M PCA to 200/~1 of a homogenate and centrifuging (12,000 g, 4 °C) for 10 rain. A gradient high performance liquid chromatographic (HPLC) analyzer, equipped with 16 electrode eoulometric detectors and a refrigerated autosampler was used for the analysis (CEAS Model 55-0650, ESA, Inc. Bedford, MA) 21. This novel HPLC system provides measurement of both retention time (RT) and ratio of peak response at the dominant sensor to the preceding and following sensors. Sample peaks were compared to standard peaks for both RT and sensor ratio measures for compound identification. Quantification was achieved by comparison of the height of peaks from the sample to those of the standards on the dominant sensor. (2) Mobile phases were prepared from the following Stock solutions: A: 4 liters of 0.1 M phosphate buffer (monobasic), containing 55 mg SDS, 100 /~1 tetrahydrofuran, adjusted with ortho-phosphorie acid to pH 3.2. B: 4 liters of 0.1 M phosphate buffer (monobasic), containing 220 mg SDS, 2000 ml absolute methanol (Optima, Fisher), and adjusted as above to pH 3.45. Working mobile phase A contained 950 ml of Stock A and 50 ml of Stock B. Working mobile phase B was identical to Stock B. The first linear gradient began with 100% A and ended with 55% A:45% B at 47 min. These conditions were held for 1 min before beginning the second gradient which ended with 100% B at 57 min. Following an additional 3 min at 100% B, the analysis was terminated and abruptly switched to the initial conditions of 100% A. Samples were injected at 12.2 rain after initiation of the first gradient. A single 15-cm, 3-/~ Cls reverse-phase column (Pt#HR-150T, ESA, Inc.) was used for separation. All chromatographic runs were made at a constant temperature of 35 °C and a 1.0 ml/min flow rate. Electrochemical readings were made at 60 mV increments, with sensor 1 equal to 0 mV and sensor 16 equal to 900 mV.
Amino acid analysis Amino acid levels were determined by HPLC with electrochemical detection as described previously 16. Aliquots of the perchloric acid extract described above were pretreated with orthophthaldialdehyde (OPA) and then injected onto an 8-g Bondpak Cls reverse-phase column (Waters Corp.). The mobile phase consisted of a 0.1-M sodium phosphate buffer, pH = 5.25 containing 34% methanol and 1.5% acetonitrile (v/v), and defivered at a flow rate of 1.8 ml/min. A coulometric electrochemical detector (Coulchem model 5100A, ESA Inc.) equipped with a conditioning cell (+80 mV) and a dual electrode analytical cell (Model 5011) set at +350 mV and +800 mV respectively was used to determine amino acid
86 100 90 80 ~-~
~
O
N
'
F
R
~ ~ O L
IBO OUlS
70 so
50
2o
0 Z UJ
~
t~ Z
8
90 80 70 60 50 40 30
[] []
EllOCK9*1:3 BLOCK1I+1~
20
10-
10 0
TRIAL BLOCKS
Fig. 1. Latency (in s) to find and climb onto the platform in the Morris water maze test during acquisition training. Each point represents the mean of 2 trials/block for needle drop control rats (CONTROL; n = 10), rats with ibotenic acid lesions of NBm (IBO; n = 10) and for rats with quisqualic acid lesions of NBm (QUIS; n = 10). All groups are significantly different from each other.
