- Email: [email protected]
Impaired spatial learning after hypergravity exposure in rats
Cognitive Brain Research 22 (2004) 94 – 100 www.elsevier.com/locate/cogbrainres
Research report
Impaired spatial learning after hypergravity exposure in rats Kenji Mitani, Arata Horii*, Takeshi Kubo Department of Otolaryngology, Osaka University Medical School, 2-2 Yamadaoka, Suita, Osaka 565-0871, Japan Accepted 9 August 2004 Available online 11 September 2004
Abstract Most astronauts experience spatial disorientation after exposure to weightlessness, indicating that constant gravity is utilized as a stable external reference during spatial cognition. We attempted to elucidate the role of constant gravity in spatial learning using a radial arm maze test on rats housed in a hypergravity environment (HG) produced by a centrifuge device. Male Wistar rats were kept in 2G linear acceleration for 2 weeks before the spatial learning task, which lasted for 10 days. The control rats were placed close to the centrifuge device but not exposed to hypergravity. Spatial learning was evaluated by the accuracy and the re-entry rate, which were the rate of correct arm entries and the rate of entries into the arms that they had already visited, respectively. Locomotor activity was measured by number of entries per minute. The number of baits the animal took per minute was also measured. The results showed that accuracy was significantly inferior and the re-entry rate was significantly higher in the HG rats than in the controls, suggesting that animals use a constant gravity as a stable external reference in spatial learning. However, these differences disappeared at 5 days later, indicating that the HG rats learned the spatial task more rapidly than the controls. Locomotor activity was higher in the HG rats and there was no difference in number of baits per minute between the HG and control animals. In conclusion, if one sensory cue necessary for spatial cognition is disturbed by gravity change, animals can subsidize with other sensory cues such as proprioceptive and motor efference copy signals through increased locomotor activities. D 2004 Elsevier B.V. All rights reserved. Theme: Motor systems and sensorimotor integration Topic: Vestibular system Keywords: Hypergravity; Locomotor activity; Radial arm maze; Rat; Spatial orientation
1. Introduction For the correct spatial cognition, both allocentric and egocentric cues are necessary [9]. The vestibular system, which is primarily a sensor of linear and angular acceleration, provides egocentric information during spatial navigation [13,19]. According to the neural mismatch theory of motion sickness, a mismatch of multisensory spatial cognitive systems would induce the spatial disorientation and vegetative symptoms in motion sickness [3,11]. Recently, this hypothesis was extended to explain the underlying mechanisms of space adaptation syndrome
* Corresponding author. Tel.: +81 6 6879 3951; fax: +81 6 6879 3959. E-mail address: [email protected] (A. Horii). 0926-6410/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.cogbrainres.2004.08.002
and peripheral vestibular vertigo [3]. In space, there is a lack of constant Earth’s gravity which stimulates otolith organs on the ground, while semicircular canals are still stimulated by head movement. This mismatch information from the otolith organs and from the semicircular canals could explain the space adaptation syndrome. Unilateral peripheral vestibular disorders could induce spatial disorientation due to a mismatch between bilateral vestibular systems. As support for this hypothesis, the vestibular system has been recognized as an important sensory system required for correct spatial cognition: it was demonstrated that normal vestibular function is necessary for the spatial firing of hippocampal bplace cellsQ, which only fire when an animal is located at specific areas in space [13,19], and also for the spatial learning tested by several behavioral tasks in animals and humans
K. Mitani et al. / Cognitive Brain Research 22 (2004) 94–100
[1,10,14,15,18,23]. These reports suggest an important role of the vestibular system in spatial cognition as well as in visual and postural stabilization during head movement. Most astronauts experience spatial disorientation and space adaptation syndrome for several days after exposure to weightlessness [3,8,12], indicating that constant gravity, mainly sensed by the otolith vestibular organs, is utilized as a stable external reference during spatial cognition. In this study, we attempted to further elucidate the role of constant gravity in spatial learning using a radial arm maze test on animals housed in a hypergravity environment (HG: 2G) produced by a centrifuge device.
2. Materials and methods All experiments were approved by the Animal Care Committee of Osaka University Medical School. Experimental protocol was summarised in a scheme (Fig. 1).
95
2.1. Animals Wistar rats (Japan SLC, Shizuoka, Japan) weighing 150– 180 g at the beginning of the experiments were used. The animals were housed individually with free access to food and water under a 12-h light/12-h dark cycle (light on at 8:00 a.m.). Before habituation to the radial arm maze, rats were handled and food-deprived to 80–90% of their freefeeding weight for at least a week (Days 2–12). 2.2. Radial arm maze The radial arm maze consisted of an octagonal central platform with a diameter of 33 cm, with eight arms 50 cm long and 13 cm wide radiating from the center. The maze had a 3-cm-high rail around it, and 24-cm-high and 25-cm-long walls at the central side of the arms to prevent rats from moving from one arm to an adjacent arm without returning to the platform. Each arm had a food well 3 cm in diameter and 0.5 cm deep at the distal end, which was not visible from the
Fig. 1. Experimental protocol. Animals were centrifuged for 2 weeks in HG group, while the control animals were just placed beside the centrifuge device for the same period. Before and after the centrifugation (or control stimulation), habituation and shaping trial and re-habituation trial was made, respectively. Between the Session 1 (Days 29–33, Test days 1–5) and Session 2 (Days 35–39, Test days 6–10) of spatial learning task, there was a 1-day interval (Day 34).
