- Email: [email protected]
Corticosteroid receptor analyses in rat and hamster brains reveal species specificity in the type I and type II receptors
J. s&mid Eiochem. Vol. 30, No. 1-6, pp. 417-420, 1988 Printed in Great Brttain. All rights reserved
0022-4731/88 $3.00+0.00 Copyright 0 1988 Pergamon Press plc
CORTICOSTEROID RECEPTOR ANALYSES IN RAT AND HAMSTER BRAINS REVEAL SPECIES SPECIFICITY IN THE TYPE I and TYPE II RECEPTORS W. SUTAN~.O*, J. M. H. M. REUL, J. A. M. VAN EEKELEN and E. R. Rudolf Magnus Institute
for Pharmacology,
Medical Faculty, University Utrecht, The Netherlands
of Utrecht,
DE KLOET
Vondellaan
6, 3521 GD
Summary-h vitro cytosol binding, receptor autoradiography with radiolabelled corticosteroid analogs, and immunocytochemistry with monoclonal antibodies have revealed the presence of two receptor systems for corticosteroids in rat and hamster brains. The type I receptor is found mainly in the hippocampal region, and in the hamster it binds cortisol (F) and corticosterone (B) with similar affinity while in the rat (a species which unlike the hamster secretes solely B) the type I receptor shows high affinity to B and not to F. The type II receptor is more widely distributed in the brain and it binds to F (hamster) or B (rat) with affinity 4-6-fold lower than to the type I. in viva, the hamster type I and II retain F much more than B while those in the rat show the opposite. In conclusion, the present study clearly indicates species-specificity in type I and type II receptor systems in these animals. Furthermore, the type I receptor displays in vivo stringent preference for retention of the animal’s predominantly circulating corticosteroid (F in hamster, in B in rat).
INTRODUCTION
It is now well established that two receptor systems for corticosteroids are present in the brain [l-3] and they are the corticosterone-preferring receptor (type I CR, a subtype of the mineralocorticoid receptor MR) localized almost exclusively in the pyramidal neurones of the hippocampus, the lateral septum and central amygdala, and the classical glucocorticoid receptor (type II, GR) with its widespread but uneven distribution in neurones and glial cells [2-41. Glucocorticoid action via each receptor type may serve different functions. Behavioural and biochemical studies suggest the involvement of type I sites in the regulation of ongoing behaviour, e.g. in the activation and synchronization of daily activities (exploration, food-seeking), and sleep-related events, while the type II site mediates the feedback action of B on stressactivated neural metabolism and circuitry. In the present study we examined the heterogeneity and specificity of corticosteroid receptors in the brain of hamsters, a species secreting both F and B with F as the predominantly circulating glucocorticoid [5-Q.
EXPERIMENTAL
Animals and chemicals
Male Chinese hamsters (Cricetulus griseus) weighing 20-25 g, and male Wistar-derived rats weighing 150-200 g were used throughout this study. They were adrenalectomized (ADX) under ether anaesthesia using a dorsal approach 2-3 days prior to experimentation. Following surgery they were housed
Proceedings of the 8th International Symposium of The Journal of Steroid Biochemistry “Recent Advances in Steroid Biochemistry” (Paris, 24-27 May 1987). *To whom correspondence should be addressed. 417
in Perspex cages (hamsters individually, rats 4-5 per cage) at 23°C under standard light conditions (14 h: 10 light-dark period) and given food and drinking water containing 0.9% NaCl, ad libitum. Blood from nonADX hamsters was collected in the morning, evening and following ether stress, for the determination of plasma F and B levels carried out by Dr Th.J. Benraad and co-workers of the University of Nijmegen, The Netherlands. In vitro cytosolic receptor and transcortin binding assays
ADX animals were perfused intracardially under Nembutal anaesthesia with 0.9% ice-cold saline, prior to which blood was collected to obtain plasma. Hippocampal dissection and preparation of cytosol were as described elsewhere [2,9]. Aliquots of diluted plasma (1:50) or cytosol were incubated with [3H]F or [3H]B each at concentration range 0.5-20.0nM except in the case of [3H]F binding to rat (0.5-40.0 nM), in the absence and presence of RU 28362 (in cytosol assays) for discrimination of Type I and II receptor as described previously [2, lo]. [3H]RU28362 (concentration range 0.5-15.0 nM) was used to measure directly binding constants to type II sites. Cytosolic protein as measured by the methods of Lowry et al.[ 111. In vivo receptor occupancy study To determine the in viva specificity of F and B to type I and type II receptors in hamster and rat, the availability of binding of labelled steroid to hippocampal cytosol was measured in ADX animals given graded doses (0.5-1O,OOO~g/1OOg body wt) of F or B 1 h prior to killing. Receptor measurement was carried out using single-point assays by incubating aliquots of cytosol with 15.0-20.0 nM of the appropriate [3H]-corticoid to determine specific binding to
418
W. SurANTo et al.
the Type I or II receptor as described earlier. The calculation of the remaining available binding sites in treated animals was based on the values measured in ADX control animals. In vitro autoradiography The corticoid binding sites were studied using the in vitro autoradiography [3, 131. The sections were dried and exposed to the [‘HI-sensitive LKB Ultrofilm for i([‘H]RU 28362), 6 ([3H]B, rat) or 12 weeks ([3H]F, hamster); each of the latter two was in the presence of 100 x RU 28362. Non-specific binding in each case was determined by including SOO-fold excess of the appropriate unlabelled steroid. A VIPER image analysis system (Gesotec, F.D.R.) equipped with a Hitachi CCTV camera was used for analysis of the autoradiogram. The image was digitized (512 x 512 pixels) and 256 grey levels could be distinguished.
