Neuroscienee Letter~, 142 (1992) 155 158 ~', 1992 Elsevier Scientific Publishers Ireland Ltd. All rights reserved 0304-3940/92/$ 05.00
155
NSL 08814
The distribution of angiotensin II ATI receptor subtype mRNA in the rat brain Bernd B u n n e m a n n ", N a o h a r u Iwai b, Rainer Metzger c, Kjell Fuxe'~ Tadashi Inagami ~ and Detlev Ganten ~ "Department ~['Histology and Neurobiology, Karolinska Institute, Stockhohn ~Sweden), ~'Department ~]Bioehemistl3', L¢mderBilt University School o/ Medicine, Nashville, TN (USA) and' Department of Pharmacology. University ~[' Heidelherg, Heidelberg and Max-DelhrtJck-('enter.[i~r Molecular Medicine. Berl#~-Buch ( FRG) (Received 24 February 1992: Revised version received 29 April 1992: Accepted 4 May 19921
Key words: Angiotensin I1 AT~ receptor subtype; mRNA: Brain; In situ hybridization; Rat The present study demonstrates the existence and regional distribution of angiotensin 11 AT~ receptor subtype mRNA expression in the rat brain by the use of in situ hybridization and RNase protection assay. Substantial expression levels in the brain have only been detected in certain distinct areas, such as the subfornical organ, the parvocellular part of the paraventricular hypothalamic nucleus, and the median preoptic nucleus. The results give further evidence for the involvement of the angiotensin II AT~ receptor subtype in the classical functions of central angiotensin ll, like blood pressure control, body fluid homeostasis and in corticotropin-releasing factor (CRF) secretion.
Ever since the demonstration of renin and angiotensin II (ANG If) in brain tissue, the interest of research has also been focused on the corresponding central ANG II receptors. [1251]ANG lI [4] or ANG II [9] antagonists like [~25I]saralasin were used as ligands, revealing a widespread and heterogeneous distribution of ANG II receptors in the brain. Differences in the ANG II binding pattern in the brain demonstrated by the use of sulfhydryl reducing agents suggested the existence o f A N G II receptor subtypes [4, 16]. The development of non-peptidic, selective and competitive antagonists for the ANG Il receptor subtypes, like DuP 753, gave final evidence for the existence of two subtypes of ANG lI receptors in the brain [5, 8, 13, 14, 19], named the ANG II AT~ and AT, receptor subtypes, lwai et al. recently reported the cloning of the rat ANG II AT~ receptor subtype cDNA [7]. This study reports on the presence of AT~ receptor subtype mRNA in the male rat brain detected by RNase protection assay and on its regional distribution demonstrated by in situ hybridization. Animals. Twelve specific pathogen-free adult male Wistar-Kyoto rats (300 g average b.wt.) on a 12 h light/ dark cycle and free access to food and water were studied
in the experiments. Animals processed for in situ hybridization (n=6) were perfused through the left cardiac ventricle with 200 ml of ice-cold 0.9% saline under sodium pentobarbital (60 mg/kg b.wt.) anaesthesia. The organs were immediately removed, snap-frozen and sectioned at a thickness of 10 ym in a cryostat. The sections were fixed for 15 min in 4% paraformaldehyde in PBS, pH 7.0. Animals processed for RNase protection assays (n=6) were sacrificed by decapitation, the organs removed immediately and frozen in liquid nitrogen. RNA probe synthesis. A 733 basepair EcoRI/Kpnl rat AT 1 receptor subtype cDNA fragment, representing positions 86-819 and subcloned into the vector pGEM 3 (Promega, Madison, WI, USA) was used for probe synthesis. Linearization with SO'I and in vitro transcription with T7 RNA polymerase resulted in a 353-nucleotidelong cRNA, whereas a 380-nucleotide-long mRNA was obtained after transcription with SP 6 RNA polymerase. Radiolabelling was performed with either [35S]c¢-UTP (Dupont NEN, Boston, MA, USA) tk~rin situ hybridization or [32p]~-UTP (Amersham, Amersham, UK) for the RNase protection assays. A 150 bp SaII/XhaI rat fl-actin cDNA subcloned into the Bluescript Sk ~ vector (Stratagene, La Jolla, CA, USA) was used for [32P]cc-UTP labelling via in vitro transcription with T7 RNA polyme-
Correspomh'nce." K. Fuxe, Department of Histology and Neurobiol-
rase.
ogy, Karolinska Institute, Box 60400, S-10401 Stockholm, Sweden. Fax: (46) (8) 337941.
