Ontogenesis of the angiotensin II (ANGII) receptor system in the duck brain

1[)

Det,elopmenla[ Brain Re.sear,';; 7~;( I;)t)4i Jil- i~ , 1994 Elsevier Science B.V. All righls reserved 0165 380¢~/~4 $1~7.IH~

BRESD 51751

Ontogenesis of the angiotensin II (ANGII) receptor system in the duck brain Andreas

R. Miiller

*, R i i d i g e r

Gerstberger

Max-Planck Institut fiir physiologische und klinische Forschung, W.G. Kerckhoff-lnstitut, Parkstrasse 1, 61231 Bad Nauheim, Germany

(Accepted 7 September 1993)

Key words: Pekin duck; Subfornical organ; Anteroventral third ventricular region; AV3V region; Choroid plexus; Paraventricular nucleus;

Limbic system; ANGII; Angiotensin II; Angiotensin receptor

The ontogenetic development of the central nervous angiotensin 1I (ANGII) receptor system in the duck was studied at embryonic days E20 and E27 and at postnatal days P3 and P14 by computerized semiquantitative autoradiography employing the receptor antagonist 125111Sar,SIIeJANGII as radioligand. For circumventricular structures involved in the sensing of brain-intrinsic (AV3V region) or blood-borne (subfornical organ, SFO) ANGII, binding sites for 125I[lSar,SIle]ANGIIwere first detectable at E27, with a steady rise in binding density up to P14. The choroid plexus of the lateral (PCVL) and third (PCVIII) cerebral ventricles responsible for cerebrospinal fluid (CSF) production were endowed with maximal ANGII receptor densities at E20 with subsequent reduction to constant medium (PCVIII) or low (PCVL) values. Besides the choroid plexus, the magnocellular paraventricular nucleus (PVN) was the only structure presenting ANGII specific binding sites at E20, although at low density. As for the SFO and AV3V region, labelling of ANGII binding sites in the PVN increased continuously during development to high values at P14. Nuclear components of the limbic system (archistriatum, amygdala and habenular complex) did not reveal specific labelling by the radioligand at E20 with constant, moderate binding densities evaluated for E27, P3 and P14. In the cluck brain, functionally related structures exhibited a homogenuous ontogenetic development of their ANGII receptor system.

INTRODUCTION Angiotensin II ( A N G I I ) , the biologically active principle of the systemic as well as brain-intrinsic reninangiotensin system (RAS), exerts n u m e r o u s actions on central nervous target sites both inside and outside the b l o o d - b r a i n barrier (BBB) not only in m a m m a l i a n but also avian species 27'29'31. Stimulation of water intake, elevation of systemic blood pressure and antidiuresis represent the major routes that centrally acting A N G I I exerts to replenish extracellular fluid ( E C F ) v o l u m e 31'39. A N G I I - l i k e immunoreactivity in n e u r o n e s of the hypothalamus, as well as the expression of A N G I I converting enzyme in the brain have b e e n amply confirmed in m a m m a l s 14,z°'38. I m m u n o c h e m i c a l studies p e r f o r m e d in the Pekin duck and quail 28, but mainly the p r e s e n c e of A N G I I in the duck third ventricular fluid with its concentration being regulated in an E C F v o l u m e - d e p e n d e n t m a n n e r 11 strongly support the existence of a central R A S also in birds.

* Corresponding author. Fax: (49)(6032) 705-211. SSDI 0165-3806(93)E0158-H

Thirst could be elicited in various avian species by intracranial injection of ANGI116"31, while only for the duck and pigeon data are available with regard to the control of a u t o n o m i c functions 3"7'3~. Intracerebroventricular (i.c.v.) application of A N G I I in the conscious duck led to an increase in systemic blood pressure and a d o s e - d e p e n d e n t release of antidiuretic h o r m o n e ( A D H ) 5. Direct central action on neuronal structures is indicated by the ineffectiveness of systemic A N G I I to elicit either of the two responses in the bird, quite different from the situation in various m a m m a l s 27'31,39. To eliminate excess salt from the ECF, the Pekin duck possesses supraorbital salt secreting glands. Both systemic and i.c.v, application of A N G I I caused an inhibition of stimulated salt gland secretion 5"s. Experiments p e r f o r m e d by Butler and coworkers 2 clearly indicate a neuronal structure outside the BBB as target for the circulating A N G I I . The high density of binding sites specific for radiolabelled A N G I I within the S F O 6"7 and electrophysiological studies using the in vitro brain