animals were measured and calculated as described in Experimental. Increasing doses of F or B result in differential occupation of type I and II receptors (Table 2). In the hamster, 50% of type I is occupied by low doses of F (0.4 pg/ 100 g body wt) but higher doses of B (67.5,&lOOg body wt). To occupy 50% of type II site, either F or B must be increased to 0.28 and 0.1 mg/lOO g body wt respectively. In the rat, low doses of B (0.9 &lOO g body wt) or very high doses of F (0.8 mg/lOO g body wt) are required to occupy 50% of type I site. To occupy 50% of type II site, 60.0 pg B or 1.4 mg F/l 00 g body wt is required. Table 2. Type I and Type II receptor occupancies in ADX animals Species
Receptor types
ED,,F
ED,,B
Hamster
Type I Type II Type I Type II
0.4 280.0 775.0 1350.0
61.5 100.0 0.9 60.0
Rat RESULTS
Heterogeneity of corticosteroid binding to hamster and rat brains The apparent binding affinity (K& in nM) and
capacity (B,,,, in fmol/mg protein) shown in Table 1, are calculated from the Scatchard and Woolf plots of the binding data. [3H]Corticoid (F in hamster, B in rat) bind to type I receptor with higher affinity (K,, 0.9 and 1.0 nM, respectively) than it does to type II (Kd 3.0 and 3.9 nM respectively). B also binds to hamster type I (Kd 0.9nM) and II (Kd 0.5 nM) with high affinity while in contrast, F binds to rat type I and II with lower affinity (Kd 2.2 and 20.1 nM respectively) than does B to these receptors. Table 1. Corticosteroid binding constants to hamster and rat hippocampi
Species Hamster Type I Type II Rat Type I Type II
[3H]Cortisol (F) Bmar. &
[3H]Corticosterone B rnax Kd
0.9 3.0
42.3 106.5
0.9 0.5
131.9 61.3
2.2 20.1
15.0 19.6
I.0 3.9
43.5 260.3
Binding constants of [3H]F and of [‘H]B binding to receptors in hamster and rat hippocampal cytosol. The apparent binding affinity (Kd, expressed as nM) and capacity (as fmollmg protein) were calculated from Scatchard [ 141 and Woolf [ 15, 161analyses of the binding data of [jH]F and [)H]B, each in the presence or absence of loo-fold excess RU 28362; and of [3H]RU 28362 (see Experimental). In all binding assays, r,,, ranged from 0.95 to 0.98.
In vivo receptor occupancy
ADX animals were given graded doses of either F or B (s.c. 0.5-1O,OOO~g/1OOg body wt) 1 h prior to sacrifice. The remaining available hippocampal cytosolic receptor sites (types I and II) in these treated
Receptor occupancy of hamster and rat hippocampal cytosol by exogenously administered F or B. ED,, (effective dose of steroid in &lo0 g body wt) is calculated as the dose of F or B required to occupy 50% of the receptor. Binding capacities were determined as described in Experimental.
In vitro autoradiography In the binding of [W]B (rat) and of [3H]F (hamster) each in the presence of RU 28362 to rat and hamster brains, the autoradiograms show clearly the binding of the corticoid to type I site in the dorsal hippocampal cell field. Within the hippocampus, type I is heterogeneously localized with levels as follows. In the rat: CA1 cell field of the dorsal subiculum= CA2=dentate gyrus (DG)>CA3=ventral hippocampus=lateral septum. In the hamster: CA3= CA2=CAl =DG>CA4>ventral hippocampus=lateral septum. The binding to the type II site (using [3H]RU 28362) reveals a different pattern of distribution in the dorsal hippocampus than type I. Type II is mainly present in the hippocampus, cortex, amygdala, thalamus and hypothalamus where particularly high density is found in the lateral septum and anterior hypothalamus. DISCUSSION The present study demonstrates the existence of two receptor systems for corticosteroids in the brains of hamster and rat. The type I receptor system has a species-specificity for the predominantly circulating glucocorticoid, i.e. F in hamster and B in rat, which is best revealed in vivo. The type II receptor specificity in vivo reflects the binding affinity in vitro which
seems to correspond to the actual presence of circulating F and B in hamster or B only in rat. In other words, the rat type II receptor shows a much higher affinity in vitro to B than F (the latter being poorly retained in vivo in this species). The hamster type II,
Species-specificity corticoid receptor Table 3. Corticosteroid binding constants to hamster and rat transcortin B Inax
Species
Ligand
& (nM)
(fmobmg protein)
Hamster
[)H]B
3.7 4.3 5.8 35.1
769.9 1377.2 8531.9 5838.0
PI-V= Rat
[3H]B [‘HIF
Bindin constants of [JH]F and of [3H]B binding to hamster and rat transcortin. The apparent binding affinity (&) and capacity (B,,,) were calculated from Scatchard and Woolf analyses of the binding data. In all binding assays, r,,, ranged from 0.92 to 0.98.