RNase protection assay. Total RNA was extracted from the tissue according to Paul et al. [11] and the
I% RNase protection assays were performed as described by Mullins et al. [10]. hi situ /l.t'bridizalion. In situ hybridization was performed as described previously [2]. Briefly, the sections were deproteinated, acetylated and prehybridized for 2 4 h at 37°C. Hybridization was carried out with 0.1 ng/ section of [35S]~-UTP-labelled AT~ cRNA at 37°C for 16 18 h. Alter washing in 0.5 x SSC/50% formamide at 48°C for 4 h, autoradiography was carried out on HyperfilmJH (Amersham, Amersham, UK) for 3 4 weeks. Specificity was determined by hybridization with [35S]c~UTP-labelled AT~ m R N A at the same specific activity and concentration as the AT~ cRNA. With the RNase protection assay we demonstrated AT~ m R N A in the hypothalamus, the medulla oblongata, and the cerebral cortex, whereas the cerebellum displayed no signal (Fig. 1). A strong in situ hybridization signal was present in the liver, the adrenal cortex and the anterior and intermediate lobes of the pituitary gland, whereas the posterior lobe showed no AT~ expression (Fig. 2). The central ATt receptor gene expression was generally very low and moderate to strong expression was restricted to certain distinct areas: the organum vasculosum of the lamina terminalis (Fig. 3b) and the subfornical organ (Fig. 3c) displayed the highest hybridization signals detected in the brain. High labelling was also present in the parvocellular, but not in the magnocellular part of the paraventricular hypothalamic nucleus and in the periventricular hypothalamic nucleus (Fig. 3d). Likewise, the median preoptic nucleus showed a high signal (data not shown). Moderate to high signals could be detected in the anterior olfactory nucleus and the internal granular layer of the olfactory bulb (Fig. 3a), as well as in the piriform cortex (Fig. 3b). A moderate labelling was present in the nucleus of the lateral olfactory tract and the suprachiasmatic nucleus, extending into the lateroanterior hypothalamic nucleus (Fig. 3c). The labelling demonstrated in the hippocampus was present in all layers, but seemed to be partially unspecific, being also weakly present in the control section (Fig. 3d,e). Weak signals were present in all layers of the cerebral cortex (Fig. 3a-f) reaching also into the amygdalo-hippocampal transition zone (Fig. 3f'). AT~ expression in the medulla was present in the area postrema, the medial part of the nucleus tractus solitarius and the dorsal motor nucleus of the vagus (Fig. 3g). The control hybridizations with the radiolabelled AT~ m R N A showed no labelling above background except in the granular layer of the cerebellum and to a very low extent in the hippocampus. The very strong signal in the granular layer of the cerebellum (Fig. 3g) was unspecific, since it also persisted in control sections (Fig. 3h). This strong unspecific
BR
MO
HY
CO
CE
rAT1 --~ (353bp)
rBAct
--.-
(150bp)
Fig. I. RNase protectionassay t\~rAT~mRNA aflc~rcgttmal dissection of the brain: rAT1, [32P]c~-UTP-labelledrat AI-~cRNA probe: rflAct, [32P]~-UTP-labelledrat fl-actin cRNA probe: MO, medulla oblongata: HY, hypothalamus: CO, cerebral cortex; CE. cerebellum. 100ag total RNA were used per area. labelling might be related to the extremely high density of nerve cells in this area. If unspecific binding occurs in these cells it might be amplified to the present intensity. The present paper gives evidence for the existence of A N G II ATI receptor subtype m R N A in the brain with
Fig. 2. Autoradiogramsof sections hybridizedwith [~sS]a-UTP-labelled AT~ cRNA (a,c,e) and [35S]cc-UTP-labelledAT~ mRNA (b,d,f). (a,b/ liver; (c,d) adrenal gland. ZG, zona glomerulosa: M, medulla; (e,f) pituitary gland. A, anterior lobe; I. intermediatelobe: P, posterior lobe.