concentration in 10/~1 of extract. Final concentration of amino acid is expressed as nanomoles per mg of tissue protein 18. Neurotensin receptor binding The number of neurotensin receptors in anterior and posterior cortex was determined by the binding of [3H]neurotensin ([3H]NT) to tissue homogenates 3°. Tissue homogenate was thawed and washed 3x in 50 mM Tris buffer (pH = 7.4) then resuspended (approx. 25 mg/ml). The washed tissues (100 ml) were added to tubes containing 800/al of reaction buffer (50 mM "Iris, 0.1% BSA, 0.2 mM Bacitracin and 1 mM EDTA) and 100 ~1 of [3H]NT (124 nM, 80.7 Ci/mmol) and incubated at 25 °C for 45 min. The reaction solution was then applied to filters presoaked in polyethylenimine (0.1%) on a Brandel cell harvester. Non-specific binding was determined by binding of [3H]NT in the presence of 10 mM cold neurotensin. Histology Three days after being placed into the formalin solution, the subcortical block was transferred to sucrose (10%). Frozen sections, 40-/~m thick, were cut from approximately 6.5 to 9.5 mm anterior to the intraaural line24 on a sliding microtome. To assess the extent of denervation, acetylcholinesterase (ACHE) staining was done according to a modification9 of the Koelle method. Cresyl violet staining was used to assess non-specific neuronal damage. Statistics The water maze data were analyzed by repeated measures ANOVA and subsequent Newman-Keuis post-hoe analysis of specific group differences. All other behavioral data and biochemical assays were analyzed by one-way ANOVA and post-hoc comparisons between groups were accomplished with NewmanKeuls test. RESULTS Water maze The I B O animals had the longest latencies for finding the platform, the Q U I S were intermediate and the controls performed the best across the 10 water maze acquisition blocks (Fig. 1). There were significant overall group and block effects on latency to find the platform (repeated measures A N O V A , P < 0.001). The interaction effect (group x block) was not significant (P > 0.1).
CONTROL
QUIS
IBO
GROUP
Fig. 2. Latency (in s) to find and climb onto the platform in the Morris water maze test before and after a 10-day retention interval. The mean latency on blocks 9 + 10 (total 4 trials) prior to the retention period are compared to mean latency on blocks 11 + 12 (total 4 trials) obtained after the retention interval for needle drop control rats (CONTROL; n = 10), rats with ibotenic acid lesions of NBm (IBO: n = 10) and for rats with quisqualic acid lesions of NBm (QUIS; n = 10). There is no increase in latency to find the platform for any of the groups after the retention interval. Bars represent S.E.M.
All 3 groups (collapsed across blocks) were significantly different from each other (NK P < 0.025). This pattern of longest latencies in the I B O group, intermediate latencies in the Q U I S group and shortest time to reach the platform for the control animals was also observed on the retention trials. The 10-day retention interval produced no change in latency to find the platform for any of the 3 groups, t-Tests on the means of the last two acquisition blocks (9 + 10) vs the first two retention blocks (11 + 12) were non-significant for all groups ( P > 0.4) (Fig. 2). A more sensitive m e a s u r e m e n t of retention may be achieved by comparing the first trial on the last acquisition day (block 10) with the first trial on the first retention day (block 11). This comparison also failed to show any effect of the retention interval on latency or distance measures. There was no difference in swim speed between the groups during water maze testing. The spatial probe data over the 90-s interval showed a significant group effect ( A N O V A P < 0.001, Fig. 3A). Controls swam significantly greater distances in a n n u l u s 2 (the platform annulus), than the Q U I S group, which swam more than the I B O animals in this annulus (NK P < 0.025). There was a tendency for the controls to swim more distance in the platform quadrant than either of the lesion groups, but this difference was not statistically significant (Fig. 3B). Passive avoidance All groups demonstrated short but comparable latencies during training. All groups demonstrated comparably long latencies during retention testing 72 h later (Fig. 4).
87 TABLE I
(B)
(A) 30
50'
Choline acetyltransferase (CHAT) activity
T
Data are expressed as nanomoles of acetylcholine formed per hour per milligram protein (mean + S.E.M.) and as percent decrease in ChAT activity relative to controls. All groups were significantly different from each other in both areas (NK P < 0.01). CON, needle drop control rats; QUIS, rats with bilateral quisqualic acid lesions of NBm; IBO, rats with bilateral ibotenic lesions of NBm.
4O IOm
Z
30
,,,~_
O,',, n-
10 w Q.
0
Cortex 0
CONTROL QUIS IBO GROUP
""%"
CONTROL QUIS
IBO
GROUP
Fig. 3. Distribution of distance swum during the 90-s spatial probe trial of the water maze test conducted after the final retention trial. Data represent the mean distance traveled in annulus 2 (A) and quadrant 3 (B) as a percent of the total distance traveled. All groups are significantly different from each other in distance swum in annulus 2, the platform location (Newman-Keuls post hoe-test, P < 0.05).
