Ontogeny of the Type 2 glucocorticoid receptor in discrete rat brain regions: an immunocytochemical study

Developmental Brain Research, 42 (1988) 119-127

119

Elsevier BRD 5(1777

Ontogeny of the Type 2 glucocorticoid receptor in discrete rat brain regions: an immunocytochemical study P. Rosenfeld 2, J.A.M. V a n E e k e l e n 1, S. Levine 2 and E . R . D e Kloet l IRudolf Magnus Institute for Pharmacology, University of Utrecht, Utrecht (The Netherlands) and 2Department of Psychiatry, Stanford University, Stanford, CA 94305 (U.S.A.) (Accepted 8 March 1988)

Key words: Glucocorticoid receptor; Immunocytochemistry; Ontogeny; Brain; Rat

The ontogeny of the Type 2 glucocorticoid receptor (GR) in the rat brain was examined using a monoclonal antibody raised against the rat liver GR. Marked changes both in the intensity and in the localization of GR immunoreactivity (GR-ir) were found to occur as a function of age and brain area examined. First, GR-ir was high perinatally and decreased to a low intensity of immunostaining around postnatal day 12 (pnd 12). Thereafter, GR-ir increased to a moderate intensity, which resembled adult levels by pnd 20 in most brain areas. Second, in some regions, such as the hippocampal CA3-4 pyramidal cell fields and the suprachiasmatic nucleus of the hypothalamus, GR-ir was only clearly present during the first postnatal week. Third, in the hippocampus, GR-ir localization showed a distinctive developmental trend towards greater compactness within the CA1-2 pyramidal cell fields and a greater restriction of immunoreactive staining to these cell fields with exclusion of the adjoining areas. Fourth, adrenalectomy reduced overall GR-immunopositive staining, which could be reversed by administration of the selective glucocorticoid agonist, RU 28362. Our results suggest that during ontogeny the glucocorticoid receptor system displays considerable plasticity. Such plasticity may provide a basis for understanding the role of glucocorticoids during brain development.

A n u m b e r of different a p p r o a c h e s have been used in the past to characterize the corticosteroid receptors in the brain during ontogeny. Thus, in vitro cytosol binding assays have d e m o n s t r a t e d changes in receptor concentrations occurring both as a function of age and regional distribution 5A7"2°. Similarly, in vivo

( C O R T ) and a l d o s t e r o n e in the rat with a high affinity. A peculiar p r o p e r t y of T y p e 1 receptors in the brain is that they m e d i a t e with stringent specificity the actions of C O R T , while a l d o s t e r o n e often acts as competitive antagonist 7"8. In the adult brain, the Type 1 r e c e p t o r (or C R ) is found p r e d o m i n a n t l y within the CA1-2 and d e n t a t e gyrus areas of the hippocampus and in the septum 16"21'27. It is involved in

tracer techniques have shown m a r k e d developmental changes in h i p p o c a m p a l microdistribution of corticosterone-labelled receptors18'28. The a b o v e - m e n t i o n e d studies assumed the existence of a single population of corticosteroid receptors. Recent research, however, has shown that there are actually two types of corticosteroid receptors in the central nervous system: T y p e 1 and T y p e 2 (refs. 12, 21). The Type 1 r e c e p t o r is the same in its primary a m i n o : z c i d sequence as the kidney mineralocorticoid r e c e p t o r 2, which binds corticosterone

the synchronization and coordination of circadian activities as well as in certain aspects of adaptive behavior, and in the subtle a d j u s t m e n t s of the H P A axis during its basal state of activity6'9. The T y p e 2 receptor (or G R ) is the classical glucocorticoid r e c e p t o r which displays highest affinity to p o t e n t synthetic glucocorticoids. Affinity for C O R T is 6 - 1 0 times lower than that shown by T y p e 1 receptors. It has a widespread localization in glial cells and neurons throughout the adult brain, especially in those neurons involved in the regulation of the autonomic, be-

INTRODUCTION

Correspondence: E.R. de Kloet, Rudolf Magnus Institute for Pharmacology, Medical Faculty, University of Utrecht, Vondellaan 6, 3521 GD Utrecht, The Netherlands. 0165-3806/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)

