- Email: [email protected]
Evidence for involvement of glucocorticoid response in the hippocampal changes in aged molarless SAMP8 mice
Behavioural Brain Research 131 (2002) 125– 129 www.elsevier.com/locate/bbr
Research report
Evidence for involvement of glucocorticoid response in the hippocampal changes in aged molarless SAMP8 mice Minoru Onozuka a,b,*, Kazuko Watanabe c, Masafumi Fujita a, Keiichi Tonosaki d, Shigeru Saito e a
Department of Anatomy (2nd Di6ision), School of Medicine, Gifu Uni6ersity, 40 Tsukasa-machi, Gifu 500 -8705, Japan b Institute for Frontier Oral Science, Kanagawa Dental College, Yokosuka 238 -8580, Japan c Department of Physiology, School of Medicine, Gifu Uni6ersity, 40 Tsukasa-machi, Gifu 500 -8705, Japan d Department of Veterinary Physiology, Faculty of Agriculture, Gifu Uni6ersity, Gifu 501 -1193, Japan e Business Center for Academic Societies Japan, Tokyo 113 -8622, Japan Received 24 April 2001; received in revised form 6 July 2001; accepted 14 August 2001
Abstract The involvement of glucocorticoid response in the hippocampal changes in aged SAMP8 mice after removal of their upper molar teeth (molarless condition) was examined using biochemical, morphological and behavioral techniques. Molarless mice showed plasma corticosterone levels to be significantly greater than those in molar-intact control mice. Pretreatment with metyrapone, which suppresses the stress-induced rise in plasma corticosterone levels, prevented the molarless condition-induced increase in plasma corticosterone levels, reduction in CA1 pyramidal neuron numbers, and impairment of spatial learning. The results suggest a link between the molarless condition and the glucocorticoid response, which may be involved in spatial learning deficits and hippocampal neuronal death in aged SAMP8 mice. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Tooth dysfunction; Corticosterone; Stress; Neuronal death; Spatial memory; Hippocampus; SAMP8 mice
1. Introduction Loss of the molar teeth induces deficits in spatial memory and acetylcholine release from the parietal cortex in aged rats [6]. In more recent studies using the aged accelerated senescence-prone mouse (SAMP8), which has been established as a murine model of accelerated senescence [3], we found that lack of the upper molars (molarless condition) results in neuronal death in the hippocampal CA1 subfield and reduced spatial learning ability in the swim maze test [11] and induces astroglial responsiveness in this subfield [12]. Previous studies have shown that basal plasma corticosterone levels in aged rats correlate significantly with * Corresponding author. Tel.: + 81-58-267-2221; fax: +81-58-2402194. E-mail address: [email protected] (M. Onozuka).
hippocampal degeneration and spatial learning deficits [8,9]. Elevated plasma corticosterone levels are found only in aged rats with spatial memory deficits and not in those with normal spatial memory [4]. Cumulative exposure to high glucocorticoid levels throughout life disrupts electrophysiological function, leading to atrophy and, ultimately, death of hippocampal neurons, all of which can cause severe cognitive deficits in hippocampal-dependent learning and memory [17]. Combined with the fact that hippocampal neuronal loss occurs both in patients with symptoms of senile dementia [14] and in animals with senile deficits of cognitive function [14,17], it is conceivable that the molarless condition-induced deficits in spatial learning and hippocampal neurons seen in aged SAMP8 may be due to the damaging effects of glucocorticoid. In this study, in order to investigate the role of glucocorticoids in these post-molarless dysfunctions, we
0166-4328/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 4 3 2 8 ( 0 1 ) 0 0 3 7 8 - 3
126
M. Onozuka et al. / Beha6ioural Brain Research 131 (2002) 125–129
measured the glucocorticoid response in aged molarless SAMP8 mice and tested the effect of manipulating corticosterone production on the neuronal and memory changes seen in these mice. Ten-month-old male SAMP8 mice, which have a median life span of 13 months [3], were used. A detailed description of the features of this strain can be found in other papers [3]. The mice were maintained under conventional conditions with free access to water and food, as described previously [11,12]. We have previously demonstrated that the molarless condition causes deficits in spatial memory and a reduction in the number of hippocampal CA1 neurons in aged SAMP8 mice, whereas none of these pathological changes are seen in molarless young-adult mice [11,12]. For this reason, only aged SAMP8 mice were used in the present study. For preparing molarless condition, the mice were deeply anesthetized with sodium pentobarbital (40 mg kg − 1, i.p.) and their upper molar teeth (maxillary molars) extracted as previously described [12]. Control animals underwent the same procedure except for the removal of the molars. Following surgery, the mice were fed pelleted food. A slight decrease in body weight and food intake in the molarless mice was seen for a few days after the operation, but these parameters were completely restored to pre-molarless levels before the experiments were performed, as in a previous repot [12]. Data were analyzed using an analysis of variance (ANOVA) with Scheffe´ post hoc test performed when appropriate. In the present study, two different experiments (Experiments 1 and 2) were carried out on control and molarless mice.
