Behavioural Brahz Research, 50 (1992) 167-175 9 1992 Elsevier Science Publishers B.V. All rights reser,'ed. 0166-4328/92]$05.00
167
BBR 01343
Limbic seizures, but not kindling, reversibly impair place learning in the Morris water maze R o b e r t K. M c N a m a r a ,
R. D u n c a n K i r k b y , G r e g o r y E. d e P a p e a n d M i c h a e l E. C o r c o r a n Department of Ps)'cholog)', University of Victoria, Vie., BC (Canada) (Received 11 March 1992) (Revised version received 20 May 1992) (Accepted 9 June 1992)
Key words: Kindling; Seizure; Medial perforant path; ttippocampus; Septum; Amygdala; Place learning; Cue learning; Reference memory; Morris water maze
We investigated tl/e effects of kindling and kindled seizures in different limbic structures on place and cue learning in the Morris water maze. The triggering of seizfires by stinmlation of the perforant path, septum, or anlygdala prior to daily training impaired place learning, but had little effect on visible platform training or swim speed. Seizures triggered by stimulation of the medial perforant path after daily training also impaired place learning. Conversely, place learning proceeded normally in rats tested 24 h after kindling triggered by stimulation of the perforant path, septum, or amygdala, indicating that kindling per se does not affect place learning. Each group was able to learn the location of a reversed platform when pretraining seizures were discontinued; and perforant path and septal kindled rats, but not amygdaloid kindled rats, were impaired at learning the location of a reversed platform when seizures were triggered before training. The results confirm previous reports that limbic seizures produce amnesia, but they contradict the finding that hippocampal kindling impairs learning on tasks sensitive to hippocampal lesions.
INTRODUCTION
Kindling involves the repeated administration of low intensity electrical stimulation that eventually leads to development of generalized seizures 7. Kindling results in a heightened state of seizure susceptibility that can last for yeast after cessation of stimulation5'7. Although the anaygdala is the most common site for kindling, stimulation of most sites in the limbic forebrain, including the hippocampus and septum, can kindle seizures 7, as can stimulation ofneocortex and paleocortex 23'27'31, certain thalamic nuclei 2, and striatum 26. In addition to their ability to support kindling, several limbic sites also subserve a mnemonic function. Lesions of the hippocampus 19'3~ or the medial seprum 8"9 impair place learning in the Morris water maze, an aversively motivated spatial memory task 18. Although amygdaloid lesions do not impair place learning in the Morris water maze 29, they do impair avoidance conditioning 14. Because these limbic structures are involved in both epileptogenesis and mnemonic
Correspondence: M.E. Corcoran, Department of Psychology, P.O. Box 3050, University of Victoria, Victoria, B.C., Canada, V8W 3P5. Fax (1)(604)721-8929.
processes, it is not surprising that seizure activity in them should have effects on memory. For example, patients with temporal lobe epilepsy frequently display chronic impairments ofmemory before, during and after epileptic episodes 6, and brain stimulation or seizure induction of limbic sites in rats produces amnesia in a variety of paradigms (see ref. 10 for a review). More specifically, seizure propagation in hippocampal field CAI produces a retrograde but not anterograde impairment of working (trial-dependent) but not reference (trial-independent) memory 12"2~ Thus seizure activity in limbic structures produces amnesia in both lmmans and infrahumans. Hippocampal kindling has been shown to both enhance and impair memory. For example, perforant path (PP)-kindled rabbits acquired a discriminative response more quickly than controls but were impaired during reversal training24. The selective impairment of reversal learning is similar to the effects of a bilateral hippocampectomy ~. Similarly, kindling with stimulation of hippocampal field CAI produced deficits in both working memory t3 and reference memory t5 on the radial arm maze. Together, these findings suggest that hippocampal kindling produces deficits in learning and memory that resemble hippocampal lesions. In the following experiment, generalized seizures
168 were induced in the hippocampus via stimulation of the PP, either before or just after daily training on a stationary platform version of the Morris water maze. From the findings discussed above ~2'2~ such seizures would be expected to have little effect on Morris maze performance in either an anterograde or retrograde direction. Moreover, the effect of PP kindling on place learning was assessed. If hippocampal kindling mimics hippocampal lesions, as suggested above, PP kindling should produce a severe impairment in place learning. Additionally, neuroanatomical specificity of seizures and kindling was assessed by stimulating the septum and the amygdala in different rats. From the lesion data, it would be expected that septal but not amygdaIoid seizures or kindling would impair place learning if nmemonic effects are mediated locally.