120 havioral and neuroendocrine responses t o stress ~3'22. It is becoming increasingly evident that the receptor system may be subject to diverse regulatory influences, beyond those exerted by the iigand itself. These may modify not only its capacity to bind to the ligand, but also its ability to translocate to the nucleus and bind to D N A 15'24. It is thus no longer sufficient to know only receptor capacity and regional distribution: information on the state of the receptor is critical if we are to understand the physiological role the receptor plays in the developing organism. With the recent availability of monoclonal antibodies raised specifically against the rat liver GR, immunocytochemical research on Type 2 glucocorticoid receptors in the brain has now become feasible. The regional distribution of G R in the brain of adult rats, observed with immunocytochemical methods, coincides largely with that obtained via autoradiography of brain sections labelled in vitro with radiolabelled steroids 1322"29. In this study we present the results of a detailed immunocytochemical analysis of the ontogeny of G R in the rat brain. The monoclonal antibody 31 applied allows visualization of the receptor conformation in its DNA-binding state (i.e. 'activated' receptor). We show that GR-immunoreactivity (GR-ir) in the intact neonatal animal changes dramatically both in intensity and localization as a function of age and area examined, in a way not predicted by prior biochemical studies. MATERIALS AND METHODS

Animals. The day after giving birth (= day 1) individually housed, multiparous Wistar rats were given 10 male one-day-old pups. From this moment on the animals were not handled in any way nor were the cages cleaned until time of testing. Food and water (0.9% saline for adrenalectomized animals) were available ad libitum. The litters were maintained under standard lighting (lights on between 06.00 and 20.00 h) and temperature (23 °C) conditions. All procedures were carried out during the afternoon. Selection of animals for any given condition within a single litter was random. For the intact condition, animals were examined at pnd 2, 4, 8, 12, 16 and 20. Adult (= 3 months) ani-

reals were used as reference. Pups were left with their mother until immediately prior to testing. At that moment 2 pups were removed, anaesthesized, perfused and their brains collected to process for immunocytochemistry. A further 2 pups were taken from the same litter, weighed and injected subcutaneously with RU 28362 (100/~g/100 g b. wt., dissolved in 2% E t O H and 98% PEG). Injected pups were returned to their home cage for an hour, after which they were anaesthesized and perfused, as above. Four of the remaining pups were sacrificed and trunk blood was collected in chilled EDTAcoated tubes for plasma-CORT RIA. Blood samples from adult animals were obtained via heart puncture immediately prior to perfusion. For the adrenalectomized (ADX) condition, pups were tested on pnd 4, 8, 12, 16 and 20. A D X was performed on all the pups and their dam 48 h prior to testing time. The operation was carried out via the dorsal approach under ether (12-, 16- and 20-dayolds and adults) or hypothermia (4- and 8-day-olds) anaesthesia. Perfusion, brain and blood collection then proceeded as described above for intact animals.

Tissue preparation and immunocytochemistry Two-, 4- and 8-day-old pups were anaesthesized by hypothermia; 12-, 16- and 20-day-old animals were anaesthesized with pentobarbital (6 mg in 0.1 ml/rat i.p.), as were adult rats (15 mg in 0.25 ml/100 g b.wt., i.p.). Fixation of the brain was performed by transaortic perfusion with 4% paraformaldehyde, 0.2% picric acid in 0.1 M phosphate buffer (pH = 7.4, 4 °C). The brains were placed in the fixative until coronal sections (50 a m ) were cut with the vibratome. Brains proceeding from 2-day-old animals were left for at least 10 days in fixative; with increasing age, progressively less time was required for adequate fixation, so that 20-day-old brains could be cut after 24 h in fixative. The sections were washed in several changes of 0.1 M phosphate buffer (pH = 7.2). The buffer was then replaced by normal horse serum (5% in phosphate buffer, containing 0. i % Triton X-100). Subsequently, the floating sections were incubated with GR-antiserum (1:1000 in phosphate buffer) for 2 days at 4 °C. Further incubation with biotinylated anti-mouse IgG (1:500 in 0.05 M Tris buffer, pH = 7.6, containing 0.1% Triton X-100) and

