Brain Research, 133 (1977) 277-289 © Elsevier/North-Holland Biomedical Press
277
EFFECTS OF H I P P O C A M P A L E L E C T R I C A L S T I M U L A T I O N ON L O N G T E R M M E M O R Y A N D ON C H O L I N E R G i C MECHANISMS IN T H R E E I N B R E D STRAINS OF MICE
ROBERT JAFFARD, ALIX EBEL, CLAUDE DESTRADE, TOM DURKIN, PAUL MANDEL and BERNARD CARDO Laboratoire de Psychophysiologie, Institut de Biologie Animale, Universitd de Bordeaux I, 33405 Talence and (A.E., T.D. and P.M.) Centre de Neurochimie, Centre National de Recherche Scientifique, 67000 Strasbourg (France)
(Accepted December 30th, 1976)
SUMMARY Two sets of experiments have been carried out in an attempt to determine the role of hippocampal cholinergic mechanisms in long-term memory storage. Three inbred strains of mice were presented with two different learning tasks in order to estimate their long-term retention abilities as well as changes in this ability after a post-trial hippocampal stimulation. In parallel experiments the enzymes involved in acetylcholine metabolism were studied under different experimental conditions. Our results indicate: (a) The capacity for long-term memory of the BALB/c line is much greater than that of either the C57BL/6 or C57BR strain. (b) Hippocampal post-trial electrical stimulation leads to an improvement of this capacity in the BALB/c strain. This phenomenon is less pronounced in C57BL/6 and non-existent in 'C57BR mice. (c) Choline acetyltransferase activity in the hippocampus is significantly higher in BALB/c than in the other two strains. In BALB/c this enzyme activity is greatly changed by the post-trial stimulation whereas in the C57BL/6 strain only a slight variation of enzyme activity is observed. No modification occurs in C57BR. The results suggest that the more active acetylcholine synthesizing enzyme in the hippocampus of BALB/c may be related to a greater acetylcholine availability, thus favoring the establishment of a long-term memory, perhaps by releasing greater amounts of acetylcholine in the hippocampus immediately after the learning session. The electrical stimulation of the hippocampus acts to magnify or accelerate this phenomenon. It is suggested that the efficiency of the stimulation would be related to the genetically determined higher cholinergic activity of the hippocampus. INTRODUCTION A large body of evidence exists to support the association of central cholinergic
278 mechanisms with behaviour2. It has particularly been suggested that central cholinergic systems may be involved in memory encoding storage, or retrievaP, v,25. The fact that hippocampus both receives important cholinergic inputs ~1 and probably plays a role in memory processes 13 suggests an hippocampal cholinergic control on some mechanisms involved in this function. Although frequently used, pharmacological studies have produced considerable disagreement about this problem so that direct neurochemical investigations have been developed23,2a. In the present experiment we have investigated the neurochemical correlates of several behavioural patterns using inbred strains of mice, as performed in previous studiesS,23,24. In the behavioural aspect of this study, experiments were performed in order to determine the long-term memory capacityis of 3 inbred strains of mice submitted to two quite different learning tasks and further, to analyse this retention ability after a post-trial electrical stimulation in the dorsal hippocampus 6. In order to investigate some of the neurochemical aspects of these problems, cholinergic mechanisms were evaluated by testing the two main enzymes involved in acetylcholine (ACh) synthesis and degradation, choline acetyltransferase (ChAc; EC 2.3.1.6) and acetylcholinesterase (ACHE; EC 3.1.1.7). These enzyme activities were measured in the dorsal hippocampus of the 3 inbred mice strains in normal conditions and ai'ter hippocampal stimulation. EXPERIMENT 1: BEHAVIOURAL DATA The first series of experiments was designed to compare learning and memory processes in the 3 inbred strains of mice. For this purpose animals were first submitted to two different learning experiments: a lever press conditioning on CRF (continuous reinforcement) and a passive avoidance test. Then, the post-trial effect of a subconvulsive electrical stimulation of the dorsal hippocampus was compared among strains in these two situations.
Methods General procedure Subjects were male mice of the BALB/c Orl, C57BL/6 Or1 and C57BR Or1 strains and were approximately 12 weeks old at the end of the experiments. At 8 weeks of age they were housed individually with ad lib. access to food and water in a constant temperature room (21 °C) maintained on a light-dark cycle (12 h-12 h). Surgery. According to a previously described procedure 3,6 animals were implanted bilaterally under general anaesthesia with bipolar electrodes in the CA1 field of dorsal hippocampus. The stereotaxic coordinates were the same for the 3 strains: 1.5 mm posterior to bregma, 1.3 mm lateral from the midline, and 1.8 mm below the calvarium. Deep anaesthesia was obtained by i.p. injections of sodium penthiobarbital (BALB/c: 100 mg/kg; C57BL/6:70 mg/kg and C57BR: 45 mg/kg.) Determination of after-discharge threshold values: this previously described
279 procedure 31 was aimed at determining the threshold intensity of a sinusoidal current (100 Hz) that produced hippocampal after-discharges. In behavioural and biochemical experiments, the intensity of stimulation was half of this threshold value (S/2)5, 8 so that there were no disturbances of hippocampal records.