CON QUIS IBO
Anterior (n = 10)
Posterior (n = 10)
ChAT
ChAT
(% Depletion)
50.5 _ 0.88 18.5+ 0.41 (63%) 22.6_+0.69 (55%)
(% Depletion)
47.0 + 1.17 21.3 + 0.57 (54%) 3 0 . 2 _ + 0 . 7 9(36%)
Monoamines Open field activity No differences were found between groups on horizontal or vertical activity measures for either the first 90 s of the trial or the total time (5 min) (Fig. 5), although there was a non-significant tendency for the IBO group to have a greater n u m b e r of horizontal crossings over the 5-min interval (P < 0.08).
Choline acetyltransferase (CHAT) Both I B O and Q U I S caused significant depletions of C h A T activity in the anterior and posterior cortical regions ( A N O V A , P < 0.001), with Q U I S causing significantly more depletion than I B O in both regions (Table I; N K P < 0.01).
The concentrations of NE, D A and 5-HT in the anterior and posterior regions of cortex of the two lesion groups were not different from controls. However, small but significant differences were found for 3 of the metabolites in anterior cortex (Table II). In this region, a significant decrease (NK, P < 0.05) from controls was observed in H V A and 5 - H I A A levels ( > 3 0 % and 15% respectively) of the I B O and Q U I S groups. A significant decrease in free M H P G levels relative to controls (26%) was seen only in the I B O group. In the posterior contex, the only significant difference observed was a lower H V A level in the I B O group (23%; data not shown). No significant differences b e t w e e n I B O and Q U I S groups were found for any of the catecholamines, indolamines, or their metabolites in either anterior or posterior cortex.
(A)
[ ] CONTROL >0 Z IM I'<[
[ ] QUIS
500
[] mo
400
T ~
300
O CC ~ Z ~ I,-
200
I,U F-
100
(n)
7 0"
T
6 0-
|
350_z
40- ~ 30"
~/
:~ 20-
d.
100
0-90
~r~ CONTROL 1 4 OUlS [] IBO ._ 1 2
1
0-300
~
INTERVAL (seconds) TRAINING
RETENTION
Fig. 4. Latency (in s) to enter the dark chamber of a passive avoidance box. The mean latency on a single training trial and a retention trial conducted 72 h after the training trial are shown for needle drop control rats (CONTROL; n = 10), rats with ibotenic acid lesions of NBm (IBO; n = 10) and for rats with quisqualic acid lesions of NBm (QUIS; n = 10). No differences were found between the groups on either the training or the retention day. Bars represent S.E.M.
• CONTROL [] OUlS IBO
g,o i ~)
8 6 4 2 0
0-90
INTERVAL
0-300 (seconds)
Fig. 5. Distribution of the number of movements in an open field activity test. Data are the mean number of block crossings in the horizontal (A) and the number of rearings in vertical (B) planes during the first 90 s and the entire 0- to 300-s interval for needle drop control rats (CONTROL; n = 10), rats with ibotenic acid lesions of NBm (IBO; n = 10) and for rats with quisqualic acid lesions of NBm (QUIS; n = 10). There was no significant difference between groups for either the 0- to 90- or 0- to 300-s intervals for horizontal or vertical movements. Bars represent S.E.M.
88 'FABLE II
"FABLE IV
Catecholamine and indolamine concentration in anterior cortex
[~H] Neurotensin binding ( B ,,,,~) in cortex
Data are expressed as n a n o g r a m s of c o m p o u n d per milligram protein (mean + S.E.M.). *P < 0.05 from controls, ( N e w m a n - K e u l s post hoc test). C O N , needle drop control rats; Q U I S , rats with bilateral quisqualic acid lesions of N B m ; IBO, rats with bilateral ibotenic lesions of NBm.