in contrast, does not discriminate between B and F. The concept of glucocorticoid receptor heterogeneity in the rat is a well-known phenomenon [ 10, 17-l 91. It was shown that the rat brain contains binding sites which bind naturally occurring and synthetic glucocorticoids with different specificity. Synthetic steroids such as RU 26988 and RU 28362 with exclusive glucocorticoid properties have since become available to resolve the distinction between the mineralocorticoid-like receptor 120,211, i.e. type I CR, and the classical glucocorticoid receptor, i.e. type II GR [2,3]. The type I receptor in both rat and hamster is localized almost exclusively in the hippo~mpus, as the automdio~ams clearly show. However, there are differences in the mineraloco~icoid-like property of type I receptor in each species. Whereas rat type I binds ALDO with high affinity, as has been shown by our laboratory and others [2,20-211, hamster type I does not bind ALDO very well and the binding of [IH]F to this receptor can only be displaced by high concentrations of ALDO (unpublished observations). In vitro binding studies show that hamster type I binds F and B with similar affinity but the in vivo receptor occupancy studies show a distinct preference for F over B (the reverse is true in the rat). Such striking differences cannot be easily explained by the intrinsic properties of type I receptor since the in vitro binding affinities, although relatively in line with the autoradiography, did not differ much for F and B. Similar affinity (K,, 3-4 nM) was also obseved for binding of F and B to transcortin (see Table 3). Such preference, therefore, may be due to the steroid’s rate of penetration into the target calls and/or the affinity of the respective receptor complex to DNA acceptor sites. It should also be noted that the in vitro Table 4. Ptasma cortisol and corticosterone levels in hamster Condition a.m. p.m. Stress
Cortisol
Corticosterone
0.8 5.5 4.8
0.014 1.550 1.210
Blood samples were taken in the morning (a.m.), evening (p.m.) and following ether stress. Plasma corticoid levels are expressed in ,I& 100 ml plasma.
419
condition of the binding assay may have masked or unmasked receptor sites that do or do not participate in the in vivo retention process [ 191although evidence for this is still lacking. This study presents a puzzling problem for the ligand specificity of type I. In the rat, type I hippocampus and kidney is identical in the in vitro binding profile and primary structure. Type I is a mineraloco~icoid receptor that displays, in the kidney, a stringent specificity for ALDO as agonist. In the brain, ALDO is an antagonist for events in neurotransmission and behaviour triggered specifically by B. In the hamster brain, even the ALDO affinity to type I is strongly reduced and F appears to be the predominantly retained steroid. Clearly, future research based on the known primary structure of the type I receptor (Evans, personal communication) will shed more light on this peculiar specificity-confe~ing mechanism underlying type I function. The type II receptor system represents the classical glucocorticoid receptor and is widely, albeit unevenly, distributed in the brain as results from autoradiography show. Immunocytochemical study also revealed a predominant distribution of immunoreactive staining for type II (type II-ir) in the CAl, CA2 (as well as CA3 and CA4 in hamster but not in rat) and dentate gyrus of the hippocampus as well as in the paraventricular, arcuate and supraoptic nuclei of the hypothalamus. F (in hamster) or B (in rat) binds to type II receptor with affinity 4-&fold lower than it does to type I. In the in vivo receptor occupation studies in the hamster similar doses of F or B (1 .O mg/lOO g body wt given SC to ADX animals) are required to occupy 80-90% of the type II receptor. In contrast, the rat type II is 80-90% occupied when ADX rats are given F at doses 10 times higher than the dose of B. This result confirms the earlier in vitro binding studies in which the rat type II binds F with very low affinity. We conclude, therefore, that the type II specificity is intrinsic to the receptor and corresponds to the animal’s circulating corticosteroid. Thus, the hamster type II, in contrast to the rat type II, does not discriminate between F and B. The present findings may help the understanding of the function of type I and II receptors in these two species. In the rat, type I receptor seems to be involved in synchronization and activation of circadian-driven brain processes [20,23], while type II, in the termination of the stress response (i.e. feedback). The question now arises if a similar function for type I and type II can be postulated in the hamster from the present binding studies. Hamsters secrete both B and F and as our data (see Table 4) and others [24] show, the latter is the primary circulating glucocorticoid at circadian peak and following a stress, indicating that F is the corticoid involved in the feedback action on stressactivated and circadian driven brain mechanism (a supporting role of B, however, cannot be excluded). Previous report from this laboratory 1221showed that very low doses of B were required to saturate the type I receptor and that the function of type I could only be
W. SUTANT~ et al.
420
revealed following ADX and the subsequent replacement of B at pg doses. Type I function was interpreted as being involved in the regulation of on-going behaviour. The present study indicates that the hamster type I displays a similar stringent specificity towards F at low doses as shown in the in vivo uptake experiment. This experiment also showed that similar doses of F and B are required to occupy the type II site. Therefore, it is likely that both steroids are involved in glucocorticoid feedback of behavioural adaptation and neuroendocrine regulation of the stress response (with F, however, being the prevalent one since it circulates in higher concentration). The present study demonstrates the presence of a heterogeneous population of brain corticoid receptors in rat and in hamster, a species which resembles man in having F as the predominant glucocorticoid. The understanding of the corticoid action via type I and II receptor systems is of great importance since glucocorticoids are among the most extensively used therapeutic agents.
in hamster
adrenal
tissue in vitro. J. steroid Biochem. 2
(1971) 307-311.