Behaviour (1) Operant behaviour
Apparatus. The apparatus already described 6 was a Skinner box in which the lever and the food cup are separated by a small partition. Procedure. All implanted animals were allowed to recuperate for 8 days following determination of after-discharge thresholds. Throughout the course of the studies, all mice (implanted or not) were maintained at 81-84 ~ of their ad lib. body weight and were given their feeding approximately 20 min after each session 5,16,18. In the first experiment reminiscence, i.e., the retention interval effect on performances was studied on non-implanted animals3,16,18. For this purpose 30 animals of each strain were randomly assigned to 3 groups of 10 subjects. Mice of the first and second groups underwent a first learning session. For BALB/c and C57BL/6 this session lasted 15 min. For C57BR, such a procedure was inadequate because these animals being quicker to associate the bar-press with reinforcement 19 were consequently overtrained at the end of this first session. Finally C57BR mice were allowed to have 16 responses, i.e., approximately the same mean number of responses registered during the first session on BALB/c (16.4 ± 0.3) and C57BL/6 (17.2 ± 0.4). The second session (30 rain in duration) was initiated 2 min 30 sec or 24 h later. A third group (naive) underwent this session without preliminary training. The second experiment was designed to compare the effect of a post-first session hippocampal stimulation on performances registered 24 h later. For this purpose, 20 implanted animals of each strain were randomly assigned to 2 groups of 10 subjects. According to the previously described procedure, they were first submitted to the learning session; 30 sec after the end of this first session the first group was submitted to hippocampal stimulation for 80 sec at S/2 (see above). The second group received no stimulation. Twenty-four hours later performances were controlled on a second 30 rain session. (2) Passive avoidance Apparatus. The training cage was a two-compartment 20 cm high box 17. The small compartment (7.5 cm × 10.5 cm) constructed of clear plexiglass was illuminated with a 40 W lamp. It was connected by a circular guillotine door (6 cm in diameter) to a large compartment (11 cm × 15 cm) constructed of black plexiglass. The floor was made of metal rods 0.2 cm in width. The footshock (50 Hz) was delivered by a constant current generator. The delay intervals were automatically recorded. Procedure. Training consisted of placing subjects in the small compartment facing away from the closed door. After 30 sec, the door was opened and the animal was allowed to cross into the large compartment. As soon as it had crossed completely (4-paw criterion), the footshock (FS) was administered. The FS used (0.3 or 0.6 mA;
280 see Table II) were sufficient to drive all animals out of the black compartment. The door was then closed and the mouse immediately removed from the small compartment. Testing consisted of again placing subjects in the small compartment and to open the door 30 sec later for 5 min. Three parameters were used: latency (L) to completely cross into the black compartment (training and test) and the time (T) spent in the black compartment during the 5 rain test session; in addition, the number of rearing (R) observed in the small compartment during the 30 sec waiting period was used as an index of CER (conditioned emotional response)iS, 17. The same 60 animals used in the first experiment and 80 newly implanted mice served as subjects. As for the 60 animals used in the first experiment, the newly implanted mice were trained in the Skinner box 10 days before the beginning of experiment on passive avoidance. According to training conditions (footshock or not: FS or NFS) and to treatment (hippocampal stimulation or not: ST or NST) mice of each strain were randomly assigned to 3 groups. As in the first experiment ST mice were submitted to an electrical stimulation of the hippocampus (80 sec at S/2) 30 sec after the end of training. All mice were tested 24 h later.
Results (1) Hippocampal after-discharge (HAD) thresholds Results of the pooled implanted mice (behavioural and biochemical experiments) are summarized in Table I. The H A D threshold values were not significantly different between strains (Student's t comparison). However, C57BR mice more frequently exhibited low threshold values: 43 ~ of them had H A D for intensities less than 15 /~A; this ratio was only 29 ~ for C57BL/6 (Z~ = 3.29; P < 0.10) and 25 ~ for BALB/c (g 2 -- 4.36; P < 0.05). The functional significance of these results will not be discussed here.