157
a
5.2
•
-I,8
b
~
0.2
c
~
-
1
.
3
f
* 6,2
d
Pt~ n
.,
-I._
g
h
-13.8
-13.8
Fig. 3. Autoradiograms of coronal sections from the rat brain hybridized with [3sS]~-UTP-labelled AT~ cRNA and [3sS]~-UTP-labelled AT~ mRNA (e,h). Note the strong labelling in the granular cell layer of the cerebellum (g) still persists in the control section (h). The numbers in the lower left corner of the figures indicate the approximate bregma levels in mm according to Paxinos and Watson [13]. AON. anterior olfaclory nucleus: AHi. amygdalo-hippocampal transition zone; AP, area postrema; IGr, internal granular layer of the olfactory bulb: LA, lateroanterior h>pothalamic nucleus: OVLT, organum vasculosum of the lamina terminalis: PaPP, parvocellular part of the paravcntricular hypothalamic nucleus: Pir, piriform cortex: SCh, suprachiasmatic nucleus: SFO, subfornical organ: Sol/10, nucleus tractus solitarius/dorsal motor nucleus of the vagus complex.
a highly regional distribution pattern. In situ hybridization revealed the regional distribution of the expression of the AT~ receptor in the adrenal gland, the liver, the pituitary, and the brain. The in situ hybridization results, taken together with the RNase protection data and the agreement with biochemical and autoradiographical binding studies assured the specificity of the present method. The available binding data on the adrenal gland [1] are in complete agreement with our findings on the presence of ATI receptor gene expression, as is the strong labelling found in the anterior lobe of the pituitary gland [14]. However, we demonstrated an equally strong signal in the intermediate lobe, where no binding has been reported [17]. The role of AT~ receptor expression in the intermediate lobe of the pituitary gland remains to be investigated. Also in the brain the expression pattern showed a high, although not a complete correlation with binding and distribution studies using subtype specific ligands [5]. Thus, in areas involved in the classical functions of the brain renin-angiotensin system (RAS) we found the dis-
tinct presence of AT~ m R N A at relatively high levels correlated with the presence of binding inhibited by DuP 753 [5]. The present paper provides evidence that the ATI receptor subtype is in fact synthesized in these areas and is therefore possibly located on neuronal cell bodies and their processes. However, in areas not related to these functions (like e.g. the superior colliculus or the medial geniculate body) which contain predominantly the AT 2 receptor subtype as shown by autoradiography demonstrating very low binding of the subtype specific ligand DuP 753 [5, 14], AT~ m R N A could not be detected (Fig. 313. It is postulated that the AT t receptor in these areas might be located on afferents. AT, receptor gene expression was, on the other hand, weakly detected in some areas not closely linked to the classical functions of A N G II, namely the cerebral cortex, the nuclei of the olfactory system and the hippocampus. The functions assigned to the AT I receptor in these areas await demonstration. The regional distribution pattern of AT] receptor mRNA revealed by this study clearly underlines the in-
15~