96
K. Mitani et al. / Cognitive Brain Research 22 (2004) 94–100
platform. The platform, arms and food wells of the maze were made of black Plexiglas. The walls and rails were made of clear Plexiglas, so that the rats could see the several extramaze cues. The apparatus was placed 39 cm above the floor of the maze room. In the maze room, various extramaze cues such as a desk, posters, a radio, a chair and doors were available and their locations were constant throughout the experiments. The radio was used as masking against noise from the outside of the maze room and as a constant auditory cue. A digital video camera was placed on the ceiling of the maze room just above the center of the platform for recording the behavior of the animals in the maze.
fourth, 7th and 10th days. The control rats (n=8) were housed individually in the same room as the centrifuge and their cages were located beside the device. There was no difference in visual information between the control and HG animals. 2.5. Re-habituation On the next day after centrifugation (Day 28), all animals received three additional re-habituation trials at 5-min interval in the radial arm maze. They were allowed to explore the maze for five min in each trial. The baits were limited to the food wells of four arms (Nos. 1, 2, 4, and 7).
2.3. Habituation and shaping 2.6. Spatial learning task After handling and food deprivation, the rats received four habituation trials (two trials per day) in the radial arm maze for two consecutive days (Days 11–12). A small piece of a corn flake was used as a reinforcer. During habituation, the rats were gradually shaped to go to the distal end of the arm and to eat a reinforcer. On the first day of habituation and shaping (Day 11), they were transported to the maze room within a small cage (holding cage), and left on the desk for 5 min in order to habituate the maze room. Reinforcers were scattered throughout the maze and each rat was placed on the platform facing towards Arm No. 1, which was directed towards the experimenter’s location. They were allowed to explore the maze for 20 min and then returned to their home cage. Four hours later, the rats were transported to the maze room again to undergo the second habituation trial. After habituation to the maze room for 5 min in the holding cage, they were released at the platform of the maze according to the procedure of the first trial, but reinforcers were available only in the arms. The rats were returned to their home cage following a 10-min free run. On the next day (Day 12), the third and fourth habituation trials were performed with a 4-h interval, for 10 min each. The reinforcers were restricted to the ends of the arms for the third trial and to only the food wells for the fourth trial. 2.4. Centrifugation Sixteen rats were randomly divided into the hypergravity (HG) and the control groups. Eight animals were exposed to hypergravity in a centrifuge device [20] for 2 weeks (Days 13–27), which started on the next day of habituation. Each animal was located individually in a centrifuge cage (252020 cm), which hung at the distal end of a turning arm (diameter: 1 m). The arms were rotated at an angular velocity of 3368/s (56 r.p.m.), so each animal was constantly kept in 2G linear acceleration perpendicular to the floor of the centrifuge cage. During centrifugation, rats could freely access food and water and were under the same light–dark cycle as before. In order to supply food and water and clean the cage, the centrifuge was stopped for about 20 min at the
The spatial learning task started on the second day after centrifugation. The rats received three spatial trials for 5 days (Session 1: Days 29–33). After a 1-day interval (Day 34), they were subjected to the same trials for another 5 days (Session 2: Days 35–39). The trial was performed in the following manner. Each rat was transferred to the maze room within the holding cage, and then left on the desk. Five minutes later, the rat was placed on the platform of the radial arm maze with its head facing towards Arm No. 1. Reinforcers were available in the food wells of Arms Nos. 1, 2, 4, and 7. The animal was allowed to run on the platform and in the arms freely until all four reinforcers were taken or 5 min had elapsed, and then was returned to the holding cage on the desk. Three trials were carried out each day, with a 5-min interval. During this interval, the surface of the maze was wiped with 70% ethanol and, if necessary, reinforcers were supplied to the food wells. After the third trial of the day, the animal was returned to the home cage. 2.7. Behavioral measures The behavior of the rats in the maze was recorded with a digital video camera and the number of arms the animals had entered in the trial was scored. An arm entry was counted if all four limbs of the rat were within an arm. The following parameters were measured: (1) accuracy, (2) reentry rate, (3) entries/min and (4) baits/min. Accuracy was defined as the percentage of correct entries in all arm entries. A correct entry was scored when the rat entered a correct arm (Nos. 1, 2, 4, or 7) and ate a reinforcer. The re-entry rate was defined as the ratio of the number of entries into an arm that the animal had previously entered in the same trial to the number of all arm entries. Entries/min was defined as the number of entries into an arm per minute. Baits/min was defined as the number of baits that the rat took per minute. 2.8. Statistical analysis Analysis of variance (ANOVA) for repeated measures was conducted to evaluate the effects of time, group, and the
K. Mitani et al. / Cognitive Brain Research 22 (2004) 94–100
97
Fig. 2. Accuracy in entering the correct arms. (A) Accuracy in entering the correct arms was inferior in the hypergravity rats (closed circles), compared to that in the control rats (open circles). (B) During the early phase of the test (Session 1, Test days 1–5), the hypergravity animals (closed column) showed lower accuracy compared to that of the control animals (open column), while there was no difference at the later phase (Session 2, Test days 6–10).
interaction of time and groups. When significance of differences was shown by ANOVA between groups, comparisons in each session were made by paired t-test or one-way ANOVA, followed by Scheffe’s test. P values under 0.05 were determined to be significant.
following the 2-weeks centrifugation period (see Section 2.5), all rats took the baits and did not defecate in the maze, suggesting that they were still habituated to the maze and the experimenter’s manipulations, even after a 2-week interval of centrifugation. 3.2. Spatial learning task
3. Results 3.1. Behavioral observation at habituation periods before and after centrifugation At the first habituation trial (see Section 2.3), the rats often froze and defecated, and exploration behavior throughout the maze was observed. At the fourth (last) trial, all the rats had been shaped to search for the baits, and defecation was not observed. At the re-habituation trial
As shown in Fig. 2A, the accuracy of the HG rats was lower than that of the control rats. ANOVA for repeated measures revealed significant effects of group [ F (1, 14)=5.059, p=0.041] and day [ F (9, 126)=3.095, p= 0.00217], but no effect of groupday interaction [ F (9, 126)=1.514, p=0.150]. We classified Test days 1–5 (Days 29–33) as Session 1 and Test days 6–10 (Days 35–39) as Session 2. When the accuracy was compared between sessions, a significant effect of groupsession interaction
Fig. 3. Re-entry rate of entering the arms that the animals had already entered. (A) The re-entry rate was higher in the hypergravity rats (closed circles) compared to that in the control rats (open circles). (B) During the early phase of the test (Session 1, Test days 1–5), the hypergravity animals (closed column) showed a higher re-entry rate compared to that of the control animals (open column), while there was no difference at the later phase (Session 2, Test days 6–10).