9. de Kloet
E. R., Wallach G. and McEwen B. S.: Differences in corticosterone and dexamethasone binding to rat brain and pituitary. Endocrinology 96 (1975)
598-609. 10. Philibert D. and Moguilewsky M.: RU 28362, a useful tool for the characterization of glucocorticoid and mineralocorticoid receptors. Proc. 65th Ann. Mtg. Endoc. Sot., San Antonio, Texas, U.S.A. (1983) Abstr. 1018, p. 335. 11. Lowry 0. H., Rosebrough N. J., Farr A. L. and Randall R. J.: Protein measurement with the Folin-phenol reagent. J. biol. Chem. 193 (195 1) 265-275. 12 van Eekelen J. A. M., Kiss J. Z., Westphal H. M. and de Kloet E. R.: lmmunocytochemical study on the regional distribution and intracellular localization of the glucocorticoid receptor in the rat brain. Brain Rex 436 (1987)
120-128. M., 13. Sarrieau A., Vial M., Philibert D., Moguilewsky Dussaillant M., McEwen B. S. and Rostene W.: In vitro binding of tritiated glucocorticoids directly on unfixed rat brain sections. J. steroid Biochem. 20 (1984) 1233-1238. G.: The attractions of proteins for small 14. Scatchard molecules and ions. Proc. N. Y. Acad. Sci. 51 (1949)
660-672.
is a IBRO/UNESCO
Post-Doctoral Fellow. The authors are grateful for the generous gifts of steroids from Roussel-UCLAF (RU 28362), Behring, Hoechst ([‘HI RU 28362) and Organon International (F and
15. Keightley D. D., Fisher R. J. and Cressie N. A. C.: The Woolf plot is more reliable than the Scatchard plot in analysing data from hormone receptor assays. J. steroid
W.
16. Keightley D. D., Fisher R. J. and Cressie N. A. C.: Properties and interpretation of the Woolf and Scatchard plots in analysing data from steroid receptor assay. J. steroid Biochem. 19 (1983) 1407-1412. 17. de Kloet E. R. and McEwen B. S.: Differences between cytosol receptor complexes with corticosterone and dexamethasone in hippocampal tissue from rat brain. Biochim. biophys. Acta 421 (1976) 124-132. receptors in the brain. In 18. Funder J.: Adrenocortical Frontiers in Neuroendocrinology (Edited by W. F. Ganong and L. Martini). Raven Press, New York (1986) pp. 169-189. 19. McEwen B. S., de Kloet E. R. and Rostene W.: Adrenal steroid receptors and actions in the nervous system. Phvsioi. Rev. 66 (1986) 1121-1188. of 20. Beaumont K. and Fenestil D. D.: Characterization rat brain aldosterone receptor reveals high affinity for corticosterone. Endocrinology 113 (1983) 2043-205 1. 21. Krozowski Z. S. and Funder J. W.: Renal mineralocorticoid receptors and hippocompal corticosterone-binding species have identical intrinsic steroid specificity. Proc.
Acknowledgements-WS
REFERENCES 1. de Kloet E. R. and Reul J. M. H. M.: Feedback action and tonic influence of corticosteroids on brain function: a concept arising from the heterogeneity of brain receptor systems. Psychoneuroendocrinology 12 (1987) 83-105. 2. Reul J. M. H. M. and de Kloet E. R.: Two receptor systems for corticosterone in rat brain: microdistribution and differential occupation. Endocrinology 117 (1985) 2505-2511. 3. Reul J. M. H. M. and de Kloet E. R.: Anantomical resolution of two types of corticosterone receptor sites in rat brain with in vitro autoradiography and computer image analysis. J. steroid Biochem. 24 (1986) 269-272. A. C., Okret S., Agnati L. F., 4. Fuxe K., Wikstrom Harfstrand 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 diencephalon using a monoclonal antibody against rat liver glucocorticoid receptor. Endocrinology 117 (1985) 1803-1812. N. E. and Grizzle W. E.: Golden Syrian 5. Dunlap hamsters: A new experimental model for adrenal compensatory hypertrophy. Endocrinology 114 (1984) 1490-1495. 6. Schindler W. J. and Knigge K. M.: Adrenal secretion by the golden hamster. Endocrinology 65 (1959) 739-747. 7. Schindler W. J. and Knigge K. M.: In vitro studies and adrenal steroidogenesis by the golden hamster. Endocri-
nology 65 (1959) 748-765. 8
Whitehouse B. J. and Vinson G. P.: Specific variation in steroid biosynthesis pathways: the formation of cortisol
Biochem. 13 (1980) 1317-1323.
natn. Acad. Sci. U.S.A. 80 (1983) 6056-6060. 22.
23
24
Bohus B. and de Kloet E. R.: Adrenal steroids and extinction behaviour: antagonism by progesterone, deoxycorticosterone and dexamethasone of a specific effect of corticosterone. Life Sci. 28 (198 1) 433-440. Bohus B., de Kloet E. R. and Veldhuis H. D.: Adrenal steroids and behavioural adaptation: relationship to brain corticoid receptors. In Current Topics in Neuroendocrinology, Adrenal Action on Brain (Edited by D. Ganten and D. W. Pfaff). Springer-Verlag, New York (1982) pp. 107-148. Albers H. E., Yogev L., Todd R. B. and Goldman B. D.: Adrenal corticoids in hamsters: role of circadian timing. Am. J. Physiol. 248 (1985) R434-R438.
