- Email: [email protected]
Spatial memory under acute cold and restraint stress
Physiology & Behavior, Vol. 64, No. 5, pp. 605– 609, 1998 © 1998 Elsevier Science Inc. All rights reserved. Printed in the U.S.A. 0031-9384/98 $19.00 1 .00
PII S0031-9384(98)00091–2
Spatial Memory Under Acute Cold and Restraint Stress MICHAEL J. STILLMAN,*†1 BARBARA SHUKITT-HALE,*† AHARON LEVY,*‡ AND HARRIS R. LIEBERMAN*2 *Military Nutrition and Biochemistry Division, United States Army Research Institute of Environmental Medicine, Natick, MA 01760-5007; †GEO-CENTERS, Inc., Newton Centre, MA 02159; and ‡Department of Pharmacology, Israel Institute for Biological Research, Ness Ziona, 70450 Israel Received 6 June 1997; Accepted 3 March 1998 STILLMAN, M.J., B. SHUKITT-HALE, A. LEVY AND H.R. LIEBERMAN. Spatial memory under acute cold and restraint stress. PHYSIOL BEHAV 64(5) 605– 609, 1998.—This study examined spatial memory as measured by radial arm maze (RAM) performance after exposure to two stress conditions and a normothermic-unrestrained control condition. Male Fischer 344 rats were trained on the win-shift RAM procedure for 7 days, by which time they achieved asymptotic performance. The next day, rats in the two stress groups were exposed to 15 min of restraint in either 37°C water (normothermic-restraint) or in 20°C water (cold-restraint). Rats were then allowed 40 min in a dry cage before being tested in the RAM. Performance was measured using the following dependent variables: number of correct out of the first eight choices, total number of choices, and time per choice. There were statistically significant effects of stress on all these variables; performance decrements were observed in both stress conditions relative to the normothermicunrestrained condition. Normothermic-restrained rats displayed less impairment than cold-restrained rats on the stress day. Performance of normothermic-restrained rats returned to baseline levels the day after stress, whereas performance for the cold-restrained rats typically did not. This study demonstrates that: 1) restraint and cold stress impair performance on a memory task; and 2) impairment extent is related to stress severity. One of the mechanisms responsible for the observed behavioral deficits under cold stress may involve altered cholinergic function, because we previously demonstrated that hippocampal acetylcholine levels also decrease in relation to the severity of cold stress. © 1998 Elsevier Science Inc. Learning
Hypothermia
Performance
Radial arm maze
Despite the heterogeneity of stressor type and duration, many studies have demonstrated that stressed animals cannot learn as readily as nonstressed animals. Restraint has been shown to impair radial arm maze (RAM) performance (8), and to disrupt shock discrimination performance (7). Similarly, cold exposure produces deficits in delayed-matching-to-sample performance, a measure of working memory in both humans and other animals (1,2,16,18,19). In addition, spatial memory impairments in the Morris water maze task were observed after cold stress (14,15,20). The majority of investigations thus far generally has examined different intensities of one type of stressor, predominantly demonstrating behavioral effects proportional to stressor intensity (e.g., 14). In this study, we examined the behavioral effects of one stressor (restraint) alone and in combination with another stressor (cold). It was hypothesized that the combination of thermal and restraint stressors would be more severe, as suggested by one of our recent neurochemical investigations in which acetylcholine (ACh) release was decreased during cold and restraint relative to normothermic restraint conditions (17). The present research was
Rats
performed to compare spatial memory performance as measured by the RAM task among cold-restrained, normothermic-restrained, and normothermic-unrestrained rats.
MATERIALS AND METHODS
Subjects A total of 33 male Fischer 344 rats (Charles River Labs, Kingston, NY, USA), weighing 250 –360 g at time of testing, were used in this study. The animals were randomly divided into three groups: twelve rats were cold-restrained, eleven were normothermic-restrained, and ten rats served as nonstressed, normothermicunrestrained controls. Rats were restricted to 1 h per day of water for 5 days before the start of RAM training and continuing throughout the study. They were housed individually in wire mesh cages, and were maintained on a 12-h light/dark cycle (lights on at 0600 hours).