121 avidin-biotin-peroxidase complex (ABC, 40/A avidin + 40 kd biotinylated peroxidase in 10 ml Tris buffer) was performed for 1 h each at room temperature. The ABC-components were obtained in kit-form from Vector Laboratories (Burlingame, CA). Between the latter incubation steps the sections were rinsed twice with Tris buffer. 3,3-Aminobenzidine tetrahydrochloride (Polysciences) was applied as the chromogen. Finally, the sections were mounted on gelatin-coated slides, dehydrated and coverslipped. The monoclonal antibody 1 G R 49/4 was generated against rat liver GR and kindly provided by Dr. H.M. Westphal (F.R.G., see ref. 31 for procedure of antibody generation). Control studies in adult as well as 2-day-old pups did not show immunoreaction when the sections were incubated with ascites fluid, nor when the primary antibody was omitted. In the adult rat, the immunoperoxidase reaction was markedly reduced after preabsorption of the antiserum with a purified GR preparation of rat liver29. Additionally, a substantial decrease in immunopositive GR-staining was observed when 2-day-old pup sections were incubated with antiserum that was preabsorbed with fresh liver cytosol. Prior to preabsorption, the cytosol fraction was incubated for 3 h with a saturating dose of CORT (20 nM), whereafter the receptor-ligand complex was separated from its unbound form by passage through a Sephadex LH20 column 21.

Statistical analysis The Student's t-test was applied for statistical comparison of plasma CORT between the intact and ADX condition at pnd 4, 8, 12, 16 and 20. RESULTS

Intact condition GR-ir was confined mostly to the cell nucleus. Cytoplasmic immunostaining was very faint, to the point of being indistinguishable from background staining. Adult patterns of staining emerged gradually as a result of two different, simultaneously occurring developmental processes: (1) changes in GR-ir per cell and/or number of cells presenting GR-ir within a given area, as evidenced by changes in the intensity of immunoreactive staining in the cell fields: and (2) cellular re-organization, as evidenced by

changes in the localization of GR-ir to, and within, the cell fields. These features are illustrated below in discrete brain areas: the hippocampus and the hypothalamus. Hippocampus. Two- and 4-day old animals presented relatively strong cell nuclear GR-ir in all the pyramidal cell fields (CA1-4) as well as in the granular cells of the dentate gyrus (DG, Fig. 1A). By 8 days of age, a distinct difference could be observed between the different pyramidal cell fields: GR-ir was moderate but still clearly visible in the CA1-2 fields and DG; cell nuclear immunostaining in the CA3-4 region, however, was markedly reduced (Fig. 1B). This distinction became even more apparent in the 12-day-old animal; whereas GR-ir was still detectable in the CA1-2 fields and in DG, CA3-4 immunostaining was indistinguishable from the background (Fig. 1C). At this age the lowest intensity of immunoreactive staining was observed; hereafter, at pnd 16, GR-ir increased again in the CA1-2 cell fields and DG. Cell nuclear immunostaining, however, remained absent from the CA3-4 areas (Fig. 1D). At pnd 20, intensity of cell nuclear staining in these areas resembled that found in the adult hippocampus (Fig. 1E). Age-related changes were also observed in the pattern and localization of GR-ir to, and within, the hippocampal cell fields. At 2 and 4 days of age, the external zone of CA1-4 showed a greater intensity of GR-ir than did the internal zone (Fig. 1A, see arrows). The dorsal limb of DG was homogeneously stained with an intensity of immunopositive staining resembling that of the external pyramidal cell zone (Fig. 1A, see arrow heads), whereas the ventral limb presented a lower intensity of GR-ir. Parallel with age, GR-ir cell nuclei presented a more compact appearance, the difference between external and internal cell layers was no longer evident, and there was a marked reduction in the number of immunostained cell nuclei found in the areas adjoining the cell fields (Fig. 1A vs B.C,D,E). Hypothalamus. At 2 days of age. the different hypothalamic nuclei could not be clearly distinguished. since cell nuclear GR-ir was found throughout the whole hypothalamus (Fig. 2A. PVN: Fig. 3A. SCN). GR-ir in these nuclei decreased slightly with age. This decrease, however, was far smaller than that observed for the surrounding areas, resulting in in-

122 creasingly distinct h y p o t h a l a m i c nuclei. This change was particularly m a n i f e s t b e t w e e n days 8 and 16 (Fig. 2B,C, PVN). A similar process was o b s e r v e d in the S C N of the a n t e r i o r h y p o t h a l a m u s . A t pnd 8, clear cell n u c l e a r i m m u n o r e a c t i v i t y was visible t h r o u g h o u t

Fig. 1. Distribution of GR-ir in the dorsal hippocampus of the neonatal rat, showing the CA 1-4 pyramidal cell fields and the dentate gyrus (DG). Photomicrographs present coronal sections of the hippocampus at pnd 2 (A), pnd 8 (B), pnd (12), pnd 16 (D) and at 3 months after birth (adult, E). Note strong cell nuclear GR-ir in the external zone of the CA1-4 (arrows) and the dorsal limb of the DG (arrow heads), x30.

the nu-

Fig. 2. Photomicrographs show GR-ir in the paraventricular nucleus of the hypothalamus (PVN) at pnd 2 (A), pnd 8 (B) and pnd 16 (C). Concomitant with age, GR-ir in the hypothalamic area adjacent to the PVN decreased substantially, x81.