11

slice technique 22'3° make this the most likely target for blood-borne ANGII. The exclusive effectiveness of centrally administered ANGII in the vicinity of the anteroventral third ventricular (AV3V) region 5 and the presence of high affinity binding sites for radiolabelled ANGII in this region as well as the ADH-synthesizing paraventricular nucleus (PVN) 6'7 suggest the AV3V region a n d / o r PVN as possible targets within the BBB, with ANGII acting as intrahypothalamic neuromodulator 31. The demonstrated regulation of ANGII receptor densities within the duck SFO 6'23, as well as recently available data concerning the ontogenetic development of the avian ADl21-system and central osmoregulation in general ~5'24 suggest detailed ontogenetic investigations of the central avian RAS. The present study therefore aimed at the development of the ANGII receptive system both inside and ouside the BBB of the Pekin duck brain. MATERIALS AND METHODS

Animals For the synchronization of Pekin duck (Anas platyrhynchos) ontogenetic development, freshly fertilized eggs were stored at room temperature for up to 5 days. Subsequently, incubation was carried out at 37.8°C with 60% (embryonic day El-E23) and at 37.0°C with 80% (embryonic day E24-E28) relative humidity, respectively (BSS300 incubator; Grumbach, Germany). Eggs were rotated automatically five times daily (El-E23) and subjected to 45-min cooling periods (20°C) at 12-h intervals (E10-E23). Hatched ducklings were housed in flocks at 25°C ambient temperature with a 12/12 h dark/light cycle and fed commercial chicken food.

Tissue preparation To determine the distribution as well as the relative density of central nervous binding sites specific for ANGII during various stages of duck ontogenesis, five brains each were rapidely dissected of animals sacrificed by decapitation at embryonic days E20 and E27 as well as 3 (P3) and 14 (P14) days after hatching. The tissues were frozen in hexane (Sigma, Germany) cooled to -60°C and mounted onto cryostat chucks (Tissue-Tek; Reichert-Jung, Germany). Serial coronal sections (20 p.m) were cut at -20°C (Frigocut-2700; Reichert-Jung, Germany), thaw-mounted onto slides coated with 10% poly-L-lysine (Sigma, Germany), dehydrated for 12 h at 4°C in a dessicator and stored at -24°C in sealed boxes.

Receptor autoradiography The ANGII receptor antagonist [1Sar,8IIe]ANGII (SARILE) was employed as ligand for quantitative receptor binding studies with duck brain sections6. The peptide was radiolabelled according to the chloramine-T method, with 10 nmol SARILE (Bissendorff Biochem., Germany) being incubated with 0.5 nmol Na12SI (Amersham, Germany) for 13 s in a final volume of 50 /xl. Monoradioiodinated [125I]SARILE was subsequently separated from unlabelled and diradioiodinated peptide as well as free Na1251 and catalysts by reversed-phase HPLC (C-18 NovaPak 5/zm; Waters-Millipore, Germany). Gradient elution was performed with acetonitrile (20-30%) (Baker, USA) containing 0.1% trifluoroacetic acid at a flow rate of 1.2 ml/min. The specific activity of [12SI]SARILE was determined according to Biirgisser I as 2000-2300 Ci/mmol using isolated rat kidney glomeruli. For semiquantitative receptor autoradiography, coronal sections of duck brains at given ontogenetic stages (E20, E27, P3, PI4) were

incubated with the radioligand as described previously 6. Slides were first preincubated for 20 min at 22°C in 30 mM Tris/HCl buffer, pH 7.40, containing 100 mM NaCl, 30 mM sucrose, 5 mM MgCI 2, 1 mM EGTA, 10 -4 M phenylmethylsulfonylfluoride (PMSF; Sigma, Germany) and 0.1% BSA (RIA grade V; Sigma, Germany). Subsequently, a 40-min incubation was carried out at 22°C with 200 pM [I25I]SARILE in the absence (total binding, TB) or presence (nonspecific binding, NSB) of 10 -6 M unlabelled SARILE. After three 2-rain washes in ice-cold BSA-free Tris/HCl buffer and a dip in cold destilled water, sections were dried under a stream of dry cold air and exposed to radiation-sensitive AgfaScopix XR3 film (AGfAGevaert, Germany) at 4°C for 4-6 weeks, with co-exposure of calibrated 1251-standards (Amersham, Germany). Consecutive serial brain sections counterstained according to Kliiver-Barrera allowed neuroanatomical identificationTM.