2. Experiment 1 Ten days after the operation, plasma corticosterone levels in the control (N = 24) and molarless (N= 24) mice were assayed, since the molarless mice clearly exhibited deficits in spatial learning in the water maze and hippocampal CA1 neurons at the post-operative 10 days [11,12]. Animals were decapitated at different times during the day (i.e. 08:00, 12:00, 16:00, 22:00, 24:00, 04:00 h) because of circadian variation in corticosterone levels [4], and blood was collected in 2.0 ml microcentrifuge tubes containing no anticoagulant. Anesthesia was not used because it can affect corticosterone levels [1]. The blood samples were immediately centrifuged at 1600× g for 3– 5 min at 4 °C, and the serum stored at −80 °C until the assay was performed. Corticosterone was measured by radioimmunoassay in the Biochemistry Labs of the BML General Laboratory (Kawagoe, Japan). As shown in Fig. 1, the corticosterone levels of the control (F(5, 18)=8.75, P B0.001) and molarless (F(5,
18)=6.28, PB 0.001) groups showed significant circadian variation, peaking in both groups at the onset of the dark period (i.e. 20:00 h). However, at four time points (i.e. 16:00, 20:00, 24:00, and 04:00 h), the molarless group had significantly [F(11, 36)= 20.76, P B 0.001] higher plasma corticosterone levels than the control group, indicating that the molarless condition causes an increased plasma corticosterone levels in the dark period.
3. Experiment 2 In order to assess the effect of the corticosterone synthesis inhibitor metyrapone on the molarless condition-induced increase in the corticosterone levels and reduction in the CA1 neurons, 1 day before the operation and every 2 days until the collection of blood, control and molarless mice received an injection of this inhibitor (150 mg kg − 1 s.c.), a dosage known to inhibit the stress-induced rise in plasma corticosterone levels and hippocampal neuronal damage [7]. Ten days after the operation, we measured plasma corticosterone levels in the vehicle-injected control (N= 10), vehicle-injected molarless (N=10) and metyrapone-treated molarless (N= 10) groups at two time points (16:00 and 20:00 h), as described above. As shown in Fig. 2a, there was no significant difference in the corticosterone levels between the vehicle-injected control and metyrapone-treated molarless groups, while a significant difference was seen between the vehicle-injected molarless group and either the vehicle-injected control group or the metyrapone-treated molarless group, (F(5, 54)= 42.942, PB 0.0001), indicating that metyrapone suppressed the molarless condition-induced increase in plasma corticosterone levels. However, it had no significant effect on plasma corticosterone levels in control mice (data not shown). We also counted neuronal numbers in the CA1 subfield of the hippocampus of these animals, according
Fig. 1. Effects of the molarless condition on plasma corticosterone levels in aged SAMP8 mice. Mean ( 9 S.E.) plasma corticosterone levels (mg dl − 1) in control and molarless mice (n = 4 for each column) at various times over a 24 h cycle (*, P B0.05 compared with controls; **, PB0.01 compared with controls). Con, control group; Mol, molarless group.
M. Onozuka et al. / Beha6ioural Brain Research 131 (2002) 125–129
127
Fig. 2. Effects of metyrapone: (a) Mean ( 9S.E., N =10 for each column) plasma corticosterone levels (mg dl − 1) in vehicle-injected control, vehicle-injected molarless, and metyrapone-injected molarless groups at 16:00 and 20:00 h (**, PB 0.01 compared with vehicle-injected controls). (b) Representative photomicrographs of Cresyl violet-stained sections showing the CA1 cell field in control (upper), molarless (middle) and metyrapone-treated molarless (lower) groups. Scale bar; 50 mm. (c) Mean ( 9 S.E., N= 7 for each column) neuron density (per mm3) in the CA1 pyramidal cell field in vehicle-injected control, vehicle-injected molarless, and metyrapone-injected molarless groups (N= 7 animals per group; *, P B 0.05). (d, e) Spatial learning in a water maze test. Ten days after surgery, the learning test was started. Mean ( 9 S.E., N =10 for each group) latency (d) or swim distance (e) to locate submerged platform in Morris swim maze in vehicle-injected control, vehicle-injected molarless, and metyrapone-injected molarless groups. Insert, Visible probe test. At the end of the maze test, animals performed a visible probe test. Con, molar-intact control group; Mol, molarless group; E.L., escape latency; CV, vehicle-injected control group; MV, vehicle-injected molarless group; MM, metyrapone-injected molarless group; Note no significant difference between groups.