123 cleus; the intensity of immunoreactive staining, however, differed per region (Fig. 3B). Well-outlined darkly immunostained cell nuclei were found along the peak of the optic chiasm and the ventral boundary of the 3rd ventricle (Fig. 3B, see arrows). Moder-

ate to low GR-ir was found in the remaining neurons of the SCN. GR-ir was, however, still clear and distinct from background immunostaining. In contrast with the PVN, maximum cell nuclear staining in the SCN was achieved at this age: at older ages a gradual

Fig. 3. Photomicrographs present GR-ir in the suprachiasnaatic nucleus of the anterior hypothalamus (SCN). At pnd 2 (A) and 8 (B), GR-ir is clearly present. Arrows in B point to darkly immunostained cell nuclei along the peak of the optic chiasm and the boundary of the 3rd ventricle. At pnd 12 (C) and 16 (D), GR-immunopositive staining disappears irreversibly. At pnd 20 (E) as in the adult intact rat (F), the SCN is devoid of GR-ir. x 170.

124 and irreversible decline in G R - i m m u n o s t a i n i n g became manifest (Fig. 3C,D). By pnd 20, G R - i r was m a r k e d l y r e d u c e d over the entire SCN (Fig. 3E). A s in the adult its localization was no longer clearly defined to the cell nuclei of the SCN neurons (Fig. 3F).

A drenalectomized condition As shown for the h i p p o c a m p u s , a d r e n a l e c t o m y ( A D X ) resulted in an overall reduction of immunoreactive staining at all ages (Fig. 4B, pnd 12). A D X animals injected with R U 28362 presented a general

Fig. 4. The effect of manipulation on GR-ir in the hippocampus of the neonatal rat (pnd 12). Photomicrographs show a low intensity of GR-ir in the intact 12-day-old pup (A), and absence of GR-ir after ADX (B). GR-ir reappeared after ADX combined with administration of RU 28362 (100/ag/100 g b.wt., s.c.), to an intensity and location which is comparable wlth the intact condition (C). ×45.

125 increase in GR-ir towards the intensity of GR-immunostaining observed in the intact animal (Fig. 4C vs 4A). Plasma corticosterone Fig. 5 shows the plasma C O R T levels measured in blood collection at sacrifice of intact as well as A D X rats. The lowest C O R T level could be observed at pnd 8, which is consistent with other reports ~4'25. Forty-eight hours after ADX, plasma C O R T concentrations were significantly decreased (P < 0.001) and close to detection limits. DISCUSSION Two main developmental processes characterize the emergence of adult GR-ir (Type 2) patterns in the intact rat. First, the number of cells presenting GR-ir, and/or the intensity of GR-ir per cell, showed pronounced age-related variations. Thus, overall intensity of cell nuclear staining in the brain was high perinatally, decreased to a low intensity of GR-ir at pnd 12, and thereafter gradually increased again to moderate immunostaining in most brain areas, but remained conspicuously absent from others (CA3-4, SCN). Second, a marked degree of cellular re-organization seemed to be occurring during development, as evidenced by the changes observed in GR-ir localCORT plclsma(ug/dl )

4

1

FL 2

[~

inlocl

4

m

FL 8

12

16

20

adult

ADX(Z'Bhrs)

Fig. 5. Plasma CORT concentrations in intact and adrenalectomized (ADX) rats, ranging in age from 2 to 20 days. The adult animal (3 months) served as a reference. From the neonatal pups, trunk blood samples were taken in the afternoon. From the adult rat, blood was collected by heart puncture prior to perfusion. Each value expresses mean = S.E.M. from 2-6 animals. At pnd 4, 8, 12, 16 and 20, plasma CORT was significantly reduced (P < 0.001) in the ADX condition, when compared with the age-matched intact condition. Intact, D; ADX (48 h), m.