volvement of the ATj receptor subtype in the ANG 11 mediated effects on blood pressure, body fluid homeostasis and neuroendocrine control. All areas containing AT~ receptor mRNA in high amounts are related to these functions. However, interestingly the AT~ receptor is highly expressed in the parvocellular but not in the magnocellular part of the paraventricular hypothalamic nucleus and also not in the supraoptic nucleus, suggesting that the effects of ANG II on vasopressin and oxytocin release are not directly mediated by the AT~ receptor subtype, since these hormones are mainly located in magnocellular neurons [15]. This finding is supported by binding studies revealing a very low density of ANG II receptors in the magnocellular part of the paraventricular hypothalamic nucleus [3] and a DuP 753 displaceable binding only in the parvocellular part [5]. On the other hand, the A N G II mediated release of CRF seems to be triggered via the AT~ receptor, since CRF is predominantly present in parvocellular neurons [18]. The present study provides conclusive evidence for the presence of ATI receptor mRNA in discrete areas of the rat brain showing a high correlation with binding patterns obtained by AT~ receptor antagonists. Taken together, these data strongly imply a major role of the AT~ receptor subtype in mediating the functions of central ANG II in blood pressure control, body fluid homeostasis and in the control of CRF secretion. This study was supported by grants of the Swedish Medical Research Council (04X-715), the Deutsche Forschungsgemeinschaft (SFB 317) and NIH Grants HL-35821 and HL-35323. 1 Balla, T., Baukal, A.J., Eng, S. and Catt, K,J., Angiotensin II receptor subtypes and biological responses in the adrenal cortex and medulla, Mol. Pharmacol., 40 (1991) 401-406. 2 Bunnemann, B., Fuxe, K., Bjelke, B. and Ganten, D., The brain renin-angiotensin system and its possible involvement in volume transmission. In K. Fuxe and L.F. Agnati (Eds.), Advances in Neuroscience, Vol. 1, Raven, New York, 1991, pp. 131 -158. 3 Castren, E. and Saavedra, J.M., Angiotensin II receptors in paraventricular nucleus, subfornical organ, and pituitary gland of hypophysectomized, adrenalectomized, and vasopressin-deficient rats. Proc. Natl. Acad. Sci. USA, 86 (1989) 725--729. 4 Gehlert. D.R., Gackenheimer, S.L. and Schober, D.A., Angiotensin I I receptor subtypes in rat brain: dithiothreitol inhibits ligand binding to AII-1 and enhances binding to AII-2, Brain Res,, 546 (1991) 161 165.
5 Gehlert~ D.R., Gackenheimei, S.L. and Sch~,,~,cr, 1).
155
NSL 08814
The distribution of angiotensin II ATI receptor subtype mRNA in the rat brain Bernd B u n n e m a n n ", N a o h a r u Iwai b, Rainer Metzger c, Kjell Fuxe'~ Tadashi Inagami ~ and Detlev Ganten ~ "Department ~['Histology and Neurobiology, Karolinska Institute, Stockhohn ~Sweden), ~'Department ~]Bioehemistl3', L¢mderBilt University School o/ Medicine, Nashville, TN (USA) and' Department of Pharmacology. University ~[' Heidelherg, Heidelberg and Max-DelhrtJck-('enter.[i~r Molecular Medicine. Berl#~-Buch ( FRG) (Received 24 February 1992: Revised version received 29 April 1992: Accepted 4 May 19921
Key words: Angiotensin I1 AT~ receptor subtype; mRNA: Brain; In situ hybridization; Rat The present study demonstrates the existence and regional distribution of angiotensin 11 AT~ receptor subtype mRNA expression in the rat brain by the use of in situ hybridization and RNase protection assay. Substantial expression levels in the brain have only been detected in certain distinct areas, such as the subfornical organ, the parvocellular part of the paraventricular hypothalamic nucleus, and the median preoptic nucleus. The results give further evidence for the involvement of the angiotensin II AT~ receptor subtype in the classical functions of central angiotensin ll, like blood pressure control, body fluid homeostasis and in corticotropin-releasing factor (CRF) secretion.
