- Email: [email protected]
Prenatal Exposure to Ethanol Disrupts Spatial Memory: Effect of the Training–Testing Delay Period
Physiology & Behavior, Vol. 64, No. 1, pp. 63– 67, 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)00019-5
Prenatal Exposure to Ethanol Disrupts Spatial Memory: Effect of the Training–Testing Delay Period DOUGLAS B. MATTHEWS1 AND PETER E. SIMSON Department of Psychology and Center for Neuroscience Research, Miami University, Oxford, OH 45056, USA Received 20 August 1997; Accepted 29 December 1997 MATTHEWS, D. B. AND P. E. SIMSON. Prenatal exposure to ethanol disrupts spatial memory: Effect of training–testing delay period. PHYSIOL BEHAV 64(1) 63– 67, 1998.—The present study investigated how variations in the period of delay between training and testing in the Morris water maze task affect the use of spatial memory in adult rats that were prenatally exposed to ethanol. Previous results utilizing the Morris water maze task have shown that prenatal, or early postnatal, exposure to ethanol produces deficits in the use of spatial memory, a type of memory that is dependent on an intact hippocampus. However, in these prior studies the delay period between the training of animals and the testing of spatial memory is typically fixed at only 1 day. In the current study, which utilized a revised training procedure within the Morris water maze task, the period of delay between training and testing was altered such that it was either 1 day or 3 days. Following the 3-day delay, different levels of prenatal exposure to ethanol impaired the use of spatial memory. In contrast, following the 1-day delay, prenatal exposure to ethanol failed to impair the use of spatial memory. The present study thus shows that prenatal exposure to ethanol differentially affects spatial memory in the Morris water maze task depending on the period of delay between training and testing. © 1998 Elsevier Science Inc. Fetal alcohol syndrome
Prenatal ethanol exposure
Hippocampus
Spatial memory
to ethanol alters hippocampal anatomy and because the hippocampus is critically involved in spatial cognitive processing, one might predict that prenatal exposure to ethanol would impair spatial cognitive processing (6,7,10,11–13,27). This is, in fact, the case. Prenatal exposure to ethanol impairs spatial learning and spatial memory in juvenile rats (6,7,13) and adult rats (10,27). One paradigm often used to investigate the effects of prenatal exposure to ethanol on spatial learning and memory is the Morris water maze task. In the Morris water maze task, a rat is released into a pool of cloudy water in which there exists an escape “island,” or platform. The escape platform can either protrude above the surface of the water, thus being visible to the rat, or be slightly submerged below the surface of the water, thus being hidden from the rat (21). When presented with a visible, protruding platform, the rat can learn the task by using either a nonspatial strategy (i.e., swim to the only visible cue in the pool) or a spatial strategy (i.e., swim to the same place in the room to find the platform). In contrast, when presented with a submerged platform, the rat can only learn to use the spatial strategy (i.e., swim to the same spatial location to find the submerged platform). Lesions to the hippocampal system impair performance on the Morris water maze task when animals can only utilize the spatial strategy, but do not impair performance when animals can utilize a nonspatial strategy (22). Similarly, prenatal exposure to ethanol selectively
EXPOSURE to ethanol throughout gestation in humans can produce an array of morphological changes and cognitive deficits that has been termed “fetal alcohol syndrome” (FAS) (3,14). Even exposure to ethanol during a part of gestation produces some cognitive deficits. This latter syndrome has been termed “fetal alcohol effect” (FAE) (9), a partial form of FAS, and is the syndrome modeled in this study. In rat models of FAS and FAE (15,19,29), exposure to ethanol during gestation produces behavioral changes, cognitive impairments, and neurological changes, as evidenced by increases in the normally suppressed motor activity in open-field tests (19), impairments of a number of other types of response inhibition (29), and reductions in brain weight and brain volume (15). Prenatal exposure to ethanol has also been shown to produce effects on the hippocampal system. For example, it decreases the number of CA1 pyramidal neurons (5), decreases both dendritic branching and the number of dendritic spines on pyramidal neurons (3), and produces aberrant mossy fiber projections from the dentate gyrus to the CA3 region (30). The hippocampal system is critical for spatial learning and memory (25). Lesions to the hippocampus or its afferents impair the learning of spatial tasks (24) while concomitantly facilitating the learning of nonspatial tasks (17,26). Hippocampal lesions also impair the use of spatial memory (22). Because prenatal exposure 1
To whom requests for reprints should be addressed. Present address: Center for Alcohol Studies, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599. E-mail: [email protected]
63
64
MATTHEWS AND SIMSON
impairs performance in the Morris water maze task when animals can only utilize a spatial strategy (10). In animals prenatally exposed to ethanol, the effect of varying the training–testing delay period in the Morris water maze task has yet to be investigated. There is reason to believe the training– testing period is an important variable in determining the effect of ethanol on spatial memory because, in another task involving spatial learning and memory, adult rats prenatally exposed to ethanol show an impairment in “spontaneous alteration” that is sensitive to a forced time delay during testing (23). Specifically, short time delays (,10 s) do not impair spontaneous alteration, but long delays (.30 s) do impair spontaneous alteration (23). Consequently, this study sought to determine whether variations in the period of delay between training and testing in the Morris water maze task alter the effect of prenatal exposure to ethanol on spatial memory in adult rats. Specifically, the study investigated whether prenatal exposure to different doses of ethanol produces deficits in spatial memory in the Morris water maze task over both a 1-day and/or a 3-day training–testing delay period. METHOD
Animals Female and male Long–Evans hooded rats (Charles River, Portage, MI), weighing approximately 225 and 250 g, respectively, were housed singly in hanging steel cages in an environmentally controlled vivarium at Wayne State University’s Mott Center. The vivarium was maintained at approximately 24°C on a 12:12 light– dark cycle (12 h with the lights on and 12 h with the lights off) with the lights on at 0700 hours. Breeding Each female rat was placed with a male, picked at random, overnight and the cages were examined the next morning for evidence of a sperm plug. If a plug was found, the dam was assumed to be pregnant and the day was marked as Gestational Day (GD) 0. Dams were weighed on GD 0, 8, and 15. Diet On GD 8, pregnant dams were assigned (counterbalanced for initial body weight) to one of four drug conditions: a No Treatment Control, an Isocalorie Control, Low EtOH (ethanol 3 g/kg), and High EtOH (5 g/kg). Ethanol, 10% w/v, was administered via oral intubation between 1000 and 1200 hours directly into the stomach of the dam. This procedure typically produces mean blood alcohol concentrations (BACs) at 30-min postintubation of 120 mg % (3.0 g/kg EtOH) and 180 mg % (5.0 g/kg EtOH) (J. H. Hannigan, personnel communication). Animals in the Isocalorie Control group received a solution in which maltose-dextrin was substituted for ethanol. Dams received the drug treatment from GD 8 to GD 20. During this time, dams were fed Purina Lab Diet Chow 5012 and the Isocalorie Control and Low EtOH animals were pair fed to the food intake of the High EtOH animals. Lab Chow was available only when animals were not intoxicated and, therefore, no significant difference in food intake was found. Pre-Experimental Treatment Beginning on GD 20, cages were inspected twice per day for pups. Following the day that pups were found, Postnatal Day (PN) 1, litters were pseudorandomly culled to seven males and three females and were cross-fostered to surrogate nonintubated control dams. On PN 1, 7, and 14, a male rat from each litter was randomly selected and removed to be a subject in a different study.
