Brain Research, 522 (1990) 227-234 Elsevier
227
BRES 15679
Muscimol injections in the medial septum impair spatial learning Jorge D. Brioni 1, Michael W. Decker 1"2, Lynette P. Gamboa 2, Ivan Izquierdo 3 and James L. McGaugh t'2 1Centerfor the Neurobiology of Learning and Memory, and eDepartmentof Psychobiology, University of California, lrvine, CA 92717 (U.S.A.) and 3Centro Da Memoria, Departamentode Bioquimica, U.E G.R.S., Porto Alegre (Brazil)
(Accepted 16 January 1990) Key words: ~'-Aminobutyric acid; Hippocampus: Muscimol; Septum; Spatial learning; Choline uptake
These experiments examined the role of GABAergic systems in modulating septohippocampal cholinergic influences on learning. Microinjections of the GABA A agonist muscimol (0.5, 1.0 or 5.0 nmol) or physiological saline were administered (0.5 pl) into the medial septum of rats via chronically implanted cannulae just prior to daily training in the Morris water maze spatial learning task. The animals received 3 training trials on each of 4 days. The escape latencies of rats trained with a submerged escape platform at a fixed location were significantly shorter than those trained with a randomly located platform. Rate of learning of the fixed location was significantly impaired in rats given pretraining muscimol injections in the medial septum at doses (1.0 and 5.0 nmol) that significantly reduced hippocampal high-affinity choline uptake (HACU). Analyses of responses on a probe trial with no pretraining injections and no platform revealed that, in comparison with controls, animals that had received muscimol prior to each training session were less likely to swim in the region where the platform had been located. The finding that muscimol-injected rats were subsequently able to learn the task when trained without muscimol injections indicates that the acquisition impairment was not due to a lasting effect of the drug injections. Our results are consistent with the view that the septal GABAergic modulation of the septohippocampal cholinergic pathway is involved in regulating the acquisition of spatial information.
INTRODUCTION There is extensive evidence indicating that, in rats, the hippocampus is involved in the acquisition of spatial information 36. Conditions, such as brain lesions, that impair the functioning of the hippocampus disrupt learning of a variety of kinds of spatial tasks including radial arm mazes, and hole boards, as well as place navigation in a water tank 2. In the water-maze task 31'32, rats rapidly learn to locate a submerged platform in the absence of local cues by using cues distal to the platform position. Such findings are consistent with the view that rat forms a spatial map of the location of extramaze cues in relation to the position of the escape platform 36. Lesions of the medial septal nucleus that eliminate theta rhythm impair spatial learning 48, and hippocampal lesions cause a profound and long-lasting impairment in place navigation 2,33. Pharmacological studies have indicated that cholinergic muscarinic antagonists impair place navigation learning 24,46. Recent findings 25 strongly suggest that the cholinergic septohippocampal pathway is critically involved in regulating spatial learning. Ibotenic acid lesions of the diagonal band/medial septal area that deplete hippocampal choline acetyltransferase (CHAT) impair spatial learning.
The cell bodies of the septohippocampal cholinergic neurons are located in the medial septum and in the nucleus of the diagonal band, and the axons project mainly via the fimbria and the dorsal fornix to the hippocampal formation 34'47. There is evidence that this system is regulated at the level of the septum by several neurochemical systems including y-aminobutyric acid ( G A B A ) , fl-endorphin, substance P, norepinephrine, dopamine and corticotropin neuropeptides: manipulations of these systems modulate hippocampal highaffinity choline uptake ( H A C U ) and the turnover rate of acetylcholine (ACh) 6,16. Intraperitoneal injections of pentobarbital, which is known to influence the G A B A ergic system, reduce in vitro H A C U , an index of cholinergic activity, and A C h turnover in the hippocampus 12'45. Further, systemic injections of muscimol, T H I P and diazepam reduce turnover rate of hippocampal ACh 5~. Intraseptal injections of phenobarbital and muscimol reduce HACU 43, and muscimol, diazepam and T H I P significantly reduce the turnover rate of ACh 4"5°. The reduction of hippocampal H A C U induced by intraseptal G A B A e r g i c drug injections has also been correlated at the behavioral level with the extinction of operant responses, and with the anticonvulsant effect of barbiturates 4"43. In view of the evidence that spatial
Correspondence: J.D. Brioni. Present address: Departamento de Farmacologia, Facultad de Ciencias Quimicas, Universidad Nacional de Cordoba, Sucursal 16-CC 61, 5016 Cordoba, Argentina.