Data are expressed as femtomoles per milligram protein ( m e a n +_ S.E.M.). C O N , needle drop control rats; Q U I S , rats with bilateral quisqualic acid lesions of N B m ; IBO, rats with bilateral ibotenic lesions of N B m Anterior (n = 10)
Posterior (n = 10)
81.31 + 5.84 82.50 + 4.45 82.22 _+ 6.73
103.05 + 10.83 101.93 _ 7.26 103.28 -+ 6.67
Anterior cortex
NE MHPG DA HVA 5-HT 5-HIAA
C O N (n = 9)
QUIS (n = 10)
I B O (n = 10)
0.991 0.046 0.078 0.054 1.993 0.947
0.945 0.040 0.077 0.037 1.921 0.797
0.921 0.034 0.075 0.032 1.900 0.771
+ + + + + +
0.044 0.003 0.010 0.004 0.097 0.055
+ 0.020 __+0.003 + 0.006 + 0.002* + 0.064 + 0.033*
+ + + + + +
0.060 0.002* 0.010 0.002* 0.097 0.033*
Amino acids There were no significant differences between groups for any of the amino acids measured in either anterior or posterior cortex (Table III, anterior cortical data; posterior cortical data not shown).
CON QUIS IBO
Both neurotoxins resulted in damage to the medial aspect of the reticular nucleus of the thalamus without apparent damage to adjacent thalamic nuclei. The basolateral nucleus of the amygdala was also affected by both toxins, but to a much greater extent by IBO than by QUIS. The hippocampus, striatum, nucleus of the lateral optic tract and the cortical-amygdala transition zone were not affected by either neurotoxin. DISCUSSION
Neurotensin binding No difference was found between groups for the Bmax of [3H]NT binding in the anterior or posterior cortex (Table IV).
Histology Histological evaluation of the tissue block containing the NBm by Cresyl violet and AChE staining showed slight cell loss and macrophage accumulation in the dorsal globus pallidus in the IBO group but almost no cell loss in the QUIS group. The magnocellular cholinergic neurons in the substantia innominata and ventral globus pallidus were affected by both toxins, but the degree of cell loss and gliosis was much less in the QUIS group.
T A B L E III A m i n o acid concentration in anterior cortex Data are expressed as n a n o m o l e s of amino acid per milligram protein
(mean + S.E.M.). CON, needle drop control rats; QUIS, rats with bilateral quisqualic acid lesions of NBm; IBO, rats with bilateral ibotenic lesions of NBm.
Aspartate Serine Glutamine Glutamate Glycine Taurine Alanine Ethanolamine GABA
C O N (n = 8)
QUIS (n = 10)
I B O (n = 9)
28.82 7.92 34.69 109.63 6.41 48.44 5.56 2.84 13.07
38.71 8.77 37.88 117.39 6.79 52.81 5.86 3.40 13.96
33.62 +_ 2.23 8.81 + 0.58 36.55 + 3.09 113.66 _+ 8.02 7.40 + 0.68 54.94 + 3.66 5.88 -+ 0.43 2.90 + 0.14 15.44 + 1.89
+ 2.05 -+ 0.46 -+ 1.72 + 7.08 +- 0.41 _+ 2.59 + 0.36 + 0.16 + 0.93
+- 4.90 + 0.543 + 2.52 -+ 7.72 -+ 0.35 + 3.30 + 0.20 +- 0.25 + 0.95
In agreement with our previous studies t9'2°, this study demonstrated that bilateral I130 lesions of the NBm produced behavioral impairment on the acquisition of a spatial memory task without causing a deficit in the retention of this task. The acquisition deficit was significantly less severe in the animals treated with QUIS. The behavioral difference cannot be attributed to motor disturbances since general motor activity, water maze speed and passive avoidance retention were not significantly affected by NBm lesion with either neurotoxin. Despite the greater behavioral impairment of the IBO group on the water maze task, ChAT depletion was greater in both anterior and posterior regions of cortex of the QUIS group. These data, in conjunction with the behavioral data, agree with previous s t u d i e s 2'26"33 and demonstrate that the extent of cortical ChAT depletion does not necessarily correlate with the degree of behavioral impairment. In the water maze task, the annulus measure of the spatial probe showed a good correlation with the latency to find the platform. The distance swum in quadrant 3 (platform quadrant; Fig. 3B) showed a tendency for both lesion groups to swim less in this quadrant than controls. Observation of the animals during the task revealed that controls had a greater tendency to circle in the area of the platform while QUIS and IBO animals tended to push off from one wall to another for pool crossings. This was supported by the annulus data which showed that the group that swam the most in annulus 2 (platform annulus) also had the shortest latency and the group that