to our recent report [19]. Briefly, after the collection of their blood, the brain was rapidly removed and immersion-fixed for 1 day at 4 °C in 4% paraformaldehyde in phosphate-buffered saline, pH 7.0, before preparing 40 mm coronal microtome sections, followed by Nissl
staining with cresyl violet. Neuron counts were performed in sections spaced 240 mm apart throughout the entire CA1 subfield of the hippocampus from one hemisphere using the optical dissector technique and a computer-controlled morphometric analyzer (Luzex-AP,
128
M. Onozuka et al. / Beha6ioural Brain Research 131 (2002) 125–129
Nikon Co., Tokyo, Japan). Neurons were only counted if they fell within the three-dimensional measuring volume of the dissector and did not overlap the two forbidden lines on the grid (the left and lower sides). They were identified by a clear nuclear profile containing a nucleolus and Nissl substance in the cell body as the focus went through the section from one grid to the other. The coefficient of error for the counting technique in the CA1 subfield is B0.10 [19]. As shown in Fig. 2b and c, metyrapone prevented the molarless-induced reduction in neuronal number in this subfield, since no significant difference in the number was seen between vehicle-injected control mice and metyrapone-injected molarless mice (group effect: F(1, 12)= 0.61, NS). We next examined the effect of metyrapone on molarless condition-induced reduction in spatial learning ability. One day before the operation and every 2 days until the end of the swim tests, molarless mice received an injection of the same dose of this inhibitor. Ten days after the operation, animals were tested in a version of the Morris swim maze, the details of which have been described in our previous reports [12]. To ensure that any observed impairments were not related to the animals’ inability to meet the visual demands of the task, at the end of the test, a visible probe test (four trials per animal) was also carried out. As shown in Fig. 2d, metyrapone prevented the increase in escape latency in the water maze test induced by the molarless condition, since a difference in latency was seen between metyrapone-injected molarless mice and vehicle-injected molarless mice (P B 0.01), but not between metyrapone-injected molarless mice and vehicle-injected control mice, F(12, 162)= 1.33. Similar findings were also seen on the swim distance, F(12, 162)= 1.14 (Fig. 2e). The present work, using biochemical, behavioral and morphological techniques in aged SAMP8 mice, provides the possibility that the molarless-induced deficits in spatial learning and hippocampal neuron numbers in aged SAMP8 mice may be related to exposure to increased corticosterone levels. In the present study, the mice showed a circadian variation in corticosterone levels, which was similar to that reported in a previous study using rats [10] showing a peak at the beginning of the dark period, when activity is generally greatest, and the lowest levels near the end of the dark period or the beginning of the light period, when rodents are least active. However, the molarless group had overall higher corticosterone levels than the control group, indicating that the molarless condition results in increased exposure to corticosterone and suggesting an impairment in hypothalamic-pituitary-adrenal negative-feedback inhibition in molarless mice. We restricted our analysis to the CA1 subfield, since, although some authors have reported that exposure to
high corticosterone levels leads to neuron loss in the hippocampal CA3 subfield or dentate gyrus [9,13,16], in our previous report, neuron damage was seen only in the CA1 subfield in the molarless condition, and not in the CA3 subfield and dentate gyrus [11]. Furthermore, it is only in the CA1 subfield of the hippocampus that the molarless condition enhances the age-related increase in the density and hypertrophy of GFAP (glial fibrous acidic protein)-labeled astrocytes [12]. In this study, we found that neuron loss in this subfield of the molarless group [11,12] was positively correlated with increased corticosterone levels. Taken together with the observations that the adrenals of molarless aged SAMP8 mice are heavier than those of age-matched molar-intact control mice [12] and that increased mastication reduces the plasma corticosterone response during novelty exposure in mice [2], a prolonged stressful response to the molarless condition and the consequent increase in exposure to corticosterone could hasten hippocampal neuron damage in this species. In a more direct proof of our theory, we tested the effect of metyrapone treatment, which is reported to block the stress-related plasma corticosterone increase [15], and found that it not only suppressed the molarless condition-induced increase in plasma corticosterone levels, but also effectively protected against the spatial memory impairment and CA1 neuron loss induced by the molarless condition. Therefore, we suggest that the deficits in spatial learning and hippocampal neurons seen in aged molarless SAMP8 mice may be due largely to exposure to corticosterone levels that are high enough to cause cell death. There is a large literature relating inflammation to hypothalamic–pituitary–adrenal axis (reviewed in [5,18]). Since the present study revealed that the molarless condition stimulates the hypothalamic-pituitaryadrenal axis, the possibility arises that the effect seen in the molarless mice are due to the inflammations in the alveolar bone. The following unpublished observations may be argued against inflammations being responsible: (1) in an experiment using aged SAMP8 mice, in which pathological examinations were carried out 10 days after extraction of the upper molar teeth, we found that root fragments left in the alveolar bone were detected in about 12% of the molarless mice. However, they did not show obvious inflammation in this region. (2) Plasma cytokine levels, such as interleukin (IL)-6 or interferon-gamma, of the molar-intact control and molarless aged mice were measured 10 days after the operation and compared with the groups. However, we failed to detect a significant difference in their levels between the control and molarless groups. (3) In aged, but not young-adult, SAMP8 mice, spatial learning deficits and cell death in the hippocampus are seen after cutting one side of their masseteric nerve thus preventing contraction of the masseter muscle. Thus, it seems
M. Onozuka et al. / Beha6ioural Brain Research 131 (2002) 125–129
that the effects seen are not due to inflammation as a result of the extraction of the molar teeth. However, the question of how removing the molar teeth could affect the hypothalamic pituitary adrenal axis negative feedback mechanism is, at this stage, unanswered. Furthermore, it is not known why the molarless condition does not result in hippocampal change in young mice and further work is required.