0006-8993/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
228 l e a r n i n g is r e g u l a t e d b y t h e c h o l i n e r g i c s e p t o h i p p o c a m p a l p a t h w a y , i n j e c t i o n s o f t h e G A B A A a g o n i s t m u s c i m o l 3' 19,26,38 a d m i n i s t e r e d to t h e m e d i a l s e p t a l r e g i o n s h o u l d alter spatial learning. The present experiments examined this i m p l i c a t i o n .
MATERIALS AND METHODS
Animals Male Sprague-Dawley rats (60 days old, 200-220 g on arrival) from Charles River Laboratories were used. They were individually housed upon arrival and maintained on a 12-h light-dark period (lights on 07.00 h) with food and water available ad libitum. They were acclimatized to laboratory conditions for 1 week before undergoing surgery.
Cannula implantation The animals were anesthetized with sodium pentobarhital (50 mg/kg, i.p.), and given atropine sulfate (0.4 mg/kg, i.p.) and a permanent stainless-steel guide cannula (23 gauge; 15 mm) was implanted stereotaxically (Kopf Instruments). The tip of the cannula was aimed at the dorsal surface of the medial septum. The stereotaxic coordinates were: AP +0.5 mm from bregma; ML 00 mm (on midline); DV -4.0 mm from dura; with the nosebar at -3.3 mm from the interaural line41. The cannula was fixed to the skull using two screws and dental acrylic. A styler was inserted into the cannula to keep it patent prior to injections. Immediately after surgery the animals received an intramuscular injection of penicillin and were maintained in a temperature-controlled chamber until recovery from anesthesia. The animals were allowed 1 week of recovery before initiation of the behavioral experiments.
lntraseptal injections The rats were gently restrained by hand, the stylets withdrawn from the guide cannula, and a 30-gauge injection needle was inserted. The injection needle was attached by a polyethylene tube to a 5 ¢tl syringe driven by a minipump. The injection needle was slightly bent so that when inserted into the guide cannula the bend would prevent further penetration, and the tip would thus protrude 2 mm beyond the tip of the guide cannula. The injection solutions (muscimol or physiological saline) were administered in a total volume of 0.5 ~1 at a rate of 0.75/A/min, a volume and rate of delivery we have found optimal in previous studies. The injection needle was retained in the guide cannulae for an additional 30 s after the injection to facilitate the diffusion of the drug. Five min after the drug injection the animals were subjected to the behavioral or biochemical studies.
High-affinity choline uptake HACU was measured using a modification of the procedure described by Sherman et al. 44 and Simon et alY. The animals were sacrificed 5 min after the intraseptal injections of muscimol or saline and the brains were rapidly removed and dissected on ice. The hippocampus was dissected out, weighed and homogenized in 20 vols. of ice-cold 0.32 M sucrose using a hand-held Teflon-on-glass homogenizer. After centrifugation of the homogenate at 1000 g for 10 rain at 2 °C, the supernatant was spun at 18,000 g for 20 min to yield a crude synaptosomal/mitochondrial pellet. The pellet was resuspended in 24 vols. of cold 0.32 M sucrose, and 50/A of this suspension was added to tubes containing 450 ktl of ice-cold incubation buffer (final dilution in mM: NaCI 126, KCI 4.75, CaCI z 1.27, Na2HPO 4 15.8, MgCI2 1.42, 2 mg/ml dextrose, pH 7.4), to which [3H]choline had been added to attain a final concentration of 0.4 ~M. Sodium independent uptake was measured in parallel using incubation buffer in which NaC1 was replaced by LiCI, and Na2HPO 4 was replaced with Tris phosphate. Tubes were then placed in a shaking water bath for 2.5 min at 37 °C, followed by cooling in an ice-water bath. After the addition of 2 ml of ice-cold sodium-free
buffer, samples were filtered over glass fiber filters (A/E, Gelman). The filters were then rinsed twice with 2 ml of ice-cold buffer and counted in 10 ml Scintiverse E (Fisher). HACU in all samples was measured in triplicate and determined by subtracting sodiumindependent uptake from total uptake. Under these conditions HACU is linear with incubation time up to 3 min, and linear with tissue concentration in the range we used TM.
