- Email: [email protected]
Tyrosine hydroxylase gene expression in the locus coeruleus of depression-model rats and rats exposed to short- and long-term forced walking stress
PII SOO24-3205(98)00183-0
ELSEVIER
Life Sciences, Vol. 62, No. 23, pp. 2o83-2092, 1998 CQpyright0 19% Jskevier science Inc. Printed in the USA. All rights refed ooz4-32os/98 519.00 t .oo
TYROSINE HYDROXYLASE GENE EXPRESSION IN THE LOCUS COERULEUS OF DEPRESSION-MODEL RATS AND RATS EXPOSED TO SHORT- AND LONG-TERM FORCED WALKING STRESS
Ping Wang, Isao Kitayamai Department
of Psychiatry,
Mie University
& Junichi Nomura School of Medicine, Tsu, Japan
(Received in final form March l3,1998) Summary Abnormal brain noradrenergic function is thought to cause depressive illnesses which are sometimes manifested or aggravated under stressful conditions. To investigate the effect of chronic stress on noradrenaline (NA) synthesis in the brain we used in sihl hybridization to examine the expression of tyrosine hydroxylase (TH) mRNA in the locus coeruleus (LC) of “depression-model rats” that exhibit reduced activity following exposure to long-term (14 days) forced walking stress (FWS) We also examined TH mRNA expression in rats stressed for 30 minutes, 3 hours and 1, 2 (short-term), 6 or 12 (long-term) days. The expression of TH mRNA increased markedly following 1 to 12 days of FWS, but not in rats exposed to FWS for 30 minutes or 3 hours. The expression also increased significantly in the depression-model rats, but not in the “spontaneous recovery rats” whose activity was restored after long-term stress. Our results suggest that N-4 synthesis remains high in the FWS-induced depression-model rats because of the high levels of TH mRNA expression in the LC. Our results also suggest that FWS is initially a mild stress but gradually becomes a severe form of unadaptable stress as reflected by delayed but persistent increases in TH mRNA expression. Key Words: walking stress, locus coeruleus, tyrosine hydroxylase mRNA, in situ hybridization, animal model, depression
Since depression often develops under conditions of stress, stress appears to cause depression (1). Lloyd (2) Paykel, and Clayton and Darvish (2 papers in reference 3) emphasized that exposure to stress leads to the development of depression, and Berkman (4) Ilfeld (5) and Paykel et al. (6) demonstrated that a strong stressor increases the risk of manifestation of depression. It is also well known in animal experiments that exposure to stress enhances noradrenaline (NA) turnover in the brain (7). The turnover of brain NA can be estimated to some extent by measurement of 3-methoxy-4-hydroxyphenylethyleneglycol (MHPG), a metabolite of NA in urine, although about 20% of the metabolite is of central origin (8). In the human, male normal volunteers showed a significant increase in urinary MHPG under visual and painful electrical stimuli (9). Furthermore, students and instructors during in-flight emergencies also showed an increase in urinary NA as well as adrenaline, and students revealed ‘Corresponding author, Isao Kitayama, M.D., Department of Psychiatry, Mie University School of Medicine, 2-l 74, Edobashi, Tsu, Mie 5 14 Japan, Tel: +81 (59) 23 l-5018, Fax: +81 (59) 23 l-5208, e-mail, [email protected]
TH mRNA in Lx:of Depression Model
Vol. 62, No. 23,1!2!%?
higher 3-methoxy-4-hydroxy-mandelic acid (VMA) and the sum of VMA plus MHPG than those of their instructors (10). These reports suggested to us that a stressor enhances NA turnover in human brain. Based on this consideration, we hypothesized the involvement of abnormal NA turnover (synthesis and metabolism) in the brain in the pathophysiology of depression (11) We examined this hypothesis by first establishing an animal model of depression by exposing rats to chronic stress (12). In a series of studies in our laboratory, we showed that the depression model induced by long-term forced walking stress (FWS) met the minimum criteria of depression (13, 14) in terms of (I) persistent inactivity during spontaneous running, fir) reversal of depressive behavior following long-term treatment with imipramine (15) and other antidepressants such as setiptiline and maprotiline, (111) hyperfunctional hypothalamic-pituitary-adrenal axis, (iti, circadian dysrhythmia of serum corticosterone (16) and (v/ disturbances of the estrous cycle in female rats (12). Affected rats also show other behaviors including motor retardation, seclusive (crouching at a comer of cage) and aggressive behavior (attacking against a forwarded stick), lack of coupling behavior, fitful sleep and hypersensitivity to light and sound (17) Persistent weight loss was also observed, but blood clots on the gastric wall, detected during the acute stage of stress, are not found in the model. Food and water intake, rectal temperature and motor responsiveness (the speed at which rats run in the drum in response to clapping sounds) return to prestress level several days after the FWS. Thus, these model animals are considered to recover from physical exhaustion within several days following FWS These features are not seen in the spontaneous recovery rats Such a model is thus suitable for the study of NA synthesis, particularly the expression of mRNA necessary for the production of tyrosine hydroxylase (TH), a rate-limiting enzyme in the biosynthesis of NA Therefore, we examined the expression of TH mRNA in the locus coeruleus (LC) of the brain in the following experimental groups of rats using ill srfrr hybridization and image analysis fr) depression-model rats: rats with reduced spontaneous activity induced by FWS for 14 days, and (II) spontaneous recovery rats rats during recovery of spontaneous activity following identical stress To examine the time-course of changes in TH mRNA under stress conditions, we also examined its expression after a short period of FWS (30 minutes, 3 hours, 1 or 2 days of forced walking) or long-term FWS (6 or 12 days of forced walking) Materials and Methods Animuls Eleven-week-old male rats were housed in freely rotating drum cages in a room of 22+2”C Lights were turned on at 7:00 and turned off at 19 00 every day The spontaneous running activity (SRA frequency of drum rotations per day) was measured in each animal over a period of 5 weeks Animals were allowed free access to water and a standard diet The experimental design was approved by the Committee for Care and Use of Laboratory Animals, Mie University School of Medicine Exposure to stress Depression-like reduced activity was induced in the rats following a period of long-term FWS according to the method described previously by our laboratory (17-19) Briefly, SRA was examined under the above conditions, then the rats were transferred into forced walking drums (which rotated automatically at 5 revolutions per minute) placed in a temperature-controlled room set at 15°C The selected room temperature did not influence the expression of TH mRNA of freely walking rats. and was specifically selected to allow easy determination of the point of exhaustion which was manifested by a drop in rectal temperature. For rats exposed to