98
K. Mitani et al. / Cognitive Brain Research 22 (2004) 94–100
Fig. 4. The number of entries per minute (A) and the number of baits per minute (B). Entries/min was higher in the hypergravity rats (A, closed circles) compared to that in the control rats (A, open circles). There was no difference in baits/min between the hypergravity rats (B, closed circles) and control rats (B, open circles).
was shown [ F (1, 14)=7.949, p=0.0137]. One-way ANOVA of the four groups divided by group and session revealed a significant group effect [ F (3, 28)=8.146, p=0.00047]. Posthoc analysis showed that the HG rats chose more incorrect arms in Session 1 ( p=0.00147), but not in Session 2 ( p=0.965) (Fig. 2B). As shown in Fig. 3A, the re-entry rate of the HG rats was higher than that of the control rats. ANOVA for repeated measures revealed a significant effect of group [ F (1, 14)=8.622, p=0.0108], but no effect of day [ F (9, 126)=1.216, p=0.291] or groupday interaction [ F (9, 126)=1.265, p=0.262]. A significant group effect was also shown on the analysis of the two sessions [ F (1, 14)=8.622, p=0.0108]. The re-entry rate of the HG rats was higher than that of the control rats in Session 1 ( p=0.00415), but not in Session 2 ( p=0.386) (Fig. 3B). 3.3. Entries/min and baits/min As shown in Fig. 4A, the HG rats entered more arms per minute than the control animals. ANOVA for repeated measures revealed significant effects of group [ F (1, 14)=4.636, p =0.0492], day [ F (9, 126)=18.323, pb0.00001] and groupday interaction [ F (9, 126)=3.347, p=0.00105]. Fig. 4B shows the number of baits taken by rats per minute (baits/min). While it seems that the HG rats took more baits/min than the control animals, there was no difference in baits/min between the HG and control animals, as revealed by ANOVA for repeated measures [ F (1, 14)=3.527, p=0.0813]. Groupday interaction was significant [ F (9, 126)=2.524, p=0.0108].
4. Discussion The present spatial learning test using a radial arm maze revealed that the accuracy for choosing a correct arm was
significantly lower in the HG rats compared to that of the control rats (Fig. 2). The re-entry rate, which is an indicative of a spatial working memory, was also higher in the HG rats compared to that of the control rats (Fig. 3). Therefore, it is suggested that the constant gravity, one of egocentric vestibular cues, is essential for the spatial learning. In contrast, it was reported that allocentric visual cues are important for the spatial reference memory rather than the spatial working memory [5]. While it is difficult to assess the spatial cognition performance and memory performance separately from the radial maze test, these results indicate that hypergravity exposure for 2 weeks impaired the spatial learning process in the rat, which is generally consistent with previous animal experiments [7,17] and astronaut data [3,8]. Previous data regarding the effects of hypergravity on the peripheral otolith organs have demonstrated neither morphological changes nor synthesis of otoconia [6,16,21], suggesting that the present impairment of spatial learning might not be attributed to the effects of hypergravity on the peripheral otolith organs. In contrast, cerebellar inhibition of the vestibular nucleus and decreased neurotransmission between vestibular hair cells and primary vestibular afferents might reduce the excess of otolith inputs by hypergravity [22]. These changes in peripheral and central vestibular neurotransmission might also contribute to impaired spatial learning through decreased vestibular inputs to areas important for spatial cognition, such as the hippocampus. Indeed, changes in hippocampal gene expression following hypergravity have recently been reported, and some of these changes could be the molecular background of spatial disorientation due to hypergravity [4]. Even though the HG rats showed impairment of spatial cognition, they could take almost the same or a relatively high amount of reinforcers in a minute (baits/min) (Fig. 4B). This was achieved by increased entries into the arms due to hyperkinetic behavior (Fig. 4A). Hyperkinetic behavior by hypergravity exposure have already been reported [2,17],
K. Mitani et al. / Cognitive Brain Research 22 (2004) 94–100
however, this is the first report to demonstrate that increased locomotor activity due to hypergravity exposure subsidized the impairment of spatial cognition and brought about no deficits in terms of getting reinforcers. The HG animals initially showed impaired spatial learning compared to the control animals, however, there was no difference in accuracy or re-entry rate between the HG animals and the control animals at the later stage (Figs. 2B, 3B). This means that the HG animals learned the spatial task more rapidly than the control animals, which is probably due to increased locomotor activity after hypergravity exposure subsidizing a loss of vestibular egocentric information by increasing information from other egocentric cues such as proprioceptive and motor efference copy signals. We cannot exclude a possibility that the spatial learning performance was limited in the control rats by a ceiling effect at the early phase of the spatial task so that the HG rats seemed to learn a spatial task more rapidly than the control animals. However, this interpretation may be unlikely because we found that both of the control and HG rats did not show a ceiling effect even 15 days after the start of the spatial task (data not shown). Therefore, the lack of deficits in getting reinforcers could have been brought about by not only an accidental increase in chances to enter the correct arm, but also by compensated spatial orientation, both of which were results of increased locomotor activities. In a radial arm maze test, animals with bilateral vestibular damage showed a similar trend regarding impaired spatial learning and final reinforcers: impaired spatial learning subsidized by increased locomotor activity resulted in no deficits in getting reinforcers [14]. Spatial firing of hippocampal place cells, which have important roles in spatial cognition, is disturbed following bilateral vestibular damage, indicating that the vestibular signal is an important egocentric cue in navigation tasks [13,18]. As in the study of bilateral vestibular damage, the HG rats subsidized their impaired spatial cognition by increased motor activities. It is suggested that animals need a constant gravity for spatial learning, to utilize as a stable external reference. If one sensory cue was disturbed (e.g., by bilateral vestibular damage or gravity change), animals subsidized it with other sensory cues, such as proprioceptive and motor efference copy signals, through increased locomotor activities.