MATERIAES AND METIIODS
Sttbjects Seventy'six male hooded rats of the Long-Evans strain served as subjects. They were housed in pairs in shoebox cages or individually in suspended stainless steel cages, with food and water available ad libitum. The vivarium was maintained on a 12:12 h light-dark cycle, and testing was conducted during the light phase of the cycle. The rats weighed approximately 400 g at the time of surgery.
Surgery Under pentobarbital anesthesia (60 mg/kg), chronic bipolar electrodes were implanted, relative to bregma, into the following sites in different rats: septum ( + 0.8 mm AP; 1.0 mm ML; 4.5 mm DV from dura) and entorhinal cortex ( - 6 . 5 mm AP; 5.0 mm ML; 3.5 mm DV from dura); bilateral amygdala ( - 2 . 8 mm AP; 5.2 mm ML; 7.3 mm DV from dura); unilateral or bilateral dentate gyrus ( - 3 . 5 mm AP; 2.0 mm ML; ~ 3 . 4 m m DV from dura) and PP ( - 7 . 9 m m AP; 4.2 mm ML; 2.9 mm DV from dura). The incisor bar was set at -3.9 for all placements. Electrodes aimed at the dentate gyrus and PP were implanted under electrophysiological guidance. As both PP and dentate gyrus electrodes were lowered, a storage oscilloscope displayed field potentials recorded from the electrode in the dentate gyrus, evoked by single square wave pulses delivered to the PP electrode (0.1 ms, 0.2 Hz, 500 #A). The positions and depths of the electrodes were adjusted to produce an evoked potential in the dentate gyrus with a population spike of a maximum amplitude and minimum threshold on the rising edge of the EPSP. Recording electrodes consisted of a single strand of
stainless-steel wire, 76 #m in diameter and coated with enamel to a total diameter of 114 #m, that was bare of insulation 0.25-0.5 mm at the tip. Serving as currentreturn and ground-reference electrodes were two skull screws, connected to uninsulated stainless-steel wire. Gold-plated Amphenol pins soldered to the ends of the electrodes were inserted into a plastic electrode pedestal, which was affixed to the skull using dental acrylic and two additional anchoring skull screws.
Kindling After a 2-week recovery period, the threshold for afterdischarge (AD) was determined. Electrical stimulation consisted of a 1-s train of constant current balanced biphasic square-wave pulses, with a pulse width of 1.0 ms and a frequency of 60/s. A series of stinmlations was delivered (l/rain) starting at 30 #A (base-topeak) and increasing in 20 #A steps until AD was elicited. Threshold was arbitrarily defined as the lowest intensity to evoke AD, and rats were stinmlated once daily at an intensity 100 llA above their threshold. EEG was recorded for 10 or more s before and at least 1 min after each stimulation, and the intensity of behavioral seizures was classified according to Racine's scale zz. When three stage 5 generalized seizures had been evoked on three consecutive days, stimulation was discontinued for rats in the kindled group (see below), for which training in the maze began 24 h after the last seizure. Rats in the bilateral kindled group received unilateral stimulation until three stage 5 seizures were elicited; 24 h later stimulation was delivered once daily to the contralateral site until an additional three stage 5 seizures were triggered, and maze training began 24 h later. Stimulation that induced generalized seizures continued to be applied to rats in the seizure group (see below) prior to daily maze testing; over the course of training these rats experienced a total of 8 generalized seizures. Behavioral testing was conducted 25-45 min after seizure induction, an interval sufficient to allow the restoration of normal EEG activity. Implanted controls received similar handling but no electrical stimulation.
Group assignment After kindling was completed, rats were separated into the following ten treatment groups: (1)PP implanted control group (unilateral electrodes implanted, n = 10; bilateral electrodes implanted, n = 6); (2) PP seizure pretraining group (n = 6); (3) PP seizure posttraining group (n = 6); (4) PP kindled group (unilateral, n = 6; bilateral, n = 6); (5)septal implanted control group (n = 5); (6) septal seizure group (n = 6); (7) septal kindled group (n=5); (8)amygdala implanted control
169 group (n=5); (9)amygdala seizure group (n=8); (10) amygdala kindled group (n = 7).