Vol. 62, No. 23,1998
TH mRNA in LC of Depre&oo Model
2085
stress for more than 24 hours, one-hour rest periods were allowed every 4 hours. During the rest periods, the rectal temperature of these animals was monitored to estimate the degree of fatigue. When the rectal temperature fell below 34°C a 24-hour or longer break was permitted until body temperature returned to > 36°C. The period of stress exposure was 30 minutes, 3 hours, 1 or 2 days in the short-term stress experiment and 6 or 12 days in the long-term stress experiment. Animals allocated to the depression-model and recovery groups were exposed to stress for 14 days. In both the shortterm and long-term stress experiments, all animals (including control rats not exposed to stress) were decapitated immediately after the end of stress exposure and their brains were removed. In experiments to develop the model system, the animals were returned to the freely rotating drum cages after the end of stress exposure and their SRA was monitored for the next 2 weeks. Percent recovery of SR4 was calculated by comparing SRA recorded over 7 days before and 3 days after (immediately before decapitation) exposure to stress. Based on the recovery rate, the animals were then divided into the depression-model group (percent recovery < 10%; n == 7) and spontaneous recovery group (percent recovery 110%; n = 6). Tissue samples from these two groups and the control group were used for in situ hybridization histochemistry The distribution of recovery rate of SRA in these animals has been recently described in detail elsewhere (l&20). In situ hybridization After removal of the rat brain, the tissue was immediately frozen on dry ice and cut into 10 coronal 10 urn thick sections including the caudal LC (bregma level: -10.05 to -10.15 mm) (21). Every alternate section was fixed in 4% paraformaldehyde and treated with 0.1 M In situ triethanolamine and 0.9% NaCl solution containing 0.25% acetic anhydride hybridization for TH mRNA was performed using an alkaline phosphatase (ALP)-labeled probe (Yatron, Tokyo), according to the method of Kiyama et al. (22). This probe was a 30 mer oligodeoxynucleotide complementary to rat TH cDNA sequence (1223-1252) and its sensitivity and specificity has been confirmed by Kiyama. The probe was combined at a concentration of approximately 5 fmol/ul with a hybridization buffer consisting of 30% deionized formamide, 1 x Denhardt’s solution, 10% dextran sulfate, 4 x SSC (3.5% NaCl, 1.76% sodium citrate) and salmon testis DNA (500 ug/rnl). After incubation at 37°C for 15 hours, the sample was immersed in 1 x SSC buffer (pH 7.0) containing 0.1 mM ZnClz, and incubated for a further one hour at 50°C. This was followed by 30 minutes incubation at room temperature, immersion for 10 minutes in buffer A [O. 1 M Tris-HCl (pH 7.5), 0.15M NaCI] and then immersion for 5 minutes in buffer B (0 1 M Tris-HCl, 0.1 M NaCl, 0.1 M MgC12) ALP color development was then induced for 5 hours using nitroblue tetrazolium (NBT) and S-bromo-4-chloro-3-indolphosphate-p-toluidine salt (BCIP) as substrates. The color development reaction was stopped by immersion of the sample in buffer C [O.Ol M Tris-HCI, 0.9% NaCl, 1 mM ethylenediaminetetraacetic acid (EDTA)]. Quantification of hybridization signals An IBAS image analyzer (Kontron, Munich) was used for quantification of hybridization signals. The density (grey values: GV) of colored signals in LC nuclei of each side was determined according to the densitometric procedure described by Kitayama et al. (23). The calculated mean optical density (OD) of 5 sections served as a quantitative measurement of TH mRNA expression. The OD of alkaline phosphatase signal (24) was converted from GV according to the equation: OD = -log (255-GV) / 255 (cf. Theoretical Explanation of IBAS) and proved to correlate well with the relative amount of mRNA expression measured using Northern analysis by Kiyama et al. (25). The relative level of TH mRNA expression in test
2086
TH mRNA in LC of Depression Model
Vol. 62, No. 23, 1998
rats was expressed as a percent change of mean OD of control rats
Data were expressed as mean T SD Inter-group differences in TH mRNA expression were examined for statistical significance using ANOVA and post hoc Scheffe’s F test. A P value less than 0.05 was regarded as statistically significant
Figure IA Tyrosine hydroxylase (TH) mRNA expression detected by alkaline phosphatase (ALP)-labeled In .szh hybridization in the locus coeruleus (LC) of rats exposed to short-term forced walking stress (FWS) C I control rat, 30M. 30-minute stress rat, 3H 3-hour stress rat. ID l-day stress rat, 2D: 2-day stress rat Bar = 100 urn Results Figure 1A shows microphotographs depicting TH mRNA expression in the LC of the shortterm stress rats and control rats The photographs are representative of the average density of TH mRNA in each group Measurement of the ievel of expression of TH mRNA indicated that the OD did not differ significantly between the control group and 30-minute or 3-hour
Vol. 62, No. 23,19!?8
TH mRNA in LC of Depression Model
Cl
M30 Figure
H3
Dl
D2
1B
Levels of TH mRNA expression presented as optical density (OD) in LC of rats exposed to the short-term FWS. OD was significantly higher in the l-day and 2-day stress rats compared with the control rats, but not in 30-minute and 3-hour stress rats. Cl: control rats (n = 7), 30M: 30-minute stress rats (n = 7), 3H: 3-hour stress rats (n = 7), 1D l-day stress rats (n = 7), 2D: 2-day stress rats (n= 7). *P < 0.05, **P < 0 01 stress groups, but increased significantly following 1 and 2 days of exposure to stress (Figure 1B). OD was higher in the 2-day stress group than in the l-day stress group, but the difference was not statistically significant.
Figure 2A TH mRNA expression
by ALP-labeled 111SI~ZIhybridization in the LC of rats exposed to the long-term FWS C2. control rat, 6D. 6-day stress rat, 12D 12day stress rat. Bar = 100 pm.