TABLE 1 Training
Testing
Delay Period
Days 1–3 Day 4 Days 5–7 Day 8 Day 9 Days 10–11 Day 12
Pups were weaned at approximately PN 21 and male rats were transported to an animal colony at Miami University. Animals were group housed in PlexiglasTM cages (two to three animals per box) until approximately 10 days before the beginning of testing. At this time, animals were housed singly and handled each day by an experimenter. There were nine rats in the No Treatment Control and High EtOH conditions and eight rats in the Isocalorie Control and Low EtOH conditions. There was no more than one animal per litter in any condition. Task and Apparatus The water tank used in the current study was a circular metal tub, painted white, that was 1.8 m in diameter and 50-cm tall. The water in the tank was 42-cm deep and was made opaque by the addition of white tempera, nontoxic, watercolor paint. The water temperature was maintained at approximately 23°C. The escape platform was made of clear PlexiglasTM, 9.75 3 9.75 3 39.5-cm high, and was submerged approximately 2.5 cm under the water’s surface. In a modification of the typical training procedure utilized by others wherein the escape platform is not visible to the animal during training, the escape platform was fitted with a dark brown wooden cap that protruded from the surface of the water by approximately 2.5 cm, thereby making the platform visible. Training and Testing Procedure The average age of the animals when training began was 86 days (ranging from 84 to 87 days), whereas the animals’ average age at the last test day was 95 days (ranging from 93 to 96 days). The entire training and testing procedure required 12 days. Animals were trained on Days 1–3, 5–7, and 9 with the platform visible to the animal. Animals were tested on Days 4, 8, and 12 with the platform submerged, and therefore not visible, to the animal. Animals were neither trained nor tested on Days 10 and 11. (See Table 1 for the experimental timeline.) The order of animal training was counterbalanced between conditions to control for changes in the test environment. On the training days, animals received four trials in which they were started from each of four unique start locations (N, S, E, W) and were trained to swim to the platform. The platform remained in a constant spatial location in the tank and, as stated earlier, was visible to the animal on all training trials. The order of the start locations was counterbalanced across days and matched between conditions, thereby producing a pseudo-Latin square design. Further, the same sequence of start locations was used on Days 1–3 and 5–7. The training that occurred on Day 9 used the start location sequence of that used on Days 1 and 5. On each training trial, the animal was released, facing the wall, into the tank and the latency of the animal to climb onto the platform was recorded. If the animal did not find the platform
PRENATAL EXPOSURE TO ETHANOL
65
FIG. 1. Mean latency for each of the seven training days. No significant difference was found between conditions. Note: The test days (Days 4, 8, and 12) are not included in the figure. Error bars denote SEM.
within 60 s, the experimenter gently guided the animal to the platform (the requirement of the experimenter to guide the animal to the platform occurred rarely, and only on the first day of training). Once the animal climbed onto the platform, the animal was allowed to remain there for 10 s before being removed and placed in a plastic holding cage. This procedure controlled for the amount of time each animal spent on the platform during training regardless of performance. Following an intertrial interval of 1 min, the animals were placed at the next start location. Testing occurred on Days 4 (24-h delay), 8 (24-h delay), and 12 (72-h delay). The start location sequence for the testing days was such that it completed the Latin square design used during training (thus the start location sequence was different from the sequences used during training). During testing, the platform remained in the same spatial location as that used during training. However, the wooden cap was removed from the platform, thereby hiding the platform from the animal. On each testing trial, the animal was released into the tank facing the wall and the latency to climb onto the platform was recorded. On the first testing day, if the animal did not find the platform within 90 s, it was gently guided to the platform as on the first day of training and then removed from the tank after 10 s. On the remaining two testing days, if the animal failed to find the platform, it was removed by placing it on the platform after 30 s had elapsed. Animals remained on the platform for 10 s before being removed and placed in a holding plastic cage. As during training, the intertrial interval was 1 min. Days 10 and 11 served as a “memory decay,” or training– testing delay, period. On these days, all animals remained in their home cages and were neither trained nor tested. RESULTS
Maternal Litter Characteristics Maternal weight gain during pregnancy was significantly different for dams in the four groups [F(3, 20) 5 4.6, p , 0.02]. Newman–Keuls comparisons revealed that dams in the High EtOH condition gained significantly less weight than those in the remaining three conditions (p , 0.05), whereas dams in the Low EtOH condition, Isocalorie Control condition, and the No Treatment Control condition had weights that did not differ significantly.