ization to, and within, the cell fields. Concomitant with increasing age, immunoreactive staining became restricted to progressively more discrete areas, such as the CA1-2 areas and dentate gyrus in the hippocampus and certain hypothalamic nuclei. Staining within the cell fields became homogeneous, and GRimmunopositive cells presented an increasingly compact appearance as maturation proceeded. The present study thus shows a remarkable shift in intensity of GR-ir staining during ontogeny. The apparent U-shaped curve in GR-immunoreactive intensity could simply reflect changes in receptor occupancy due to differences in ligand availability, since our monoclonal antibody recognizes preferentially the 'activated' GR. Circulating C O R T is high around the time of birth, and then decreases precipitously, remaining at low and fairly constant levels until around pnd 14. Interestingly, during this period of low basal levels of plasma CORT, our recent in vitro cytosol binding assay showed a developmental pattern for G R opposite to that found for GR-ir 23. The soluble GR-capacity, 24 h after A D X , appeared low around birth, and increased gradually thereafter. At pnd 20, however, Bmax was still substantially below adult levels. Thus, the number of receptors present neonatally does not coincide with the observed changes in GR-ir. Moreover, although, the receptor is capable of binding injected R U 28362 after A D X which resuits in a subsequent characteristic enhanced immunoreactivity 29, it is noteworthy, that the intensity in GR-ir does not exceed the degree of immunostaining in the intact age-matched pup. These data, taken together, could imply that GR, present in the cytoplasm and capable of binding ligands, may not always be able to translocate to the cell nucleus. There are some suggestive antecedents in the literature, where cytosolic GR-translocation-inhibiting 'factors' have been hypothesized to exist during ontogeny in the rat pituitary 24. Similarly, Litwack et al.~5 have demonstrated the existence of a low molecular weight substance in cytosol of a liver G R preparation, which inhibited receptor complex activation. It remains to be established whether similar factors exist in the brain of the neonatal (and adult) rat. As remarkable as the above-discussed changes in intensity of immunoreactive staining, was the observation of distinct GR-ir in the CA3-4 areas of the hippocampus and in the SCN. Cell nuclear staining in

126 these regions showed a progressive and irreversible decrease over the first 1-3 weeks of life, and was undetectable in the adult. The cause of this disappearance is at present unknown. However, studies in adult rats have shown that G R capacity is sensitive to presynaptic activity. Thus, G R induction in the target region has been found to occur following lesion of the commissura-associational fibers of the hippocampus ~° and denervation of the intermediate lobe of the pituitary 1'26. In the hippocampus, at least, the induction is temporary. Receptor levels return to presurgery levels approximately 4 weeks after lesion, possibly correlating with the re-innervation of the receptor-containing neurons l°. This would suggest that functional neural input suppresses the expression of GR. The disappearance of GR-ir in the CA3-4 region and in the SCN coincides with a period of intense synaptogenesis in these areas. It is tempting to speculate that, in the neonate, in discrete brain regions, suppression of the functional receptor may occur as a result of the establishment of synaptic contacts. The physiological implications of the early post-natal presence of GR-ir in the CA3-4 regions and SCN remain obscure. In the SCN it has been hypothesized that G R could mediate the effect of C O R T of maternal origin on the synchronization of the infant's circadian rhythmicity prior to entrainment of the infant's oscillation by the light-dark cycle 3°. Some 'scattered' GR-ir was observed perinatally, which decreased with age until virtually disappearing. This nuclear staining could correspond, for example, to pyramidal cells that have not yet migrated to their final position within the hippocampus. Alternatively, the more diffused pattern of staining could reflect temporary expression of GR-ir by other types of cells. The changes in distribution of immunoreactive staining both to, and within, the cell fields, may be related to timing of neurogenesis and synaptogenesis. Turner 28, using high resolution autoradiography, has shown that, in the hippocampus, cells must have been 'in position' for some time before differentiation is sufficient to allow significant nuclear retention of CORT. Although her measurements probably reflected mostly Type 1 binding, the principle may be true for both types of receptor. We observed a gradient of immunoreactive staining across the depth of