Figure 2A shows TH mRNA expression in the LC of rats exposed to stress for longer periods. TH mRNA expression (presented as OD) in both the 6- and 12-day stress groups was
2087
TH mRNA in LC of Depression Model
Vol. 62, No. 23,1998
significantly higher than in the control group The OD was slightly higher in the 12-day stress group than in the 6-day stress group, but the difference was not statistically significant (Figure 2B)
C2
D6
D12
Figure 2B Levels of TH mRNA expression presented as OD in LC of rats exposed to long-term FWS OD increased significantly in the 6-day and 12-day stress rats compared to the control rats C2- control rats (n = 7) 6D. 6-day stress rats (n = 7) 12D: 12-day stress rats (n = 7) *P c. 0 05
Figure 3A shows TH mRNA expression in the LC of depression-model and spontaneous recovery rats, 2 weeks after FWS for 14 days OD was significantly higher in the depressionmodel group than in the control group. OD in the recovery group was slightly higher than in the control group, but the difference was not significant (Figure 3B)
Figure 3A TH mRNA expression by ALP-labeled IN SIIU hybridization in LC of a depression-model rat and a spontaneous recovery rat 2 weeks after long-term FWS C3: control rat, M. depression-model rat, R. spontaneous recovery rat Bar= 100 urn
Vol. 62, No. 23,1998
TH mRNA ia LC of Depression Model
Figure 3B Levels of TH mRNA expression presented as OD in LC of depression-model rats and spontaneous recovery rats. OD was significantly higher in the depression-model rats compared to control rats, but not in spontaneous recovery rats. C3: control rats (n = 7), M: depression-model rats (n = 7) R: spontaneous recovery rats (n = 6) *P < 0.05 An inverse correlation (r= -0.693) between expression (represented by OD) of TH mRNA and recovery rate of SRA was present when data of 13 depression model and recovery rats were pooled together These results suggest that rats showing increased NA synthesis are more likely to develop depression. Discussion In this study, we examined TH mRNA expression by in situ hybridization histochemistry and performed densitometric image analysis of the LC in rats exposed to short-term FWS (30 minutes, 3 hours, 1 or 2 days) and long-term FWS (6 or 12 days). We also examined TH mRNA expression in depression-model rats with low activity and in spontaneous recovery rats following exposure to long-term (14 days) FWS. Exposure to stress for short periods (3 hours) did not change TH mRNA expression. However, a marked increase in TH mRNA expression was noted following 24 hour-stress Smith et al. (26) reported that TH mRNA expression in the LC increased for 3 or 4 hours after a one hour application of foot-shock or immobilization stress. These responses were different from our observation showing a lack of change in TH mRNA levels after 3 hours exposure to FWS. This discrepancy may be due to either differences in type, duration or severity of stress, or difference in stress sensitivity or tolerance of animals. We think that the severity of stress induced by a short period of forced walking is milder than that produced by electroshock or immobilization. When our animals were exposed to the stress for prolonged periods (6 or 12 days), however, TH mRNA expression remained elevated at the level noted after stress for 1 or 2 days. The persistent increase in TH mRNA expression in these rats is similar to that reported following exposure to cold stress (27) and repeated immobilization stress (28) but different from that reported for repeated foot-shock or immobilization (26) which showed a return of TH mRNA expression to pre-stress levels. Therefore, it is likely that long-term FWS employed in this study and immobilization stress may represent an unadaptable stress in rats. This unadaptable chronic stress seems to be experimentally useful for inducing behavioral depression in animals.
2mJ
TH mRNA in u3 of Depression Model
Vol. 62, No. 23,19!%
In the depression-model rats in which the percent recovery of SRA two weeks after the end of a 14-day exposure to stress was less than 10%. the level of TH mRNA expression remained significantly higher than in the control group, though they were lower than in 2- 12 day stress rats In the spontaneous recovery rats in which the percent recovery two weeks after the end of exposure to stress was over IO%, TH mRNA expression was less than that of the depression-model rats and not different from the control level There was also a inverse correlation between expression of TH mRNA and recovery rate of SR.4 in the model and recovery rats, suggesting that rats showing increased NA synthesis are prone to develop into the depression model, and vice ver.sa~ NA synthesis can reverse to normal when spontaneous activity normalizes after the end of stress We also postulate that the NA synthesis in LC may remain high in the state of depression even two weeks after stress, and this increase may be attributed to an over-excitation of animals. Since it was difficult to discriminate the model rats from the recovery group several days after the end of FWS, we could not determine whether the increased levels of TH mRbIA expression decreased after stress in the recovery rats and persisted in the model rats From a therapeutic point of view, our results may be potentially useful for the development of new pharmacological agents which relieve stress or have antidepressive efficacy. using substances that reduce the expression of TH mRNA In a recent study from our laboratory (16). we demonstrated in the depression-model rats a state of hypetfimctional hypothalamo-pituitary-adrenal axis, particularly a marked increase of the expression of corticotropin-releasing factor (CRF) mRNA as well as arginine vasopressin (AVP) mRNA in the paraventricular nucleus (16) Alonso ri a/. (29) reported that the ascending noradrenergic neurons in the brain accelerated the synthesis and release of hypothalamic CRF and AVP, and Calogero and coileagues (30) stated that CRF hyperhmction was mediated by (x.- and tx:-adrenergic receptors In this regard, the increased synthesis and release of NA in the ascending NA neurons of the model are likely to be involved in the hyperfunction of the HPA-axis, i.e. the stress-sensitive noradrenergic function may facilitate the synthesis and release of CRF and .4VP in the depression-modei animals On the other hand, recent studies have shown that LC noradrenergic neurons are activated by CRF in terms of both firing rate (3 1, 32) and noradrenaline release (32) These findings suggest that either changes in CRF or noradrenergic function may be the primary abnormality in our depressionmodel rats, but the exact abnormality remains to be elucidated Not only noradrenaline but also serotonin or acetylcholine are known to have effects on the synthesis or release of hypothalamic CRF In our previous reports using this animal model of depression, catecholamine (CA) levels in the ascending noradrenergic neurons, such as those in the LC, were examined histochemically using the Falck-Hillarp procedure These studies showed increased fluorescence intensity of C.4 in the areas of the brain examined (15, 17, 33) Changes in fluorescence intensity detected in those studies paralleled the levels of TH mRN,4 expression in the present study, and both related to the length of stress exposure, and inversely correlated with the recovery rate of X.4 of model and recovery groups We have also shown elevated NA levels in the LC of depression-model rats, as measured by high-performance liquid chromatography (34) Moreover, immunoreactivity of TH, as determined by the immunohistochemical method with TH antibody and image analysis, increases in the LC following short-term and long-term FWS, and tends to increase in the depression-model group (unpublished data). In view of these findings, it appears that the high expression of TH mRNA corresponds with an elevated brain noradrenergic function concomitant with increased I\‘.4 synthesis in model rats
Vol. 62, No. 23,1998
TH mRNA io IL of Depression Model
2091
In conclusion, our results showed that more than 3 hours of stress are required before a change in TH mRNA expression occurs in the LC. Longer periods of exposure (6 or 12 days) result in a persistent elevation of TH mRNA expression in LC nuclei. In the depression-model rats, in which spontaneous activity continued to be reduced after long-term exposure to FWS, the level of TH mRNA expression was significantly higher than in control rats, In the spontaneous recovery rats, TH mRNA was lower than in model rats. Our results suggest that the NA synthesis in the LC neurons remains elevated in the depression-model rats even two weeks after long-term FWS, but returned to baseline levels in spontaneous recovery rats. Acknowledgments The authors are indebted to Dr. S. Nakase and other members of Department of Psychiatry, Mie University School of Medicine, for their valuable advice concerning in situ hybridization.