Apparatus and procedure The Morris water maze consisted of a circular pool (diameter: 150 cm, height: 45 cm), with a featureless white inner surface. The pool was filled to a height of 25 cm with 22 ~ ( + 1 ~ water, in which 1,500 ml of powdered skim milk was dissolved. The hidden escape platform was a clear Plexiglas stand (13 x 13 cm) submerged 2 cm below the water surface so that it was invisible at water level. The visible platform was a black stand (13 x 13 cm) that protruded 5 cm above the surface of the water.
so that rats in the kindled group during acquisition now received pretraining seizures and rats in the seizure group during acquisition no longer received pretraining stimulation. After the final trial on the last day of reversal training, a 60-s probe trial was given. For both phases of the experiment, swim speed was calculated by dividing distance swum (cm) by escape latency.
Histology At the competition of behavioral testing, rats were anesthetized with an overdose of sodium pentobarbital, and the brains were perfused with normal saline followed by 10~ formalin. Sections 80/1m thick were taken and stained with Cresyl violet.
Acquisition phase Statistical analysis
During acquisition, the submerged escape platform was located in the center of the northwest quadrant. All groups were given 4 trials per day for 8 consecutive days. On each trial, the rat was placed in the water facing the pool "~vall at one of 4 randomly determined starting locations (north' south, east or west pole). The rat's swim path and escape latency were measured with a video tracking system (Chromotrack; SD Instruments). Once the rat had located the platform, it was permitted to remain on it for 10 s. If the rat did not locate the platform within 60 s, it was gently guided to it and permitted to remain on it for 10 s. After each trial, the rat was returned to a holding cage positioned 90 cm under a 250 W brooding lamp (for warmth) and was allowed to remain there for the intertrial interval, approximately 3 min. After the final trial on the last day of initial training, a probe trial (postacquisition) was given to assess the strength and accuracy of initial learning in the control and pretrial seizure groups. In the posttrial groups, the probe trial was given on the following day. Rats were required to swim in the pool without an escape platform for 60 s. All rats were released from the southern pole and the time spent in each quadrant was recorded. The following day, a cue task was given in which rats were required to navigate to a visible platform located in the center of a different quadrant on each trial (to prevent the accurate use of spatial cues). Escape latencies were recorded.
Table I shows the mean number of ADs required to kindle three stage 5 seizures, the mean AD thresholds, and the mean duration of stage 5 seizures for the seizure and kindled groups at each stimulation site. The AD threshold was significantly higher in the PP group than in the septal and amygdaloid groups (Ps
167
BBR 01343
Limbic seizures, but not kindling, reversibly impair place learning in the Morris water maze R o b e r t K. M c N a m a r a ,
R. D u n c a n K i r k b y , G r e g o r y E. d e P a p e a n d M i c h a e l E. C o r c o r a n Department of Ps)'cholog)', University of Victoria, Vie., BC (Canada) (Received 11 March 1992) (Revised version received 20 May 1992) (Accepted 9 June 1992)
Key words: Kindling; Seizure; Medial perforant path; ttippocampus; Septum; Amygdala; Place learning; Cue learning; Reference memory; Morris water maze
We investigated tl/e effects of kindling and kindled seizures in different limbic structures on place and cue learning in the Morris water maze. The triggering of seizfires by stinmlation of the perforant path, septum, or anlygdala prior to daily training impaired place learning, but had little effect on visible platform training or swim speed. Seizures triggered by stimulation of the medial perforant path after daily training also impaired place learning. Conversely, place learning proceeded normally in rats tested 24 h after kindling triggered by stimulation of the perforant path, septum, or amygdala, indicating that kindling per se does not affect place learning. Each group was able to learn the location of a reversed platform when pretraining seizures were discontinued; and perforant path and septal kindled rats, but not amygdaloid kindled rats, were impaired at learning the location of a reversed platform when seizures were triggered before training. The results confirm previous reports that limbic seizures produce amnesia, but they contradict the finding that hippocampal kindling impairs learning on tasks sensitive to hippocampal lesions.