Data et~aluation As a semiquantitative measure of ligand-reeeptor interaction, optical densities (OD) in the radiation-sensitive films as induced by y-emission from various brain regions labelled with [~2~I]SARILE were analyzed using a computerized image analysis system (RAG 200; BioCom, France). For each animal and neuroanatomical brain structure of interest, OD values were obtained from 4-8 consecutive brain sections for both TB and NSB each, with subsequent determination of the respective OD values for specific binding (TB--NSB). For comparison between single experiments, OD values were converted into relative optical densities (ROD is defined as OD of neuronal structure divided by OD of reference standard) with the reference standard representing 1.5 nCi/mg protein. Results obtained from measurements in four ontogenetic groups (E20, E27, P3. P14) and eight different brain structures were expressed as means_+ S.E.M. Data were analyzed by the one-way Wilcoxon-Mann-Whitney U-test. P < 0.05 was considered statistically significant.

RESULTS Employing the monoradioiodinated ANGII receptor antagonist [~25I]SARILE with high specific activity as ligand for ANGII specific binding sites within the duck central nervous system, neuronal structures both inside and outside the BBB forming morphological or functional entities could be labelled. Specificity of the radioligand binding was determined in the presence of 10 -6 M unlabelled peptide resulting in low values for optical densities not different from the background in the processed film autoradiograms, as determined via image analysis (Fig. 1). With ANGII playing a major role in the central control of autonomic functions, the ontogenetic development of ANGII receptor expression was determined at embryonic days E20, E27 as well as 3 (P3) and 14 (P14) days after hatching. Semiquantitative receptor autoradiography was performed in three neuroanatomical systems, the circumventricular organs (CVOs) lacking a pronounced BBB including the choroid plexus, the paraventricular nucleus of the hypothalamus and the periventricular zone, as well as structures of the limbic system. Of the third ventricular CVOs with suggested sensory functions for monitoring the composition of the systemic extracellular space, only the SFO, but not the organum vasculosum laminae terminalis (OVLT)

12

Fig. 1. Receptor autoradiogram of izsI[ISar,Slle]ANGll ([I:51JSARILE) binding to transverse brain sections of a 3-day-old duckling (ontogenetic stage P3). Serial tissue sections (20 ~zm) were incubated with 200 pM [IeSl]SARILE in the absence (A) (total binding) or presence (B) (non-specific binding) of 10 6 M unlabelled SARILE. Bar represents 2.3 ram. Number of experiments: n = 5.

showed specific labelling with [~25I]SARILE (Fig. 2). Expression of putative ANGI1 receptors in the SFO proved undetectable at embryonic day E20, with moderate binding density present at embryonic day E27 (Figs. 3,4). After hatching, the concentration of expressed ANGII receptors with ligand binding activity increased steadily to high values as derived from relative optical density measurements (Figs. 3,4). Circumventricular structures exclusively described for nonmammalian vertebrates such as the lateral septal organ or the prominent paraventricular organ could not be labelled with the radioligand for any of the four ontogenetic stages investigated. In contrast, the choroid plexus of the lateral and ventral ventricles (PCVL resp. PCVIII) were heavily endowed with ANGII specific binding sites during rather early phases of embryonic development (E20), with a subsequent decline of labelling densities to constant moderate (PCVIII) or low

Fig. 2. Regional distribution of t25l[iSar,S IIe]ANGII ([t251]SARILE) specific binding sites in the central nervous system of a 3-day-old Pekin duck (ontogenetic stage P3). Receptor autoradiograms were obtained from transverse brain tissue sections (20 /xm) incubated with 200 pM [t2sl]SARILE. The rostrocaudal sequence of autoradiograms reads as follows: A-E. Neuroanatomical abbreviations: A, Archistriatum; AV3V, anteroventral third ventricular region; H, habenular complex; Hp, hippocampus; PCVIII, choroid plexus of the third ventricle; PCVL, choroid plexus of the lateral ventricle; PVN, paraventricular nucleus; SFO, subfornical organ; Tn, nucleus taeniae; VIII, third cerebral ventricle. Bar represents 2 mm (A,B) or 1.5 mm (C-E). Number of experiments: n = 5.