the pyramidal layer in the CA regions in 2- and 4-dayold animals; the darker cell nuclear staining corresponded to the external layer, where the older cells are found2Q Similarly, the granule cells of the dorsal limb of DG develop sooner than those in the ventral limb; they were also found to present darker GR-ir. Neuronal time of origin, however, cannot be the only explanation; pyramidal cells are formed prenatally, whereas a large proportion of the granule cells of the dentate gyrus have a postnatal time of origin 4. Nevertheless, what remains true is that to a large extent the establishment of permanent connections in these areas is postnatal. Transsynaptic modulation of G R expression may thus underlie the patterns of GR-ir observed. The question arises as to what functional implication this modulation of GR-expression in general has'? It is known that neonatal administration of high doses of C O R T inhibits cell proliferation, synaptogenesis, axonal growth, etc.3Q At a behavioral level, similar doses produce long-term/permanent effects on adaptive behavior, including certain features of the stressresponse assumed to be related to hippocampal function 19. Consequently, high C O R T levels are detrimental, whereas low C O R T levels seem required for normal growth and differentiation processes postnatally. If most C O R T actions are mediated via its receptors, the observed changes in GR-localization neonatally would thus account for the perplexing variety of effects obtained by glucocorticoid administration at different ages. In summary, we have presented an immunocytochemical characterization of GR-ontogeny in the brain of the intact rat. This approach has allowed us to detect profound changes both in the intensity of GR-ir and in its localization to specific brain areas.

ACKNOWLEDGEMENTS The authors are grateful for the generous gifts of the monoclonal antibody 1 G R 49/4 (Dr. H.M. Westphal, Marburg, F.R.G.) and RU 28362 (RousselUCLAF Pharmaceutical Co., Romainville, France). This work was supported by the Dutch Foundation for Medical and Health Research M E D I G O N (to J.A.M.v.E.).

127 REFERENCES 1 Antakly, T., Sasaki, A., Liotta, A.S., Palkovits, M. and Krieger, D.T., Induced expression of the glucocorticoid receptor in the rat intermediate pituitary lobe, Science, 229 (1985) 277-279. 2 Arriza, J.L., Weinberger, C., Cerelli, G., Glaser, T.M., Handelin, B.L., Housman, D.E. and Evans, R.M., Cloning of human mineralocorticoid receptor complementary DNA: structural and functional kinship with the glucocorticoid receptor, Science, 237 (1987) 268-275. 3 Bohn, M.C., Glucocorticoid induced teratologies of the nervous system. In J. Yanai (Ed.), Neurobehavioral Teratologies of the Nervous System, Elsevier, Amsterdam, 1984, pp. 365-387. 4 Bohn, M.C., Granule cell genesis in the hippocampus of rats treated neonatally with hydrocortisone, Neuroscience, 5 (1980) 2003-2012. 5 Clayton, C.J., Grosser, B.I. and Stevens, W., The ontogeny of corticosterone and dexamethasone receptors in the rat brain, Brain Res., 134 (1977) 445-453. 6 Dallman, M.F., Regulation of ACTH secretion: variation on a theme of B, Prog. Horm. Res., 43 (1987) 113-173. 7 De Kloet, E.R., Versteeg, D.H.G. and Kovacs, G.L., Aldosterone blocks the response to corticosterone in the raphe-hippocampal serotonin system, Brain Res., 264 (1983) 323-327. 8 De Kloet, E.R., Sijbesma, H. and Reul, J.M.H.M., Selective control by corticosterone of serotonin receptor capacity in the raphe-hippocampal system, Neuroendocrinology, 42 (1986) 513-522. 9 De Kloet, E.R. and Reul, J.M.H.M., Feedback action and tonic influence of corticosteroids on brain function: a concept arising from heterogeneity of brain receptor systems, Psychoneuroendocrinology, 12 (1987) 83-105. 10 De Ronde, F.S.W., De Kloet, E.R. and Nyakas, C., Corticosteroid receptor plasticity and recovery of a deficient hippocampus associated behavior after unilateral (dorsal) hippocampectomy, Brain Res., 374 (1986) 219-226. 11 Doupe, A.J. and Patterson, P.H., Glucocorticoids and the developing nervous system. In D. Ganten and D. Pfaff (Eds.), Current Topics in Neuroendocrinology, Vol. 2, Springer-Verlag, Berlin, 1982, pp. 23-43. 12 Funder, J.W. and Sheppard, K., Adrenocortical steroids and the brain, Annu. Rev. Physiol., 49 (1987) 397-411. 13 Fuxe, K., Wikstr6m, A.C., Okret, S., Agnati, L.F., H~irfstrand, F., Yu, Z.Y., Granholm, L., Zoli, M., Vale, W. and Gustafsson, J.A., Mapping of glucocorticoid receptor immunoreactive neurons in the rat tel- and diencephaIon using a monoclonal antibody against rat liver glucocorticoid receptors, Endocrinology, 117 (1985) 1803-1812. 14 Henning, S.J., Plasma concentrations of total and free corticosterone during development in the rat, Am. J. Physiol., 235 (1978) E451-E456. 15 Litwack, G., Schmidt, T.J., Miller-Diener, A., Webb, M., Bodine, P., Barnett, C.A., Platt, D. and Baidridge, R.C., Steroid-receptor activation: the glucocorticoid receptor as a model system. In G.P. Chrousos, D.L. Loriaux and M.B. Lipsett (Eds.), Advances in Experimental Medicine and Biology, Vol. 196, Plenum, New York, 1986, pp. 11-22. 16 McEwen, B.S., De Kloet, E.R. and Rostene, W., Adrenal