INTRODUCTION
Kindling involves the repeated administration of low intensity electrical stimulation that eventually leads to development of generalized seizures 7. Kindling results in a heightened state of seizure susceptibility that can last for yeast after cessation of stimulation5'7. Although the anaygdala is the most common site for kindling, stimulation of most sites in the limbic forebrain, including the hippocampus and septum, can kindle seizures 7, as can stimulation ofneocortex and paleocortex 23'27'31, certain thalamic nuclei 2, and striatum 26. In addition to their ability to support kindling, several limbic sites also subserve a mnemonic function. Lesions of the hippocampus 19'3~ or the medial seprum 8"9 impair place learning in the Morris water maze, an aversively motivated spatial memory task 18. Although amygdaloid lesions do not impair place learning in the Morris water maze 29, they do impair avoidance conditioning 14. Because these limbic structures are involved in both epileptogenesis and mnemonic
Correspondence: M.E. Corcoran, Department of Psychology, P.O. Box 3050, University of Victoria, Victoria, B.C., Canada, V8W 3P5. Fax (1)(604)721-8929.
processes, it is not surprising that seizure activity in them should have effects on memory. For example, patients with temporal lobe epilepsy frequently display chronic impairments ofmemory before, during and after epileptic episodes 6, and brain stimulation or seizure induction of limbic sites in rats produces amnesia in a variety of paradigms (see ref. 10 for a review). More specifically, seizure propagation in hippocampal field CAI produces a retrograde but not anterograde impairment of working (trial-dependent) but not reference (trial-independent) memory 12"2~ Thus seizure activity in limbic structures produces amnesia in both lmmans and infrahumans. Hippocampal kindling has been shown to both enhance and impair memory. For example, perforant path (PP)-kindled rabbits acquired a discriminative response more quickly than controls but were impaired during reversal training24. The selective impairment of reversal learning is similar to the effects of a bilateral hippocampectomy ~. Similarly, kindling with stimulation of hippocampal field CAI produced deficits in both working memory t3 and reference memory t5 on the radial arm maze. Together, these findings suggest that hippocampal kindling produces deficits in learning and memory that resemble hippocampal lesions. In the following experiment, generalized seizures
168 were induced in the hippocampus via stimulation of the PP, either before or just after daily training on a stationary platform version of the Morris water maze. From the findings discussed above ~2'2~ such seizures would be expected to have little effect on Morris maze performance in either an anterograde or retrograde direction. Moreover, the effect of PP kindling on place learning was assessed. If hippocampal kindling mimics hippocampal lesions, as suggested above, PP kindling should produce a severe impairment in place learning. Additionally, neuroanatomical specificity of seizures and kindling was assessed by stimulating the septum and the amygdala in different rats. From the lesion data, it would be expected that septal but not amygdaIoid seizures or kindling would impair place learning if nmemonic effects are mediated locally.
MATERIAES AND METIIODS
Sttbjects Seventy'six male hooded rats of the Long-Evans strain served as subjects. They were housed in pairs in shoebox cages or individually in suspended stainless steel cages, with food and water available ad libitum. The vivarium was maintained on a 12:12 h light-dark cycle, and testing was conducted during the light phase of the cycle. The rats weighed approximately 400 g at the time of surgery.
Surgery Under pentobarbital anesthesia (60 mg/kg), chronic bipolar electrodes were implanted, relative to bregma, into the following sites in different rats: septum ( + 0.8 mm AP; 1.0 mm ML; 4.5 mm DV from dura) and entorhinal cortex ( - 6 . 5 mm AP; 5.0 mm ML; 3.5 mm DV from dura); bilateral amygdala ( - 2 . 8 mm AP; 5.2 mm ML; 7.3 mm DV from dura); unilateral or bilateral dentate gyrus ( - 3 . 5 mm AP; 2.0 mm ML; ~ 3 . 4 m m DV from dura) and PP ( - 7 . 9 m m AP; 4.2 mm ML; 2.9 mm DV from dura). The incisor bar was set at -3.9 for all placements. Electrodes aimed at the dentate gyrus and PP were implanted under electrophysiological guidance. As both PP and dentate gyrus electrodes were lowered, a storage oscilloscope displayed field potentials recorded from the electrode in the dentate gyrus, evoked by single square wave pulses delivered to the PP electrode (0.1 ms, 0.2 Hz, 500 #A). The positions and depths of the electrodes were adjusted to produce an evoked potential in the dentate gyrus with a population spike of a maximum amplitude and minimum threshold on the rising edge of the EPSP. Recording electrodes consisted of a single strand of
stainless-steel wire, 76 #m in diameter and coated with enamel to a total diameter of 114 #m, that was bare of insulation 0.25-0.5 mm at the tip. Serving as currentreturn and ground-reference electrodes were two skull screws, connected to uninsulated stainless-steel wire. Gold-plated Amphenol pins soldered to the ends of the electrodes were inserted into a plastic electrode pedestal, which was affixed to the skull using dental acrylic and two additional anchoring skull screws.