13 (PCVL) values both before and after hatching (Figs. 3,4). Besides the choroid plexus, the magnocellular component of the PVN synthesizing the antidiuretic hormone proved to be the only neuronal structure showing binding sites specifically labelled by [~25I]SARILE in animals 8 day before hatching (E20). The low density of these PVN-located binding sites increased steadily during the following stages of development reaching rather high values for the 2-week-old ducklings (P14) (Figs. 2,4). Comparable to the continuous rise in SFOintrinsic A N G I I receptor density, the anteroventral component of the periventricular brain parenchyma inside the BBB, the AV3V region, exhibited a development-dependent augmentation in ANGII receptor expression with undetectable ligand binding at embryonic day E20 (Figs. 2,4). The major components of the avian limbic system, the archistriatum and the nucleus taeniae, structures homologuous to the mammalian amygdala, as well as the habenular complex were void of specific labelling by the radioligand at embryonic day E20 as indicated (Fig. 4). Revealing an ontogenetic pattern different from those described for the CVOs or the PVN and AV3V region, ANGII receptors could subsequently be marked for all other stages (E27, P3, P14) (Figs. 2,4) at elevated but constant levels. With low to moderate

A

ligand binding in the archistriatum and the nucleus taeniae for E27 and P3, only low ANGII labelling could be observed in the habenular complex. Values remained constant throughout further developmental stages except for the nucleus taeniae, where in three out of five experiments maximal relative optical densities were determined in the 2-week-old animals (Figs. 2,4). The small-sized, dorsally located hippocampal area and the median and lateral septal nuclei, both belonging to the limbic system, could not be labelled specifically. DISCUSSION Several components of the central RAS have been identified in the fetal rat brain during late gestation, including angiotensinogen m R N A and ANGII specific binding sites 929'36, with data concerning the latter reported just recently25'35. At 2 days as well as 2, 4 and 7 weeks of postnatal age, detailed studies revealed three major ontogenetic patterns with regard to ANGII receptor expression in various functional neuroanatomical systems of the rat brain 2s. The present study performed for the central nervous system of the Pekin duck as a representative of the avian class included both pre- and postnatal stages of ontogenesis to follow the distribution and development of ANGII receptive

B

,~ ,,,,

,# - ' %

C

PCV

....

S F O

.......

Fig. 3. Ontogenetic development of the specific radiolabeUing of the subfornical organ (SFO) and the choroid plexus of the lateral ventricle (PCVL) in the duck brain with 200 pM 125111Sar,8IIe]ANGIIin transverse duck brain sections (20 tzm) at embryonicday E20 (A) and E27 (B), as well as at day P3 (C) and P14 (D) after hatching. Bar represents 1 mm. Number of experiments: n = 5.

14

120 100

0

120

AV3V

100

80

80

60

60

40

40

20

20

0

0

x'2°I 8 o ~ 100

I

PVN

120

40 20 0

o

120

120 100

Tn

100

80

80

60

60

E-.<

40

40

U3 I

20

20

L~

{3

~

H*

0

vi

120

120 100

A

100

60

4 20 0 Q_ o3

,Jl

80

60

2

SFO

PCVL

100 80

8O

60

4O 2O 0

40 20 0

E20 E27

P3 P14 Ontogenetic

~20

E27

P3 P14

stage

Fig. 4. Ontogenetic development of 1251[ISar,Slle]ANOll ([1251]SARILE) specific binding sites in the central nervous system of the Pekin duck, employing semiquantification of optical binding densities by computerized image analysis. Ontogenetic stages: E20, embryonic day 20; E27, embryonic day 27; P3, postnatal day 3; P14, postnatal day 14. Structures: A, Archistriatum; AV3V, anteroventral third ventricular region; H, habenular complex; PCVIII, choroid plexus of the third ventricle; PCVL, choroid plexus of the lateral ventricle; PVN, paraventricular nucleus; SFO, subfornical organ; Tn, nucleus

taeniae. Results represent means_+S.E.M. Number of experiments: n = 5. * P < 0.05, older vs. younger ontogenetic stage, U-test.

elements in neuronal structures preferentially involved in autonomic control functions. The use of the ANGII receptor antagonist [125I]SARILE as radioligand for putative central nervous ANGII binding sites in the duck has been validated in biological assay systems such as the control of avian salt gland secretion v''3 and adrenal aldosterone release lz. Three distinct patterns of ANGII receptor densities could be traced within the duck central nervous system. Neuronal regions in approximate vicinity to the third cerebral ventricle (VIII) including the SFO outside the BBB, the AV3V region within the BBB and the PVN comprised the first group of hypothalamic structures revealing a uniform pattern of ANGII receptor development. Being detectable (PVN) or non-detectable (SFO, AV3V) during very early stages (E20) of ontogenesis, ANGII receptor density increased steadily dur-