Ever since the demonstration of renin and angiotensin II (ANG If) in brain tissue, the interest of research has also been focused on the corresponding central ANG II receptors. [1251]ANG lI [4] or ANG II [9] antagonists like [~25I]saralasin were used as ligands, revealing a widespread and heterogeneous distribution of ANG II receptors in the brain. Differences in the ANG II binding pattern in the brain demonstrated by the use of sulfhydryl reducing agents suggested the existence o f A N G II receptor subtypes [4, 16]. The development of non-peptidic, selective and competitive antagonists for the ANG Il receptor subtypes, like DuP 753, gave final evidence for the existence of two subtypes of ANG lI receptors in the brain [5, 8, 13, 14, 19], named the ANG II AT~ and AT, receptor subtypes, lwai et al. recently reported the cloning of the rat ANG II AT~ receptor subtype cDNA [7]. This study reports on the presence of AT~ receptor subtype mRNA in the male rat brain detected by RNase protection assay and on its regional distribution demonstrated by in situ hybridization. Animals. Twelve specific pathogen-free adult male Wistar-Kyoto rats (300 g average b.wt.) on a 12 h light/ dark cycle and free access to food and water were studied
in the experiments. Animals processed for in situ hybridization (n=6) were perfused through the left cardiac ventricle with 200 ml of ice-cold 0.9% saline under sodium pentobarbital (60 mg/kg b.wt.) anaesthesia. The organs were immediately removed, snap-frozen and sectioned at a thickness of 10 ym in a cryostat. The sections were fixed for 15 min in 4% paraformaldehyde in PBS, pH 7.0. Animals processed for RNase protection assays (n=6) were sacrificed by decapitation, the organs removed immediately and frozen in liquid nitrogen. RNA probe synthesis. A 733 basepair EcoRI/Kpnl rat AT 1 receptor subtype cDNA fragment, representing positions 86-819 and subcloned into the vector pGEM 3 (Promega, Madison, WI, USA) was used for probe synthesis. Linearization with SO'I and in vitro transcription with T7 RNA polymerase resulted in a 353-nucleotidelong cRNA, whereas a 380-nucleotide-long mRNA was obtained after transcription with SP 6 RNA polymerase. Radiolabelling was performed with either [35S]c¢-UTP (Dupont NEN, Boston, MA, USA) tk~rin situ hybridization or [32p]~-UTP (Amersham, Amersham, UK) for the RNase protection assays. A 150 bp SaII/XhaI rat fl-actin cDNA subcloned into the Bluescript Sk ~ vector (Stratagene, La Jolla, CA, USA) was used for [32P]cc-UTP labelling via in vitro transcription with T7 RNA polyme-
Correspomh'nce." K. Fuxe, Department of Histology and Neurobiol-
rase.
ogy, Karolinska Institute, Box 60400, S-10401 Stockholm, Sweden. Fax: (46) (8) 337941.
RNase protection assay. Total RNA was extracted from the tissue according to Paul et al. [11] and the
I% RNase protection assays were performed as described by Mullins et al. [10]. hi situ /l.t'bridizalion. In situ hybridization was performed as described previously [2]. Briefly, the sections were deproteinated, acetylated and prehybridized for 2 4 h at 37°C. Hybridization was carried out with 0.1 ng/ section of [35S]~-UTP-labelled AT~ cRNA at 37°C for 16 18 h. Alter washing in 0.5 x SSC/50% formamide at 48°C for 4 h, autoradiography was carried out on HyperfilmJH (Amersham, Amersham, UK) for 3 4 weeks. Specificity was determined by hybridization with [35S]c~UTP-labelled AT~ m R N A at the same specific activity and concentration as the AT~ cRNA. With the RNase protection assay we demonstrated AT~ m R N A in the hypothalamus, the medulla oblongata, and the cerebral cortex, whereas the cerebellum displayed no signal (Fig. 1). A strong in situ hybridization signal was present in the liver, the adrenal cortex