Litter size, gender ratio, and the mean birthweight of the male pups did not significantly differ between conditions. Training Days 1–3. Animals in all conditions demonstrated significant learning during the first 3 days of training. A two-way ANOVA for mixed designs utilizing harmonic means (16) for analysis based on repeated measures revealed that subjects’ latencies to climb onto the visible platform significantly decreased as a function of the number of training days [F(2, 60)5 113.11, p , 0.001]. In addition, no significant difference was found between conditions [F(3, 30) 5 0.63, p . 0.50), and no significant interaction (Days 3 Condition) was found [F(6,60) 5 0.39, p . 0.75]. Days 5–7. Animals in all conditions also showed significant learning during the second 3-day block of training. A two-way ANOVA for mixed designs utilizing harmonic means (16) for analysis based on repeated measures revealed that subjects’ latencies to climb onto the platform continued to decrease as a function of the number of training days [F(2, 60) 5 17.13, p , 0.001]. Once again, no significant difference was found between conditions [F(3, 30) 5 0.28, p . 0.75] no significant interaction (Days 3 Condition) was found [F(6, 60) 5 0.80, p . 0.50] Day 9. No significant difference in latency to find the platform existed between conditions on the last training day. A one-way ANOVA revealed no significant difference between conditions [F(3, 30) 5 0.30, p . 0.10; see Fig. 1]. Testing The mean swim latencies for each group on the three testing days (Days 4, 8, and 12) were analyzed independently. Prenatal exposure to ethanol did not significantly impair performance on either the first or second test day but, following the 2-day training– testing period, did significantly impair animals’ performance. Specifically, a one-way ANOVA between conditions showed no significant difference in performance on the first day of testing [F(3, 30)5 0.26, p . 0.15] or on the second day of testing [F(3, 30) 5 1.59, p . 0.05]. However, on the last day of testing, a one-way ANOVA revealed a significant difference in performance [F(3, 30) 5 3.51, p , 0.025]. t-Tests conducted on latency data
66
MATTHEWS AND SIMSON
FIG. 2. Mean latency for each of the three test days. No significant difference between conditions was found on the first or second test day, but a significant impairment in spatial memory was found between conditions on the last test day. Note: Day 12 followed the 2-day training–testing delay period. Error bars denote SEM.
from the last day of testing revealed that No Treatment Control animals and Isocalorie Control animals did not perform at a level significantly different from one another (t 5 0.29, df (16), p . 0.15), but animals in both conditions performed significantly better than animals in the High EtOH condition (t 5 2.80, df (16), p , 0.01; t 5 2.65, df (16), p , 0.01, respectively). Animals in the Low EtOH condition did not perform at a level significantly different from animals in either the No Treatment Control condition, the Isocalorie Control condition, or the High EtOH condition (t 5 0.99, 1.10, and 1.60, respectively; all df (16), p . 0.10; see Fig. 2). To investigate if performance within a group differed due to the repeated testing, one-way ANOVAs were conducted on the test scores for each group. It was found that performance for animals in the No Treatment Control, Isocalorie Control, or High EtOH condition did not differ as a function of the test trials (one-way ANOVA, F 5 2.59, 2.21, and 1.17, p . 0.05, respectively). However, performance in the Low EtOH group did differ as a function of the test trial [one-way ANOVA, F(2, 24) 5 4.89, p , 0.05]. Specifically, animals performed significantly better on the second 24-h delay test trial than they did on the first 24-h delay test trial (t 5 2.60, df (16), p , 0.02) but they performed significantly worse on the 72-h delay test trial compared to the second 24-h delay test trial (t 5 2.35, df (16), p , 0.05) (data not shown). DISCUSSION
This study demonstrates that exposure to ethanol during gestation produces an impairment in spatial memory that is sensitive to the interval between training and testing. This is evidenced by the fact that ethanol impaired spatial memory when the interval between training and testing was 72 h, but failed to impair spatial memory when the interval between training and testing was 24. Although prenatal exposure to ethanol impaired performance on a spatial task, it did not impair performance when animals could learn the task using a nonspatial strategy. That is, no significant
difference was found between any conditions when animals were trained to swim to the visible platform. Thus, it appears that prenatal exposure to ethanol produces a selective impairment of spatial, but not nonspatial, memory in this task. The deficit in performance of the spatial memory task cannot be due to a deficit in either motivation to escape or gross motor function produced by prenatal exposure to ethanol. This is evidenced by the fact that during the nonspatial training period and during the first two spatial memory tests, no significant differences in performance were found. Therefore, it appears that all subjects, regardless of the level of ethanol to which they were exposed prenatally, were motivated to escape and were not impaired in their ability to swim. The present study supports and extends earlier reports (23) that prenatal exposure to ethanol impairs spontaneous alteration at long delays (.30 s) in adult rats but does not impair spontaneous alteration at short delays (,10 s) in adult rats. Specifically, the current results demonstrate that an effect of the delay period is also observed in the Morris water maze task and can occur at much longer delays than previously reported. It has been proposed that the primary effect of prenatal exposure to ethanol is to produce a delay in development (1). This hypothesis has been supported by studies demonstrating that animals exposed prenatally to ethanol appear to “recover” from impairments when they are tested as adults (8,13). In contrast, it has also been hypothesized that animals exposed prenatally to ethanol develop compensatory strategies as adults that are not adequate when animals are tested under challenging conditions (28). The present results support the latter hypothesis in that no impairment in spatial memory was found on the first or second test day (the test days which followed a 24-h memory decay period), whereas a significant impairment in spatial memory was found on the last test day (the test day which followed a 72-h memory decay period). That is, it may well be that the 72-h training–testing delay period posed more of a challenge to the animal than the 24-h training–testing delay period, and thus it is on the last test day that
PRENATAL EXPOSURE TO ETHANOL
67
an impairment of spatial memory by prenatal ethanol exposure is obtained. These results support previous research demonstrating that prenatal exposure to ethanol impairs rats’ performance on spatial tasks (6,7,10,13,27) and extend these earlier results by demonstrating that this impairment is selective in that performance on a nonspatial task is not impaired. These behavioral results parallel anatomical and electrophysiological evidence that prenatal exposure to ethanol alters the structure and function of the hippocampus (3,5,30), one brain site necessary for successful performance on spatial tasks (25). For example, hippocampal pyramidal neurons are generated between Gestational Days 16 –20 (4), and these neurons have been shown to process spatial information (18,20). Hence, the selective impairment in the use of spatial memory found in this study might
be due to the offspring being exposed to ethanol when the hippocampal pyramidal neurons were generating. ACKNOWLEDGEMENTS
This work was supported in part by a predoctoral award to D. B. M. from the National Institute on Alcohol Abuse and Alcoholism (AA-05414), a FIRST Award to P. E. S. from the National Institute on Alcohol Abuse and Alcoholism (AA-09079), an award to P.E.S. from the Alcoholic Beverage Medical Research Foundation, and a center grant from the National Institute on Alcohol Abuse and Alcoholism (AA-07606) to the Fetal Alcohol Research Center, Detroit, MI. The authors thank Eric Brush for help in data collection, Aaron White for assistance with data analysis, and Dr. Phillip Best for use of his equipment, helpful discussion, and edits on early drafts of the manuscript. Finally, the authors thank Drs. Robert Berman and John Hannigan for their assistance with the generation of the offspring.