1
Michael J. Stillman, PhD, is currently at DendWrite Communications, 33 Dinsmore Avenue, Suite 602, Framingham, MA 01702, USA Requests for reprints should be addressed to Dr. Harris R. Lieberman, Military Nutrition and Biochemistry Division, United States Army Research Institute of Environmental Medicine, Natick, MA 01760-5007, USA. E-mail: [email protected] 2
605
606 RAM Spatial memory was assessed in the present research via performance on the RAM (11). Working memory is required while the rat is selecting the next arm to visit, apparently “keeping track of” which arms were previously selected. Various control experiments (11,12) have eliminated extraneous nonmemory processes (e.g., odor cues) as well as intramaze cues as confounding factors and strongly suggest that the rat performs the RAM test by utilizing extramaze cues. One possibility is that the rat builds a cognitive map of the environment to learn the maze, and utilizes its working memory to monitor which arms have been visited. The win-shift procedure, employed in this study, requires the rat to retrieve the reward at the end of each of the eight arms without returning to any previously visited arm because visited arms are not rebaited; an intact hippocampus is necessary for accurate performance in the win-shift paradigm (13). The Plexiglas RAM consisted of eight equally spaced arms (70-cm long, 12-cm wide, 15-cm high) radiating from an octagonal central area 30 cm in diameter. A Plexiglas ring surrounded the central area, preventing the rat from entering any of the arms. An overhead pulley and rope enabled the experimenter to raise the ring and allow the rat to leave the central area and select from any of the eight arms. Various extramaze cues were present in the testing room, including posters, a laboratory cart, and miscellaneous laboratory equipment. Each maze arm contained a small dish at its end that was baited with 0.25 mL of 0.30% saccharin solution. Procedure Rats were water restricted for 5 days (1 h/day) before RAM training. To avoid taste aversion to a novel stimulus during maze training, the saccharin solution was introduced in the home cage on Days 3 and 4 of the water-restriction schedule. For the last 2 of these 5 days each rat was allowed 10 min to explore the RAM, after which training began. The saccharin solution was scattered about the maze during these last 2 days, to facilitate exploration of the maze. Rats were tested for 9 days in the RAM, including 7 training days, 1 Stress Day, and 1 Poststress day. Each RAM trial commenced when the rat was placed in the center of the maze for 15 s and then permitted free access to all of the baited arms. In accordance with the win-shift procedure (in which visited arms are not rebaited), reentries into previously visited arms were scored as errors. Each trial was terminated when all eight arms were visited, when sixteen entries were made, or at the end of 600 s, whichever came first. The animal was determined to have made a choice once it moved from the central area and all four paws were located in a maze arm. The following measures were recorded by experimenter observation: the order of entry into the maze arms, the number of revisits, the total number of choices, the time to complete a session, and the number of correct entries out of the first eight choices. Time per choice was calculated as total time for completion of the session divided by the total number of entries. Stress Day occurred after the 7 training days. Stressed rats were placed in a small, custom-designed, metal and Velcro restraining apparatus during stressor exposure. Acute stress was produced by immersing the rat up to the neck for 15 min in a water bath (RC20, Lauda) filled with either 20 6 2°C water for the cold-restraint condition, or with 37 6 2°C water for the normothermic-restraint condition. Previous research in our laboratory showed that core body temperature decreased to approximately 30°C after 15 min after this cold stress procedure (17). Rats were then towel dried and placed in a holding container for 40 min before RAM testing.
STILLMAN ET AL. Normothermic-unrestrained rats remained in the holding container for 55 min before RAM testing. All rats underwent final RAM testing 24 h later under no-stress conditions (Poststress Day) to determine the extent of performance recovery. Statistical Analysis The following measures served as dependent variables: number of correct entries in the first eight choices, total number of choices, and time per choice. Each measure was analyzed by a two-way ANOVA using stress condition as the between-subjects factor and day (Day 7, Stress Day, and Poststress Day) as the within-subject factor. All statistical tests utilized a p , 0.05 criterion for hypothesis testing. Post hoc comparisons were performed, when appropriate, using Duncan’s multiple range test. RESULTS
All subjects improved performance to asymptotic levels during the 7 training days. Overall, performance decrements were observed on Stress Day in both stress conditions relative to the normothermic-unrestrained condition; moreover, the performance of cold-restrained rats was inferior to that of the normothermicrestrained rats. Performance returned to baseline (Day 7) levels the day after stress for the normothermic-restrained group, but typically not for the cold-restrained group. The number of correct out of the first eight choices for Day 7, Stress Day, and Poststress Day is shown in Fig. 1. ANOVA indicated a significant condition effect [F(2, 30) 5 19.41, p , 0.001], a significant day effect [F(2, 60) 5 51.73, p , 0.001], and a significant condition-by-day interaction [F(4, 60) 5 15.01, p , 0.001]. Post hoc analyses revealed that number correct performance on Stress Day in the cold-restrained (p , 0.01) and normothermic-restrained (p , 0.01) groups was lower than that of the normothermic-unrestrained group; also, performance was lower in the cold-restrained group (p , 0.01) compared to the normothermic-restrained group. On Poststress Day, performance of the coldrestrained rats (p , 0.05) was still impaired relative to the other groups, which were not different from each other. There were no significant differences between the groups on Day 7. Over time, post hoc analyses for both stress groups for the number of correct out of the first eight choices revealed that performance on Stress Day was significantly impaired compared to that of Day 7 (p , 0.01) and Poststress day (p , 0.01). However, performance on Poststress Day for the cold-restrained group was still inferior to that of Day 7 (p , 0.05), whereas Poststress Day performance for the normothermic restrained rats was not different from Day 7 levels. Performance did not change over time in the normothermic-unrestrained group. The total number of choices for Day 7, Stress Day, and Poststress Day is shown in Fig. 2. ANOVA indicated a significant condition effect [F(2, 30) 5 19.92, p , 0.001], a significant day effect [F(2, 60) 5 42.32, p , 0.001], and a significant conditionby-day interaction [F(4, 60) 5 16.58, p , 0.001]. Post hoc analyses revealed that choice performance on Stress Day in the cold-restrained (p , 0.01) and normothermic-restrained (p , 0.01) groups was again lower than that of the normothermic-unrestrained group; also, the cold-restrained group (p , 0.01) performance was impaired relative to the normothermic-restrained group. On Poststress Day, performance of the cold-restrained rats was impaired relative to the normothermic-unrestrained (p , 0.05) and normothermic-restrained (p , 0.01) groups, which were not different from each other. Again, there were no significant differences between the groups on Day 7. Over time, post hoc analyses for both stress groups for the total number of choices revealed that Stress Day performance was
MEMORY AFTER ACUTE STRESS
607
FIG. 1. Number of correct out of the first eight choices (means 1 SEM) in the RAM on Training Day 7, Stress Day, and Poststress Day for three different groups: normothermic-unrestrained (n 5 10), normothermic-restrained (n 5 11), and cold-restrained (n 5 12).
significantly impaired compared to that of Day 7 (p , 0.01) and Poststress day (p , 0.01). However, performance on Poststress Day for the cold-restrained group was still inferior to that of Day 7 (p , 0.05), whereas Poststress Day performance for the normothermic restrained rats was not different from Day 7 levels. Performance did not change over time in the normothermic-unrestrained group. The time per choice for Day 7, Stress Day, and Poststress Day
is shown in Fig. 3. ANOVA indicated a significant condition effect [F(2, 30) 5 12.07, p , 0.001], a significant day effect [F(2, 60) 5 40.48, p , 0.001], and a significant condition-by-day interaction [F(4, 60) 5 9.82, p , 0.001]. Post hoc analyses revealed that time per choice performance on Stress Day in the cold-restrained (p , 0.01) and normothermic-restrained (p , 0.01) groups was longer than that of the normothermic-unrestrained group; also, the coldrestrained group (p , 0.01) performance was impaired relative to
FIG. 2. Total number of choices (means 1 SEM) in the RAM on Training Day 7, Stress Day, and Poststress Day for three different groups: normothermic-unrestrained (n 5 10), normothermic-restrained (n 5 11), and cold-restrained (n 5 12).