Kindling After a 2-week recovery period, the threshold for afterdischarge (AD) was determined. Electrical stimulation consisted of a 1-s train of constant current balanced biphasic square-wave pulses, with a pulse width of 1.0 ms and a frequency of 60/s. A series of stinmlations was delivered (l/rain) starting at 30 #A (base-topeak) and increasing in 20 #A steps until AD was elicited. Threshold was arbitrarily defined as the lowest intensity to evoke AD, and rats were stinmlated once daily at an intensity 100 llA above their threshold. EEG was recorded for 10 or more s before and at least 1 min after each stimulation, and the intensity of behavioral seizures was classified according to Racine's scale zz. When three stage 5 generalized seizures had been evoked on three consecutive days, stimulation was discontinued for rats in the kindled group (see below), for which training in the maze began 24 h after the last seizure. Rats in the bilateral kindled group received unilateral stimulation until three stage 5 seizures were elicited; 24 h later stimulation was delivered once daily to the contralateral site until an additional three stage 5 seizures were triggered, and maze training began 24 h later. Stimulation that induced generalized seizures continued to be applied to rats in the seizure group (see below) prior to daily maze testing; over the course of training these rats experienced a total of 8 generalized seizures. Behavioral testing was conducted 25-45 min after seizure induction, an interval sufficient to allow the restoration of normal EEG activity. Implanted controls received similar handling but no electrical stimulation.
Group assignment After kindling was completed, rats were separated into the following ten treatment groups: (1)PP implanted control group (unilateral electrodes implanted, n = 10; bilateral electrodes implanted, n = 6); (2) PP seizure pretraining group (n = 6); (3) PP seizure posttraining group (n = 6); (4) PP kindled group (unilateral, n = 6; bilateral, n = 6); (5)septal implanted control group (n = 5); (6) septal seizure group (n = 6); (7) septal kindled group (n=5); (8)amygdala implanted control
169 group (n=5); (9)amygdala seizure group (n=8); (10) amygdala kindled group (n = 7).
Apparatus and procedure The Morris water maze consisted of a circular pool (diameter: 150 cm, height: 45 cm), with a featureless white inner surface. The pool was filled to a height of 25 cm with 22 ~ ( + 1 ~ water, in which 1,500 ml of powdered skim milk was dissolved. The hidden escape platform was a clear Plexiglas stand (13 x 13 cm) submerged 2 cm below the water surface so that it was invisible at water level. The visible platform was a black stand (13 x 13 cm) that protruded 5 cm above the surface of the water.
so that rats in the kindled group during acquisition now received pretraining seizures and rats in the seizure group during acquisition no longer received pretraining stimulation. After the final trial on the last day of reversal training, a 60-s probe trial was given. For both phases of the experiment, swim speed was calculated by dividing distance swum (cm) by escape latency.
Histology At the competition of behavioral testing, rats were anesthetized with an overdose of sodium pentobarbital, and the brains were perfused with normal saline followed by 10~ formalin. Sections 80/1m thick were taken and stained with Cresyl violet.
Acquisition phase Statistical analysis
During acquisition, the submerged escape platform was located in the center of the northwest quadrant. All groups were given 4 trials per day for 8 consecutive days. On each trial, the rat was placed in the water facing the pool "~vall at one of 4 randomly determined starting locations (north' south, east or west pole). The rat's swim path and escape latency were measured with a video tracking system (Chromotrack; SD Instruments). Once the rat had located the platform, it was permitted to remain on it for 10 s. If the rat did not locate the platform within 60 s, it was gently guided to it and permitted to remain on it for 10 s. After each trial, the rat was returned to a holding cage positioned 90 cm under a 250 W brooding lamp (for warmth) and was allowed to remain there for the intertrial interval, approximately 3 min. After the final trial on the last day of initial training, a probe trial (postacquisition) was given to assess the strength and accuracy of initial learning in the control and pretrial seizure groups. In the posttrial groups, the probe trial was given on the following day. Rats were required to swim in the pool without an escape platform for 60 s. All rats were released from the southern pole and the time spent in each quadrant was recorded. The following day, a cue task was given in which rats were required to navigate to a visible platform located in the center of a different quadrant on each trial (to prevent the accurate use of spatial cues). Escape latencies were recorded.
Table I shows the mean number of ADs required to kindle three stage 5 seizures, the mean AD thresholds, and the mean duration of stage 5 seizures for the seizure and kindled groups at each stimulation site. The AD threshold was significantly higher in the PP group than in the septal and amygdaloid groups (Ps