and the anterior and intermediate lobes of the pituitary gland, whereas the posterior lobe showed no AT~ expression (Fig. 2). The central ATt receptor gene expression was generally very low and moderate to strong expression was restricted to certain distinct areas: the organum vasculosum of the lamina terminalis (Fig. 3b) and the subfornical organ (Fig. 3c) displayed the highest hybridization signals detected in the brain. High labelling was also present in the parvocellular, but not in the magnocellular part of the paraventricular hypothalamic nucleus and in the periventricular hypothalamic nucleus (Fig. 3d). Likewise, the median preoptic nucleus showed a high signal (data not shown). Moderate to high signals could be detected in the anterior olfactory nucleus and the internal granular layer of the olfactory bulb (Fig. 3a), as well as in the piriform cortex (Fig. 3b). A moderate labelling was present in the nucleus of the lateral olfactory tract and the suprachiasmatic nucleus, extending into the lateroanterior hypothalamic nucleus (Fig. 3c). The labelling demonstrated in the hippocampus was present in all layers, but seemed to be partially unspecific, being also weakly present in the control section (Fig. 3d,e). Weak signals were present in all layers of the cerebral cortex (Fig. 3a-f) reaching also into the amygdalo-hippocampal transition zone (Fig. 3f'). AT~ expression in the medulla was present in the area postrema, the medial part of the nucleus tractus solitarius and the dorsal motor nucleus of the vagus (Fig. 3g). The control hybridizations with the radiolabelled AT~ m R N A showed no labelling above background except in the granular layer of the cerebellum and to a very low extent in the hippocampus. The very strong signal in the granular layer of the cerebellum (Fig. 3g) was unspecific, since it also persisted in control sections (Fig. 3h). This strong unspecific
BR
MO
HY
CO
CE
rAT1 --~ (353bp)
rBAct
--.-
(150bp)
Fig. I. RNase protectionassay t\~rAT~mRNA aflc~rcgttmal dissection of the brain: rAT1, [32P]c~-UTP-labelledrat AI-~cRNA probe: rflAct, [32P]~-UTP-labelledrat fl-actin cRNA probe: MO, medulla oblongata: HY, hypothalamus: CO, cerebral cortex; CE. cerebellum. 100ag total RNA were used per area. labelling might be related to the extremely high density of nerve cells in this area. If unspecific binding occurs in these cells it might be amplified to the present intensity. The present paper gives evidence for the existence of A N G II ATI receptor subtype m R N A in the brain with
Fig. 2. Autoradiogramsof sections hybridizedwith [~sS]a-UTP-labelled AT~ cRNA (a,c,e) and [35S]cc-UTP-labelledAT~ mRNA (b,d,f). (a,b/ liver; (c,d) adrenal gland. ZG, zona glomerulosa: M, medulla; (e,f) pituitary gland. A, anterior lobe; I. intermediatelobe: P, posterior lobe.
157
a
5.2
•
-I,8
b
~
0.2
c
~
-
1
.
3
f
* 6,2
d
Pt~ n
.,
-I._
g
h
-13.8
-13.8
Fig. 3. Autoradiograms of coronal sections from the rat brain hybridized with [3sS]~-UTP-labelled AT~ cRNA and [3sS]~-UTP-labelled AT~ mRNA (e,h). Note the strong labelling in the granular cell layer of the cerebellum (g) still persists in the control section (h). The numbers in the lower left corner of the figures indicate the approximate bregma levels in mm according to Paxinos and Watson [13]. AON. anterior olfaclory nucleus: AHi. amygdalo-hippocampal transition zone; AP, area postrema; IGr, internal granular layer of the olfactory bulb: LA, lateroanterior h>pothalamic nucleus: OVLT, organum vasculosum of the lamina terminalis: PaPP, parvocellular part of the paravcntricular hypothalamic nucleus: Pir, piriform cortex: SCh, suprachiasmatic nucleus: SFO, subfornical organ: Sol/10, nucleus tractus solitarius/dorsal motor nucleus of the vagus complex.