89 swam the least in annulus 2 had the longest latency (Fig. 1 vs Fig. 3A). The possibility that IBO and QUIS could differ with respect to the cholinergic neuronal subpopulation affected in the NBm is supported by evidence that IBO and QUIS bind to different subclasses of glutamate receptors with IBO preferring the N M D A subtype and QUIS utilizing the quisqualate subtype 3'4. However, another study measuring second messenger effects indicates that the two toxins may act at the same receptor with different affinities 29. If the behavioral impairment is based purely on a NBm cholinergic lesion, then by subtracting the depletion caused by QUIS in this study (54--63%) from the total cortical innervation by NBm neurons (70%) 14, it can be concluded that less than 15% of the NBMcortical cholinergic input is left to explain the difference between IBO and QUIS. That is, less than 15% of the cortical projecting cholinergic neurons in the NBm must be responsible for the behavioral differences between the two neurotoxins if this hypothesis is true. Although interactions between the noradrenergic, dopaminergic, serotonergic and cholinergic systems have been previously demonstrated in a variety of models 7' 11,12.25.28,32, changes in cortical levels of these substances do not explain the more severe behavioral impairments produced by IBO. Levels of each of the neurotransmitters were unchanged between all groups. Significant differences from control levels were noted for some of the monoamine metabolites, but these metabolites were not different between the two lesion groups. Similarly, cortical amino acid levels were not different in the two populations. However, it should be noted that the amino acid levels in extracts of cortex reflect the metabolic as well as the neurotransmitter pools of these compounds. Measures of in vivo release of the monoamines, amino acids and acetylcholine using microdialysis techniques are needed to more accurately assess the utilization and turnover of these transmitters following IBO and QUIS lesions. It is also possible that IBO may have a greater tendency to diffuse to other structures such as the globus pallidus and amygdala. This difference was suggested by studies in which both neurotoxins were injected into the lateral hypothalamus and different diffusion patterns were seen 8. Other s t u d i e s 2'26 have found that in some cases after IBO infusion to the NBm, surrounding structures showed cell loss and gliosis. Our results showed slight damage to dorsal globus pallidus with more REFERENCES 1 Coyle, J.T., Molliver, M.E. and Kuhar M.J., In situ injection of kainic acid: a new method for selectively lesioning neuronal cell bodies while sparing axons of passage, J. Comp. Neurol., 180 (1978) 301-324.
extensive damage to the thalamic reticular nucleus and the basolateral nucleus of the amygdala with IBO as compared to QUIS. Contrary to reports from other labs 2"5A°, this study, as well as our previous studies 15'19, failed to detect any differences between IBO and controls or between QUIS and control animals in passive avoidance training or in retention at 72 h after shock. A ceiling effect is possible but our titration studies indicate that controls are just at optimal retention (unpublished observations) and therefore if IBO or QUIS produces an impairment of passive avoidance it would be very subtle. It is possible that studies demonstrating a passive avoidance deficit involve a more extensive lesion and damage to structures other than the NBm as seen in previous work from our laboratory 31. Other groups have also demonstrated a passive avoidance impairment with lesions of the dorsolateral globus pallidus 5. A final possibility is that since several groups chose a 24-h retention interval, they may be testing their groups before consolidation has been completely established. In conclusion, although a wide variety of cortical neurochemical systems were studied, none were able to explain the behavioral difference between the IBO and QUIS groups on water maze acquisition. Cortical ChAT levels showed a large decrease after both IBO and QUIS lesion of the NBm (Table I), as did HVA (Table II), but neither corresponded with the order of behavioral impairment caused by the two toxins. Further, no change was seen in cortical levels of DA, NE, 5-HT, amino acids or neurotensin binding. At present, the reason for the milder behavioral deficit when QUIS is used as the neurotoxin is unknown. It is possible that IBO destroyed other unknown projections to the cortex that were not examined. Alternately, important NBm pathways projecting to non-cortical structures may have been damaged by the IBO but not the QUIS lesion. Observation of damage to adjacent areas (amygdala, reticular nucleus of the thalamus) after IBO infusion may also have contributed to the behavioral difference.