608
STILLMAN ET AL.
FIG. 3. Time per choice (means 1 SEM) in the RAM on Training Day 7, Stress Day, and Poststress Day for three different groups: cold-restrained (n 5 12), normothermic-restrained (n 5 11), and normothermicunrestrained (n 5 10).
the normothermic-restrained group. On Poststress Day, performance of the cold-restrained rats was impaired relative to only the normothermic-restrained (p , 0.05) group; however, the normothermic-restrained and normothermic-unrestrained groups were not different from each other. Once again, there were no significant differences between the groups on Day 7. Over time, post hoc analyses for both stress groups for the total number of choices revealed that Stress Day performance was significantly impaired compared to that of Day 7 (p , 0.01) and Poststress Day (p , 0.01). Performance on the Poststress Day for the cold-restrained and normothermic-restrained groups was not different from Day 7 levels. Performance did not change over time in the normothermic-unrestrained group. DISCUSSION
These experiments demonstrate that exposure to either normothermic-restraint stress or cold-restraint stress produces moderate and severe impairments, respectively, on performance in a spatial memory (RAM) task. After either stressor, rats displayed increases in time per choice, as well as decreases in the number correct of the first eight choices and the total number of choices. The results of the present study suggest that this procedure (15-min restraint, 40-min “time out,” 10-min RAM testing) may be sensitive to differential effects of stressor severity. One possible explanation for impaired spatial memory performance after stressor exposure may be altered central cholinergic neurotransmission. Our previous investigation showed that acute stress altered hippocampal cholinergic neurotransmission by utilizing in vivo microdialysis to evaluate hippocampal extracellular ACh and choline levels in male Fischer 344 rats before, during, and after an 80-min exposure to two different stress conditions (17). In cold-restrained rats, ACh levels significantly declined during cold stress relative to both normothermic-restrained and normothermic-freely moving rats. By the end of the cold exposure period and after removal from cold, ACh levels had returned to near-baseline values. Normothermic-restrained rats had levels sim-
ilar to those of normothermic-freely moving rats, except for a marked increase in ACh after removal from restraint. Cold-restrained rats displayed a gradual elevation in choline levels during cold stress, followed by a gradual decline after stress termination, whereas both normothermic-restrained and normothermic-freely moving rats displayed gradual decreases during the microdialysis session. Relative to normothermic-unrestrained rats, the decrease in ACh levels was greatest in cold-restrained rats and intermediate in normothermic-restrained rats. Spatial memory is strongly dependent on the hippocampus, a structure in which ACh is a major neurotransmitter (10). Therefore, a stress-induced decrease in hippocampal ACh may be responsible for the impaired spatial memory performance observed in this study. Further evidence for this hypothesis is that both neurochemistry and behavior seem related to stressor severity, with more severe stressor exposure effecting greater decrements in both measures. The present results are in agreement with those demonstrating memory performance decrements after cold exposure. Rats that were trained on a Morris water maze task for 6 days and then cooled to a core temperature of 28°C or 30°C displayed higher escape latencies and higher mean swim distances than normothermic controls; moreover, these performance impairments were reversed to control values upon artificial rewarming (15). Several studies demonstrated that exposure to 2°C ambient temperature, which did not decrease colonic temperature, impaired performance on a delayed matching task when rats were given a 16-s delay between stimuli (1,16,18,19). Performance on a memory task in humans, as measured by fact recall, was decreased after 1-h immersion in 4.4°C water without detectable impairments in other cognitive measures such as vigilance and reasoning (3). Also, exposure to 5°C ambient temperature led to decreased performance on a working memory task in humans (18). Subjects immersed up to the neck in 8°C water, which decreased core temperature, showed decrements in complex, but not simple, cognitive tasks involving working memory (6). Performance on spatial tasks seems to be correlated with the
MEMORY AFTER ACUTE STRESS
609
intensity of hypothermia. Rats trained under normothermic conditions then tested under cold stress (resulting in 30°C colonic temperature) showed much lower escape latencies in the Morris water maze than rats exposed to more severe (,25°C colonic temperature) hypothermia (14). Our results also agree with studies that show behavioral impairment after restraint. The present data for acute restraint are in accord with a previous study that demonstrated impaired RAM performance after chronic (6 h/day for 3 weeks) restraint (8). Other investigations have demonstrated impaired shock discrimination performance (7) and increased escape latency from shock acquisition in rats (4) after restraint stress. Mice that were immobilized immediately after training in a passive avoidance paradigm showed impaired retention performance, although this was not observed in all strains tested (5). The stress-induced behavioral effects in this study may be due to a direct effect on brain temperature. Although not measured in the present study, it has been shown that exposure to ambient cold resulted in decreases in both memory and hippocampal temperature, suggesting that central cooling may be responsible for performance decrements (1). It may be that a decrease in brain temperature slows hippocampal synaptic transmission, which would impair spatial memory (9). There are other mechanisms that could account for the results obtained in this study. When placed in the maze, stressed rats typically would select two or three arms, traversing them at a normal pace, and then remain in one location until 10 min had elapsed. Because the stressed rats only made two to three arm choices, they did not make more errors than nonstressed control rats (i.e., they usually did not enter the same arm). It is uncertain why the stressed rats stopped traversing the maze after seemingly normal performance; it may be due to a combination of psychological (e.g., motivation and attention), neurophysiological (e.g., decreased neuronal conductance), and physical impairments (e.g.,