Behavioral studies Apparatus. The water maze was a circular, galvanized-steel tank measuring 1.83 m in diameter and 0.58 m in height, filled to a depth of 20 cm with 25 °C water. Attached to the rim of the tank along 1.70 m of its circumference was a white strip extending 16 cm from the rim that served as a salient cue. Other cues including a cabinet, a poster on the wall and the experimenter were also available in the environment surrounding the tank. Four points equally spaced around the perimeter of the tank were arbitrarily selected as starting locations. On this basis, the tank was divided into 4 equal quadrants in a clockwise order (target, adjacent, opposite and adjacent). Located in the center of one of these quadrants was a 12 x 14 cm rectangular Plexiglas platform 19 cm in height (i.e. its surface was 1 cm below the water level). The platform remained in a fixed location throughout the training period of the place-trained animals. Following each day of training the tank was drained and cleaned. Fixed location training procedure. Injections were administered as described above 5 min prior to each day's training session. A trial began when the rat was placed in the pool facing the wall of the tank at one of the starting positions which varied from trial to trial in a quasirandom order. The rat was allowed to swim until it located and climbed onto the escape platform. Rats were gently guided to the platform if they failed to locate it within 90 s. The rat remained on the platform for 30 s before being removed. Two additional trials were given with 5-10 s intertrial intervals. Following the third trial the rat was returned to its home cage. The rats received a total of 12 trials during the 4 consecutive days of training. Escape latencies were measured as the time from release in the pool until escape onto the platform. All trials were videotaped through a wide angle lens attached to a camera mounted above the tank. Random vs place training. This experiment was conducted in order to examine the acquisition of spatial learning under our training conditions, and no drugs were administered. Two training conditions were used. Random-trained rats were trained to locate a submerged platform whose location was varied in a quasirandom order from trial to trial among the 4 possible locations of the platform. As the platform was never in the same location on two consecutive trials, the rat could not use the spatial environment to locate it. Place-trained animals were required to locate a submerged platform that was not visible to the rat, but was always located in the same position with respect to the spatial environment, thereby allowing the rat to determine the platform location by using the distal cues. Animals were placed in the pool at the designated starting locations, and their starting position was varied from trial to trial in a quasirandom order. The training was conducted according to the general procedures described above. Free-swim trial. This 60-s trial was carried out 4 days after the completion of the 4 training sessions. No drug or saline injections were administered prior to the test. The four possible positions of the platform and the limits of the 4 quadrants were marked on the video screen to indicate their exact surface area. The escape platform was removed from the tank and the rat was placed in the pool for 60 s. From video tapes made during this free swim it was possible to calculate 3 measures of spatial behavior: (1) latency to cross the target platform location - - the time in seconds to cross the target platform marked in the screen; (2) quadrant time - - the number of seconds spent by the rat in the target quadrant where the platform was formerly placed; and (3) platform crossings - - the number of times the rat traversed the former position of the escape platform as well as the crossings made on the 3 remaining platform positions (one for each of the remaining quadrants). These 3 measures were used to assess the degree of spatial bias of the rats
229 toward the target quadrant during the free swim. Retraining. After the 60-s free-swim trial, the platform was placed in its former location and the rats were guided to it and allowed to remain on it for 30 s. Immediately afterwards, they were given 3 retraining trials, according to the general training procedures, to determine whether the prior drug injection procedures produced tissue damage that would have significant behavioral consequences in this task.
Drugs Muscimol (Sigma Chemical Co.) was dissolved in saline solution and injected in doses of 0.05, 0.5, 1.0 and 5.0 nmol. Choline chloride was purchased from Sigma Chemical Co. and [3H]choline chloride (80.0 Ci/mmol) was purchased from New England Nuclear.
Histology One week after the behavioral studies the cannulae placements were verified histologically. The rats were anesthetized with an overdose of sodium pentobarbital and perfused through the heart with saline solution followed by 10% formaldehyde solution. Slices (40/~m) indicating the position of the cannulae were stained with Cresyl violet. One rat with incorrect placement of the cannula was excluded from the analysis. Cannula placement of animals that were sacrificed for biochemical studies on the hippocampus after muscimol injection in the medial septum, were also verified histologically.
Statistics The data were analyzed by one-way or two-way analysis of variance (ANOVA) followed by the Fisher PLSD test for individual mean comparisons.