References 1. 2. 3. 4. 5. 6. 7 8. 9 10. 11. 12. 13. 14. 15. 16.
17. 18.
H. ANISMAN, I, Neurobiology of Mood Disorders, R.M. Post and J.C Ballenger (Eds), 407-43 1, William & Wilkins, Baltimore (1984). C. LLOYD, Arch. Gen. Psychiatry 37 541-548 (1980). J. E. BARNETT, R. M. ROSE and G. L. KLERMAN, Stress and mental disorder, 7186 and 12 1- 136, Raven Press, New York (1979) P. BERKMAN, J. Health Sot. Behav. 12 35-45 (1971). F W. Jr. ILFELD, Am. J. Psychiatry 134 161-166 (1977). E. S. PAYKEL, J. K. MYERS, M. N. DIENNELT, G. L. KLERMAN, J. J. LINDENTHAL, M. P. PEPPER, Arch. Gel>. Psychiatry 21 753-760 (1969). E.A. STONE, Catecholamines andBehavior. Vol. 2, A.J. Friedhoff (Ed), 31-72 Plenum, New York (1975). J. G. FILSER, W. E. MUELLER and H. BECKMANN, Br. J. Psychiatry 148 95-97 (1986). GOODWIN, G. MUSCETTOLA and F. K. M. S. BUCHSBAUM, Neuropsychobiology 7 2 12-224 (1981). G.S. KRAHENBUHL, J. HARRIS, R. D. MALCHOW and J. R. STERN, Aviation Space and Environmental Medicine 56 576-580 (1985). J.J. SCHLDKRAUT, Am. J. Psychiatry, 122 509-522 (1965). I. KITAYAMA, INOUE and J. NOMURA, K N. HATOTANI, Psychoneuroendocrinology 4 155-l 72 (1979) W.T. MCKINNEY, Animal models m psychiatry and neurology., I. Hanin and E. Usdin (Eds), 117-126, Pergamon Press, Oxford (1977). P. WILLNER, Psychopharmacology 83 l-16 (1984). I. KITAYAMA S. MURASE, M. KOISHIZAWA, S. KAWAGUCHI and J NOMLTRA, Jpn J. Psychopharmacol. 7 433-436 (1987). I. KITAYAMA, S. NAKASE, S. KAWAGUCHI, R. NAKASE, J. ONO, K. HAMANAKA, H. SOYA and J. NOMURA, Neurobiology of Depression and Related Disorders, J. Nomura (Ed.), Mie University Press, Tsu (in press). I. KITAYAMA, N. HATOTANI and J. NOMURA, Neurobiology of Periodic Psychoses, N. Hatotani and J. Nomura (Eds), 169-183, Igaku-shoin, Tokyo (1983). I. KITAYAMA, T. YAGA, T. KAYAHARA, K. NAKANO, S. MURASE, M. OTANI, and J. NOMURA, Biol. Psychiatry 42 687-696 (1997).
2m2
19.
20.
21 22. 23
24. 25. 26. 27. 28. 29. 30. 31 32. 33.
34
TH mRNA ia LC of Depression Model
Vol. 62, No. 23,1998
I. KITAYAMA, P. WANG, K. ZHANG, J. ZENG, Y. UESUGI and J. NOMURA, Neurobiology of Depression and Related Disorders, J. Nomura (Ed.), Mie University Press, Tsu (in press). K. HAMANAKA, H. SOYA, H. YOSHIZATO, S. NAKASE, J. ONO, K. INUI, K. ZHANG, R. OKUYAMA, Y. ISHIKAWA, I. KITAYAMA and J. NOMURA, J Neuroendocrinology (in press). G. PAXINOS and C. WATSON, 7Jre Rat Brain zn Stereotaxic Coordinates. 2nd Ed., Academic Press, New York (1986) H. KIYAMA P.C. EMSON, J. RUTH and C MORGAN, Mol. Brain Res 7 213-219 (1990). I. KITAYAMA, A. CINTRA, A.M JANSON, K. FUXE, L.F. AGNATI, P. ENEROTH, M. ARONSSON, A HARFSTRAND, H.W.M. STEINBUSH, T.J. VISSER, M. GOLDSTEIN, W. VALE and J.A. GUSTAFSSON, J. Neural Transm. 77 93-130 (1989). C J. F VAN NOORDEN and G. N. JONGES, Histochem. J. 19 94-102 (1987) H. KIYAMA, P C. EMSON and M. TOHYAMA, Neurosci. Res. 9 1-21 (1990) M.A. SMITH, L S. BRADY, J GLOWA, P W. GOLD and M. HERKENHAM, Brain Res. 544 26-32 (1991). F RICHARD, N. FAUCON-BIGUET, R. LABATUT, D. ROLLET, J. MALLET and M. BUDA, J. Neurosci. Res. 20 32-37 (1988). E. MAMALAKI, R. KVETNANSKY, L S BRADY, P. W. GOLD and M. HERKENHAM, J. Neuroendocrinol. 4 689-699 (1992) G. ALONSO, A. SZAFARCZYK, M BALMEFREZOL and I. ASSENMACHER, Brain Res. 397 297-307 (1986) A. E. CALOGERO, W. T GALLUCCI, G. P CHROUSOS and G. P. GOLD, J. Clin. Invest. 82 839-846 (1988). M. K. BORSODY and J. M. WEISS, Brain Res. 724 149-168 (1996). .4. L. CURTIS, S. M. LECHNER, L. A. PAVCOVICH and R. J. VALENTINO, J Pharmacol. Exp. Ther. 281 163-172 (1997). I. KITAYAMA, M. KOISHIZAWA, J NOMURA, N. HATOTANI and I. NAGATSU, Stress:: 7he Role of Catecholamines and Other Neurotransmitters, E. Usdin, R. Kvetnansky and J. Axelrod (Eds ), 325-135, Gordon and Breach, New York (1984). I. KITAYAMA, P WANG, K.YAMASHITA, M. HARADA, S. MURASE, M. OTANI, T. NAKAMURA, and J. NOMURA, Neurobzologv of Depression and Related Disorders, J Nomura (Ed.), Mie University Press, Tsu (in press).