REFERENCES 1. Abel, E. L. In utero alcohol exposure and developmental delay of response inhibition. Alcohol Clin. Exp. Res. 6:369 –376; 1982. 2. Abel, E. L. Fetal alcohol syndrome and fetal alcohol effects. New York: Plenum Press; 1984. 3. Abel, E. L.; Jacobson, S.; Sherwin, B. T. In utero alcohol exposure produced functional and structural damage. Neurobehav. Toxicol. Teratol. 5:363–366; 1984. 4. Altman, J.; Bayer, S. A. Prolonged sojourn of developing pyramidal cells in the intermediate zone of the hippocampus and their settling in the stratum pyramidale. J. Comp. Neurol. 310:343–364; 1990. 5. Barnes, D.; Walker, D. Prenatal ethanol exposure permanently reduces the number of pyramidal neurons in rat hippocampus. Dev. Brain Res. 1:333–340; 1981. 6. Blanchard, B. A.; Pilati, M. L.; Hannigan, J. H. The role of stress and age in spatial navigation deficits following prenatal exposure to ethanol. Psychobiology. 18:48 –54; 1990. 7. Blanchard, B. A.; Riley, E. P.; Hannigan, J. H. Deficits on a spatial navigation task following prenatal exposure to ethanol. Neurotoxicol. Teratol. 9:253–258; 1987. 8. Bond, N. W. Prenatal alcohol exposure in rodents: A review of its effects on offspring activity and learning ability. Aust. J. Psychol. 33:331–344; 1981. 9. Clarren, S. K.; Smith, D. W. The fetal alcohol syndrome. New Engl. J. Med. 298:1063–1067; 1978. 10. Gianoulakis, C. Rats exposed prenatally to alcohol exhibit impairment in spatial navigation test. Behav. Brain Res. 36:217–228; 1990. 11. Goodlett, C. R.; Kelly, S. J.; West, J. R. Early postnatal alcohol exposure that produces high blood alcohol levels impairs development of spatial navigation learning. Psychobiology. 15:64 –74; 1987. 12. Goodlett, C. R.; Peterson, S. D. Sex differences in vulnerability to developmental spatial learning deficits induced by limited binge alcohol exposure in neonatal rats. Neurobiology of Learning and Memory. 64:265–275; 1995. 13. Hall, J. L.; Church, M. W.; Berman, R. F. Radial arm maze deficits in rats exposed to alcohol during midgestation. Psychobiology. 22:181– 185; 1993. 14. Jones, K. L.; Smith, D. W. Recognition of the fetal alcohol syndrome in early infancy. Lancet. 2:999 –1001; 1973. 15. Kelly, S. J.; Pierce, D. R.; West, J. R. Microencephaly and hyperactivity in adult rats can be induced by neonatal exposure to high blood alcohol concentrations. Exp. Neurol. 96:580 –593; 1987.
16. Keppel, G. Design and analysis: A researcher’s handbook (3rd ed). Englewood Cliffs, NJ: Prentice Hall; 1991. 17. Matthews, D. B.; Best, P. J. Fimbria/fornix lesions facilitate the learning of a non-spatial response task. Psychonom. Bull. Rev. 2:113– 116; 1995. 18. Matthews, D. B.; Simson, P.; Best, P. Ethanol alters spatial processing of hippocampal place cells: A mechanism for impaired navigation when intoxicated. Alcohol Clin. Exp. Res. 20:404 – 407; 1996. 19. Meyer, L. S.; Riley, E. P. Behavioral teratology of alcohol. In: Riley, E. P.; Vorhees, C. V., eds. Handbook of behavioral teratology. New York: Plenum Press, 1986:101–140. 20. Miller, V. M.; Best, P. J. Spatial correlates of hippocampal unit activity are altered by lesions of the fornix and entorhinal cortex. Brain Res. 194:311–323; 1980. 21. Morris, R. G. M. Spatial localization does not require the presence of local cues. Learn. Mem. 12:239 –260; 1981. 22. Morris, R. G. M.; Garrud, P.; Rawlins, J. N. P.; O’Keefe, J. Place navigation impaired in rats with hippocampal lesions. Nature. 297: 681– 683; 1982. 23. Nagahara, A. H.; Handa, R. J. Fetal alcohol exposure produces delaydependent memory deficits in juvenile and adult rats. Alcohol Clin. Exp. Res. 21:710 –715; 1997. 24. O’Keefe, J.; Nadel, L.; Keightley, S.; Kill, D. Fornix lesions selectively abolish place learning in the rat. Exp. Neurol. 48:152–166; 1975. 25. O’Keffe, J.; Nadel, L. The hippocampus as a cognitive map. Oxford: Oxford University Press; 1978. 26. 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. Neurol. 9:1465–1472; 1993. 27. Reyes, E.; Wolfe, J.; Savage, D. The effects of prenatal alcohol exposure on radial arm maze performance in adult rats. Physiol. Behav. 46:45– 48; 1989. 28. Riley, E. P. The long-term behavioral effects of prenatal alcohol exposure in rats. Alcohol Clin. Exp. Res. 14:670 – 673; 1990. 29. Riley, E. P.; Lochry, E. A.; Shapiro, N. R. Lack of response inhibition in rats prenatally exposed to alcohol. Psychopharmacology. 62:47–52; 1979. 30. West, J.; Hodges, C.; Black, A. Prenatal exposure to ethanol alters the organization of hippocampal mossy fibers in rats. Science. 221:957– 959; 1981.