< 0.0001) and no interaction (Fl1,154 = 1.2; n.s.). A relevant observation from this set of data that can be compared with the rest of the experiments is that random-trained animals always average more than 30 s to locate the escape platform, while place-trained animals were able to escape in less than 15 s by the fourth day of training (trial 10 = 13.8 s). A n analysis of the data by days show the same results, with a training effect (F1,14 = 10.6; P < 0.01), a days effect (F3,42
=
16.9; P
227
BRES 15679
Muscimol injections in the medial septum impair spatial learning Jorge D. Brioni 1, Michael W. Decker 1"2, Lynette P. Gamboa 2, Ivan Izquierdo 3 and James L. McGaugh t'2 1Centerfor the Neurobiology of Learning and Memory, and eDepartmentof Psychobiology, University of California, lrvine, CA 92717 (U.S.A.) and 3Centro Da Memoria, Departamentode Bioquimica, U.E G.R.S., Porto Alegre (Brazil)
(Accepted 16 January 1990) Key words: ~'-Aminobutyric acid; Hippocampus: Muscimol; Septum; Spatial learning; Choline uptake
These experiments examined the role of GABAergic systems in modulating septohippocampal cholinergic influences on learning. Microinjections of the GABA A agonist muscimol (0.5, 1.0 or 5.0 nmol) or physiological saline were administered (0.5 pl) into the medial septum of rats via chronically implanted cannulae just prior to daily training in the Morris water maze spatial learning task. The animals received 3 training trials on each of 4 days. The escape latencies of rats trained with a submerged escape platform at a fixed location were significantly shorter than those trained with a randomly located platform. Rate of learning of the fixed location was significantly impaired in rats given pretraining muscimol injections in the medial septum at doses (1.0 and 5.0 nmol) that significantly reduced hippocampal high-affinity choline uptake (HACU). Analyses of responses on a probe trial with no pretraining injections and no platform revealed that, in comparison with controls, animals that had received muscimol prior to each training session were less likely to swim in the region where the platform had been located. The finding that muscimol-injected rats were subsequently able to learn the task when trained without muscimol injections indicates that the acquisition impairment was not due to a lasting effect of the drug injections. Our results are consistent with the view that the septal GABAergic modulation of the septohippocampal cholinergic pathway is involved in regulating the acquisition of spatial information.
INTRODUCTION There is extensive evidence indicating that, in rats, the hippocampus is involved in the acquisition of spatial information 36. Conditions, such as brain lesions, that impair the functioning of the hippocampus disrupt learning of a variety of kinds of spatial tasks including radial arm mazes, and hole boards, as well as place navigation in a water tank 2. In the water-maze task 31'32, rats rapidly learn to locate a submerged platform in the absence of local cues by using cues distal to the platform position. Such findings are consistent with the view that rat forms a spatial map of the location of extramaze cues in relation to the position of the escape platform 36. Lesions of the medial septal nucleus that eliminate theta rhythm impair spatial learning 48, and hippocampal lesions cause a profound and long-lasting impairment in place navigation 2,33. Pharmacological studies have indicated that cholinergic muscarinic antagonists impair place navigation learning 24,46. Recent findings 25 strongly suggest that the cholinergic septohippocampal pathway is critically involved in regulating spatial learning. Ibotenic acid lesions of the diagonal band/medial septal area that deplete hippocampal choline acetyltransferase (CHAT) impair spatial learning.
The cell bodies of the septohippocampal cholinergic neurons are located in the medial septum and in the nucleus of the diagonal band, and the axons project mainly via the fimbria and the dorsal fornix to the hippocampal formation 34'47. There is evidence that this system is regulated at the level of the septum by several neurochemical systems including y-aminobutyric acid ( G A B A ) , fl-endorphin, substance P, norepinephrine, dopamine and corticotropin neuropeptides: manipulations of these systems modulate hippocampal highaffinity choline uptake ( H A C U ) and the turnover rate of acetylcholine (ACh) 6,16. Intraperitoneal injections of pentobarbital, which is known to influence the G A B A ergic system, reduce in vitro H A C U , an index of cholinergic activity, and A C h turnover in the hippocampus 12'45. Further, systemic injections of muscimol, T H I P and diazepam reduce turnover rate of hippocampal ACh 5~. Intraseptal injections of phenobarbital and muscimol reduce HACU 43, and muscimol, diazepam and T H I P significantly reduce the turnover rate of ACh 4"5°. The reduction of hippocampal H A C U induced by intraseptal G A B A e r g i c drug injections has also been correlated at the behavioral level with the extinction of operant responses, and with the anticonvulsant effect of barbiturates 4"43. In view of the evidence that spatial
Correspondence: J.D. Brioni. Present address: Departamento de Farmacologia, Facultad de Ciencias Quimicas, Universidad Nacional de Cordoba, Sucursal 16-CC 61, 5016 Cordoba, Argentina.