steroid receptors in the nervous system, Physiol. Rev., 66 (1986) 1121-1189. 17 Meaney, M.J., Sapolsky, R.M. and McEwen, B.S., The development of the glucocorticoid receptor system in the rat limbic brain. I. Ontogeny and autoregulation, Dev. Brain Res., 18 (1985) 159-164. 18 Meaney, M.J., Sapolsky, R.M. and McEwen, B.S., The development of the glucocorticoid receptor system in the rat limbic brain. II. An autoradiographic study, Dev. Brain Res., 18 (1985) 165-168. 19 Meaney, M.J., Aitken, D., Bodnoff, S.R., Iny, L.J., Tatarewicz, J.E. and Sapolsky, R.M., Early postnatal handling alters glucocorticoid receptor concentrations in selected brain regions, Behav. Neurosci., 99 (1985) 765-770. 20 Olpe, H. and McEwen, B.S., Glucocorticoid binding to receptor-like proteins in rat brain and pituitary: ontogenetic and experimentally induced changes, Brain Res., 105 (1976) 121-128. 21 Reul, J.M.H.M. and De Kloet, E.R., Two receptor systems for corticosteroids in the brain: microdistribution and differential occupation, Endocrinology, 117 (1985) 2505-2511. 22 Reul, J.M.H.M. and De Kloet, E.R., Anatomical resolution of two types of corticosterone receptor sites in rat brain with ih vitro autoradiography and computerized image analysis, J. Steroid Biochem., 24 (1986) 269-272. 23 Rosenfeld, P., Sutanto, W., Levine, S. and De Kloet, E.R., Ontogeny of Type 1 and Type 2 corticosteroid receptors in the rat hippocampus, Dev. Brain Res., 42 (1988) 113-118. 24 Sakly, M. and Koch, B., Ontogenesis of the glucocorticoid receptor in the anterior pituitary gland: transient dissociation among cytoplasmic receptor density, nuclear uptake and region of corticotropic activity, Endocrinology, 108 (1981) 591-596. 25 Sapolsky, R.M. and Meaney, M.J., Maturation of the adrenal stress response: neuroendocrine control mechanisms and the stress hyporesponsive period, Brain Res. Rev., 11 (1986) 65-76. 26 Seger, M., Van Eekelen, J.A.M., Kiss, J.Z., Burbach, J.P.H. and De Kloet, E.R., Stimulation of pro-opiomelanocortin gene expression by glucocorticoids in the denervated rat intermediate pituitary gland, Neuroendocrinology, 47 (1988) 350-357. 27 Stumpf, W.E. and Sar, M., Steroid hormone target cells in the extrahypothalamic brain stem and cervical spinal cord: neuroendocrine significance, J. Steroid Biochem., 11 (1979) 801-807. 28 Turner, B.B., Ontogeny of glucocorticoid binding in the rodent brain, Am. Zool., 18 (1978) 461-475. 29 Van Eekelen, J.A.M., Kiss, J.Z., Westphal, H.M. and De Kloet, E.R., Immunocytochemical study of the intraceUular localization of the Type 2 glucocorticoid receptor in the rat brain, Brain Res., 436 (1987) 120-128. 30 Van Eekelen, J.A.M., Rosenfeld, P., Levine, S., Westphal, H.M. and De Kloet, E.R., Postnatal disappearance of glucocorticoid receptor immunoreactivity in the suprachiasmatic nucleus of the rat, Neurosci. Res. Comrnun., 3 (1987) 129-133. 31 Westphal, H.M., Moldenhauer, G. and Beato, M., Monoclonal antibodies to the rat liver glucocorticoid receptor, EMBOJ., 1 (1982) 1467-1471.