[8]
[9]
[10]
[11]
Acknowledgements This research was supported by an Esso Research Grant for Women. We also express our deep gratitude to K. Hasegawa for his encouragement and cooperation during the course of this work.
[12]
[13]
References [14] [1] Coleman MA, Garland T, Marler CA, Newton SS, Swallow JG, Carter PA. Glucocorticoid response to forced exercise in laboratory house mice (Mus domesticus). Physol Behav 1998;63:279 – 85. [2] Hennessy MB, Foy T. Nonedible material elicits chewing and reduces the plasma corticosterone response during novelty exposure in mice. Behav Neurosci 1987;101:237 –45. [3] Hosokawa M, Umezawa M, Higuchi K, Takeda T. Interventions of senescence in SAM mice. J Anti-Aging Med 1998;1:27 – 37. [4] Issa AM, Rowe W, Gauthier S, Meaney MJ. Hypothalamic-pituitary-adrenal activity in aged, cognitively impaired and cognitively unimpaired rats. J Neurosci 1990;10:3247 –54. [5] Kapcala LP, Chautard T, Eskay RL. The protective role of the hypothalamic–pituitary–adrenal axis against lethality produced by immune, infectious, and inflammatory stress. Ann New York Acad Sci 1995;771:419 –37. [6] Kato T, Usami T, Noda Y, Hasegawa M, Ueda M, Nabeshima T. The effect of the loss of molar teeth on spatial memory and acetylcholine release from the parietal cortex in aged rats. Behav Brain Res 1997;83:239 –42. [7] Krugers HJ, Maslam S, Korf J, Joe¨ ls M. The corticosterone synthesis inhibitor metyrapone prevents hypoxia/ischemia-in-
[15]
[16]
[17]
[18]
[19]
129
duced loss of synaptic function in the rat hippocampus. Stroke 1998;2:237 – 49. Landfield PW, Baskin RK, Pitler TA. Brain aging correlates: retardation by hormonal-pharmacological treatments. Science 1981;214:581 – 3. Landfield PW, Waymire JC, Lynch G. Hippocampal aging and adrenocorticoids: quantitative correlations. Science 1978;202:1098 – 102. Manning JM, Bronson FH. Suppression of puberty in rats by exercise: effects on hormone levels and reversal with GnRH infusion. Am J Physiol 1991;260:R717 – 23. Onozuka M, Watanabe K, Mirbod SM, Ozono S, Nishiyama K, Karasawa N, Nagatsu I. Reduced mastication stimulates impairment of spatial memory and degeneration of hippocampal neurons in aged SAMP8. Brain Res 1999;826:148 – 53. Onozuka M, Watanabe K, Nagasaki S, Jiang Y, Ozono S, Nishiyama K, Kawase T, Karasawa N, Nagatsu I. Impairment of spatial memory and changes in astroglial responsiveness following loss of molar teeth in aged SAMP8 mice. Behav Brain Res 2000;108:145 – 55. Sapolsky RM, Krey LC, McEwen BS. Prolonged glucocorticoid exposure reduces hippocampal neuron number: implications for aging. J Neurosci 1985;5:1222 – 7. Sirevaag AM, Black JE, Greenough WT. Astrocyte hypertrophy in the dentate gyrus of young male rats reflects variation of individual stress rather than group environmental complexity manipulations. Exp Neurol 1991;111:74 – 9. Smith-Swintoski VL, Pettigrew LC, Saplosky RM, Phares C, Craddock SD, Brooke SM, Mattson MP. Metyrapone, an inhibitor of glucocorticoid production, reduces brain injury induced by focal and global ischemia and seizures. J Cereb Blood Flow Metab 1996;16:585 – 98. Sousa N, Madeira MD, Paula-Barbosa MM. Effects of corticosterone treatment and rehabilitation on the hippocampal formation of neonatal and adult rats. An unbiased stereological study. Brain Res 1998;794:199 – 210. Squire LR. Memory and the hippocampus: a synthesis from findings with rats, monkeys, and humans. Psychol Rev 1992;99:195 – 231. Turnbull AV, Lee S, Rivier C. Mechanisms of hypothalamic –pituitary – adrenal axis stimulation by immune signals in the adult rat. Ann New York Acad Sci 1998;840:434 – 43. Watanabe K, Tonosaki K, Kawase T, Karasawa N, Nagatsu I, Fujita M, Onozuka M. Evidence for involvement of dysfunctional teeth in the senile process in the hippocampus of SAMP8 mice. Exp Gerontol 2001;36:283 – 95.