0006-8993/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
228 l e a r n i n g is r e g u l a t e d b y t h e c h o l i n e r g i c s e p t o h i p p o c a m p a l p a t h w a y , i n j e c t i o n s o f t h e G A B A A a g o n i s t m u s c i m o l 3' 19,26,38 a d m i n i s t e r e d to t h e m e d i a l s e p t a l r e g i o n s h o u l d alter spatial learning. The present experiments examined this i m p l i c a t i o n .
MATERIALS AND METHODS
Animals Male Sprague-Dawley rats (60 days old, 200-220 g on arrival) from Charles River Laboratories were used. They were individually housed upon arrival and maintained on a 12-h light-dark period (lights on 07.00 h) with food and water available ad libitum. They were acclimatized to laboratory conditions for 1 week before undergoing surgery.
Cannula implantation The animals were anesthetized with sodium pentobarhital (50 mg/kg, i.p.), and given atropine sulfate (0.4 mg/kg, i.p.) and a permanent stainless-steel guide cannula (23 gauge; 15 mm) was implanted stereotaxically (Kopf Instruments). The tip of the cannula was aimed at the dorsal surface of the medial septum. The stereotaxic coordinates were: AP +0.5 mm from bregma; ML 00 mm (on midline); DV -4.0 mm from dura; with the nosebar at -3.3 mm from the interaural line41. The cannula was fixed to the skull using two screws and dental acrylic. A styler was inserted into the cannula to keep it patent prior to injections. Immediately after surgery the animals received an intramuscular injection of penicillin and were maintained in a temperature-controlled chamber until recovery from anesthesia. The animals were allowed 1 week of recovery before initiation of the behavioral experiments.
lntraseptal injections The rats were gently restrained by hand, the stylets withdrawn from the guide cannula, and a 30-gauge injection needle was inserted. The injection needle was attached by a polyethylene tube to a 5 ¢tl syringe driven by a minipump. The injection needle was slightly bent so that when inserted into the guide cannula the bend would prevent further penetration, and the tip would thus protrude 2 mm beyond the tip of the guide cannula. The injection solutions (muscimol or physiological saline) were administered in a total volume of 0.5 ~1 at a rate of 0.75/A/min, a volume and rate of delivery we have found optimal in previous studies. The injection needle was retained in the guide cannulae for an additional 30 s after the injection to facilitate the diffusion of the drug. Five min after the drug injection the animals were subjected to the behavioral or biochemical studies.
High-affinity choline uptake HACU was measured using a modification of the procedure described by Sherman et al. 44 and Simon et alY. The animals were sacrificed 5 min after the intraseptal injections of muscimol or saline and the brains were rapidly removed and dissected on ice. The hippocampus was dissected out, weighed and homogenized in 20 vols. of ice-cold 0.32 M sucrose using a hand-held Teflon-on-glass homogenizer. After centrifugation of the homogenate at 1000 g for 10 rain at 2 °C, the supernatant was spun at 18,000 g for 20 min to yield a crude synaptosomal/mitochondrial pellet. The pellet was resuspended in 24 vols. of cold 0.32 M sucrose, and 50/A of this suspension was added to tubes containing 450 ktl of ice-cold incubation buffer (final dilution in mM: NaCI 126, KCI 4.75, CaCI z 1.27, Na2HPO 4 15.8, MgCI2 1.42, 2 mg/ml dextrose, pH 7.4), to which [3H]choline had been added to attain a final concentration of 0.4 ~M. Sodium independent uptake was measured in parallel using incubation buffer in which NaC1 was replaced by LiCI, and Na2HPO 4 was replaced with Tris phosphate. Tubes were then placed in a shaking water bath for 2.5 min at 37 °C, followed by cooling in an ice-water bath. After the addition of 2 ml of ice-cold sodium-free
buffer, samples were filtered over glass fiber filters (A/E, Gelman). The filters were then rinsed twice with 2 ml of ice-cold buffer and counted in 10 ml Scintiverse E (Fisher). HACU in all samples was measured in triplicate and determined by subtracting sodiumindependent uptake from total uptake. Under these conditions HACU is linear with incubation time up to 3 min, and linear with tissue concentration in the range we used TM.