PII S0031-9384(98)00019-5
Prenatal Exposure to Ethanol Disrupts Spatial Memory: Effect of the Training–Testing Delay Period DOUGLAS B. MATTHEWS1 AND PETER E. SIMSON Department of Psychology and Center for Neuroscience Research, Miami University, Oxford, OH 45056, USA Received 20 August 1997; Accepted 29 December 1997 MATTHEWS, D. B. AND P. E. SIMSON. Prenatal exposure to ethanol disrupts spatial memory: Effect of training–testing delay period. PHYSIOL BEHAV 64(1) 63– 67, 1998.—The present study investigated how variations in the period of delay between training and testing in the Morris water maze task affect the use of spatial memory in adult rats that were prenatally exposed to ethanol. Previous results utilizing the Morris water maze task have shown that prenatal, or early postnatal, exposure to ethanol produces deficits in the use of spatial memory, a type of memory that is dependent on an intact hippocampus. However, in these prior studies the delay period between the training of animals and the testing of spatial memory is typically fixed at only 1 day. In the current study, which utilized a revised training procedure within the Morris water maze task, the period of delay between training and testing was altered such that it was either 1 day or 3 days. Following the 3-day delay, different levels of prenatal exposure to ethanol impaired the use of spatial memory. In contrast, following the 1-day delay, prenatal exposure to ethanol failed to impair the use of spatial memory. The present study thus shows that prenatal exposure to ethanol differentially affects spatial memory in the Morris water maze task depending on the period of delay between training and testing. © 1998 Elsevier Science Inc. Fetal alcohol syndrome
Prenatal ethanol exposure
Hippocampus
Spatial memory
to ethanol alters hippocampal anatomy and because the hippocampus is critically involved in spatial cognitive processing, one might predict that prenatal exposure to ethanol would impair spatial cognitive processing (6,7,10,11–13,27). This is, in fact, the case. Prenatal exposure to ethanol impairs spatial learning and spatial memory in juvenile rats (6,7,13) and adult rats (10,27). One paradigm often used to investigate the effects of prenatal exposure to ethanol on spatial learning and memory is the Morris water maze task. In the Morris water maze task, a rat is released into a pool of cloudy water in which there exists an escape “island,” or platform. The escape platform can either protrude above the surface of the water, thus being visible to the rat, or be slightly submerged below the surface of the water, thus being hidden from the rat (21). When presented with a visible, protruding platform, the rat can learn the task by using either a nonspatial strategy (i.e., swim to the only visible cue in the pool) or a spatial strategy (i.e., swim to the same place in the room to find the platform). In contrast, when presented with a submerged platform, the rat can only learn to use the spatial strategy (i.e., swim to the same spatial location to find the submerged platform). Lesions to the hippocampal system impair performance on the Morris water maze task when animals can only utilize the spatial strategy, but do not impair performance when animals can utilize a nonspatial strategy (22). Similarly, prenatal exposure to ethanol selectively
EXPOSURE to ethanol throughout gestation in humans can produce an array of morphological changes and cognitive deficits that has been termed “fetal alcohol syndrome” (FAS) (3,14). Even exposure to ethanol during a part of gestation produces some cognitive deficits. This latter syndrome has been termed “fetal alcohol effect” (FAE) (9), a partial form of FAS, and is the syndrome modeled in this study. In rat models of FAS and FAE (15,19,29), exposure to ethanol during gestation produces behavioral changes, cognitive impairments, and neurological changes, as evidenced by increases in the normally suppressed motor activity in open-field tests (19), impairments of a number of other types of response inhibition (29), and reductions in brain weight and brain volume (15). Prenatal exposure to ethanol has also been shown to produce effects on the hippocampal system. For example, it decreases the number of CA1 pyramidal neurons (5), decreases both dendritic branching and the number of dendritic spines on pyramidal neurons (3), and produces aberrant mossy fiber projections from the dentate gyrus to the CA3 region (30). The hippocampal system is critical for spatial learning and memory (25). Lesions to the hippocampus or its afferents impair the learning of spatial tasks (24) while concomitantly facilitating the learning of nonspatial tasks (17,26). Hippocampal lesions also impair the use of spatial memory (22). Because prenatal exposure 1
To whom requests for reprints should be addressed. Present address: Center for Alcohol Studies, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599. E-mail: [email protected]
63
64
MATTHEWS AND SIMSON
impairs performance in the Morris water maze task when animals can only utilize a spatial strategy (10). In animals prenatally exposed to ethanol, the effect of varying the training–testing delay period in the Morris water maze task has yet to be investigated. There is reason to believe the training– testing period is an important variable in determining the effect of ethanol on spatial memory because, in another task involving spatial learning and memory, adult rats prenatally exposed to ethanol show an impairment in “spontaneous alteration” that is sensitive to a forced time delay during testing (23). Specifically, short time delays (,10 s) do not impair spontaneous alteration, but long delays (.30 s) do impair spontaneous alteration (23). Consequently, this study sought to determine whether variations in the period of delay between training and testing in the Morris water maze task alter the effect of prenatal exposure to ethanol on spatial memory in adult rats. Specifically, the study investigated whether prenatal exposure to different doses of ethanol produces deficits in spatial memory in the Morris water maze task over both a 1-day and/or a 3-day training–testing delay period. METHOD
Animals Female and male Long–Evans hooded rats (Charles River, Portage, MI), weighing approximately 225 and 250 g, respectively, were housed singly in hanging steel cages in an environmentally controlled vivarium at Wayne State University’s Mott Center. The vivarium was maintained at approximately 24°C on a 12:12 light– dark cycle (12 h with the lights on and 12 h with the lights off) with the lights on at 0700 hours. Breeding Each female rat was placed with a male, picked at random, overnight and the cages were examined the next morning for evidence of a sperm plug. If a plug was found, the dam was assumed to be pregnant and the day was marked as Gestational Day (GD) 0. Dams were weighed on GD 0, 8, and 15. Diet On GD 8, pregnant dams were assigned (counterbalanced for initial body weight) to one of four drug conditions: a No Treatment Control, an Isocalorie Control, Low EtOH (ethanol 3 g/kg), and High EtOH (5 g/kg). Ethanol, 10% w/v, was administered via oral intubation between 1000 and 1200 hours directly into the stomach of the dam. This procedure typically produces mean blood alcohol concentrations (BACs) at 30-min postintubation of 120 mg % (3.0 g/kg EtOH) and 180 mg % (5.0 g/kg EtOH) (J. H. Hannigan, personnel communication). Animals in the Isocalorie Control group received a solution in which maltose-dextrin was substituted for ethanol. Dams received the drug treatment from GD 8 to GD 20. During this time, dams were fed Purina Lab Diet Chow 5012 and the Isocalorie Control and Low EtOH animals were pair fed to the food intake of the High EtOH animals. Lab Chow was available only when animals were not intoxicated and, therefore, no significant difference in food intake was found. Pre-Experimental Treatment Beginning on GD 20, cages were inspected twice per day for pups. Following the day that pups were found, Postnatal Day (PN) 1, litters were pseudorandomly culled to seven males and three females and were cross-fostered to surrogate nonintubated control dams. On PN 1, 7, and 14, a male rat from each litter was randomly selected and removed to be a subject in a different study.