0022-4731/88 $3.00+0.00 Copyright 0 1988 Pergamon Press plc
CORTICOSTEROID RECEPTOR ANALYSES IN RAT AND HAMSTER BRAINS REVEAL SPECIES SPECIFICITY IN THE TYPE I and TYPE II RECEPTORS W. SUTAN~.O*, J. M. H. M. REUL, J. A. M. VAN EEKELEN and E. R. Rudolf Magnus Institute
for Pharmacology,
Medical Faculty, University Utrecht, The Netherlands
of Utrecht,
DE KLOET
Vondellaan
6, 3521 GD
Summary-h vitro cytosol binding, receptor autoradiography with radiolabelled corticosteroid analogs, and immunocytochemistry with monoclonal antibodies have revealed the presence of two receptor systems for corticosteroids in rat and hamster brains. The type I receptor is found mainly in the hippocampal region, and in the hamster it binds cortisol (F) and corticosterone (B) with similar affinity while in the rat (a species which unlike the hamster secretes solely B) the type I receptor shows high affinity to B and not to F. The type II receptor is more widely distributed in the brain and it binds to F (hamster) or B (rat) with affinity 4-6-fold lower than to the type I. in viva, the hamster type I and II retain F much more than B while those in the rat show the opposite. In conclusion, the present study clearly indicates species-specificity in type I and type II receptor systems in these animals. Furthermore, the type I receptor displays in vivo stringent preference for retention of the animal’s predominantly circulating corticosteroid (F in hamster, in B in rat).
INTRODUCTION
It is now well established that two receptor systems for corticosteroids are present in the brain [l-3] and they are the corticosterone-preferring receptor (type I CR, a subtype of the mineralocorticoid receptor MR) localized almost exclusively in the pyramidal neurones of the hippocampus, the lateral septum and central amygdala, and the classical glucocorticoid receptor (type II, GR) with its widespread but uneven distribution in neurones and glial cells [2-41. Glucocorticoid action via each receptor type may serve different functions. Behavioural and biochemical studies suggest the involvement of type I sites in the regulation of ongoing behaviour, e.g. in the activation and synchronization of daily activities (exploration, food-seeking), and sleep-related events, while the type II site mediates the feedback action of B on stressactivated neural metabolism and circuitry. In the present study we examined the heterogeneity and specificity of corticosteroid receptors in the brain of hamsters, a species secreting both F and B with F as the predominantly circulating glucocorticoid [5-Q.
EXPERIMENTAL
Animals and chemicals
Male Chinese hamsters (Cricetulus griseus) weighing 20-25 g, and male Wistar-derived rats weighing 150-200 g were used throughout this study. They were adrenalectomized (ADX) under ether anaesthesia using a dorsal approach 2-3 days prior to experimentation. Following surgery they were housed
Proceedings of the 8th International Symposium of The Journal of Steroid Biochemistry “Recent Advances in Steroid Biochemistry” (Paris, 24-27 May 1987). *To whom correspondence should be addressed. 417
in Perspex cages (hamsters individually, rats 4-5 per cage) at 23°C under standard light conditions (14 h: 10 light-dark period) and given food and drinking water containing 0.9% NaCl, ad libitum. Blood from nonADX hamsters was collected in the morning, evening and following ether stress, for the determination of plasma F and B levels carried out by Dr Th.J. Benraad and co-workers of the University of Nijmegen, The Netherlands. In vitro cytosolic receptor and transcortin binding assays
ADX animals were perfused intracardially under Nembutal anaesthesia with 0.9% ice-cold saline, prior to which blood was collected to obtain plasma. Hippocampal dissection and preparation of cytosol were as described elsewhere [2,9]. Aliquots of diluted plasma (1:50) or cytosol were incubated with [3H]F or [3H]B each at concentration range 0.5-20.0nM except in the case of [3H]F binding to rat (0.5-40.0 nM), in the absence and presence of RU 28362 (in cytosol assays) for discrimination of Type I and II receptor as described previously [2, lo]. [3H]RU28362 (concentration range 0.5-15.0 nM) was used to measure directly binding constants to type II sites. Cytosolic protein as measured by the methods of Lowry et al.[ 111. In vivo receptor occupancy study To determine the in viva specificity of F and B to type I and type II receptors in hamster and rat, the availability of binding of labelled steroid to hippocampal cytosol was measured in ADX animals given graded doses (0.5-1O,OOO~g/1OOg body wt) of F or B 1 h prior to killing. Receptor measurement was carried out using single-point assays by incubating aliquots of cytosol with 15.0-20.0 nM of the appropriate [3H]-corticoid to determine specific binding to
418
W. SurANTo et al.
the Type I or II receptor as described earlier. The calculation of the remaining available binding sites in treated animals was based on the values measured in ADX control animals. In vitro autoradiography The corticoid binding sites were studied using the in vitro autoradiography [3, 131. The sections were dried and exposed to the [‘HI-sensitive LKB Ultrofilm for i([‘H]RU 28362), 6 ([3H]B, rat) or 12 weeks ([3H]F, hamster); each of the latter two was in the presence of 100 x RU 28362. Non-specific binding in each case was determined by including SOO-fold excess of the appropriate unlabelled steroid. A VIPER image analysis system (Gesotec, F.D.R.) equipped with a Hitachi CCTV camera was used for analysis of the autoradiogram. The image was digitized (512 x 512 pixels) and 256 grey levels could be distinguished.