Acknowledgements This study was partly supported by Grants-in-Aid for scientific research from the Ministry of Education, Culture, Sports, Science and Technology of Japan for AH and by a research grant for intractable diseases (vestibular disorders) from the Ministry of Health, Labor and Welfare of Japan for TK. We thank Prof. A. Yamatodani for a valuable suggestion on the radial arm maze test and Drs. A. Uno and A. Nakagawa for helpful discussion.
99
References [1] A. Berthoz, I. Israel, P. Georges-Francois, R. Grasso, T. Tsuzuku, Spatial memory of body linear displacement: what is being stored? Science 269 (1995) 95 – 98. [2] V. Bouet, Y. Gahery, M. Lacour, Behavioural changes induced by early and long-term gravito-inertial force modification in the rat, Behav. Brain Res. 17 (2003) 97 – 104. [3] T. Brandt, Vertigo: Its Multisensory Syndromes, 2nd ed., SpringerVerlag, London, 1999. [4] A. Del Signore, S. Mandillo, A. Rizzo, E. Di Mauro, A. Mele, R. Negri, A. Oliverio, P. Paggi, Hippocampal gene expression is modulated by hypergravity, Eur. J. Neurosci. 19 (2004) 667 – 677. [5] E. Jakubowska-Dogru, U. Gumusbas, F. Kara, Individual variation in the spatial reference and working memory assessed under allothetic and idiothetic orientation cues in rat, Acta Neurobiol. Exp. (Wars) 63 (2003) 17 – 23. [6] D.J. Lim, J.A. Stith, C.W. Stockwell, J. Oyama, Observations on saccules of rats exposed to long-term hypergaravity, Aerosp. Med. 45 (1974) 705 – 710. [7] S. Mandillo, A. Del Signore, P. Paggi, N. Francia, D. Santucci, A. Mele, A. Oliverio, Effects of acute and repeated daily exposure to hypergravity on spatial learning in mice, Neurosci. Lett. 336 (2003) 147 – 150. [8] A.B. Newberg, Changes in the central nervous system and their clinical correlates during long-term spaceflight, Aviat. Space Environ. Med. 65 (1994) 562 – 572. [9] J. O’Keefe, L. Nadel, The Hippocamous as A Cognitive Map, Clarendon Press, Oxford, 1978. [10] P. Peruch, L. Borel, F. Gaunet, G. Thinus-Blanc, J. Magnan, M. Lacour, Spatial performance of unilateral vestibular defective patients in nonvisual versus visual navigation, J. Vestibular Res. 9 (1999) 37 – 47. [11] J.T. Reason, Motion sickness adaptation: a neural mismatch model, J. R. Soc. Med. 71 (1978) 819 – 829. [12] M.F. Reschke, J.J. Bloomberg, D.L. Harm, W.H. Paloski, C. Layne, V. McDonald, Posture, locomotion, spatial orientation, and motion sickness as a function of space flight, Brain Res. Rev. 28 (1998) 102 – 117. [13] N.A. Russell, A. Horii, P.F. Smith, C.L. Darlington, D.K. Bilkey, Long-term effects of permanent vestibular lesions on hippocampal spatial firing, J. Neurosci. 23 (2003) 6490 – 6498. [14] N.A. Russell, A. Horii, P.F. Smith, C.L. Darlington, D.K. Bilkey, Bilateral peripheral vestibular lesions produce long-term changes in spatial learning in the rat, J. Vestibular Res. 13 (2003) 9 – 16. [15] H.N.P.M. Sondag, H.A.A. deJong, W.J. Oosterveld, Effect of prolonged hypergravity on the vestibular system: a behavioural study, ORL J. Oto-rhinolaryngol. Relat. Spec. 57 (1995) 189 – 193. [16] H.N.P.M. Sondag, H.A.A. deJong, J. van Marle, W.J. Oostervel, Effects of sustained acceleration on the morphological properties of otoconia in hamsters, Acta Oto-Laryngol. (Stockh.) 115 (1995) 227 – 230. [17] H.N.P.M. Sondag, H.A.A. deJong, W.J. Oosterveld, Altered behaviour of hamsters by prolonged hypergravity: adaptation to 2.5 G and re-adaptation to 1 G, Acta Oto-Laryngol. (Stockh.) 116 (1996) 192 – 197. [18] R.W. Stackman, A.M. Herbert, Rats with lesions of the vestibular system require a visual landmark for spatial navigation, Behav. Brain Res. 128 (2002) 27 – 40. [19] R.W. Stackman, A.S. Clark, J.S. Taube, Hippocampal spatial representations require vestibular input, Hippocampus 12 (2002) 291 – 303. [20] N. Takeda, A. Horii, A. Uno, M. Morita, T. Mochizuki, A. Yamatodani, T. Kubo, A ground-based animal model of space adaptation syndrome, J. Vestibular Res. 6 (1996) 403 – 409.
100
K. Mitani et al. / Cognitive Brain Research 22 (2004) 94–100
[21] Y. Uno, A. Horii, M. Umemoto, T. Hasegawa, K. Doi, A. Uno, T. Takemura, T. Kubo, Effects of hypergravity on morphology and osteopontin expression in the rat otolith organs, J. Vestibular Res. 10 (2000) 283 – 289. [22] Y. Uno, A. Horii, A. Uno, Y. Fuse, M. Fukushima, K. Doi, T. Kubo, Quantitative changes in mRNA expression of glutamate receptors in
the rat peripheral and central vestibular systems following hypergravity, J. Neurochem. 81 (2002) 1308 – 1317. [23] D.G. Wallace, D.J. Hines, S.M. Pellis, I.Q. Whishaw, Vestibular information is required for dead reckoning in the rat, J. Neurosci. (2002) 10009 – 10017.