impaired joint and muscle function) (6). Physical impairment, the only observable parameter in this study, was unlikely to have been a major cause of the performance deficit, due to the normal speed and gait observed during initial choice selection; i.e., the stressors produced no detectable physical impairments during behavioral testing. Furthermore, even though the stressed rats did not commit more errors, they may have stopped responding due to impaired memory such as forgetting the task or decreased attention to the task. However, because these mechanisms were unobservable in the present study, further research is needed to determine if one of these other explanations may be the reason for the reduction in activity after stressor exposure. In summary, the present study demonstrated impaired performance on a memory-based task after cold-restraint and normothermic-restraint conditions. As observed previously in the neurochemical investigations, the magnitude of the behavioral effect also seems related to the stress severity, in that cold-restraint produced larger behavioral deficits than did normothermic-restraint. This decreased performance was not likely a result of gross physical insult, as rats seemed to traverse the maze with a gait similar to that observed during nonstressed conditions. ACKNOWLEDGEMENTS
Grateful appreciation is extended to Bryan P. Coffey, Sergeant Jennifer Seymour, and Specialist Jason Irwin for their assistance in data collection. In conducting the research described in this report, the investigators adhered to the “Guide for the Care and Use of Laboratory Animals,” as prepared by the Committee on Care and Use of Laboratory Animals of the Institute of Laboratory Animal Resources, National Research Council. The views, opinions, and/or findings contained in this report are those of the authors and should not be construed as an official Department of the Army position, policy, or decision, unless so designated by other official documentation.
REFERENCES 1. Ahlers, S. T.; Thomas, J. R.; Berkey, D. L. Hippocampal and body temperature changes in rats during delayed matching-to-sample performance in a cold environment. Physiol. Behav. 50:1013–1018; 1991. 2. Ahlers, S. T.; Thomas, J. R.; Schrot, J; Shurtleff, D. Tyrosine and glucose modulation of cognitive deficits resulting from cold stress. In: Marriott, B.M., ed. Food components to enhance performance. Washington, DC: National Academy Press; 1994:301–320. 3. Baddeley, A. D.; Cuccaro, W. J.; Egstrom, G. H.; Weltman, G.; Willis, M. A. Cognitive efficiency of divers working in cold water. Hum. Factors 17:446 – 454; 1975. 4. Bracewell, R. J.; Black, A. H. The effects of restraint and noncontingent preshock on subsequent escape learning in the rat. Learning Motiv. 5:53– 69; 1974. 5. Castellano, C.; Puglisi-Allegra, S. Strain-dependent modulation of memory by stress in mice. Behav. Neural Biol. 38:133–138; 1983. 6. Giesbrecht, G. G.; Arnett, J. L.; Vela, E.; Bristow, G. K. Effect of task complexity on mental performance during immersion hypothermia. Aviat. Space Environ. Med. 64:206 –211; 1993. 7. Grilly, D. M.; Dugovics, J. P. Effects of immobilization stress on shock discrimination performance in rats. Physiol. Behav. 29:1077– 1081; 1982. 8. Luine, V.; Villegas, M.; Martinez, C.; McEwen, B. S. Repeated stress causes reversible impairments of spatial memory performance. Brain Res. 639:167–170; 1994. 9. Moser, E. I.; Andersen, P. Conserved spatial learning in cooled rats despite slowing of dentate field potentials. J. Neurosci. 14:4458 – 4466; 1994. 10. O’Keefe, J.; Nadel, L. The hippocampus as a cognitive map. New York: Oxford University Press; 1978.
11. Olton, D. S. The radial arm maze as a tool in behavioral pharmacology. Physiol. Behav. 40:793–797; 1987. 12. Olton, D. S.; Samuelson, R. J. Remembrance of places passed: Spatial memory in rats. J. Exp. Psychol. Anim. Behav. Process. 2:97–116; 1976. 13. Packard, M. G.; Hirsh, R.; White, N. M. Differential effects of fornix and caudate nucleus lesions on two radial maze tasks: Evidence for multiple memory systems. J. Neurosci. 9:1465–1472; 1989. 14. Panakhova, E.; Buresova, O.; Bures, J. The effect of hypothermia on the rat’s spatial memory in the water tank task. Behav. Neural Biol. 42:191–196; 1984. 15. Rauch, T. M.; Welch, D. I.; Gallego, L. Hypothermia impairs performance in the Morris water maze. Physiol. Behav. 46:315– 320; 1989. 16. Shurtleff, D.; Thomas, J. R.; Ahlers, S. T.; Schrot, J. Tyrosine ameliorates a cold-induced delayed matching-to-sample performance decrement in rats. Psychopharmacology 112:228 –232; 1993. 17. Stillman, M. J.; Shukitt-Hale, B.; Coffey, B. P.; Levy, A.; Lieberman, H. R. In vivo hippocampal acetylcholine release during exposure to acute stress. Stress 1:191–199; 1997. 18. Thomas, J. R.; Ahlers, S. T.; House, J. F.; Schrot, J. Repeated exposure to moderate cold impairs matching-to-sample performance. Aviat. Space Environ. Med. 60:1063–1067; 1989. 19. Thomas, J. R.; Ahlers, S. T.; Schrot, J. Cold-induced impairment of delayed matching in rats. Behav. Neural Biol. 55:19 –30; 1991. 20. Vanderwolf, C. H. Effects of water temperature and core temperature on rat’s performance in a swim-to-platform test. Behav. Brain Res. 44:105–106; 1991.