TABLE 1 Training
Testing
Delay Period
Days 1–3 Day 4 Days 5–7 Day 8 Day 9 Days 10–11 Day 12
Pups were weaned at approximately PN 21 and male rats were transported to an animal colony at Miami University. Animals were group housed in PlexiglasTM cages (two to three animals per box) until approximately 10 days before the beginning of testing. At this time, animals were housed singly and handled each day by an experimenter. There were nine rats in the No Treatment Control and High EtOH conditions and eight rats in the Isocalorie Control and Low EtOH conditions. There was no more than one animal per litter in any condition. Task and Apparatus The water tank used in the current study was a circular metal tub, painted white, that was 1.8 m in diameter and 50-cm tall. The water in the tank was 42-cm deep and was made opaque by the addition of white tempera, nontoxic, watercolor paint. The water temperature was maintained at approximately 23°C. The escape platform was made of clear PlexiglasTM, 9.75 3 9.75 3 39.5-cm high, and was submerged approximately 2.5 cm under the water’s surface. In a modification of the typical training procedure utilized by others wherein the escape platform is not visible to the animal during training, the escape platform was fitted with a dark brown wooden cap that protruded from the surface of the water by approximately 2.5 cm, thereby making the platform visible. Training and Testing Procedure The average age of the animals when training began was 86 days (ranging from 84 to 87 days), whereas the animals’ average age at the last test day was 95 days (ranging from 93 to 96 days). The entire training and testing procedure required 12 days. Animals were trained on Days 1–3, 5–7, and 9 with the platform visible to the animal. Animals were tested on Days 4, 8, and 12 with the platform submerged, and therefore not visible, to the animal. Animals were neither trained nor tested on Days 10 and 11. (See Table 1 for the experimental timeline.) The order of animal training was counterbalanced between conditions to control for changes in the test environment. On the training days, animals received four trials in which they were started from each of four unique start locations (N, S, E, W) and were trained to swim to the platform. The platform remained in a constant spatial location in the tank and, as stated earlier, was visible to the animal on all training trials. The order of the start locations was counterbalanced across days and matched between conditions, thereby producing a pseudo-Latin square design. Further, the same sequence of start locations was used on Days 1–3 and 5–7. The training that occurred on Day 9 used the start location sequence of that used on Days 1 and 5. On each training trial, the animal was released, facing the wall, into the tank and the latency of the animal to climb onto the platform was recorded. If the animal did not find the platform
PRENATAL EXPOSURE TO ETHANOL
65
FIG. 1. Mean latency for each of the seven training days. No significant difference was found between conditions. Note: The test days (Days 4, 8, and 12) are not included in the figure. Error bars denote SEM.
within 60 s, the experimenter gently guided the animal to the platform (the requirement of the experimenter to guide the animal to the platform occurred rarely, and only on the first day of training). Once the animal climbed onto the platform, the animal was allowed to remain there for 10 s before being removed and placed in a plastic holding cage. This procedure controlled for the amount of time each animal spent on the platform during training regardless of performance. Following an intertrial interval of 1 min, the animals were placed at the next start location. Testing occurred on Days 4 (24-h delay), 8 (24-h delay), and 12 (72-h delay). The start location sequence for the testing days was such that it completed the Latin square design used during training (thus the start location sequence was different from the sequences used during training). During testing, the platform remained in the same spatial location as that used during training. However, the wooden cap was removed from the platform, thereby hiding the platform from the animal. On each testing trial, the animal was released into the tank facing the wall and the latency to climb onto the platform was recorded. On the first testing day, if the animal did not find the platform within 90 s, it was gently guided to the platform as on the first day of training and then removed from the tank after 10 s. On the remaining two testing days, if the animal failed to find the platform, it was removed by placing it on the platform after 30 s had elapsed. Animals remained on the platform for 10 s before being removed and placed in a holding plastic cage. As during training, the intertrial interval was 1 min. Days 10 and 11 served as a “memory decay,” or training– testing delay, period. On these days, all animals remained in their home cages and were neither trained nor tested. RESULTS
Maternal Litter Characteristics Maternal weight gain during pregnancy was significantly different for dams in the four groups [F(3, 20) 5 4.6, p , 0.02]. Newman–Keuls comparisons revealed that dams in the High EtOH condition gained significantly less weight than those in the remaining three conditions (p , 0.05), whereas dams in the Low EtOH condition, Isocalorie Control condition, and the No Treatment Control condition had weights that did not differ significantly.