a highly regional distribution pattern. In situ hybridization revealed the regional distribution of the expression of the AT~ receptor in the adrenal gland, the liver, the pituitary, and the brain. The in situ hybridization results, taken together with the RNase protection data and the agreement with biochemical and autoradiographical binding studies assured the specificity of the present method. The available binding data on the adrenal gland [1] are in complete agreement with our findings on the presence of ATI receptor gene expression, as is the strong labelling found in the anterior lobe of the pituitary gland [14]. However, we demonstrated an equally strong signal in the intermediate lobe, where no binding has been reported [17]. The role of AT~ receptor expression in the intermediate lobe of the pituitary gland remains to be investigated. Also in the brain the expression pattern showed a high, although not a complete correlation with binding and distribution studies using subtype specific ligands [5]. Thus, in areas involved in the classical functions of the brain renin-angiotensin system (RAS) we found the dis-
tinct presence of AT~ m R N A at relatively high levels correlated with the presence of binding inhibited by DuP 753 [5]. The present paper provides evidence that the ATI receptor subtype is in fact synthesized in these areas and is therefore possibly located on neuronal cell bodies and their processes. However, in areas not related to these functions (like e.g. the superior colliculus or the medial geniculate body) which contain predominantly the AT 2 receptor subtype as shown by autoradiography demonstrating very low binding of the subtype specific ligand DuP 753 [5, 14], AT~ m R N A could not be detected (Fig. 313. It is postulated that the AT t receptor in these areas might be located on afferents. AT, receptor gene expression was, on the other hand, weakly detected in some areas not closely linked to the classical functions of A N G II, namely the cerebral cortex, the nuclei of the olfactory system and the hippocampus. The functions assigned to the AT I receptor in these areas await demonstration. The regional distribution pattern of AT] receptor mRNA revealed by this study clearly underlines the in-
15~
volvement of the ATj receptor subtype in the ANG 11 mediated effects on blood pressure, body fluid homeostasis and neuroendocrine control. All areas containing AT~ receptor mRNA in high amounts are related to these functions. However, interestingly the AT~ receptor is highly expressed in the parvocellular but not in the magnocellular part of the paraventricular hypothalamic nucleus and also not in the supraoptic nucleus, suggesting that the effects of ANG II on vasopressin and oxytocin release are not directly mediated by the AT~ receptor subtype, since these hormones are mainly located in magnocellular neurons [15]. This finding is supported by binding studies revealing a very low density of ANG II receptors in the magnocellular part of the paraventricular hypothalamic nucleus [3] and a DuP 753 displaceable binding only in the parvocellular part [5]. On the other hand, the A N G II mediated release of CRF seems to be triggered via the AT~ receptor, since CRF is predominantly present in parvocellular neurons [18]. The present study provides conclusive evidence for the presence of ATI receptor mRNA in discrete areas of the rat brain showing a high correlation with binding patterns obtained by AT~ receptor antagonists. Taken together, these data strongly imply a major role of the AT~ receptor subtype in mediating the functions of central ANG II in blood pressure control, body fluid homeostasis and in the control of CRF secretion. This study was supported by grants of the Swedish Medical Research Council (04X-715), the Deutsche Forschungsgemeinschaft (SFB 317) and NIH Grants HL-35821 and HL-35323. 1 Balla, T., Baukal, A.J., Eng, S. and Catt, K,J., Angiotensin II receptor subtypes and biological responses in the adrenal cortex and medulla, Mol. Pharmacol., 40 (1991) 401-406. 2 Bunnemann, B., Fuxe, K., Bjelke, B. and Ganten, D., The brain renin-angiotensin system and its possible involvement in volume transmission. In K. Fuxe and L.F. Agnati (Eds.), Advances in Neuroscience, Vol. 1, Raven, New York, 1991, pp. 131 -158. 3 Castren, E. and Saavedra, J.M., Angiotensin II receptors in paraventricular nucleus, subfornical organ, and pituitary gland of hypophysectomized, adrenalectomized, and vasopressin-deficient rats. Proc. Natl. Acad. Sci. USA, 86 (1989) 725--729. 4 Gehlert. D.R., Gackenheimer, S.L. and Schober, D.A., Angiotensin I I receptor subtypes in rat brain: dithiothreitol inhibits ligand binding to AII-1 and enhances binding to AII-2, Brain Res,, 546 (1991) 161 165.
5 Gehlert~ D.R., Gackenheimei, S.L. and Sch~,,~,cr, 1).