ELSEVIER
Life Sciences, Vol. 62, No. 23, pp. 2o83-2092, 1998 CQpyright0 19% Jskevier science Inc. Printed in the USA. All rights refed ooz4-32os/98 519.00 t .oo
TYROSINE HYDROXYLASE GENE EXPRESSION IN THE LOCUS COERULEUS OF DEPRESSION-MODEL RATS AND RATS EXPOSED TO SHORT- AND LONG-TERM FORCED WALKING STRESS
Ping Wang, Isao Kitayamai Department
of Psychiatry,
Mie University
& Junichi Nomura School of Medicine, Tsu, Japan
(Received in final form March l3,1998) Summary Abnormal brain noradrenergic function is thought to cause depressive illnesses which are sometimes manifested or aggravated under stressful conditions. To investigate the effect of chronic stress on noradrenaline (NA) synthesis in the brain we used in sihl hybridization to examine the expression of tyrosine hydroxylase (TH) mRNA in the locus coeruleus (LC) of “depression-model rats” that exhibit reduced activity following exposure to long-term (14 days) forced walking stress (FWS) We also examined TH mRNA expression in rats stressed for 30 minutes, 3 hours and 1, 2 (short-term), 6 or 12 (long-term) days. The expression of TH mRNA increased markedly following 1 to 12 days of FWS, but not in rats exposed to FWS for 30 minutes or 3 hours. The expression also increased significantly in the depression-model rats, but not in the “spontaneous recovery rats” whose activity was restored after long-term stress. Our results suggest that N-4 synthesis remains high in the FWS-induced depression-model rats because of the high levels of TH mRNA expression in the LC. Our results also suggest that FWS is initially a mild stress but gradually becomes a severe form of unadaptable stress as reflected by delayed but persistent increases in TH mRNA expression. Key Words: walking stress, locus coeruleus, tyrosine hydroxylase mRNA, in situ hybridization, animal model, depression
Since depression often develops under conditions of stress, stress appears to cause depression (1). Lloyd (2) Paykel, and Clayton and Darvish (2 papers in reference 3) emphasized that exposure to stress leads to the development of depression, and Berkman (4) Ilfeld (5) and Paykel et al. (6) demonstrated that a strong stressor increases the risk of manifestation of depression. It is also well known in animal experiments that exposure to stress enhances noradrenaline (NA) turnover in the brain (7). The turnover of brain NA can be estimated to some extent by measurement of 3-methoxy-4-hydroxyphenylethyleneglycol (MHPG), a metabolite of NA in urine, although about 20% of the metabolite is of central origin (8). In the human, male normal volunteers showed a significant increase in urinary MHPG under visual and painful electrical stimuli (9). Furthermore, students and instructors during in-flight emergencies also showed an increase in urinary NA as well as adrenaline, and students revealed ‘Corresponding author, Isao Kitayama, M.D., Department of Psychiatry, Mie University School of Medicine, 2-l 74, Edobashi, Tsu, Mie 5 14 Japan, Tel: +81 (59) 23 l-5018, Fax: +81 (59) 23 l-5208, e-mail, [email protected]
TH mRNA in Lx:of Depression Model
Vol. 62, No. 23,1!2!%?
higher 3-methoxy-4-hydroxy-mandelic acid (VMA) and the sum of VMA plus MHPG than those of their instructors (10). These reports suggested to us that a stressor enhances NA turnover in human brain. Based on this consideration, we hypothesized the involvement of abnormal NA turnover (synthesis and metabolism) in the brain in the pathophysiology of depression (11) We examined this hypothesis by first establishing an animal model of depression by exposing rats to chronic stress (12). In a series of studies in our laboratory, we showed that the depression model induced by long-term forced walking stress (FWS) met the minimum criteria of depression (13, 14) in terms of (I) persistent inactivity during spontaneous running, fir) reversal of depressive behavior following long-term treatment with imipramine (15) and other antidepressants such as setiptiline and maprotiline, (111) hyperfunctional hypothalamic-pituitary-adrenal axis, (iti, circadian dysrhythmia of serum corticosterone (16) and (v/ disturbances of the estrous cycle in female rats (12). Affected rats also show other behaviors including motor retardation, seclusive (crouching at a comer of cage) and aggressive behavior (attacking against a forwarded stick), lack of coupling behavior, fitful sleep and hypersensitivity to light and sound (17) Persistent weight loss was also observed, but blood clots on the gastric wall, detected during the acute stage of stress, are not found in the model. Food and water intake, rectal temperature and motor responsiveness (the speed at which rats run in the drum in response to clapping sounds) return to prestress level several days after the FWS. Thus, these model animals are considered to recover from physical exhaustion within several days following FWS These features are not seen in the spontaneous recovery rats Such a model is thus suitable for the study of NA synthesis, particularly the expression of mRNA necessary for the production of tyrosine hydroxylase (TH), a rate-limiting enzyme in the biosynthesis of NA Therefore, we examined the expression of TH mRNA in the locus coeruleus (LC) of the brain in the following experimental groups of rats using ill srfrr hybridization and image analysis fr) depression-model rats: rats with reduced spontaneous activity induced by FWS for 14 days, and (II) spontaneous recovery rats rats during recovery of spontaneous activity following identical stress To examine the time-course of changes in TH mRNA under stress conditions, we also examined its expression after a short period of FWS (30 minutes, 3 hours, 1 or 2 days of forced walking) or long-term FWS (6 or 12 days of forced walking) Materials and Methods Animuls Eleven-week-old male rats were housed in freely rotating drum cages in a room of 22+2”C Lights were turned on at 7:00 and turned off at 19 00 every day The spontaneous running activity (SRA frequency of drum rotations per day) was measured in each animal over a period of 5 weeks Animals were allowed free access to water and a standard diet The experimental design was approved by the Committee for Care and Use of Laboratory Animals, Mie University School of Medicine Exposure to stress Depression-like reduced activity was induced in the rats following a period of long-term FWS according to the method described previously by our laboratory (17-19) Briefly, SRA was examined under the above conditions, then the rats were transferred into forced walking drums (which rotated automatically at 5 revolutions per minute) placed in a temperature-controlled room set at 15°C The selected room temperature did not influence the expression of TH mRNA of freely walking rats. and was specifically selected to allow easy determination of the point of exhaustion which was manifested by a drop in rectal temperature. For rats exposed to