PII S0031-9384(98)00091–2
Spatial Memory Under Acute Cold and Restraint Stress MICHAEL J. STILLMAN,*†1 BARBARA SHUKITT-HALE,*† AHARON LEVY,*‡ AND HARRIS R. LIEBERMAN*2 *Military Nutrition and Biochemistry Division, United States Army Research Institute of Environmental Medicine, Natick, MA 01760-5007; †GEO-CENTERS, Inc., Newton Centre, MA 02159; and ‡Department of Pharmacology, Israel Institute for Biological Research, Ness Ziona, 70450 Israel Received 6 June 1997; Accepted 3 March 1998 STILLMAN, M.J., B. SHUKITT-HALE, A. LEVY AND H.R. LIEBERMAN. Spatial memory under acute cold and restraint stress. PHYSIOL BEHAV 64(5) 605– 609, 1998.—This study examined spatial memory as measured by radial arm maze (RAM) performance after exposure to two stress conditions and a normothermic-unrestrained control condition. Male Fischer 344 rats were trained on the win-shift RAM procedure for 7 days, by which time they achieved asymptotic performance. The next day, rats in the two stress groups were exposed to 15 min of restraint in either 37°C water (normothermic-restraint) or in 20°C water (cold-restraint). Rats were then allowed 40 min in a dry cage before being tested in the RAM. Performance was measured using the following dependent variables: number of correct out of the first eight choices, total number of choices, and time per choice. There were statistically significant effects of stress on all these variables; performance decrements were observed in both stress conditions relative to the normothermicunrestrained condition. Normothermic-restrained rats displayed less impairment than cold-restrained rats on the stress day. Performance of normothermic-restrained rats returned to baseline levels the day after stress, whereas performance for the cold-restrained rats typically did not. This study demonstrates that: 1) restraint and cold stress impair performance on a memory task; and 2) impairment extent is related to stress severity. One of the mechanisms responsible for the observed behavioral deficits under cold stress may involve altered cholinergic function, because we previously demonstrated that hippocampal acetylcholine levels also decrease in relation to the severity of cold stress. © 1998 Elsevier Science Inc. Learning
Hypothermia
Performance
Radial arm maze
Despite the heterogeneity of stressor type and duration, many studies have demonstrated that stressed animals cannot learn as readily as nonstressed animals. Restraint has been shown to impair radial arm maze (RAM) performance (8), and to disrupt shock discrimination performance (7). Similarly, cold exposure produces deficits in delayed-matching-to-sample performance, a measure of working memory in both humans and other animals (1,2,16,18,19). In addition, spatial memory impairments in the Morris water maze task were observed after cold stress (14,15,20). The majority of investigations thus far generally has examined different intensities of one type of stressor, predominantly demonstrating behavioral effects proportional to stressor intensity (e.g., 14). In this study, we examined the behavioral effects of one stressor (restraint) alone and in combination with another stressor (cold). It was hypothesized that the combination of thermal and restraint stressors would be more severe, as suggested by one of our recent neurochemical investigations in which acetylcholine (ACh) release was decreased during cold and restraint relative to normothermic restraint conditions (17). The present research was
Rats
performed to compare spatial memory performance as measured by the RAM task among cold-restrained, normothermic-restrained, and normothermic-unrestrained rats.
MATERIALS AND METHODS
Subjects A total of 33 male Fischer 344 rats (Charles River Labs, Kingston, NY, USA), weighing 250 –360 g at time of testing, were used in this study. The animals were randomly divided into three groups: twelve rats were cold-restrained, eleven were normothermic-restrained, and ten rats served as nonstressed, normothermicunrestrained controls. Rats were restricted to 1 h per day of water for 5 days before the start of RAM training and continuing throughout the study. They were housed individually in wire mesh cages, and were maintained on a 12-h light/dark cycle (lights on at 0600 hours).
1
Michael J. Stillman, PhD, is currently at DendWrite Communications, 33 Dinsmore Avenue, Suite 602, Framingham, MA 01702, USA Requests for reprints should be addressed to Dr. Harris R. Lieberman, Military Nutrition and Biochemistry Division, United States Army Research Institute of Environmental Medicine, Natick, MA 01760-5007, USA. E-mail: [email protected] 2
605
606 RAM Spatial memory was assessed in the present research via performance on the RAM (11). Working memory is required while the rat is selecting the next arm to visit, apparently “keeping track of” which arms were previously selected. Various control experiments (11,12) have eliminated extraneous nonmemory processes (e.g., odor cues) as well as intramaze cues as confounding factors and strongly suggest that the rat performs the RAM test by utilizing extramaze cues. One possibility is that the rat builds a cognitive map of the environment to learn the maze, and utilizes its working memory to monitor which arms have been visited. The win-shift procedure, employed in this study, requires the rat to retrieve the reward at the end of each of the eight arms without returning to any previously visited arm because visited arms are not rebaited; an intact hippocampus is necessary for accurate performance in the win-shift paradigm (13). The Plexiglas RAM consisted of eight equally spaced arms (70-cm long, 12-cm wide, 15-cm high) radiating from an octagonal central area 30 cm in diameter. A Plexiglas ring surrounded the central area, preventing the rat from entering any of the arms. An overhead pulley and rope enabled the experimenter to raise the ring and allow the rat to leave the central area and select from any of the eight arms. Various extramaze cues were present in the testing room, including posters, a laboratory cart, and miscellaneous laboratory equipment. Each maze arm contained a small dish at its end that was baited with 0.25 mL of 0.30% saccharin solution. Procedure Rats were water restricted for 5 days (1 h/day) before RAM training. To avoid taste aversion to a novel stimulus during maze training, the saccharin solution was introduced in the home cage on Days 3 and 4 of the water-restriction schedule. For the last 2 of these 5 days each rat was allowed 10 min to explore the RAM, after which training began. The saccharin solution was scattered about the maze during these last 2 days, to facilitate exploration of the maze. Rats were tested for 9 days in the RAM, including 7 training days, 1 Stress Day, and 1 Poststress day. Each RAM trial commenced when the rat was placed in the center of the maze for 15 s and then permitted free access to all of the baited arms. In accordance with the win-shift procedure (in which visited arms are not rebaited), reentries into previously visited arms were scored as errors. Each trial was terminated when all eight arms were visited, when sixteen entries were made, or at the end of 600 s, whichever came first. The animal was determined to have made a choice once it moved from the central area and all four paws were located in a maze arm. The following measures were recorded by experimenter observation: the order of entry into the maze arms, the number of revisits, the total number of choices, the time to complete a session, and the number of correct entries out of the first eight choices. Time per choice was calculated as total time for completion of the session divided by the total number of entries. Stress Day occurred after the 7 training days. Stressed rats were placed in a small, custom-designed, metal and Velcro restraining apparatus during stressor exposure. Acute stress was produced by immersing the rat up to the neck for 15 min in a water bath (RC20, Lauda) filled with either 20 6 2°C water for the cold-restraint condition, or with 37 6 2°C water for the normothermic-restraint condition. Previous research in our laboratory showed that core body temperature decreased to approximately 30°C after 15 min after this cold stress procedure (17). Rats were then towel dried and placed in a holding container for 40 min before RAM testing.