Research report
Evidence for involvement of glucocorticoid response in the hippocampal changes in aged molarless SAMP8 mice Minoru Onozuka a,b,*, Kazuko Watanabe c, Masafumi Fujita a, Keiichi Tonosaki d, Shigeru Saito e a
Department of Anatomy (2nd Di6ision), School of Medicine, Gifu Uni6ersity, 40 Tsukasa-machi, Gifu 500 -8705, Japan b Institute for Frontier Oral Science, Kanagawa Dental College, Yokosuka 238 -8580, Japan c Department of Physiology, School of Medicine, Gifu Uni6ersity, 40 Tsukasa-machi, Gifu 500 -8705, Japan d Department of Veterinary Physiology, Faculty of Agriculture, Gifu Uni6ersity, Gifu 501 -1193, Japan e Business Center for Academic Societies Japan, Tokyo 113 -8622, Japan Received 24 April 2001; received in revised form 6 July 2001; accepted 14 August 2001
Abstract The involvement of glucocorticoid response in the hippocampal changes in aged SAMP8 mice after removal of their upper molar teeth (molarless condition) was examined using biochemical, morphological and behavioral techniques. Molarless mice showed plasma corticosterone levels to be significantly greater than those in molar-intact control mice. Pretreatment with metyrapone, which suppresses the stress-induced rise in plasma corticosterone levels, prevented the molarless condition-induced increase in plasma corticosterone levels, reduction in CA1 pyramidal neuron numbers, and impairment of spatial learning. The results suggest a link between the molarless condition and the glucocorticoid response, which may be involved in spatial learning deficits and hippocampal neuronal death in aged SAMP8 mice. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Tooth dysfunction; Corticosterone; Stress; Neuronal death; Spatial memory; Hippocampus; SAMP8 mice
1. Introduction Loss of the molar teeth induces deficits in spatial memory and acetylcholine release from the parietal cortex in aged rats [6]. In more recent studies using the aged accelerated senescence-prone mouse (SAMP8), which has been established as a murine model of accelerated senescence [3], we found that lack of the upper molars (molarless condition) results in neuronal death in the hippocampal CA1 subfield and reduced spatial learning ability in the swim maze test [11] and induces astroglial responsiveness in this subfield [12]. Previous studies have shown that basal plasma corticosterone levels in aged rats correlate significantly with * Corresponding author. Tel.: + 81-58-267-2221; fax: +81-58-2402194. E-mail address: [email protected] (M. Onozuka).
hippocampal degeneration and spatial learning deficits [8,9]. Elevated plasma corticosterone levels are found only in aged rats with spatial memory deficits and not in those with normal spatial memory [4]. Cumulative exposure to high glucocorticoid levels throughout life disrupts electrophysiological function, leading to atrophy and, ultimately, death of hippocampal neurons, all of which can cause severe cognitive deficits in hippocampal-dependent learning and memory [17]. Combined with the fact that hippocampal neuronal loss occurs both in patients with symptoms of senile dementia [14] and in animals with senile deficits of cognitive function [14,17], it is conceivable that the molarless condition-induced deficits in spatial learning and hippocampal neurons seen in aged SAMP8 may be due to the damaging effects of glucocorticoid. In this study, in order to investigate the role of glucocorticoids in these post-molarless dysfunctions, we
0166-4328/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 4 3 2 8 ( 0 1 ) 0 0 3 7 8 - 3
126
M. Onozuka et al. / Beha6ioural Brain Research 131 (2002) 125–129
measured the glucocorticoid response in aged molarless SAMP8 mice and tested the effect of manipulating corticosterone production on the neuronal and memory changes seen in these mice. Ten-month-old male SAMP8 mice, which have a median life span of 13 months [3], were used. A detailed description of the features of this strain can be found in other papers [3]. The mice were maintained under conventional conditions with free access to water and food, as described previously [11,12]. We have previously demonstrated that the molarless condition causes deficits in spatial memory and a reduction in the number of hippocampal CA1 neurons in aged SAMP8 mice, whereas none of these pathological changes are seen in molarless young-adult mice [11,12]. For this reason, only aged SAMP8 mice were used in the present study. For preparing molarless condition, the mice were deeply anesthetized with sodium pentobarbital (40 mg kg − 1, i.p.) and their upper molar teeth (maxillary molars) extracted as previously described [12]. Control animals underwent the same procedure except for the removal of the molars. Following surgery, the mice were fed pelleted food. A slight decrease in body weight and food intake in the molarless mice was seen for a few days after the operation, but these parameters were completely restored to pre-molarless levels before the experiments were performed, as in a previous repot [12]. Data were analyzed using an analysis of variance (ANOVA) with Scheffe´ post hoc test performed when appropriate. In the present study, two different experiments (Experiments 1 and 2) were carried out on control and molarless mice.