Vol. 62, No. 23,1998
TH mRNA in LC of Depre&oo Model
2085
stress for more than 24 hours, one-hour rest periods were allowed every 4 hours. During the rest periods, the rectal temperature of these animals was monitored to estimate the degree of fatigue. When the rectal temperature fell below 34°C a 24-hour or longer break was permitted until body temperature returned to > 36°C. The period of stress exposure was 30 minutes, 3 hours, 1 or 2 days in the short-term stress experiment and 6 or 12 days in the long-term stress experiment. Animals allocated to the depression-model and recovery groups were exposed to stress for 14 days. In both the shortterm and long-term stress experiments, all animals (including control rats not exposed to stress) were decapitated immediately after the end of stress exposure and their brains were removed. In experiments to develop the model system, the animals were returned to the freely rotating drum cages after the end of stress exposure and their SRA was monitored for the next 2 weeks. Percent recovery of SR4 was calculated by comparing SRA recorded over 7 days before and 3 days after (immediately before decapitation) exposure to stress. Based on the recovery rate, the animals were then divided into the depression-model group (percent recovery < 10%; n == 7) and spontaneous recovery group (percent recovery 110%; n = 6). Tissue samples from these two groups and the control group were used for in situ hybridization histochemistry The distribution of recovery rate of SRA in these animals has been recently described in detail elsewhere (l&20). In situ hybridization After removal of the rat brain, the tissue was immediately frozen on dry ice and cut into 10 coronal 10 urn thick sections including the caudal LC (bregma level: -10.05 to -10.15 mm) (21). Every alternate section was fixed in 4% paraformaldehyde and treated with 0.1 M In situ triethanolamine and 0.9% NaCl solution containing 0.25% acetic anhydride hybridization for TH mRNA was performed using an alkaline phosphatase (ALP)-labeled probe (Yatron, Tokyo), according to the method of Kiyama et al. (22). This probe was a 30 mer oligodeoxynucleotide complementary to rat TH cDNA sequence (1223-1252) and its sensitivity and specificity has been confirmed by Kiyama. The probe was combined at a concentration of approximately 5 fmol/ul with a hybridization buffer consisting of 30% deionized formamide, 1 x Denhardt’s solution, 10% dextran sulfate, 4 x SSC (3.5% NaCl, 1.76% sodium citrate) and salmon testis DNA (500 ug/rnl). After incubation at 37°C for 15 hours, the sample was immersed in 1 x SSC buffer (pH 7.0) containing 0.1 mM ZnClz, and incubated for a further one hour at 50°C. This was followed by 30 minutes incubation at room temperature, immersion for 10 minutes in buffer A [O. 1 M Tris-HCl (pH 7.5), 0.15M NaCI] and then immersion for 5 minutes in buffer B (0 1 M Tris-HCl, 0.1 M NaCl, 0.1 M MgC12) ALP color development was then induced for 5 hours using nitroblue tetrazolium (NBT) and S-bromo-4-chloro-3-indolphosphate-p-toluidine salt (BCIP) as substrates. The color development reaction was stopped by immersion of the sample in buffer C [O.Ol M Tris-HCI, 0.9% NaCl, 1 mM ethylenediaminetetraacetic acid (EDTA)]. Quantification of hybridization signals An IBAS image analyzer (Kontron, Munich) was used for quantification of hybridization signals. The density (grey values: GV) of colored signals in LC nuclei of each side was determined according to the densitometric procedure described by Kitayama et al. (23). The calculated mean optical density (OD) of 5 sections served as a quantitative measurement of TH mRNA expression. The OD of alkaline phosphatase signal (24) was converted from GV according to the equation: OD = -log (255-GV) / 255 (cf. Theoretical Explanation of IBAS) and proved to correlate well with the relative amount of mRNA expression measured using Northern analysis by Kiyama et al. (25). The relative level of TH mRNA expression in test
2086
TH mRNA in LC of Depression Model
Vol. 62, No. 23, 1998
rats was expressed as a percent change of mean OD of control rats
Data were expressed as mean T SD Inter-group differences in TH mRNA expression were examined for statistical significance using ANOVA and post hoc Scheffe’s F test. A P value less than 0.05 was regarded as statistically significant
Figure IA Tyrosine hydroxylase (TH) mRNA expression detected by alkaline phosphatase (ALP)-labeled In .szh hybridization in the locus coeruleus (LC) of rats exposed to short-term forced walking stress (FWS) C I control rat, 30M. 30-minute stress rat, 3H 3-hour stress rat. ID l-day stress rat, 2D: 2-day stress rat Bar = 100 urn Results Figure 1A shows microphotographs depicting TH mRNA expression in the LC of the shortterm stress rats and control rats The photographs are representative of the average density of TH mRNA in each group Measurement of the ievel of expression of TH mRNA indicated that the OD did not differ significantly between the control group and 30-minute or 3-hour
Vol. 62, No. 23,19!?8
TH mRNA in LC of Depression Model
Cl
M30 Figure
H3
Dl
D2
1B
Levels of TH mRNA expression presented as optical density (OD) in LC of rats exposed to the short-term FWS. OD was significantly higher in the l-day and 2-day stress rats compared with the control rats, but not in 30-minute and 3-hour stress rats. Cl: control rats (n = 7), 30M: 30-minute stress rats (n = 7), 3H: 3-hour stress rats (n = 7), 1D l-day stress rats (n = 7), 2D: 2-day stress rats (n= 7). *P < 0.05, **P < 0 01 stress groups, but increased significantly following 1 and 2 days of exposure to stress (Figure 1B). OD was higher in the 2-day stress group than in the l-day stress group, but the difference was not statistically significant.
Figure 2A TH mRNA expression
by ALP-labeled 111SI~ZIhybridization in the LC of rats exposed to the long-term FWS C2. control rat, 6D. 6-day stress rat, 12D 12day stress rat. Bar = 100 pm.