Research report
Impaired spatial learning after hypergravity exposure in rats Kenji Mitani, Arata Horii*, Takeshi Kubo Department of Otolaryngology, Osaka University Medical School, 2-2 Yamadaoka, Suita, Osaka 565-0871, Japan Accepted 9 August 2004 Available online 11 September 2004
Abstract Most astronauts experience spatial disorientation after exposure to weightlessness, indicating that constant gravity is utilized as a stable external reference during spatial cognition. We attempted to elucidate the role of constant gravity in spatial learning using a radial arm maze test on rats housed in a hypergravity environment (HG) produced by a centrifuge device. Male Wistar rats were kept in 2G linear acceleration for 2 weeks before the spatial learning task, which lasted for 10 days. The control rats were placed close to the centrifuge device but not exposed to hypergravity. Spatial learning was evaluated by the accuracy and the re-entry rate, which were the rate of correct arm entries and the rate of entries into the arms that they had already visited, respectively. Locomotor activity was measured by number of entries per minute. The number of baits the animal took per minute was also measured. The results showed that accuracy was significantly inferior and the re-entry rate was significantly higher in the HG rats than in the controls, suggesting that animals use a constant gravity as a stable external reference in spatial learning. However, these differences disappeared at 5 days later, indicating that the HG rats learned the spatial task more rapidly than the controls. Locomotor activity was higher in the HG rats and there was no difference in number of baits per minute between the HG and control animals. In conclusion, if one sensory cue necessary for spatial cognition is disturbed by gravity change, animals can subsidize with other sensory cues such as proprioceptive and motor efference copy signals through increased locomotor activities. D 2004 Elsevier B.V. All rights reserved. Theme: Motor systems and sensorimotor integration Topic: Vestibular system Keywords: Hypergravity; Locomotor activity; Radial arm maze; Rat; Spatial orientation
1. Introduction For the correct spatial cognition, both allocentric and egocentric cues are necessary [9]. The vestibular system, which is primarily a sensor of linear and angular acceleration, provides egocentric information during spatial navigation [13,19]. According to the neural mismatch theory of motion sickness, a mismatch of multisensory spatial cognitive systems would induce the spatial disorientation and vegetative symptoms in motion sickness [3,11]. Recently, this hypothesis was extended to explain the underlying mechanisms of space adaptation syndrome
* Corresponding author. Tel.: +81 6 6879 3951; fax: +81 6 6879 3959. E-mail address: [email protected] (A. Horii). 0926-6410/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.cogbrainres.2004.08.002
and peripheral vestibular vertigo [3]. In space, there is a lack of constant Earth’s gravity which stimulates otolith organs on the ground, while semicircular canals are still stimulated by head movement. This mismatch information from the otolith organs and from the semicircular canals could explain the space adaptation syndrome. Unilateral peripheral vestibular disorders could induce spatial disorientation due to a mismatch between bilateral vestibular systems. As support for this hypothesis, the vestibular system has been recognized as an important sensory system required for correct spatial cognition: it was demonstrated that normal vestibular function is necessary for the spatial firing of hippocampal bplace cellsQ, which only fire when an animal is located at specific areas in space [13,19], and also for the spatial learning tested by several behavioral tasks in animals and humans
K. Mitani et al. / Cognitive Brain Research 22 (2004) 94–100
[1,10,14,15,18,23]. These reports suggest an important role of the vestibular system in spatial cognition as well as in visual and postural stabilization during head movement. Most astronauts experience spatial disorientation and space adaptation syndrome for several days after exposure to weightlessness [3,8,12], indicating that constant gravity, mainly sensed by the otolith vestibular organs, is utilized as a stable external reference during spatial cognition. In this study, we attempted to further elucidate the role of constant gravity in spatial learning using a radial arm maze test on animals housed in a hypergravity environment (HG: 2G) produced by a centrifuge device.
2. Materials and methods All experiments were approved by the Animal Care Committee of Osaka University Medical School. Experimental protocol was summarised in a scheme (Fig. 1).
95
2.1. Animals Wistar rats (Japan SLC, Shizuoka, Japan) weighing 150– 180 g at the beginning of the experiments were used. The animals were housed individually with free access to food and water under a 12-h light/12-h dark cycle (light on at 8:00 a.m.). Before habituation to the radial arm maze, rats were handled and food-deprived to 80–90% of their freefeeding weight for at least a week (Days 2–12). 2.2. Radial arm maze The radial arm maze consisted of an octagonal central platform with a diameter of 33 cm, with eight arms 50 cm long and 13 cm wide radiating from the center. The maze had a 3-cm-high rail around it, and 24-cm-high and 25-cm-long walls at the central side of the arms to prevent rats from moving from one arm to an adjacent arm without returning to the platform. Each arm had a food well 3 cm in diameter and 0.5 cm deep at the distal end, which was not visible from the
Fig. 1. Experimental protocol. Animals were centrifuged for 2 weeks in HG group, while the control animals were just placed beside the centrifuge device for the same period. Before and after the centrifugation (or control stimulation), habituation and shaping trial and re-habituation trial was made, respectively. Between the Session 1 (Days 29–33, Test days 1–5) and Session 2 (Days 35–39, Test days 6–10) of spatial learning task, there was a 1-day interval (Day 34).
96
K. Mitani et al. / Cognitive Brain Research 22 (2004) 94–100
platform. The platform, arms and food wells of the maze were made of black Plexiglas. The walls and rails were made of clear Plexiglas, so that the rats could see the several extramaze cues. The apparatus was placed 39 cm above the floor of the maze room. In the maze room, various extramaze cues such as a desk, posters, a radio, a chair and doors were available and their locations were constant throughout the experiments. The radio was used as masking against noise from the outside of the maze room and as a constant auditory cue. A digital video camera was placed on the ceiling of the maze room just above the center of the platform for recording the behavior of the animals in the maze.
fourth, 7th and 10th days. The control rats (n=8) were housed individually in the same room as the centrifuge and their cages were located beside the device. There was no difference in visual information between the control and HG animals. 2.5. Re-habituation On the next day after centrifugation (Day 28), all animals received three additional re-habituation trials at 5-min interval in the radial arm maze. They were allowed to explore the maze for five min in each trial. The baits were limited to the food wells of four arms (Nos. 1, 2, 4, and 7).