STILLMAN ET AL. Normothermic-unrestrained rats remained in the holding container for 55 min before RAM testing. All rats underwent final RAM testing 24 h later under no-stress conditions (Poststress Day) to determine the extent of performance recovery. Statistical Analysis The following measures served as dependent variables: number of correct entries in the first eight choices, total number of choices, and time per choice. Each measure was analyzed by a two-way ANOVA using stress condition as the between-subjects factor and day (Day 7, Stress Day, and Poststress Day) as the within-subject factor. All statistical tests utilized a p , 0.05 criterion for hypothesis testing. Post hoc comparisons were performed, when appropriate, using Duncan’s multiple range test. RESULTS
All subjects improved performance to asymptotic levels during the 7 training days. Overall, performance decrements were observed on Stress Day in both stress conditions relative to the normothermic-unrestrained condition; moreover, the performance of cold-restrained rats was inferior to that of the normothermicrestrained rats. Performance returned to baseline (Day 7) levels the day after stress for the normothermic-restrained group, but typically not for the cold-restrained group. The number of correct out of the first eight choices for Day 7, Stress Day, and Poststress Day is shown in Fig. 1. ANOVA indicated a significant condition effect [F(2, 30) 5 19.41, p , 0.001], a significant day effect [F(2, 60) 5 51.73, p , 0.001], and a significant condition-by-day interaction [F(4, 60) 5 15.01, p , 0.001]. Post hoc analyses revealed that number correct performance on Stress Day in the cold-restrained (p , 0.01) and normothermic-restrained (p , 0.01) groups was lower than that of the normothermic-unrestrained group; also, performance was lower in the cold-restrained group (p , 0.01) compared to the normothermic-restrained group. On Poststress Day, performance of the coldrestrained rats (p , 0.05) was still impaired relative to the other groups, which were not different from each other. There were no significant differences between the groups on Day 7. Over time, post hoc analyses for both stress groups for the number of correct out of the first eight choices revealed that performance on Stress Day was significantly impaired compared to that of Day 7 (p , 0.01) and Poststress day (p , 0.01). However, performance on Poststress Day for the cold-restrained group was still inferior to that of Day 7 (p , 0.05), whereas Poststress Day performance for the normothermic restrained rats was not different from Day 7 levels. Performance did not change over time in the normothermic-unrestrained group. The total number of choices for Day 7, Stress Day, and Poststress Day is shown in Fig. 2. ANOVA indicated a significant condition effect [F(2, 30) 5 19.92, p , 0.001], a significant day effect [F(2, 60) 5 42.32, p , 0.001], and a significant conditionby-day interaction [F(4, 60) 5 16.58, p , 0.001]. Post hoc analyses revealed that choice performance on Stress Day in the cold-restrained (p , 0.01) and normothermic-restrained (p , 0.01) groups was again lower than that of the normothermic-unrestrained group; also, the cold-restrained group (p , 0.01) performance was impaired relative to the normothermic-restrained group. On Poststress Day, performance of the cold-restrained rats was impaired relative to the normothermic-unrestrained (p , 0.05) and normothermic-restrained (p , 0.01) groups, which were not different from each other. Again, there were no significant differences between the groups on Day 7. Over time, post hoc analyses for both stress groups for the total number of choices revealed that Stress Day performance was
MEMORY AFTER ACUTE STRESS
607
FIG. 1. Number of correct out of the first eight choices (means 1 SEM) in the RAM on Training Day 7, Stress Day, and Poststress Day for three different groups: normothermic-unrestrained (n 5 10), normothermic-restrained (n 5 11), and cold-restrained (n 5 12).
significantly impaired compared to that of Day 7 (p , 0.01) and Poststress day (p , 0.01). However, performance on Poststress Day for the cold-restrained group was still inferior to that of Day 7 (p , 0.05), whereas Poststress Day performance for the normothermic restrained rats was not different from Day 7 levels. Performance did not change over time in the normothermic-unrestrained group. The time per choice for Day 7, Stress Day, and Poststress Day
is shown in Fig. 3. ANOVA indicated a significant condition effect [F(2, 30) 5 12.07, p , 0.001], a significant day effect [F(2, 60) 5 40.48, p , 0.001], and a significant condition-by-day interaction [F(4, 60) 5 9.82, p , 0.001]. Post hoc analyses revealed that time per choice performance on Stress Day in the cold-restrained (p , 0.01) and normothermic-restrained (p , 0.01) groups was longer than that of the normothermic-unrestrained group; also, the coldrestrained group (p , 0.01) performance was impaired relative to
FIG. 2. Total number of choices (means 1 SEM) in the RAM on Training Day 7, Stress Day, and Poststress Day for three different groups: normothermic-unrestrained (n 5 10), normothermic-restrained (n 5 11), and cold-restrained (n 5 12).