Litter size, gender ratio, and the mean birthweight of the male pups did not significantly differ between conditions. Training Days 1–3. Animals in all conditions demonstrated significant learning during the first 3 days of training. A two-way ANOVA for mixed designs utilizing harmonic means (16) for analysis based on repeated measures revealed that subjects’ latencies to climb onto the visible platform significantly decreased as a function of the number of training days [F(2, 60)5 113.11, p , 0.001]. In addition, no significant difference was found between conditions [F(3, 30) 5 0.63, p . 0.50), and no significant interaction (Days 3 Condition) was found [F(6,60) 5 0.39, p . 0.75]. Days 5–7. Animals in all conditions also showed significant learning during the second 3-day block of training. A two-way ANOVA for mixed designs utilizing harmonic means (16) for analysis based on repeated measures revealed that subjects’ latencies to climb onto the platform continued to decrease as a function of the number of training days [F(2, 60) 5 17.13, p , 0.001]. Once again, no significant difference was found between conditions [F(3, 30) 5 0.28, p . 0.75] no significant interaction (Days 3 Condition) was found [F(6, 60) 5 0.80, p . 0.50] Day 9. No significant difference in latency to find the platform existed between conditions on the last training day. A one-way ANOVA revealed no significant difference between conditions [F(3, 30) 5 0.30, p . 0.10; see Fig. 1]. Testing The mean swim latencies for each group on the three testing days (Days 4, 8, and 12) were analyzed independently. Prenatal exposure to ethanol did not significantly impair performance on either the first or second test day but, following the 2-day training– testing period, did significantly impair animals’ performance. Specifically, a one-way ANOVA between conditions showed no significant difference in performance on the first day of testing [F(3, 30)5 0.26, p . 0.15] or on the second day of testing [F(3, 30) 5 1.59, p . 0.05]. However, on the last day of testing, a one-way ANOVA revealed a significant difference in performance [F(3, 30) 5 3.51, p , 0.025]. t-Tests conducted on latency data
66
MATTHEWS AND SIMSON
FIG. 2. Mean latency for each of the three test days. No significant difference between conditions was found on the first or second test day, but a significant impairment in spatial memory was found between conditions on the last test day. Note: Day 12 followed the 2-day training–testing delay period. Error bars denote SEM.
from the last day of testing revealed that No Treatment Control animals and Isocalorie Control animals did not perform at a level significantly different from one another (t 5 0.29, df (16), p . 0.15), but animals in both conditions performed significantly better than animals in the High EtOH condition (t 5 2.80, df (16), p , 0.01; t 5 2.65, df (16), p , 0.01, respectively). Animals in the Low EtOH condition did not perform at a level significantly different from animals in either the No Treatment Control condition, the Isocalorie Control condition, or the High EtOH condition (t 5 0.99, 1.10, and 1.60, respectively; all df (16), p . 0.10; see Fig. 2). To investigate if performance within a group differed due to the repeated testing, one-way ANOVAs were conducted on the test scores for each group. It was found that performance for animals in the No Treatment Control, Isocalorie Control, or High EtOH condition did not differ as a function of the test trials (one-way ANOVA, F 5 2.59, 2.21, and 1.17, p . 0.05, respectively). However, performance in the Low EtOH group did differ as a function of the test trial [one-way ANOVA, F(2, 24) 5 4.89, p , 0.05]. Specifically, animals performed significantly better on the second 24-h delay test trial than they did on the first 24-h delay test trial (t 5 2.60, df (16), p , 0.02) but they performed significantly worse on the 72-h delay test trial compared to the second 24-h delay test trial (t 5 2.35, df (16), p , 0.05) (data not shown). DISCUSSION
This study demonstrates that exposure to ethanol during gestation produces an impairment in spatial memory that is sensitive to the interval between training and testing. This is evidenced by the fact that ethanol impaired spatial memory when the interval between training and testing was 72 h, but failed to impair spatial memory when the interval between training and testing was 24. Although prenatal exposure to ethanol impaired performance on a spatial task, it did not impair performance when animals could learn the task using a nonspatial strategy. That is, no significant
difference was found between any conditions when animals were trained to swim to the visible platform. Thus, it appears that prenatal exposure to ethanol produces a selective impairment of spatial, but not nonspatial, memory in this task. The deficit in performance of the spatial memory task cannot be due to a deficit in either motivation to escape or gross motor function produced by prenatal exposure to ethanol. This is evidenced by the fact that during the nonspatial training period and during the first two spatial memory tests, no significant differences in performance were found. Therefore, it appears that all subjects, regardless of the level of ethanol to which they were exposed prenatally, were motivated to escape and were not impaired in their ability to swim. The present study supports and extends earlier reports (23) that prenatal exposure to ethanol impairs spontaneous alteration at long delays (.30 s) in adult rats but does not impair spontaneous alteration at short delays (,10 s) in adult rats. Specifically, the current results demonstrate that an effect of the delay period is also observed in the Morris water maze task and can occur at much longer delays than previously reported. It has been proposed that the primary effect of prenatal exposure to ethanol is to produce a delay in development (1). This hypothesis has been supported by studies demonstrating that animals exposed prenatally to ethanol appear to “recover” from impairments when they are tested as adults (8,13). In contrast, it has also been hypothesized that animals exposed prenatally to ethanol develop compensatory strategies as adults that are not adequate when animals are tested under challenging conditions (28). The present results support the latter hypothesis in that no impairment in spatial memory was found on the first or second test day (the test days which followed a 24-h memory decay period), whereas a significant impairment in spatial memory was found on the last test day (the test day which followed a 72-h memory decay period). That is, it may well be that the 72-h training–testing delay period posed more of a challenge to the animal than the 24-h training–testing delay period, and thus it is on the last test day that
PRENATAL EXPOSURE TO ETHANOL
67
an impairment of spatial memory by prenatal ethanol exposure is obtained. These results support previous research demonstrating that prenatal exposure to ethanol impairs rats’ performance on spatial tasks (6,7,10,13,27) and extend these earlier results by demonstrating that this impairment is selective in that performance on a nonspatial task is not impaired. These behavioral results parallel anatomical and electrophysiological evidence that prenatal exposure to ethanol alters the structure and function of the hippocampus (3,5,30), one brain site necessary for successful performance on spatial tasks (25). For example, hippocampal pyramidal neurons are generated between Gestational Days 16 –20 (4), and these neurons have been shown to process spatial information (18,20). Hence, the selective impairment in the use of spatial memory found in this study might
be due to the offspring being exposed to ethanol when the hippocampal pyramidal neurons were generating. ACKNOWLEDGEMENTS
This work was supported in part by a predoctoral award to D. B. M. from the National Institute on Alcohol Abuse and Alcoholism (AA-05414), a FIRST Award to P. E. S. from the National Institute on Alcohol Abuse and Alcoholism (AA-09079), an award to P.E.S. from the Alcoholic Beverage Medical Research Foundation, and a center grant from the National Institute on Alcohol Abuse and Alcoholism (AA-07606) to the Fetal Alcohol Research Center, Detroit, MI. The authors thank Eric Brush for help in data collection, Aaron White for assistance with data analysis, and Dr. Phillip Best for use of his equipment, helpful discussion, and edits on early drafts of the manuscript. Finally, the authors thank Drs. Robert Berman and John Hannigan for their assistance with the generation of the offspring.