Acknowledgements. The authors gratefully acknowledge the assistance of Andrew Chen and Mark Wardlow in the neurochemical assays and Barbara Reader in the preparation of this manuscript. This work was supported by funds from the VA Merit Program (P.J.L. and L.J.T.), an NSF grant to L.J.T. and NIH/DRR ($10 RRO 4754-01 to P.J.L.).
2 Dunnett, S.B., Whishaw, I.Q., Jones, G.H. and Bunch, S.T., Behavioral, biochemical and histochemical effects of different neurotoxic amino acids injected into nucleus basalis magnocellularis of rats, Neuroscience, 20 (1987) 653-669. 3 Fagg, G.E. and Matus, A., Selective association of N-methyl aspartate and quisqualate types of L-glutamate receptor with
90
4 5
6 7 8
9 10 11
12
13 14 15
16
17
18 19
brain postsynaptic densities, Proc. Natl. Acad. Sci. U.S.A., 81 (1984) 6876-6880. Fagg, G.E., Foster, A.C. and Ganong, A.H., Excitatory amino acid synaptic mechanisms and neurological function, Trends Pharmacol. Sci., 7 (1986) 357-363. Flicker, C., Dean, R.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. Fonnum, E, Radiochemical micro assays for the determination of choline acetyltransferase and acetylcholinesterase activities, Biochem. J., 115 (1969) 465-472. Harrell, L.E. and Parsons, D.S., The role of cortical sympathetic ingrowth in the behavioral effects of nucleus basalis magnocellularis lesions, Brain Research, 474 (1988) 353-358. Hastings, M.N., Winn, P. and Dunnett, S.B., Neurotoxic amino acid lesions of the lateral hypothalamus: a parametric comparison of the effects of ibotenate, N-methyi-o.L-aspartate and quisqualate in the rat, Brain Research, 360 (1985) 248-256. Hedreen, J.C., Bacon, J.C. and Price, D.L., A modified method to visualize acetylcholinesterase-containingaxons, J. Histochem, Cytochem., 33 (1985) 134-140. Hepler, D.J., Wenk, G.L., Cribbs, B.L., Olton, D.S. and Coyle, J.T., Memory impairments following basal forebrain lesions, Brain Research, 346 (1985) 8-14. Hortnagl, H., Potter, P.E. and Hanin, I., Effect of cholinergic deficit induced by ethylcholine aziridinium (AF64A) on noradrenergic and dopaminergic parameters in rat brain, Brain Research, 421 (1987) 75-84. Hortnagl, H., Potter, P.E., Kindel, G. and Hanin, I., Noradrenaline depletion protects cholinergic neurons in rat hippocampus against AF64A-induced damage, J. Neurosci. Methods, 27 (1989) 103-108. Jarvik, M.E. and Kopp, R., An improved one-trial passive avoidance learning situation, Psychol. Rep., 21 (1967) 221-224. Johnston, M.V., McKinney, M. and Coyle, J.T., Neocortical cholinergic innervation: a description of extrinsic and intrinsic components in the rat, Exp. Brain Res., 43 (1981) 159-172. Kwo-On-Yuen, P.E, Mandel, R., Chen, A.D. and Thai, L.J., Tetrahydroaminoacridine improves the spatial acquisition deficit produced by nucleus basalis lesions in rats, Exp. Neurol., 108 (1990) 221-228. Langlais, P.J., Mair, R.G., Anderson, C.D. and McEntee, W.J., Long-lasting changes in regional brain amino acids and monoamines in recovered pyrithiamine treated rats, Neurochem. Res., 13 (1988) 1199-1206. Lo Conte, G., Bartolini, L., Casamenti, E, Marconcini-Pepeu, I. and Pepeu, G., Lesions of cholinergic forebrain nuclei: changes in avoidance behavior and scopolamine actions, Pharmacol. Biochem. Behav., 17 (1982) 933-937. Lowry, O.H., Rosenberg, N.J., Farr, A.L. and Randall, R.J., Protein measurement with the Folin phenol reagent, J. Biol. Chem., 193 (1951) 265-275. Mandel, R.J., Gage, EH. and Thai, L.J., Spatial learning in rats: correlation with cortical choline acetyltransferase and improvement with NGF following NBM damage, Exp. NeuroL, 104 (1989) 208-217.