Behavioral studies Apparatus. The water maze was a circular, galvanized-steel tank measuring 1.83 m in diameter and 0.58 m in height, filled to a depth of 20 cm with 25 °C water. Attached to the rim of the tank along 1.70 m of its circumference was a white strip extending 16 cm from the rim that served as a salient cue. Other cues including a cabinet, a poster on the wall and the experimenter were also available in the environment surrounding the tank. Four points equally spaced around the perimeter of the tank were arbitrarily selected as starting locations. On this basis, the tank was divided into 4 equal quadrants in a clockwise order (target, adjacent, opposite and adjacent). Located in the center of one of these quadrants was a 12 x 14 cm rectangular Plexiglas platform 19 cm in height (i.e. its surface was 1 cm below the water level). The platform remained in a fixed location throughout the training period of the place-trained animals. Following each day of training the tank was drained and cleaned. Fixed location training procedure. Injections were administered as described above 5 min prior to each day's training session. A trial began when the rat was placed in the pool facing the wall of the tank at one of the starting positions which varied from trial to trial in a quasirandom order. The rat was allowed to swim until it located and climbed onto the escape platform. Rats were gently guided to the platform if they failed to locate it within 90 s. The rat remained on the platform for 30 s before being removed. Two additional trials were given with 5-10 s intertrial intervals. Following the third trial the rat was returned to its home cage. The rats received a total of 12 trials during the 4 consecutive days of training. Escape latencies were measured as the time from release in the pool until escape onto the platform. All trials were videotaped through a wide angle lens attached to a camera mounted above the tank. Random vs place training. This experiment was conducted in order to examine the acquisition of spatial learning under our training conditions, and no drugs were administered. Two training conditions were used. Random-trained rats were trained to locate a submerged platform whose location was varied in a quasirandom order from trial to trial among the 4 possible locations of the platform. As the platform was never in the same location on two consecutive trials, the rat could not use the spatial environment to locate it. Place-trained animals were required to locate a submerged platform that was not visible to the rat, but was always located in the same position with respect to the spatial environment, thereby allowing the rat to determine the platform location by using the distal cues. Animals were placed in the pool at the designated starting locations, and their starting position was varied from trial to trial in a quasirandom order. The training was conducted according to the general procedures described above. Free-swim trial. This 60-s trial was carried out 4 days after the completion of the 4 training sessions. No drug or saline injections were administered prior to the test. The four possible positions of the platform and the limits of the 4 quadrants were marked on the video screen to indicate their exact surface area. The escape platform was removed from the tank and the rat was placed in the pool for 60 s. From video tapes made during this free swim it was possible to calculate 3 measures of spatial behavior: (1) latency to cross the target platform location - - the time in seconds to cross the target platform marked in the screen; (2) quadrant time - - the number of seconds spent by the rat in the target quadrant where the platform was formerly placed; and (3) platform crossings - - the number of times the rat traversed the former position of the escape platform as well as the crossings made on the 3 remaining platform positions (one for each of the remaining quadrants). These 3 measures were used to assess the degree of spatial bias of the rats
229 toward the target quadrant during the free swim. Retraining. After the 60-s free-swim trial, the platform was placed in its former location and the rats were guided to it and allowed to remain on it for 30 s. Immediately afterwards, they were given 3 retraining trials, according to the general training procedures, to determine whether the prior drug injection procedures produced tissue damage that would have significant behavioral consequences in this task.
Drugs Muscimol (Sigma Chemical Co.) was dissolved in saline solution and injected in doses of 0.05, 0.5, 1.0 and 5.0 nmol. Choline chloride was purchased from Sigma Chemical Co. and [3H]choline chloride (80.0 Ci/mmol) was purchased from New England Nuclear.
Histology One week after the behavioral studies the cannulae placements were verified histologically. The rats were anesthetized with an overdose of sodium pentobarbital and perfused through the heart with saline solution followed by 10% formaldehyde solution. Slices (40/~m) indicating the position of the cannulae were stained with Cresyl violet. One rat with incorrect placement of the cannula was excluded from the analysis. Cannula placement of animals that were sacrificed for biochemical studies on the hippocampus after muscimol injection in the medial septum, were also verified histologically.
Statistics The data were analyzed by one-way or two-way analysis of variance (ANOVA) followed by the Fisher PLSD test for individual mean comparisons.
< 0.0001) and no interaction (Fl1,154 = 1.2; n.s.). A relevant observation from this set of data that can be compared with the rest of the experiments is that random-trained animals always average more than 30 s to locate the escape platform, while place-trained animals were able to escape in less than 15 s by the fourth day of training (trial 10 = 13.8 s). A n analysis of the data by days show the same results, with a training effect (F1,14 = 10.6; P < 0.01), a days effect (F3,42
=
16.9; P