608
STILLMAN ET AL.
FIG. 3. Time per choice (means 1 SEM) in the RAM on Training Day 7, Stress Day, and Poststress Day for three different groups: cold-restrained (n 5 12), normothermic-restrained (n 5 11), and normothermicunrestrained (n 5 10).
the normothermic-restrained group. On Poststress Day, performance of the cold-restrained rats was impaired relative to only the normothermic-restrained (p , 0.05) group; however, the normothermic-restrained and normothermic-unrestrained groups were not different from each other. Once again, there were no significant differences between the groups on Day 7. Over time, post hoc analyses for both stress groups for the total number of choices revealed that Stress Day performance was significantly impaired compared to that of Day 7 (p , 0.01) and Poststress Day (p , 0.01). Performance on the Poststress Day for the cold-restrained and normothermic-restrained groups was not different from Day 7 levels. Performance did not change over time in the normothermic-unrestrained group. DISCUSSION
These experiments demonstrate that exposure to either normothermic-restraint stress or cold-restraint stress produces moderate and severe impairments, respectively, on performance in a spatial memory (RAM) task. After either stressor, rats displayed increases in time per choice, as well as decreases in the number correct of the first eight choices and the total number of choices. The results of the present study suggest that this procedure (15-min restraint, 40-min “time out,” 10-min RAM testing) may be sensitive to differential effects of stressor severity. One possible explanation for impaired spatial memory performance after stressor exposure may be altered central cholinergic neurotransmission. Our previous investigation showed that acute stress altered hippocampal cholinergic neurotransmission by utilizing in vivo microdialysis to evaluate hippocampal extracellular ACh and choline levels in male Fischer 344 rats before, during, and after an 80-min exposure to two different stress conditions (17). In cold-restrained rats, ACh levels significantly declined during cold stress relative to both normothermic-restrained and normothermic-freely moving rats. By the end of the cold exposure period and after removal from cold, ACh levels had returned to near-baseline values. Normothermic-restrained rats had levels sim-
ilar to those of normothermic-freely moving rats, except for a marked increase in ACh after removal from restraint. Cold-restrained rats displayed a gradual elevation in choline levels during cold stress, followed by a gradual decline after stress termination, whereas both normothermic-restrained and normothermic-freely moving rats displayed gradual decreases during the microdialysis session. Relative to normothermic-unrestrained rats, the decrease in ACh levels was greatest in cold-restrained rats and intermediate in normothermic-restrained rats. Spatial memory is strongly dependent on the hippocampus, a structure in which ACh is a major neurotransmitter (10). Therefore, a stress-induced decrease in hippocampal ACh may be responsible for the impaired spatial memory performance observed in this study. Further evidence for this hypothesis is that both neurochemistry and behavior seem related to stressor severity, with more severe stressor exposure effecting greater decrements in both measures. The present results are in agreement with those demonstrating memory performance decrements after cold exposure. Rats that were trained on a Morris water maze task for 6 days and then cooled to a core temperature of 28°C or 30°C displayed higher escape latencies and higher mean swim distances than normothermic controls; moreover, these performance impairments were reversed to control values upon artificial rewarming (15). Several studies demonstrated that exposure to 2°C ambient temperature, which did not decrease colonic temperature, impaired performance on a delayed matching task when rats were given a 16-s delay between stimuli (1,16,18,19). Performance on a memory task in humans, as measured by fact recall, was decreased after 1-h immersion in 4.4°C water without detectable impairments in other cognitive measures such as vigilance and reasoning (3). Also, exposure to 5°C ambient temperature led to decreased performance on a working memory task in humans (18). Subjects immersed up to the neck in 8°C water, which decreased core temperature, showed decrements in complex, but not simple, cognitive tasks involving working memory (6). Performance on spatial tasks seems to be correlated with the
MEMORY AFTER ACUTE STRESS
609
intensity of hypothermia. Rats trained under normothermic conditions then tested under cold stress (resulting in 30°C colonic temperature) showed much lower escape latencies in the Morris water maze than rats exposed to more severe (,25°C colonic temperature) hypothermia (14). Our results also agree with studies that show behavioral impairment after restraint. The present data for acute restraint are in accord with a previous study that demonstrated impaired RAM performance after chronic (6 h/day for 3 weeks) restraint (8). Other investigations have demonstrated impaired shock discrimination performance (7) and increased escape latency from shock acquisition in rats (4) after restraint stress. Mice that were immobilized immediately after training in a passive avoidance paradigm showed impaired retention performance, although this was not observed in all strains tested (5). The stress-induced behavioral effects in this study may be due to a direct effect on brain temperature. Although not measured in the present study, it has been shown that exposure to ambient cold resulted in decreases in both memory and hippocampal temperature, suggesting that central cooling may be responsible for performance decrements (1). It may be that a decrease in brain temperature slows hippocampal synaptic transmission, which would impair spatial memory (9). There are other mechanisms that could account for the results obtained in this study. When placed in the maze, stressed rats typically would select two or three arms, traversing them at a normal pace, and then remain in one location until 10 min had elapsed. Because the stressed rats only made two to three arm choices, they did not make more errors than nonstressed control rats (i.e., they usually did not enter the same arm). It is uncertain why the stressed rats stopped traversing the maze after seemingly normal performance; it may be due to a combination of psychological (e.g., motivation and attention), neurophysiological (e.g., decreased neuronal conductance), and physical impairments (e.g.,