20 Mandel, R.J., Gage, EH. and Thai, L.J., Enhanced detection of nucleus basalis magnocellularis lesion-induced spatial learning deficit in rats by modification of training regimen, Behav. Brain Res., 31 (1989) 221-229. 21 Matson, W.R., Gamache, P.G., Beal, M.E and Bird, E.D., EC array sensor concepts and data, Life Sci., 41 (1987) 905-908. 22 Murray, C.L. and Fibiger, H.C., Learning and memory deficits after lesions of the nucleus basalis magnocellularis: reversal by physostigmine, Neuroscience, 14 (1985) 1025-1032. 23 Nakamura, S., Nakagawa, Y., Kawai, M., Tohyama, M. and Ishihara, T., AF64A (ethylcholine aziridinum ion)-induced basal forebrain lesion impairs maze performance, Behav. Brain Res., 29 (1988) 119-126. 24 Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic Press, New York, 1986. 25 Robbins, T.W. and Everitt, B.J., Comparative functions of the central noradrenergic, dopaminergic and cholinergic systems, Neuropharmacology, 26 (1987) 893-901. 26 Robbins, T.W., Everitt, B.J., Ryan, C.N., Marston, H.M., Jones, G.H. and Page, K.J., Comparative effects of quisqualic and ibotenic acid-induced lesions of the substantia innominata and globus pallidus on the acquisition of a conditional visual discrimination: differential effects on cholinergic mechanisms, Neuroscience, 28 (1989) 337-352. 27 Salamone, J.D., Beart, P.M., Aipert, J.E. and Iversen, S.D., Impairment in T-maze reinforced alternation performance following nucleus basalis magnocellularis lesions in rats, Behav. Brain Res., 13 (1984) 63-70. 28 Sara, S.J., Noradrenergic-cholinergic interaction: its possible role in memory dysfunction associated with senile dementia, Arch. Gerontol. Geriatr., Suppl. 1 (1989)99-108. 29 Schoepp, D.D. and Johnson, B.G., Excitatory amino acid agonist-antagonist interactions at 2-amino-4-phosphonobutyric acid-sensitive quisqualate receptors coupled to phosphoinositide hydrolysis in slices of rat hippocampus, J. Neurochem., 50 (1988) 1605-1613. 30 Schotte, A. and Laduron, P.M., Different postnatal ontogeny of two [3H]neurotensin binding sites in rat brain, Brain Research, 408 (1987) 326-328. 31 Thai, L.J., Dokla, C.P.J. and Armstrong, D.M., Nucleus basalis magnocellularis lesions: lack of biochemical and immunocytochemical recovery and effect of cholinesterase inhibitors on passive avoidance, Behav. Neurosci., 102 (1988) 852-860. 32 Wenk, G., Hughey, D., Boundy, V., Kim, A., Walker, L. and Olton, D., Neurotransmitters and memory: role of cholinergic, serotonergic and noradrenergic systems, Behav. Neurosci., 101 (1987) 325-332. 33 Wenk, G.L., Markowska, A.L. and Olton, D.S., Basal forebrain lesions and memory: alterations in neurotensin, not acetylcholine, may cause amnesia, Behav. Neurosci., 103 (1989) 765-769. 34 Whishaw, I.Q., O'Connor, W.T. and 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 (1985) 103-115.