(2) Operant conditioning Results of the first experiment are summarised in Fig. 1. First session: a trend analysis 1° showed no differences between performances of BALB/c and C57BL/6 mice (F(1,38) = 2.49; P > 0.10). Within this session, C57BR are quicker to associate the lever-press with food so that responses during the last 5 min are significantly superior to those of BALB/c (t(18) ~ 5.00; P
277
EFFECTS OF H I P P O C A M P A L E L E C T R I C A L S T I M U L A T I O N ON L O N G T E R M M E M O R Y A N D ON C H O L I N E R G i C MECHANISMS IN T H R E E I N B R E D STRAINS OF MICE
ROBERT JAFFARD, ALIX EBEL, CLAUDE DESTRADE, TOM DURKIN, PAUL MANDEL and BERNARD CARDO Laboratoire de Psychophysiologie, Institut de Biologie Animale, Universitd de Bordeaux I, 33405 Talence and (A.E., T.D. and P.M.) Centre de Neurochimie, Centre National de Recherche Scientifique, 67000 Strasbourg (France)
(Accepted December 30th, 1976)
SUMMARY Two sets of experiments have been carried out in an attempt to determine the role of hippocampal cholinergic mechanisms in long-term memory storage. Three inbred strains of mice were presented with two different learning tasks in order to estimate their long-term retention abilities as well as changes in this ability after a post-trial hippocampal stimulation. In parallel experiments the enzymes involved in acetylcholine metabolism were studied under different experimental conditions. Our results indicate: (a) The capacity for long-term memory of the BALB/c line is much greater than that of either the C57BL/6 or C57BR strain. (b) Hippocampal post-trial electrical stimulation leads to an improvement of this capacity in the BALB/c strain. This phenomenon is less pronounced in C57BL/6 and non-existent in 'C57BR mice. (c) Choline acetyltransferase activity in the hippocampus is significantly higher in BALB/c than in the other two strains. In BALB/c this enzyme activity is greatly changed by the post-trial stimulation whereas in the C57BL/6 strain only a slight variation of enzyme activity is observed. No modification occurs in C57BR. The results suggest that the more active acetylcholine synthesizing enzyme in the hippocampus of BALB/c may be related to a greater acetylcholine availability, thus favoring the establishment of a long-term memory, perhaps by releasing greater amounts of acetylcholine in the hippocampus immediately after the learning session. The electrical stimulation of the hippocampus acts to magnify or accelerate this phenomenon. It is suggested that the efficiency of the stimulation would be related to the genetically determined higher cholinergic activity of the hippocampus. INTRODUCTION A large body of evidence exists to support the association of central cholinergic
278 mechanisms with behaviour2. It has particularly been suggested that central cholinergic systems may be involved in memory encoding storage, or retrievaP, v,25. The fact that hippocampus both receives important cholinergic inputs ~1 and probably plays a role in memory processes 13 suggests an hippocampal cholinergic control on some mechanisms involved in this function. Although frequently used, pharmacological studies have produced considerable disagreement about this problem so that direct neurochemical investigations have been developed23,2a. In the present experiment we have investigated the neurochemical correlates of several behavioural patterns using inbred strains of mice, as performed in previous studiesS,23,24. In the behavioural aspect of this study, experiments were performed in order to determine the long-term memory capacityis of 3 inbred strains of mice submitted to two quite different learning tasks and further, to analyse this retention ability after a post-trial electrical stimulation in the dorsal hippocampus 6. In order to investigate some of the neurochemical aspects of these problems, cholinergic mechanisms were evaluated by testing the two main enzymes involved in acetylcholine (ACh) synthesis and degradation, choline acetyltransferase (ChAc; EC 2.3.1.6) and acetylcholinesterase (ACHE; EC 3.1.1.7). These enzyme activities were measured in the dorsal hippocampus of the 3 inbred mice strains in normal conditions and ai'ter hippocampal stimulation. EXPERIMENT 1: BEHAVIOURAL DATA The first series of experiments was designed to compare learning and memory processes in the 3 inbred strains of mice. For this purpose animals were first submitted to two different learning experiments: a lever press conditioning on CRF (continuous reinforcement) and a passive avoidance test. Then, the post-trial effect of a subconvulsive electrical stimulation of the dorsal hippocampus was compared among strains in these two situations.
Methods General procedure Subjects were male mice of the BALB/c Orl, C57BL/6 Or1 and C57BR Or1 strains and were approximately 12 weeks old at the end of the experiments. At 8 weeks of age they were housed individually with ad lib. access to food and water in a constant temperature room (21 °C) maintained on a light-dark cycle (12 h-12 h). Surgery. According to a previously described procedure 3,6 animals were implanted bilaterally under general anaesthesia with bipolar electrodes in the CA1 field of dorsal hippocampus. The stereotaxic coordinates were the same for the 3 strains: 1.5 mm posterior to bregma, 1.3 mm lateral from the midline, and 1.8 mm below the calvarium. Deep anaesthesia was obtained by i.p. injections of sodium penthiobarbital (BALB/c: 100 mg/kg; C57BL/6:70 mg/kg and C57BR: 45 mg/kg.) Determination of after-discharge threshold values: this previously described
279 procedure 31 was aimed at determining the threshold intensity of a sinusoidal current (100 Hz) that produced hippocampal after-discharges. In behavioural and biochemical experiments, the intensity of stimulation was half of this threshold value (S/2)5, 8 so that there were no disturbances of hippocampal records.