2. Experiment 1 Ten days after the operation, plasma corticosterone levels in the control (N = 24) and molarless (N= 24) mice were assayed, since the molarless mice clearly exhibited deficits in spatial learning in the water maze and hippocampal CA1 neurons at the post-operative 10 days [11,12]. Animals were decapitated at different times during the day (i.e. 08:00, 12:00, 16:00, 22:00, 24:00, 04:00 h) because of circadian variation in corticosterone levels [4], and blood was collected in 2.0 ml microcentrifuge tubes containing no anticoagulant. Anesthesia was not used because it can affect corticosterone levels [1]. The blood samples were immediately centrifuged at 1600× g for 3– 5 min at 4 °C, and the serum stored at −80 °C until the assay was performed. Corticosterone was measured by radioimmunoassay in the Biochemistry Labs of the BML General Laboratory (Kawagoe, Japan). As shown in Fig. 1, the corticosterone levels of the control (F(5, 18)=8.75, P B0.001) and molarless (F(5,
18)=6.28, PB 0.001) groups showed significant circadian variation, peaking in both groups at the onset of the dark period (i.e. 20:00 h). However, at four time points (i.e. 16:00, 20:00, 24:00, and 04:00 h), the molarless group had significantly [F(11, 36)= 20.76, P B 0.001] higher plasma corticosterone levels than the control group, indicating that the molarless condition causes an increased plasma corticosterone levels in the dark period.
3. Experiment 2 In order to assess the effect of the corticosterone synthesis inhibitor metyrapone on the molarless condition-induced increase in the corticosterone levels and reduction in the CA1 neurons, 1 day before the operation and every 2 days until the collection of blood, control and molarless mice received an injection of this inhibitor (150 mg kg − 1 s.c.), a dosage known to inhibit the stress-induced rise in plasma corticosterone levels and hippocampal neuronal damage [7]. Ten days after the operation, we measured plasma corticosterone levels in the vehicle-injected control (N= 10), vehicle-injected molarless (N=10) and metyrapone-treated molarless (N= 10) groups at two time points (16:00 and 20:00 h), as described above. As shown in Fig. 2a, there was no significant difference in the corticosterone levels between the vehicle-injected control and metyrapone-treated molarless groups, while a significant difference was seen between the vehicle-injected molarless group and either the vehicle-injected control group or the metyrapone-treated molarless group, (F(5, 54)= 42.942, PB 0.0001), indicating that metyrapone suppressed the molarless condition-induced increase in plasma corticosterone levels. However, it had no significant effect on plasma corticosterone levels in control mice (data not shown). We also counted neuronal numbers in the CA1 subfield of the hippocampus of these animals, according
Fig. 1. Effects of the molarless condition on plasma corticosterone levels in aged SAMP8 mice. Mean ( 9 S.E.) plasma corticosterone levels (mg dl − 1) in control and molarless mice (n = 4 for each column) at various times over a 24 h cycle (*, P B0.05 compared with controls; **, PB0.01 compared with controls). Con, control group; Mol, molarless group.
M. Onozuka et al. / Beha6ioural Brain Research 131 (2002) 125–129
127
Fig. 2. Effects of metyrapone: (a) Mean ( 9S.E., N =10 for each column) plasma corticosterone levels (mg dl − 1) in vehicle-injected control, vehicle-injected molarless, and metyrapone-injected molarless groups at 16:00 and 20:00 h (**, PB 0.01 compared with vehicle-injected controls). (b) Representative photomicrographs of Cresyl violet-stained sections showing the CA1 cell field in control (upper), molarless (middle) and metyrapone-treated molarless (lower) groups. Scale bar; 50 mm. (c) Mean ( 9 S.E., N= 7 for each column) neuron density (per mm3) in the CA1 pyramidal cell field in vehicle-injected control, vehicle-injected molarless, and metyrapone-injected molarless groups (N= 7 animals per group; *, P B 0.05). (d, e) Spatial learning in a water maze test. Ten days after surgery, the learning test was started. Mean ( 9 S.E., N =10 for each group) latency (d) or swim distance (e) to locate submerged platform in Morris swim maze in vehicle-injected control, vehicle-injected molarless, and metyrapone-injected molarless groups. Insert, Visible probe test. At the end of the maze test, animals performed a visible probe test. Con, molar-intact control group; Mol, molarless group; E.L., escape latency; CV, vehicle-injected control group; MV, vehicle-injected molarless group; MM, metyrapone-injected molarless group; Note no significant difference between groups.