2.3. Habituation and shaping 2.6. Spatial learning task After handling and food deprivation, the rats received four habituation trials (two trials per day) in the radial arm maze for two consecutive days (Days 11–12). A small piece of a corn flake was used as a reinforcer. During habituation, the rats were gradually shaped to go to the distal end of the arm and to eat a reinforcer. On the first day of habituation and shaping (Day 11), they were transported to the maze room within a small cage (holding cage), and left on the desk for 5 min in order to habituate the maze room. Reinforcers were scattered throughout the maze and each rat was placed on the platform facing towards Arm No. 1, which was directed towards the experimenter’s location. They were allowed to explore the maze for 20 min and then returned to their home cage. Four hours later, the rats were transported to the maze room again to undergo the second habituation trial. After habituation to the maze room for 5 min in the holding cage, they were released at the platform of the maze according to the procedure of the first trial, but reinforcers were available only in the arms. The rats were returned to their home cage following a 10-min free run. On the next day (Day 12), the third and fourth habituation trials were performed with a 4-h interval, for 10 min each. The reinforcers were restricted to the ends of the arms for the third trial and to only the food wells for the fourth trial. 2.4. Centrifugation Sixteen rats were randomly divided into the hypergravity (HG) and the control groups. Eight animals were exposed to hypergravity in a centrifuge device [20] for 2 weeks (Days 13–27), which started on the next day of habituation. Each animal was located individually in a centrifuge cage (252020 cm), which hung at the distal end of a turning arm (diameter: 1 m). The arms were rotated at an angular velocity of 3368/s (56 r.p.m.), so each animal was constantly kept in 2G linear acceleration perpendicular to the floor of the centrifuge cage. During centrifugation, rats could freely access food and water and were under the same light–dark cycle as before. In order to supply food and water and clean the cage, the centrifuge was stopped for about 20 min at the
The spatial learning task started on the second day after centrifugation. The rats received three spatial trials for 5 days (Session 1: Days 29–33). After a 1-day interval (Day 34), they were subjected to the same trials for another 5 days (Session 2: Days 35–39). The trial was performed in the following manner. Each rat was transferred to the maze room within the holding cage, and then left on the desk. Five minutes later, the rat was placed on the platform of the radial arm maze with its head facing towards Arm No. 1. Reinforcers were available in the food wells of Arms Nos. 1, 2, 4, and 7. The animal was allowed to run on the platform and in the arms freely until all four reinforcers were taken or 5 min had elapsed, and then was returned to the holding cage on the desk. Three trials were carried out each day, with a 5-min interval. During this interval, the surface of the maze was wiped with 70% ethanol and, if necessary, reinforcers were supplied to the food wells. After the third trial of the day, the animal was returned to the home cage. 2.7. Behavioral measures The behavior of the rats in the maze was recorded with a digital video camera and the number of arms the animals had entered in the trial was scored. An arm entry was counted if all four limbs of the rat were within an arm. The following parameters were measured: (1) accuracy, (2) reentry rate, (3) entries/min and (4) baits/min. Accuracy was defined as the percentage of correct entries in all arm entries. A correct entry was scored when the rat entered a correct arm (Nos. 1, 2, 4, or 7) and ate a reinforcer. The re-entry rate was defined as the ratio of the number of entries into an arm that the animal had previously entered in the same trial to the number of all arm entries. Entries/min was defined as the number of entries into an arm per minute. Baits/min was defined as the number of baits that the rat took per minute. 2.8. Statistical analysis Analysis of variance (ANOVA) for repeated measures was conducted to evaluate the effects of time, group, and the
K. Mitani et al. / Cognitive Brain Research 22 (2004) 94–100
97
Fig. 2. Accuracy in entering the correct arms. (A) Accuracy in entering the correct arms was inferior in the hypergravity rats (closed circles), compared to that in the control rats (open circles). (B) During the early phase of the test (Session 1, Test days 1–5), the hypergravity animals (closed column) showed lower accuracy compared to that of the control animals (open column), while there was no difference at the later phase (Session 2, Test days 6–10).
interaction of time and groups. When significance of differences was shown by ANOVA between groups, comparisons in each session were made by paired t-test or one-way ANOVA, followed by Scheffe’s test. P values under 0.05 were determined to be significant.
following the 2-weeks centrifugation period (see Section 2.5), all rats took the baits and did not defecate in the maze, suggesting that they were still habituated to the maze and the experimenter’s manipulations, even after a 2-week interval of centrifugation. 3.2. Spatial learning task
3. Results 3.1. Behavioral observation at habituation periods before and after centrifugation At the first habituation trial (see Section 2.3), the rats often froze and defecated, and exploration behavior throughout the maze was observed. At the fourth (last) trial, all the rats had been shaped to search for the baits, and defecation was not observed. At the re-habituation trial
As shown in Fig. 2A, the accuracy of the HG rats was lower than that of the control rats. ANOVA for repeated measures revealed significant effects of group [ F (1, 14)=5.059, p=0.041] and day [ F (9, 126)=3.095, p= 0.00217], but no effect of groupday interaction [ F (9, 126)=1.514, p=0.150]. We classified Test days 1–5 (Days 29–33) as Session 1 and Test days 6–10 (Days 35–39) as Session 2. When the accuracy was compared between sessions, a significant effect of groupsession interaction
Fig. 3. Re-entry rate of entering the arms that the animals had already entered. (A) The re-entry rate was higher in the hypergravity rats (closed circles) compared to that in the control rats (open circles). (B) During the early phase of the test (Session 1, Test days 1–5), the hypergravity animals (closed column) showed a higher re-entry rate compared to that of the control animals (open column), while there was no difference at the later phase (Session 2, Test days 6–10).