animals were measured and calculated as described in Experimental. Increasing doses of F or B result in differential occupation of type I and II receptors (Table 2). In the hamster, 50% of type I is occupied by low doses of F (0.4 pg/ 100 g body wt) but higher doses of B (67.5,&lOOg body wt). To occupy 50% of type II site, either F or B must be increased to 0.28 and 0.1 mg/lOO g body wt respectively. In the rat, low doses of B (0.9 &lOO g body wt) or very high doses of F (0.8 mg/lOO g body wt) are required to occupy 50% of type I site. To occupy 50% of type II site, 60.0 pg B or 1.4 mg F/l 00 g body wt is required. Table 2. Type I and Type II receptor occupancies in ADX animals Species
Receptor types
ED,,F
ED,,B
Hamster
Type I Type II Type I Type II
0.4 280.0 775.0 1350.0
61.5 100.0 0.9 60.0
Rat RESULTS
Heterogeneity of corticosteroid binding to hamster and rat brains The apparent binding affinity (K& in nM) and
capacity (B,,,, in fmol/mg protein) shown in Table 1, are calculated from the Scatchard and Woolf plots of the binding data. [3H]Corticoid (F in hamster, B in rat) bind to type I receptor with higher affinity (K,, 0.9 and 1.0 nM, respectively) than it does to type II (Kd 3.0 and 3.9 nM respectively). B also binds to hamster type I (Kd 0.9nM) and II (Kd 0.5 nM) with high affinity while in contrast, F binds to rat type I and II with lower affinity (Kd 2.2 and 20.1 nM respectively) than does B to these receptors. Table 1. Corticosteroid binding constants to hamster and rat hippocampi
Species Hamster Type I Type II Rat Type I Type II
[3H]Cortisol (F) Bmar. &
[3H]Corticosterone B rnax Kd
0.9 3.0
42.3 106.5
0.9 0.5
131.9 61.3
2.2 20.1
15.0 19.6
I.0 3.9
43.5 260.3
Binding constants of [3H]F and of [‘H]B binding to receptors in hamster and rat hippocampal cytosol. The apparent binding affinity (Kd, expressed as nM) and capacity (as fmollmg protein) were calculated from Scatchard [ 141 and Woolf [ 15, 161analyses of the binding data of [jH]F and [)H]B, each in the presence or absence of loo-fold excess RU 28362; and of [3H]RU 28362 (see Experimental). In all binding assays, r,,, ranged from 0.95 to 0.98.
In vivo receptor occupancy
ADX animals were given graded doses of either F or B (s.c. 0.5-1O,OOO~g/1OOg body wt) 1 h prior to sacrifice. The remaining available hippocampal cytosolic receptor sites (types I and II) in these treated
Receptor occupancy of hamster and rat hippocampal cytosol by exogenously administered F or B. ED,, (effective dose of steroid in &lo0 g body wt) is calculated as the dose of F or B required to occupy 50% of the receptor. Binding capacities were determined as described in Experimental.
In vitro autoradiography In the binding of [W]B (rat) and of [3H]F (hamster) each in the presence of RU 28362 to rat and hamster brains, the autoradiograms show clearly the binding of the corticoid to type I site in the dorsal hippocampal cell field. Within the hippocampus, type I is heterogeneously localized with levels as follows. In the rat: CA1 cell field of the dorsal subiculum= CA2=dentate gyrus (DG)>CA3=ventral hippocampus=lateral septum. In the hamster: CA3= CA2=CAl =DG>CA4>ventral hippocampus=lateral septum. The binding to the type II site (using [3H]RU 28362) reveals a different pattern of distribution in the dorsal hippocampus than type I. Type II is mainly present in the hippocampus, cortex, amygdala, thalamus and hypothalamus where particularly high density is found in the lateral septum and anterior hypothalamus. DISCUSSION The present study demonstrates the existence of two receptor systems for corticosteroids in the brains of hamster and rat. The type I receptor system has a species-specificity for the predominantly circulating glucocorticoid, i.e. F in hamster and B in rat, which is best revealed in vivo. The type II receptor specificity in vivo reflects the binding affinity in vitro which
seems to correspond to the actual presence of circulating F and B in hamster or B only in rat. In other words, the rat type II receptor shows a much higher affinity in vitro to B than F (the latter being poorly retained in vivo in this species). The hamster type II,
Species-specificity corticoid receptor Table 3. Corticosteroid binding constants to hamster and rat transcortin B Inax
Species
Ligand
& (nM)
(fmobmg protein)
Hamster
[)H]B
3.7 4.3 5.8 35.1
769.9 1377.2 8531.9 5838.0
PI-V= Rat
[3H]B [‘HIF
Bindin constants of [JH]F and of [3H]B binding to hamster and rat transcortin. The apparent binding affinity (&) and capacity (B,,,) were calculated from Scatchard and Woolf analyses of the binding data. In all binding assays, r,,, ranged from 0.92 to 0.98.