impaired joint and muscle function) (6). Physical impairment, the only observable parameter in this study, was unlikely to have been a major cause of the performance deficit, due to the normal speed and gait observed during initial choice selection; i.e., the stressors produced no detectable physical impairments during behavioral testing. Furthermore, even though the stressed rats did not commit more errors, they may have stopped responding due to impaired memory such as forgetting the task or decreased attention to the task. However, because these mechanisms were unobservable in the present study, further research is needed to determine if one of these other explanations may be the reason for the reduction in activity after stressor exposure. In summary, the present study demonstrated impaired performance on a memory-based task after cold-restraint and normothermic-restraint conditions. As observed previously in the neurochemical investigations, the magnitude of the behavioral effect also seems related to the stress severity, in that cold-restraint produced larger behavioral deficits than did normothermic-restraint. This decreased performance was not likely a result of gross physical insult, as rats seemed to traverse the maze with a gait similar to that observed during nonstressed conditions. ACKNOWLEDGEMENTS
Grateful appreciation is extended to Bryan P. Coffey, Sergeant Jennifer Seymour, and Specialist Jason Irwin for their assistance in data collection. In conducting the research described in this report, the investigators adhered to the “Guide for the Care and Use of Laboratory Animals,” as prepared by the Committee on Care and Use of Laboratory Animals of the Institute of Laboratory Animal Resources, National Research Council. The views, opinions, and/or findings contained in this report are those of the authors and should not be construed as an official Department of the Army position, policy, or decision, unless so designated by other official documentation.
REFERENCES 1. Ahlers, S. T.; Thomas, J. R.; Berkey, D. L. Hippocampal and body temperature changes in rats during delayed matching-to-sample performance in a cold environment. Physiol. Behav. 50:1013–1018; 1991. 2. Ahlers, S. T.; Thomas, J. R.; Schrot, J; Shurtleff, D. Tyrosine and glucose modulation of cognitive deficits resulting from cold stress. In: Marriott, B.M., ed. Food components to enhance performance. Washington, DC: National Academy Press; 1994:301–320. 3. Baddeley, A. D.; Cuccaro, W. J.; Egstrom, G. H.; Weltman, G.; Willis, M. A. Cognitive efficiency of divers working in cold water. Hum. Factors 17:446 – 454; 1975. 4. Bracewell, R. J.; Black, A. H. The effects of restraint and noncontingent preshock on subsequent escape learning in the rat. Learning Motiv. 5:53– 69; 1974. 5. Castellano, C.; Puglisi-Allegra, S. Strain-dependent modulation of memory by stress in mice. Behav. Neural Biol. 38:133–138; 1983. 6. Giesbrecht, G. G.; Arnett, J. L.; Vela, E.; Bristow, G. K. Effect of task complexity on mental performance during immersion hypothermia. Aviat. Space Environ. Med. 64:206 –211; 1993. 7. Grilly, D. M.; Dugovics, J. P. Effects of immobilization stress on shock discrimination performance in rats. Physiol. Behav. 29:1077– 1081; 1982. 8. Luine, V.; Villegas, M.; Martinez, C.; McEwen, B. S. Repeated stress causes reversible impairments of spatial memory performance. Brain Res. 639:167–170; 1994. 9. Moser, E. I.; Andersen, P. Conserved spatial learning in cooled rats despite slowing of dentate field potentials. J. Neurosci. 14:4458 – 4466; 1994. 10. O’Keefe, J.; Nadel, L. The hippocampus as a cognitive map. New York: Oxford University Press; 1978.
11. Olton, D. S. The radial arm maze as a tool in behavioral pharmacology. Physiol. Behav. 40:793–797; 1987. 12. Olton, D. S.; Samuelson, R. J. Remembrance of places passed: Spatial memory in rats. J. Exp. Psychol. Anim. Behav. Process. 2:97–116; 1976. 13. Packard, M. G.; Hirsh, R.; White, N. M. Differential effects of fornix and caudate nucleus lesions on two radial maze tasks: Evidence for multiple memory systems. J. Neurosci. 9:1465–1472; 1989. 14. Panakhova, E.; Buresova, O.; Bures, J. The effect of hypothermia on the rat’s spatial memory in the water tank task. Behav. Neural Biol. 42:191–196; 1984. 15. Rauch, T. M.; Welch, D. I.; Gallego, L. Hypothermia impairs performance in the Morris water maze. Physiol. Behav. 46:315– 320; 1989. 16. Shurtleff, D.; Thomas, J. R.; Ahlers, S. T.; Schrot, J. Tyrosine ameliorates a cold-induced delayed matching-to-sample performance decrement in rats. Psychopharmacology 112:228 –232; 1993. 17. Stillman, M. J.; Shukitt-Hale, B.; Coffey, B. P.; Levy, A.; Lieberman, H. R. In vivo hippocampal acetylcholine release during exposure to acute stress. Stress 1:191–199; 1997. 18. Thomas, J. R.; Ahlers, S. T.; House, J. F.; Schrot, J. Repeated exposure to moderate cold impairs matching-to-sample performance. Aviat. Space Environ. Med. 60:1063–1067; 1989. 19. Thomas, J. R.; Ahlers, S. T.; Schrot, J. Cold-induced impairment of delayed matching in rats. Behav. Neural Biol. 55:19 –30; 1991. 20. Vanderwolf, C. H. Effects of water temperature and core temperature on rat’s performance in a swim-to-platform test. Behav. Brain Res. 44:105–106; 1991.