Behaviour (1) Operant behaviour
Apparatus. The apparatus already described 6 was a Skinner box in which the lever and the food cup are separated by a small partition. Procedure. All implanted animals were allowed to recuperate for 8 days following determination of after-discharge thresholds. Throughout the course of the studies, all mice (implanted or not) were maintained at 81-84 ~ of their ad lib. body weight and were given their feeding approximately 20 min after each session 5,16,18. In the first experiment reminiscence, i.e., the retention interval effect on performances was studied on non-implanted animals3,16,18. For this purpose 30 animals of each strain were randomly assigned to 3 groups of 10 subjects. Mice of the first and second groups underwent a first learning session. For BALB/c and C57BL/6 this session lasted 15 min. For C57BR, such a procedure was inadequate because these animals being quicker to associate the bar-press with reinforcement 19 were consequently overtrained at the end of this first session. Finally C57BR mice were allowed to have 16 responses, i.e., approximately the same mean number of responses registered during the first session on BALB/c (16.4 ± 0.3) and C57BL/6 (17.2 ± 0.4). The second session (30 rain in duration) was initiated 2 min 30 sec or 24 h later. A third group (naive) underwent this session without preliminary training. The second experiment was designed to compare the effect of a post-first session hippocampal stimulation on performances registered 24 h later. For this purpose, 20 implanted animals of each strain were randomly assigned to 2 groups of 10 subjects. According to the previously described procedure, they were first submitted to the learning session; 30 sec after the end of this first session the first group was submitted to hippocampal stimulation for 80 sec at S/2 (see above). The second group received no stimulation. Twenty-four hours later performances were controlled on a second 30 rain session. (2) Passive avoidance Apparatus. The training cage was a two-compartment 20 cm high box 17. The small compartment (7.5 cm × 10.5 cm) constructed of clear plexiglass was illuminated with a 40 W lamp. It was connected by a circular guillotine door (6 cm in diameter) to a large compartment (11 cm × 15 cm) constructed of black plexiglass. The floor was made of metal rods 0.2 cm in width. The footshock (50 Hz) was delivered by a constant current generator. The delay intervals were automatically recorded. Procedure. Training consisted of placing subjects in the small compartment facing away from the closed door. After 30 sec, the door was opened and the animal was allowed to cross into the large compartment. As soon as it had crossed completely (4-paw criterion), the footshock (FS) was administered. The FS used (0.3 or 0.6 mA;
280 see Table II) were sufficient to drive all animals out of the black compartment. The door was then closed and the mouse immediately removed from the small compartment. Testing consisted of again placing subjects in the small compartment and to open the door 30 sec later for 5 min. Three parameters were used: latency (L) to completely cross into the black compartment (training and test) and the time (T) spent in the black compartment during the 5 rain test session; in addition, the number of rearing (R) observed in the small compartment during the 30 sec waiting period was used as an index of CER (conditioned emotional response)iS, 17. The same 60 animals used in the first experiment and 80 newly implanted mice served as subjects. As for the 60 animals used in the first experiment, the newly implanted mice were trained in the Skinner box 10 days before the beginning of experiment on passive avoidance. According to training conditions (footshock or not: FS or NFS) and to treatment (hippocampal stimulation or not: ST or NST) mice of each strain were randomly assigned to 3 groups. As in the first experiment ST mice were submitted to an electrical stimulation of the hippocampus (80 sec at S/2) 30 sec after the end of training. All mice were tested 24 h later.
Results (1) Hippocampal after-discharge (HAD) thresholds Results of the pooled implanted mice (behavioural and biochemical experiments) are summarized in Table I. The H A D threshold values were not significantly different between strains (Student's t comparison). However, C57BR mice more frequently exhibited low threshold values: 43 ~ of them had H A D for intensities less than 15 /~A; this ratio was only 29 ~ for C57BL/6 (Z~ = 3.29; P < 0.10) and 25 ~ for BALB/c (g 2 -- 4.36; P < 0.05). The functional significance of these results will not be discussed here.
(2) Operant conditioning Results of the first experiment are summarised in Fig. 1. First session: a trend analysis 1° showed no differences between performances of BALB/c and C57BL/6 mice (F(1,38) = 2.49; P > 0.10). Within this session, C57BR are quicker to associate the lever-press with food so that responses during the last 5 min are significantly superior to those of BALB/c (t(18) ~ 5.00; P
![[Effects of hippocampal stimulation on reminiscence in two inbred strains of mice].](/assets/img/hold.png)