98
K. Mitani et al. / Cognitive Brain Research 22 (2004) 94–100
Fig. 4. The number of entries per minute (A) and the number of baits per minute (B). Entries/min was higher in the hypergravity rats (A, closed circles) compared to that in the control rats (A, open circles). There was no difference in baits/min between the hypergravity rats (B, closed circles) and control rats (B, open circles).
was shown [ F (1, 14)=7.949, p=0.0137]. One-way ANOVA of the four groups divided by group and session revealed a significant group effect [ F (3, 28)=8.146, p=0.00047]. Posthoc analysis showed that the HG rats chose more incorrect arms in Session 1 ( p=0.00147), but not in Session 2 ( p=0.965) (Fig. 2B). As shown in Fig. 3A, the re-entry rate of the HG rats was higher than that of the control rats. ANOVA for repeated measures revealed a significant effect of group [ F (1, 14)=8.622, p=0.0108], but no effect of day [ F (9, 126)=1.216, p=0.291] or groupday interaction [ F (9, 126)=1.265, p=0.262]. A significant group effect was also shown on the analysis of the two sessions [ F (1, 14)=8.622, p=0.0108]. The re-entry rate of the HG rats was higher than that of the control rats in Session 1 ( p=0.00415), but not in Session 2 ( p=0.386) (Fig. 3B). 3.3. Entries/min and baits/min As shown in Fig. 4A, the HG rats entered more arms per minute than the control animals. ANOVA for repeated measures revealed significant effects of group [ F (1, 14)=4.636, p =0.0492], day [ F (9, 126)=18.323, pb0.00001] and groupday interaction [ F (9, 126)=3.347, p=0.00105]. Fig. 4B shows the number of baits taken by rats per minute (baits/min). While it seems that the HG rats took more baits/min than the control animals, there was no difference in baits/min between the HG and control animals, as revealed by ANOVA for repeated measures [ F (1, 14)=3.527, p=0.0813]. Groupday interaction was significant [ F (9, 126)=2.524, p=0.0108].
4. Discussion The present spatial learning test using a radial arm maze revealed that the accuracy for choosing a correct arm was
significantly lower in the HG rats compared to that of the control rats (Fig. 2). The re-entry rate, which is an indicative of a spatial working memory, was also higher in the HG rats compared to that of the control rats (Fig. 3). Therefore, it is suggested that the constant gravity, one of egocentric vestibular cues, is essential for the spatial learning. In contrast, it was reported that allocentric visual cues are important for the spatial reference memory rather than the spatial working memory [5]. While it is difficult to assess the spatial cognition performance and memory performance separately from the radial maze test, these results indicate that hypergravity exposure for 2 weeks impaired the spatial learning process in the rat, which is generally consistent with previous animal experiments [7,17] and astronaut data [3,8]. Previous data regarding the effects of hypergravity on the peripheral otolith organs have demonstrated neither morphological changes nor synthesis of otoconia [6,16,21], suggesting that the present impairment of spatial learning might not be attributed to the effects of hypergravity on the peripheral otolith organs. In contrast, cerebellar inhibition of the vestibular nucleus and decreased neurotransmission between vestibular hair cells and primary vestibular afferents might reduce the excess of otolith inputs by hypergravity [22]. These changes in peripheral and central vestibular neurotransmission might also contribute to impaired spatial learning through decreased vestibular inputs to areas important for spatial cognition, such as the hippocampus. Indeed, changes in hippocampal gene expression following hypergravity have recently been reported, and some of these changes could be the molecular background of spatial disorientation due to hypergravity [4]. Even though the HG rats showed impairment of spatial cognition, they could take almost the same or a relatively high amount of reinforcers in a minute (baits/min) (Fig. 4B). This was achieved by increased entries into the arms due to hyperkinetic behavior (Fig. 4A). Hyperkinetic behavior by hypergravity exposure have already been reported [2,17],
K. Mitani et al. / Cognitive Brain Research 22 (2004) 94–100
however, this is the first report to demonstrate that increased locomotor activity due to hypergravity exposure subsidized the impairment of spatial cognition and brought about no deficits in terms of getting reinforcers. The HG animals initially showed impaired spatial learning compared to the control animals, however, there was no difference in accuracy or re-entry rate between the HG animals and the control animals at the later stage (Figs. 2B, 3B). This means that the HG animals learned the spatial task more rapidly than the control animals, which is probably due to increased locomotor activity after hypergravity exposure subsidizing a loss of vestibular egocentric information by increasing information from other egocentric cues such as proprioceptive and motor efference copy signals. We cannot exclude a possibility that the spatial learning performance was limited in the control rats by a ceiling effect at the early phase of the spatial task so that the HG rats seemed to learn a spatial task more rapidly than the control animals. However, this interpretation may be unlikely because we found that both of the control and HG rats did not show a ceiling effect even 15 days after the start of the spatial task (data not shown). Therefore, the lack of deficits in getting reinforcers could have been brought about by not only an accidental increase in chances to enter the correct arm, but also by compensated spatial orientation, both of which were results of increased locomotor activities. In a radial arm maze test, animals with bilateral vestibular damage showed a similar trend regarding impaired spatial learning and final reinforcers: impaired spatial learning subsidized by increased locomotor activity resulted in no deficits in getting reinforcers [14]. Spatial firing of hippocampal place cells, which have important roles in spatial cognition, is disturbed following bilateral vestibular damage, indicating that the vestibular signal is an important egocentric cue in navigation tasks [13,18]. As in the study of bilateral vestibular damage, the HG rats subsidized their impaired spatial cognition by increased motor activities. It is suggested that animals need a constant gravity for spatial learning, to utilize as a stable external reference. If one sensory cue was disturbed (e.g., by bilateral vestibular damage or gravity change), animals subsidized it with other sensory cues, such as proprioceptive and motor efference copy signals, through increased locomotor activities.