in contrast, does not discriminate between B and F. The concept of glucocorticoid receptor heterogeneity in the rat is a well-known phenomenon [ 10, 17-l 91. It was shown that the rat brain contains binding sites which bind naturally occurring and synthetic glucocorticoids with different specificity. Synthetic steroids such as RU 26988 and RU 28362 with exclusive glucocorticoid properties have since become available to resolve the distinction between the mineralocorticoid-like receptor 120,211, i.e. type I CR, and the classical glucocorticoid receptor, i.e. type II GR [2,3]. The type I receptor in both rat and hamster is localized almost exclusively in the hippo~mpus, as the automdio~ams clearly show. However, there are differences in the mineraloco~icoid-like property of type I receptor in each species. Whereas rat type I binds ALDO with high affinity, as has been shown by our laboratory and others [2,20-211, hamster type I does not bind ALDO very well and the binding of [IH]F to this receptor can only be displaced by high concentrations of ALDO (unpublished observations). In vitro binding studies show that hamster type I binds F and B with similar affinity but the in vivo receptor occupancy studies show a distinct preference for F over B (the reverse is true in the rat). Such striking differences cannot be easily explained by the intrinsic properties of type I receptor since the in vitro binding affinities, although relatively in line with the autoradiography, did not differ much for F and B. Similar affinity (K,, 3-4 nM) was also obseved for binding of F and B to transcortin (see Table 3). Such preference, therefore, may be due to the steroid’s rate of penetration into the target calls and/or the affinity of the respective receptor complex to DNA acceptor sites. It should also be noted that the in vitro Table 4. Ptasma cortisol and corticosterone levels in hamster Condition a.m. p.m. Stress
Cortisol
Corticosterone
0.8 5.5 4.8
0.014 1.550 1.210
Blood samples were taken in the morning (a.m.), evening (p.m.) and following ether stress. Plasma corticoid levels are expressed in ,I& 100 ml plasma.
419
condition of the binding assay may have masked or unmasked receptor sites that do or do not participate in the in vivo retention process [ 191although evidence for this is still lacking. This study presents a puzzling problem for the ligand specificity of type I. In the rat, type I hippocampus and kidney is identical in the in vitro binding profile and primary structure. Type I is a mineraloco~icoid receptor that displays, in the kidney, a stringent specificity for ALDO as agonist. In the brain, ALDO is an antagonist for events in neurotransmission and behaviour triggered specifically by B. In the hamster brain, even the ALDO affinity to type I is strongly reduced and F appears to be the predominantly retained steroid. Clearly, future research based on the known primary structure of the type I receptor (Evans, personal communication) will shed more light on this peculiar specificity-confe~ing mechanism underlying type I function. The type II receptor system represents the classical glucocorticoid receptor and is widely, albeit unevenly, distributed in the brain as results from autoradiography show. Immunocytochemical study also revealed a predominant distribution of immunoreactive staining for type II (type II-ir) in the CAl, CA2 (as well as CA3 and CA4 in hamster but not in rat) and dentate gyrus of the hippocampus as well as in the paraventricular, arcuate and supraoptic nuclei of the hypothalamus. F (in hamster) or B (in rat) binds to type II receptor with affinity 4-&fold lower than it does to type I. In the in vivo receptor occupation studies in the hamster similar doses of F or B (1 .O mg/lOO g body wt given SC to ADX animals) are required to occupy 80-90% of the type II receptor. In contrast, the rat type II is 80-90% occupied when ADX rats are given F at doses 10 times higher than the dose of B. This result confirms the earlier in vitro binding studies in which the rat type II binds F with very low affinity. We conclude, therefore, that the type II specificity is intrinsic to the receptor and corresponds to the animal’s circulating corticosteroid. Thus, the hamster type II, in contrast to the rat type II, does not discriminate between F and B. The present findings may help the understanding of the function of type I and II receptors in these two species. In the rat, type I receptor seems to be involved in synchronization and activation of circadian-driven brain processes [20,23], while type II, in the termination of the stress response (i.e. feedback). The question now arises if a similar function for type I and type II can be postulated in the hamster from the present binding studies. Hamsters secrete both B and F and as our data (see Table 4) and others [24] show, the latter is the primary circulating glucocorticoid at circadian peak and following a stress, indicating that F is the corticoid involved in the feedback action on stressactivated and circadian driven brain mechanism (a supporting role of B, however, cannot be excluded). Previous report from this laboratory 1221showed that very low doses of B were required to saturate the type I receptor and that the function of type I could only be
W. SUTANT~ et al.
420
revealed following ADX and the subsequent replacement of B at pg doses. Type I function was interpreted as being involved in the regulation of on-going behaviour. The present study indicates that the hamster type I displays a similar stringent specificity towards F at low doses as shown in the in vivo uptake experiment. This experiment also showed that similar doses of F and B are required to occupy the type II site. Therefore, it is likely that both steroids are involved in glucocorticoid feedback of behavioural adaptation and neuroendocrine regulation of the stress response (with F, however, being the prevalent one since it circulates in higher concentration). The present study demonstrates the presence of a heterogeneous population of brain corticoid receptors in rat and in hamster, a species which resembles man in having F as the predominant glucocorticoid. The understanding of the corticoid action via type I and II receptor systems is of great importance since glucocorticoids are among the most extensively used therapeutic agents.
in hamster
adrenal
tissue in vitro. J. steroid Biochem. 2
(1971) 307-311.