to our recent report [19]. Briefly, after the collection of their blood, the brain was rapidly removed and immersion-fixed for 1 day at 4 °C in 4% paraformaldehyde in phosphate-buffered saline, pH 7.0, before preparing 40 mm coronal microtome sections, followed by Nissl
staining with cresyl violet. Neuron counts were performed in sections spaced 240 mm apart throughout the entire CA1 subfield of the hippocampus from one hemisphere using the optical dissector technique and a computer-controlled morphometric analyzer (Luzex-AP,
128
M. Onozuka et al. / Beha6ioural Brain Research 131 (2002) 125–129
Nikon Co., Tokyo, Japan). Neurons were only counted if they fell within the three-dimensional measuring volume of the dissector and did not overlap the two forbidden lines on the grid (the left and lower sides). They were identified by a clear nuclear profile containing a nucleolus and Nissl substance in the cell body as the focus went through the section from one grid to the other. The coefficient of error for the counting technique in the CA1 subfield is B0.10 [19]. As shown in Fig. 2b and c, metyrapone prevented the molarless-induced reduction in neuronal number in this subfield, since no significant difference in the number was seen between vehicle-injected control mice and metyrapone-injected molarless mice (group effect: F(1, 12)= 0.61, NS). We next examined the effect of metyrapone on molarless condition-induced reduction in spatial learning ability. One day before the operation and every 2 days until the end of the swim tests, molarless mice received an injection of the same dose of this inhibitor. Ten days after the operation, animals were tested in a version of the Morris swim maze, the details of which have been described in our previous reports [12]. To ensure that any observed impairments were not related to the animals’ inability to meet the visual demands of the task, at the end of the test, a visible probe test (four trials per animal) was also carried out. As shown in Fig. 2d, metyrapone prevented the increase in escape latency in the water maze test induced by the molarless condition, since a difference in latency was seen between metyrapone-injected molarless mice and vehicle-injected molarless mice (P B 0.01), but not between metyrapone-injected molarless mice and vehicle-injected control mice, F(12, 162)= 1.33. Similar findings were also seen on the swim distance, F(12, 162)= 1.14 (Fig. 2e). The present work, using biochemical, behavioral and morphological techniques in aged SAMP8 mice, provides the possibility that the molarless-induced deficits in spatial learning and hippocampal neuron numbers in aged SAMP8 mice may be related to exposure to increased corticosterone levels. In the present study, the mice showed a circadian variation in corticosterone levels, which was similar to that reported in a previous study using rats [10] showing a peak at the beginning of the dark period, when activity is generally greatest, and the lowest levels near the end of the dark period or the beginning of the light period, when rodents are least active. However, the molarless group had overall higher corticosterone levels than the control group, indicating that the molarless condition results in increased exposure to corticosterone and suggesting an impairment in hypothalamic-pituitary-adrenal negative-feedback inhibition in molarless mice. We restricted our analysis to the CA1 subfield, since, although some authors have reported that exposure to
high corticosterone levels leads to neuron loss in the hippocampal CA3 subfield or dentate gyrus [9,13,16], in our previous report, neuron damage was seen only in the CA1 subfield in the molarless condition, and not in the CA3 subfield and dentate gyrus [11]. Furthermore, it is only in the CA1 subfield of the hippocampus that the molarless condition enhances the age-related increase in the density and hypertrophy of GFAP (glial fibrous acidic protein)-labeled astrocytes [12]. In this study, we found that neuron loss in this subfield of the molarless group [11,12] was positively correlated with increased corticosterone levels. Taken together with the observations that the adrenals of molarless aged SAMP8 mice are heavier than those of age-matched molar-intact control mice [12] and that increased mastication reduces the plasma corticosterone response during novelty exposure in mice [2], a prolonged stressful response to the molarless condition and the consequent increase in exposure to corticosterone could hasten hippocampal neuron damage in this species. In a more direct proof of our theory, we tested the effect of metyrapone treatment, which is reported to block the stress-related plasma corticosterone increase [15], and found that it not only suppressed the molarless condition-induced increase in plasma corticosterone levels, but also effectively protected against the spatial memory impairment and CA1 neuron loss induced by the molarless condition. Therefore, we suggest that the deficits in spatial learning and hippocampal neurons seen in aged molarless SAMP8 mice may be due largely to exposure to corticosterone levels that are high enough to cause cell death. There is a large literature relating inflammation to hypothalamic–pituitary–adrenal axis (reviewed in [5,18]). Since the present study revealed that the molarless condition stimulates the hypothalamic-pituitaryadrenal axis, the possibility arises that the effect seen in the molarless mice are due to the inflammations in the alveolar bone. The following unpublished observations may be argued against inflammations being responsible: (1) in an experiment using aged SAMP8 mice, in which pathological examinations were carried out 10 days after extraction of the upper molar teeth, we found that root fragments left in the alveolar bone were detected in about 12% of the molarless mice. However, they did not show obvious inflammation in this region. (2) Plasma cytokine levels, such as interleukin (IL)-6 or interferon-gamma, of the molar-intact control and molarless aged mice were measured 10 days after the operation and compared with the groups. However, we failed to detect a significant difference in their levels between the control and molarless groups. (3) In aged, but not young-adult, SAMP8 mice, spatial learning deficits and cell death in the hippocampus are seen after cutting one side of their masseteric nerve thus preventing contraction of the masseter muscle. Thus, it seems
M. Onozuka et al. / Beha6ioural Brain Research 131 (2002) 125–129
that the effects seen are not due to inflammation as a result of the extraction of the molar teeth. However, the question of how removing the molar teeth could affect the hypothalamic pituitary adrenal axis negative feedback mechanism is, at this stage, unanswered. Furthermore, it is not known why the molarless condition does not result in hippocampal change in young mice and further work is required.