Figure 2A shows TH mRNA expression in the LC of rats exposed to stress for longer periods. TH mRNA expression (presented as OD) in both the 6- and 12-day stress groups was
2087
TH mRNA in LC of Depression Model
Vol. 62, No. 23,1998
significantly higher than in the control group The OD was slightly higher in the 12-day stress group than in the 6-day stress group, but the difference was not statistically significant (Figure 2B)
C2
D6
D12
Figure 2B Levels of TH mRNA expression presented as OD in LC of rats exposed to long-term FWS OD increased significantly in the 6-day and 12-day stress rats compared to the control rats C2- control rats (n = 7) 6D. 6-day stress rats (n = 7) 12D: 12-day stress rats (n = 7) *P c. 0 05
Figure 3A shows TH mRNA expression in the LC of depression-model and spontaneous recovery rats, 2 weeks after FWS for 14 days OD was significantly higher in the depressionmodel group than in the control group. OD in the recovery group was slightly higher than in the control group, but the difference was not significant (Figure 3B)
Figure 3A TH mRNA expression by ALP-labeled IN SIIU hybridization in LC of a depression-model rat and a spontaneous recovery rat 2 weeks after long-term FWS C3: control rat, M. depression-model rat, R. spontaneous recovery rat Bar= 100 urn
Vol. 62, No. 23,1998
TH mRNA ia LC of Depression Model
Figure 3B Levels of TH mRNA expression presented as OD in LC of depression-model rats and spontaneous recovery rats. OD was significantly higher in the depression-model rats compared to control rats, but not in spontaneous recovery rats. C3: control rats (n = 7), M: depression-model rats (n = 7) R: spontaneous recovery rats (n = 6) *P < 0.05 An inverse correlation (r= -0.693) between expression (represented by OD) of TH mRNA and recovery rate of SRA was present when data of 13 depression model and recovery rats were pooled together These results suggest that rats showing increased NA synthesis are more likely to develop depression. Discussion In this study, we examined TH mRNA expression by in situ hybridization histochemistry and performed densitometric image analysis of the LC in rats exposed to short-term FWS (30 minutes, 3 hours, 1 or 2 days) and long-term FWS (6 or 12 days). We also examined TH mRNA expression in depression-model rats with low activity and in spontaneous recovery rats following exposure to long-term (14 days) FWS. Exposure to stress for short periods (3 hours) did not change TH mRNA expression. However, a marked increase in TH mRNA expression was noted following 24 hour-stress Smith et al. (26) reported that TH mRNA expression in the LC increased for 3 or 4 hours after a one hour application of foot-shock or immobilization stress. These responses were different from our observation showing a lack of change in TH mRNA levels after 3 hours exposure to FWS. This discrepancy may be due to either differences in type, duration or severity of stress, or difference in stress sensitivity or tolerance of animals. We think that the severity of stress induced by a short period of forced walking is milder than that produced by electroshock or immobilization. When our animals were exposed to the stress for prolonged periods (6 or 12 days), however, TH mRNA expression remained elevated at the level noted after stress for 1 or 2 days. The persistent increase in TH mRNA expression in these rats is similar to that reported following exposure to cold stress (27) and repeated immobilization stress (28) but different from that reported for repeated foot-shock or immobilization (26) which showed a return of TH mRNA expression to pre-stress levels. Therefore, it is likely that long-term FWS employed in this study and immobilization stress may represent an unadaptable stress in rats. This unadaptable chronic stress seems to be experimentally useful for inducing behavioral depression in animals.
2mJ
TH mRNA in u3 of Depression Model
Vol. 62, No. 23,19!%
In the depression-model rats in which the percent recovery of SRA two weeks after the end of a 14-day exposure to stress was less than 10%. the level of TH mRNA expression remained significantly higher than in the control group, though they were lower than in 2- 12 day stress rats In the spontaneous recovery rats in which the percent recovery two weeks after the end of exposure to stress was over IO%, TH mRNA expression was less than that of the depression-model rats and not different from the control level There was also a inverse correlation between expression of TH mRNA and recovery rate of SR.4 in the model and recovery rats, suggesting that rats showing increased NA synthesis are prone to develop into the depression model, and vice ver.sa~ NA synthesis can reverse to normal when spontaneous activity normalizes after the end of stress We also postulate that the NA synthesis in LC may remain high in the state of depression even two weeks after stress, and this increase may be attributed to an over-excitation of animals. Since it was difficult to discriminate the model rats from the recovery group several days after the end of FWS, we could not determine whether the increased levels of TH mRbIA expression decreased after stress in the recovery rats and persisted in the model rats From a therapeutic point of view, our results may be potentially useful for the development of new pharmacological agents which relieve stress or have antidepressive efficacy. using substances that reduce the expression of TH mRNA In a recent study from our laboratory (16). we demonstrated in the depression-model rats a state of hypetfimctional hypothalamo-pituitary-adrenal axis, particularly a marked increase of the expression of corticotropin-releasing factor (CRF) mRNA as well as arginine vasopressin (AVP) mRNA in the paraventricular nucleus (16) Alonso ri a/. (29) reported that the ascending noradrenergic neurons in the brain accelerated the synthesis and release of hypothalamic CRF and AVP, and Calogero and coileagues (30) stated that CRF hyperhmction was mediated by (x.- and tx:-adrenergic receptors In this regard, the increased synthesis and release of NA in the ascending NA neurons of the model are likely to be involved in the hyperfunction of the HPA-axis, i.e. the stress-sensitive noradrenergic function may facilitate the synthesis and release of CRF and .4VP in the depression-modei animals On the other hand, recent studies have shown that LC noradrenergic neurons are activated by CRF in terms of both firing rate (3 1, 32) and noradrenaline release (32) These findings suggest that either changes in CRF or noradrenergic function may be the primary abnormality in our depressionmodel rats, but the exact abnormality remains to be elucidated Not only noradrenaline but also serotonin or acetylcholine are known to have effects on the synthesis or release of hypothalamic CRF In our previous reports using this animal model of depression, catecholamine (CA) levels in the ascending noradrenergic neurons, such as those in the LC, were examined histochemically using the Falck-Hillarp procedure These studies showed increased fluorescence intensity of C.4 in the areas of the brain examined (15, 17, 33) Changes in fluorescence intensity detected in those studies paralleled the levels of TH mRN,4 expression in the present study, and both related to the length of stress exposure, and inversely correlated with the recovery rate of X.4 of model and recovery groups We have also shown elevated NA levels in the LC of depression-model rats, as measured by high-performance liquid chromatography (34) Moreover, immunoreactivity of TH, as determined by the immunohistochemical method with TH antibody and image analysis, increases in the LC following short-term and long-term FWS, and tends to increase in the depression-model group (unpublished data). In view of these findings, it appears that the high expression of TH mRNA corresponds with an elevated brain noradrenergic function concomitant with increased I\‘.4 synthesis in model rats
Vol. 62, No. 23,1998
TH mRNA io IL of Depression Model
2091
In conclusion, our results showed that more than 3 hours of stress are required before a change in TH mRNA expression occurs in the LC. Longer periods of exposure (6 or 12 days) result in a persistent elevation of TH mRNA expression in LC nuclei. In the depression-model rats, in which spontaneous activity continued to be reduced after long-term exposure to FWS, the level of TH mRNA expression was significantly higher than in control rats, In the spontaneous recovery rats, TH mRNA was lower than in model rats. Our results suggest that the NA synthesis in the LC neurons remains elevated in the depression-model rats even two weeks after long-term FWS, but returned to baseline levels in spontaneous recovery rats. Acknowledgments The authors are indebted to Dr. S. Nakase and other members of Department of Psychiatry, Mie University School of Medicine, for their valuable advice concerning in situ hybridization.