9. de Kloet
E. R., Wallach G. and McEwen B. S.: Differences in corticosterone and dexamethasone binding to rat brain and pituitary. Endocrinology 96 (1975)
598-609. 10. Philibert D. and Moguilewsky M.: RU 28362, a useful tool for the characterization of glucocorticoid and mineralocorticoid receptors. Proc. 65th Ann. Mtg. Endoc. Sot., San Antonio, Texas, U.S.A. (1983) Abstr. 1018, p. 335. 11. Lowry 0. H., Rosebrough N. J., Farr A. L. and Randall R. J.: Protein measurement with the Folin-phenol reagent. J. biol. Chem. 193 (195 1) 265-275. 12 van Eekelen J. A. M., Kiss J. Z., Westphal H. M. and de Kloet E. R.: lmmunocytochemical study on the regional distribution and intracellular localization of the glucocorticoid receptor in the rat brain. Brain Rex 436 (1987)
120-128. M., 13. Sarrieau A., Vial M., Philibert D., Moguilewsky Dussaillant M., McEwen B. S. and Rostene W.: In vitro binding of tritiated glucocorticoids directly on unfixed rat brain sections. J. steroid Biochem. 20 (1984) 1233-1238. G.: The attractions of proteins for small 14. Scatchard molecules and ions. Proc. N. Y. Acad. Sci. 51 (1949)
660-672.
is a IBRO/UNESCO
Post-Doctoral Fellow. The authors are grateful for the generous gifts of steroids from Roussel-UCLAF (RU 28362), Behring, Hoechst ([‘HI RU 28362) and Organon International (F and
15. Keightley D. D., Fisher R. J. and Cressie N. A. C.: The Woolf plot is more reliable than the Scatchard plot in analysing data from hormone receptor assays. J. steroid
W.
16. Keightley D. D., Fisher R. J. and Cressie N. A. C.: Properties and interpretation of the Woolf and Scatchard plots in analysing data from steroid receptor assay. J. steroid Biochem. 19 (1983) 1407-1412. 17. de Kloet E. R. and McEwen B. S.: Differences between cytosol receptor complexes with corticosterone and dexamethasone in hippocampal tissue from rat brain. Biochim. biophys. Acta 421 (1976) 124-132. receptors in the brain. In 18. Funder J.: Adrenocortical Frontiers in Neuroendocrinology (Edited by W. F. Ganong and L. Martini). Raven Press, New York (1986) pp. 169-189. 19. McEwen B. S., de Kloet E. R. and Rostene W.: Adrenal steroid receptors and actions in the nervous system. Phvsioi. Rev. 66 (1986) 1121-1188. of 20. Beaumont K. and Fenestil D. D.: Characterization rat brain aldosterone receptor reveals high affinity for corticosterone. Endocrinology 113 (1983) 2043-205 1. 21. Krozowski Z. S. and Funder J. W.: Renal mineralocorticoid receptors and hippocompal corticosterone-binding species have identical intrinsic steroid specificity. Proc.
Acknowledgements-WS
REFERENCES 1. de Kloet E. R. and Reul J. M. H. M.: Feedback action and tonic influence of corticosteroids on brain function: a concept arising from the heterogeneity of brain receptor systems. Psychoneuroendocrinology 12 (1987) 83-105. 2. Reul J. M. H. M. and de Kloet E. R.: Two receptor systems for corticosterone in rat brain: microdistribution and differential occupation. Endocrinology 117 (1985) 2505-2511. 3. Reul J. M. H. M. and de Kloet E. R.: Anantomical resolution of two types of corticosterone receptor sites in rat brain with in vitro autoradiography and computer image analysis. J. steroid Biochem. 24 (1986) 269-272. A. C., Okret S., Agnati L. F., 4. Fuxe K., Wikstrom Harfstrand 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 diencephalon using a monoclonal antibody against rat liver glucocorticoid receptor. Endocrinology 117 (1985) 1803-1812. N. E. and Grizzle W. E.: Golden Syrian 5. Dunlap hamsters: A new experimental model for adrenal compensatory hypertrophy. Endocrinology 114 (1984) 1490-1495. 6. Schindler W. J. and Knigge K. M.: Adrenal secretion by the golden hamster. Endocrinology 65 (1959) 739-747. 7. Schindler W. J. and Knigge K. M.: In vitro studies and adrenal steroidogenesis by the golden hamster. Endocri-
nology 65 (1959) 748-765. 8
Whitehouse B. J. and Vinson G. P.: Specific variation in steroid biosynthesis pathways: the formation of cortisol
Biochem. 13 (1980) 1317-1323.
natn. Acad. Sci. U.S.A. 80 (1983) 6056-6060. 22.
23
24
Bohus B. and de Kloet E. R.: Adrenal steroids and extinction behaviour: antagonism by progesterone, deoxycorticosterone and dexamethasone of a specific effect of corticosterone. Life Sci. 28 (198 1) 433-440. Bohus B., de Kloet E. R. and Veldhuis H. D.: Adrenal steroids and behavioural adaptation: relationship to brain corticoid receptors. In Current Topics in Neuroendocrinology, Adrenal Action on Brain (Edited by D. Ganten and D. W. Pfaff). Springer-Verlag, New York (1982) pp. 107-148. Albers H. E., Yogev L., Todd R. B. and Goldman B. D.: Adrenal corticoids in hamsters: role of circadian timing. Am. J. Physiol. 248 (1985) R434-R438.