Acknowledgements This study was partly supported by Grants-in-Aid for scientific research from the Ministry of Education, Culture, Sports, Science and Technology of Japan for AH and by a research grant for intractable diseases (vestibular disorders) from the Ministry of Health, Labor and Welfare of Japan for TK. We thank Prof. A. Yamatodani for a valuable suggestion on the radial arm maze test and Drs. A. Uno and A. Nakagawa for helpful discussion.
99
References [1] A. Berthoz, I. Israel, P. Georges-Francois, R. Grasso, T. Tsuzuku, Spatial memory of body linear displacement: what is being stored? Science 269 (1995) 95 – 98. [2] V. Bouet, Y. Gahery, M. Lacour, Behavioural changes induced by early and long-term gravito-inertial force modification in the rat, Behav. Brain Res. 17 (2003) 97 – 104. [3] T. Brandt, Vertigo: Its Multisensory Syndromes, 2nd ed., SpringerVerlag, London, 1999. [4] A. Del Signore, S. Mandillo, A. Rizzo, E. Di Mauro, A. Mele, R. Negri, A. Oliverio, P. Paggi, Hippocampal gene expression is modulated by hypergravity, Eur. J. Neurosci. 19 (2004) 667 – 677. [5] E. Jakubowska-Dogru, U. Gumusbas, F. Kara, Individual variation in the spatial reference and working memory assessed under allothetic and idiothetic orientation cues in rat, Acta Neurobiol. Exp. (Wars) 63 (2003) 17 – 23. [6] D.J. Lim, J.A. Stith, C.W. Stockwell, J. Oyama, Observations on saccules of rats exposed to long-term hypergaravity, Aerosp. Med. 45 (1974) 705 – 710. [7] S. Mandillo, A. Del Signore, P. Paggi, N. Francia, D. Santucci, A. Mele, A. Oliverio, Effects of acute and repeated daily exposure to hypergravity on spatial learning in mice, Neurosci. Lett. 336 (2003) 147 – 150. [8] A.B. Newberg, Changes in the central nervous system and their clinical correlates during long-term spaceflight, Aviat. Space Environ. Med. 65 (1994) 562 – 572. [9] J. O’Keefe, L. Nadel, The Hippocamous as A Cognitive Map, Clarendon Press, Oxford, 1978. [10] P. Peruch, L. Borel, F. Gaunet, G. Thinus-Blanc, J. Magnan, M. Lacour, Spatial performance of unilateral vestibular defective patients in nonvisual versus visual navigation, J. Vestibular Res. 9 (1999) 37 – 47. [11] J.T. Reason, Motion sickness adaptation: a neural mismatch model, J. R. Soc. Med. 71 (1978) 819 – 829. [12] M.F. Reschke, J.J. Bloomberg, D.L. Harm, W.H. Paloski, C. Layne, V. McDonald, Posture, locomotion, spatial orientation, and motion sickness as a function of space flight, Brain Res. Rev. 28 (1998) 102 – 117. [13] N.A. Russell, A. Horii, P.F. Smith, C.L. Darlington, D.K. Bilkey, Long-term effects of permanent vestibular lesions on hippocampal spatial firing, J. Neurosci. 23 (2003) 6490 – 6498. [14] N.A. Russell, A. Horii, P.F. Smith, C.L. Darlington, D.K. Bilkey, Bilateral peripheral vestibular lesions produce long-term changes in spatial learning in the rat, J. Vestibular Res. 13 (2003) 9 – 16. [15] H.N.P.M. Sondag, H.A.A. deJong, W.J. Oosterveld, Effect of prolonged hypergravity on the vestibular system: a behavioural study, ORL J. Oto-rhinolaryngol. Relat. Spec. 57 (1995) 189 – 193. [16] H.N.P.M. Sondag, H.A.A. deJong, J. van Marle, W.J. Oostervel, Effects of sustained acceleration on the morphological properties of otoconia in hamsters, Acta Oto-Laryngol. (Stockh.) 115 (1995) 227 – 230. [17] H.N.P.M. Sondag, H.A.A. deJong, W.J. Oosterveld, Altered behaviour of hamsters by prolonged hypergravity: adaptation to 2.5 G and re-adaptation to 1 G, Acta Oto-Laryngol. (Stockh.) 116 (1996) 192 – 197. [18] R.W. Stackman, A.M. Herbert, Rats with lesions of the vestibular system require a visual landmark for spatial navigation, Behav. Brain Res. 128 (2002) 27 – 40. [19] R.W. Stackman, A.S. Clark, J.S. Taube, Hippocampal spatial representations require vestibular input, Hippocampus 12 (2002) 291 – 303. [20] N. Takeda, A. Horii, A. Uno, M. Morita, T. Mochizuki, A. Yamatodani, T. Kubo, A ground-based animal model of space adaptation syndrome, J. Vestibular Res. 6 (1996) 403 – 409.
100
K. Mitani et al. / Cognitive Brain Research 22 (2004) 94–100
[21] Y. Uno, A. Horii, M. Umemoto, T. Hasegawa, K. Doi, A. Uno, T. Takemura, T. Kubo, Effects of hypergravity on morphology and osteopontin expression in the rat otolith organs, J. Vestibular Res. 10 (2000) 283 – 289. [22] Y. Uno, A. Horii, A. Uno, Y. Fuse, M. Fukushima, K. Doi, T. Kubo, Quantitative changes in mRNA expression of glutamate receptors in
the rat peripheral and central vestibular systems following hypergravity, J. Neurochem. 81 (2002) 1308 – 1317. [23] D.G. Wallace, D.J. Hines, S.M. Pellis, I.Q. Whishaw, Vestibular information is required for dead reckoning in the rat, J. Neurosci. (2002) 10009 – 10017.