[8]
[9]
[10]
[11]
Acknowledgements This research was supported by an Esso Research Grant for Women. We also express our deep gratitude to K. Hasegawa for his encouragement and cooperation during the course of this work.
[12]
[13]
References [14] [1] Coleman MA, Garland T, Marler CA, Newton SS, Swallow JG, Carter PA. Glucocorticoid response to forced exercise in laboratory house mice (Mus domesticus). Physol Behav 1998;63:279 – 85. [2] Hennessy MB, Foy T. Nonedible material elicits chewing and reduces the plasma corticosterone response during novelty exposure in mice. Behav Neurosci 1987;101:237 –45. [3] Hosokawa M, Umezawa M, Higuchi K, Takeda T. Interventions of senescence in SAM mice. J Anti-Aging Med 1998;1:27 – 37. [4] Issa AM, Rowe W, Gauthier S, Meaney MJ. Hypothalamic-pituitary-adrenal activity in aged, cognitively impaired and cognitively unimpaired rats. J Neurosci 1990;10:3247 –54. [5] Kapcala LP, Chautard T, Eskay RL. The protective role of the hypothalamic–pituitary–adrenal axis against lethality produced by immune, infectious, and inflammatory stress. Ann New York Acad Sci 1995;771:419 –37. [6] Kato T, Usami T, Noda Y, Hasegawa M, Ueda M, Nabeshima T. The effect of the loss of molar teeth on spatial memory and acetylcholine release from the parietal cortex in aged rats. Behav Brain Res 1997;83:239 –42. [7] Krugers HJ, Maslam S, Korf J, Joe¨ ls M. The corticosterone synthesis inhibitor metyrapone prevents hypoxia/ischemia-in-
[15]
[16]
[17]
[18]
[19]
129
duced loss of synaptic function in the rat hippocampus. Stroke 1998;2:237 – 49. Landfield PW, Baskin RK, Pitler TA. Brain aging correlates: retardation by hormonal-pharmacological treatments. Science 1981;214:581 – 3. Landfield PW, Waymire JC, Lynch G. Hippocampal aging and adrenocorticoids: quantitative correlations. Science 1978;202:1098 – 102. Manning JM, Bronson FH. Suppression of puberty in rats by exercise: effects on hormone levels and reversal with GnRH infusion. Am J Physiol 1991;260:R717 – 23. Onozuka M, Watanabe K, Mirbod SM, Ozono S, Nishiyama K, Karasawa N, Nagatsu I. Reduced mastication stimulates impairment of spatial memory and degeneration of hippocampal neurons in aged SAMP8. Brain Res 1999;826:148 – 53. Onozuka M, Watanabe K, Nagasaki S, Jiang Y, Ozono S, Nishiyama K, Kawase T, Karasawa N, Nagatsu I. Impairment of spatial memory and changes in astroglial responsiveness following loss of molar teeth in aged SAMP8 mice. Behav Brain Res 2000;108:145 – 55. Sapolsky RM, Krey LC, McEwen BS. Prolonged glucocorticoid exposure reduces hippocampal neuron number: implications for aging. J Neurosci 1985;5:1222 – 7. Sirevaag AM, Black JE, Greenough WT. Astrocyte hypertrophy in the dentate gyrus of young male rats reflects variation of individual stress rather than group environmental complexity manipulations. Exp Neurol 1991;111:74 – 9. Smith-Swintoski VL, Pettigrew LC, Saplosky RM, Phares C, Craddock SD, Brooke SM, Mattson MP. Metyrapone, an inhibitor of glucocorticoid production, reduces brain injury induced by focal and global ischemia and seizures. J Cereb Blood Flow Metab 1996;16:585 – 98. Sousa N, Madeira MD, Paula-Barbosa MM. Effects of corticosterone treatment and rehabilitation on the hippocampal formation of neonatal and adult rats. An unbiased stereological study. Brain Res 1998;794:199 – 210. Squire LR. Memory and the hippocampus: a synthesis from findings with rats, monkeys, and humans. Psychol Rev 1992;99:195 – 231. Turnbull AV, Lee S, Rivier C. Mechanisms of hypothalamic –pituitary – adrenal axis stimulation by immune signals in the adult rat. Ann New York Acad Sci 1998;840:434 – 43. Watanabe K, Tonosaki K, Kawase T, Karasawa N, Nagatsu I, Fujita M, Onozuka M. Evidence for involvement of dysfunctional teeth in the senile process in the hippocampus of SAMP8 mice. Exp Gerontol 2001;36:283 – 95.