REFERENCES 1. Abel, E. L. In utero alcohol exposure and developmental delay of response inhibition. Alcohol Clin. Exp. Res. 6:369 –376; 1982. 2. Abel, E. L. Fetal alcohol syndrome and fetal alcohol effects. New York: Plenum Press; 1984. 3. Abel, E. L.; Jacobson, S.; Sherwin, B. T. In utero alcohol exposure produced functional and structural damage. Neurobehav. Toxicol. Teratol. 5:363–366; 1984. 4. Altman, J.; Bayer, S. A. Prolonged sojourn of developing pyramidal cells in the intermediate zone of the hippocampus and their settling in the stratum pyramidale. J. Comp. Neurol. 310:343–364; 1990. 5. Barnes, D.; Walker, D. Prenatal ethanol exposure permanently reduces the number of pyramidal neurons in rat hippocampus. Dev. Brain Res. 1:333–340; 1981. 6. Blanchard, B. A.; Pilati, M. L.; Hannigan, J. H. The role of stress and age in spatial navigation deficits following prenatal exposure to ethanol. Psychobiology. 18:48 –54; 1990. 7. Blanchard, B. A.; Riley, E. P.; Hannigan, J. H. Deficits on a spatial navigation task following prenatal exposure to ethanol. Neurotoxicol. Teratol. 9:253–258; 1987. 8. Bond, N. W. Prenatal alcohol exposure in rodents: A review of its effects on offspring activity and learning ability. Aust. J. Psychol. 33:331–344; 1981. 9. Clarren, S. K.; Smith, D. W. The fetal alcohol syndrome. New Engl. J. Med. 298:1063–1067; 1978. 10. Gianoulakis, C. Rats exposed prenatally to alcohol exhibit impairment in spatial navigation test. Behav. Brain Res. 36:217–228; 1990. 11. Goodlett, C. R.; Kelly, S. J.; West, J. R. Early postnatal alcohol exposure that produces high blood alcohol levels impairs development of spatial navigation learning. Psychobiology. 15:64 –74; 1987. 12. Goodlett, C. R.; Peterson, S. D. Sex differences in vulnerability to developmental spatial learning deficits induced by limited binge alcohol exposure in neonatal rats. Neurobiology of Learning and Memory. 64:265–275; 1995. 13. Hall, J. L.; Church, M. W.; Berman, R. F. Radial arm maze deficits in rats exposed to alcohol during midgestation. Psychobiology. 22:181– 185; 1993. 14. Jones, K. L.; Smith, D. W. Recognition of the fetal alcohol syndrome in early infancy. Lancet. 2:999 –1001; 1973. 15. Kelly, S. J.; Pierce, D. R.; West, J. R. Microencephaly and hyperactivity in adult rats can be induced by neonatal exposure to high blood alcohol concentrations. Exp. Neurol. 96:580 –593; 1987.
16. Keppel, G. Design and analysis: A researcher’s handbook (3rd ed). Englewood Cliffs, NJ: Prentice Hall; 1991. 17. Matthews, D. B.; Best, P. J. Fimbria/fornix lesions facilitate the learning of a non-spatial response task. Psychonom. Bull. Rev. 2:113– 116; 1995. 18. Matthews, D. B.; Simson, P.; Best, P. Ethanol alters spatial processing of hippocampal place cells: A mechanism for impaired navigation when intoxicated. Alcohol Clin. Exp. Res. 20:404 – 407; 1996. 19. Meyer, L. S.; Riley, E. P. Behavioral teratology of alcohol. In: Riley, E. P.; Vorhees, C. V., eds. Handbook of behavioral teratology. New York: Plenum Press, 1986:101–140. 20. Miller, V. M.; Best, P. J. Spatial correlates of hippocampal unit activity are altered by lesions of the fornix and entorhinal cortex. Brain Res. 194:311–323; 1980. 21. Morris, R. G. M. Spatial localization does not require the presence of local cues. Learn. Mem. 12:239 –260; 1981. 22. Morris, R. G. M.; Garrud, P.; Rawlins, J. N. P.; O’Keefe, J. Place navigation impaired in rats with hippocampal lesions. Nature. 297: 681– 683; 1982. 23. Nagahara, A. H.; Handa, R. J. Fetal alcohol exposure produces delaydependent memory deficits in juvenile and adult rats. Alcohol Clin. Exp. Res. 21:710 –715; 1997. 24. O’Keefe, J.; Nadel, L.; Keightley, S.; Kill, D. Fornix lesions selectively abolish place learning in the rat. Exp. Neurol. 48:152–166; 1975. 25. O’Keffe, J.; Nadel, L. The hippocampus as a cognitive map. Oxford: Oxford University Press; 1978. 26. 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. Neurol. 9:1465–1472; 1993. 27. Reyes, E.; Wolfe, J.; Savage, D. The effects of prenatal alcohol exposure on radial arm maze performance in adult rats. Physiol. Behav. 46:45– 48; 1989. 28. Riley, E. P. The long-term behavioral effects of prenatal alcohol exposure in rats. Alcohol Clin. Exp. Res. 14:670 – 673; 1990. 29. Riley, E. P.; Lochry, E. A.; Shapiro, N. R. Lack of response inhibition in rats prenatally exposed to alcohol. Psychopharmacology. 62:47–52; 1979. 30. West, J.; Hodges, C.; Black, A. Prenatal exposure to ethanol alters the organization of hippocampal mossy fibers in rats. Science. 221:957– 959; 1981.