References 1. 2. 3. 4. 5. 6. 7 8. 9 10. 11. 12. 13. 14. 15. 16.
17. 18.
H. ANISMAN, I, Neurobiology of Mood Disorders, R.M. Post and J.C Ballenger (Eds), 407-43 1, William & Wilkins, Baltimore (1984). C. LLOYD, Arch. Gen. Psychiatry 37 541-548 (1980). J. E. BARNETT, R. M. ROSE and G. L. KLERMAN, Stress and mental disorder, 7186 and 12 1- 136, Raven Press, New York (1979) P. BERKMAN, J. Health Sot. Behav. 12 35-45 (1971). F W. Jr. ILFELD, Am. J. Psychiatry 134 161-166 (1977). E. S. PAYKEL, J. K. MYERS, M. N. DIENNELT, G. L. KLERMAN, J. J. LINDENTHAL, M. P. PEPPER, Arch. Gel>. Psychiatry 21 753-760 (1969). E.A. STONE, Catecholamines andBehavior. Vol. 2, A.J. Friedhoff (Ed), 31-72 Plenum, New York (1975). J. G. FILSER, W. E. MUELLER and H. BECKMANN, Br. J. Psychiatry 148 95-97 (1986). GOODWIN, G. MUSCETTOLA and F. K. M. S. BUCHSBAUM, Neuropsychobiology 7 2 12-224 (1981). G.S. KRAHENBUHL, J. HARRIS, R. D. MALCHOW and J. R. STERN, Aviation Space and Environmental Medicine 56 576-580 (1985). J.J. SCHLDKRAUT, Am. J. Psychiatry, 122 509-522 (1965). I. KITAYAMA, INOUE and J. NOMURA, K N. HATOTANI, Psychoneuroendocrinology 4 155-l 72 (1979) W.T. MCKINNEY, Animal models m psychiatry and neurology., I. Hanin and E. Usdin (Eds), 117-126, Pergamon Press, Oxford (1977). P. WILLNER, Psychopharmacology 83 l-16 (1984). I. KITAYAMA S. MURASE, M. KOISHIZAWA, S. KAWAGUCHI and J NOMLTRA, Jpn J. Psychopharmacol. 7 433-436 (1987). I. KITAYAMA, S. NAKASE, S. KAWAGUCHI, R. NAKASE, J. ONO, K. HAMANAKA, H. SOYA and J. NOMURA, Neurobiology of Depression and Related Disorders, J. Nomura (Ed.), Mie University Press, Tsu (in press). I. KITAYAMA, N. HATOTANI and J. NOMURA, Neurobiology of Periodic Psychoses, N. Hatotani and J. Nomura (Eds), 169-183, Igaku-shoin, Tokyo (1983). I. KITAYAMA, T. YAGA, T. KAYAHARA, K. NAKANO, S. MURASE, M. OTANI, and J. NOMURA, Biol. Psychiatry 42 687-696 (1997).
2m2
19.
20.
21 22. 23
24. 25. 26. 27. 28. 29. 30. 31 32. 33.
34
TH mRNA ia LC of Depression Model
Vol. 62, No. 23,1998
I. KITAYAMA, P. WANG, K. ZHANG, J. ZENG, Y. UESUGI and J. NOMURA, Neurobiology of Depression and Related Disorders, J. Nomura (Ed.), Mie University Press, Tsu (in press). K. HAMANAKA, H. SOYA, H. YOSHIZATO, S. NAKASE, J. ONO, K. INUI, K. ZHANG, R. OKUYAMA, Y. ISHIKAWA, I. KITAYAMA and J. NOMURA, J Neuroendocrinology (in press). G. PAXINOS and C. WATSON, 7Jre Rat Brain zn Stereotaxic Coordinates. 2nd Ed., Academic Press, New York (1986) H. KIYAMA P.C. EMSON, J. RUTH and C MORGAN, Mol. Brain Res 7 213-219 (1990). I. KITAYAMA, A. CINTRA, A.M JANSON, K. FUXE, L.F. AGNATI, P. ENEROTH, M. ARONSSON, A HARFSTRAND, H.W.M. STEINBUSH, T.J. VISSER, M. GOLDSTEIN, W. VALE and J.A. GUSTAFSSON, J. Neural Transm. 77 93-130 (1989). C J. F VAN NOORDEN and G. N. JONGES, Histochem. J. 19 94-102 (1987) H. KIYAMA, P C. EMSON and M. TOHYAMA, Neurosci. Res. 9 1-21 (1990) M.A. SMITH, L S. BRADY, J GLOWA, P W. GOLD and M. HERKENHAM, Brain Res. 544 26-32 (1991). F RICHARD, N. FAUCON-BIGUET, R. LABATUT, D. ROLLET, J. MALLET and M. BUDA, J. Neurosci. Res. 20 32-37 (1988). E. MAMALAKI, R. KVETNANSKY, L S BRADY, P. W. GOLD and M. HERKENHAM, J. Neuroendocrinol. 4 689-699 (1992) G. ALONSO, A. SZAFARCZYK, M BALMEFREZOL and I. ASSENMACHER, Brain Res. 397 297-307 (1986) A. E. CALOGERO, W. T GALLUCCI, G. P CHROUSOS and G. P. GOLD, J. Clin. Invest. 82 839-846 (1988). M. K. BORSODY and J. M. WEISS, Brain Res. 724 149-168 (1996). .4. L. CURTIS, S. M. LECHNER, L. A. PAVCOVICH and R. J. VALENTINO, J Pharmacol. Exp. Ther. 281 163-172 (1997). I. KITAYAMA, M. KOISHIZAWA, J NOMURA, N. HATOTANI and I. NAGATSU, Stress:: 7he Role of Catecholamines and Other Neurotransmitters, E. Usdin, R. Kvetnansky and J. Axelrod (Eds ), 325-135, Gordon and Breach, New York (1984). I. KITAYAMA, P WANG, K.YAMASHITA, M. HARADA, S. MURASE, M. OTANI, T. NAKAMURA, and J. NOMURA, Neurobzologv of Depression and Related Disorders, J Nomura (Ed.), Mie University Press, Tsu (in press).