ing subsequent pre- and postnatal periods in all of these structures involved in the regulation of body fluid homeostasis. The early appearance of ANGII binding sites in the avian PVN synthesizing the antidiuretic hormone AVT might reflect direct angiotensinergic control of AVT synthesis a n d / o r release during the second half of avian prenatal development. This is supported by studies reporting expression of the avian hypothalamic AVT system using in situ hybridization in the chicken during the later period of prenatal development 24. With regard to functionality of the AVT system in the chicken, an osmotic challenge was able to trigger AVT release in vivo at embryonic days El5 or E l 6 as demonstrated in elegant experiments by Klempt and coworkers 15. Additional indirect modulation of PVN neuronal activity and subsequent AVT release in the duckling might be due to ANGII induced activation of neurones within (AV3V) or outside (SFO, OVLT) the BBB. The OVLT, SFO as well as the AV3V region are connected to the PVN of the duck brain comparable to the situation in various mammalian species as indicated in studies employing neuroanatomical tracing techniques 17'26. The present semiquantitative autoradiographic study clearly indicates that in both neuronal structures associated with body fluid homeostasis the maturation of the ANGII receptive system parallels the findings described for the development of PVN-intrinsic A N G I I specific binding sites. With regard to the AV3V region, physiological significance of ANGII receptors is indicated by stimulated water intake in the pigeon and quail after local stereotaxically controlled ANGII injections 3'16. In the duck, ANGII applied into the anteroventral section of the third cerebral ventricle caused release of AVT into the blood stream, possibly via AV3V interaction with subsequent activation of magnocellular PVN neurones 5. Electrophysiological studies revealed that neurones recorded within the AV3V region and the magnocellular portion of the PVN respond to the application of ANGII in vitro, although the number of positive cells proved to be low 22. The duck O V L T closely associated with the AV3V region but outside the BBB did not show positive labelling by [~zSI]SARILE for all ontogenetic stages tested as well as for adult animals and none out of ten neurones investigated electrophysiologically responded to ANGII with an increased firing rate while showing sensitivity to changes in extracellular sodium concentration 26. The duck SFO proved to possess ANGII specific binding sites already before hatching and revealed an age-dependent upregulation of its ANGII receptive system. The marked responsiveness of SFO neurones

15

to ANGII in the adult duck has been demonstrated via electrical recordings in vitro 22'3°. Being directly accessible to blood-borne ANGII, only the induction of drinking and the inhibition of salt gland secretion can be associated with central osmoregulatory actions of blood-borne ANGII 2'5'31'34, whereas possible effects on AVT release and blood pressure control remain debatable 3~. In addition, however, ANGII might act as an excitatory central nervous transmitter to relay information from the SFO to the PVN and vice versa, as proposed for the mammalian brain 2°. With physiological data concerning the maturation of the avian ANGII system in the SFO not being available, recently performed ontogenetic studies in the rat concerning the development of single components of a brain-intrinsic RAS such as the expression and distribution of ANGII binding sites supported the time-course of RAS-induced salt appetite in rat pubs ~9. Ontogenetic studies of ANGII receptor density in the postnatal rat, however, revealed results different from the situation in the duck SFO. In the rat SFO, high concentrations of ANGII binding sites were present from the second postnatal day on, with optical densities remaining either constant 25 or showing even a time-dependent reduction during the following 8 weeks of development 35. Similar observations were reported for the rat PVN 25"35 A rather different ontogenetic pattern for the expression of duck central ANGII specific binding sites could be obtained for the choroid plexus of the lateral as well as third cerebral ventricles. All plexus showed maximal binding of the radioligand already at the earliest embryonic stage under investigation, with subsequent reduction to constant low or medium receptor densities. Semiquantitative receptor autoradiography for the rat brain choroid plexus revealed constant optical densities from the second postnatal day to the 7th week 2s or even through all ontogenetic stages from late fetal gestation (El8) until adulthood 35. MiUan and coworkers 25 reported high ANGII receptor densities, whereas Tsutsumi and Saavedra detected only low densities of binding sites in fetal, 2- and 8-week-old rats, characterized as ANGII receptor subtype AT~ by the use of non-peptidergic antagonists 35'36. With regard to a putative angiotensinergic role in mammalian plexus function, RAS components such as renin, converting enzyme or high affinity ANGII receptors have been identified within rat choroid p l e x u s 29"32'33. An inhibitory role of ANGII on CSF production is indicated by increased cerebrospinal fluid flow during intracerebroventricular perfusion with an inhibitor of angiotensin-converting enzyme 3v and reduced blood flow of the choroid plexus during the intravenous infusion

of ANGI121. The high density of ANGII specific binding sites in all choroid plexus of the rat 25 and duck brain during early stages of ontogenesis might therefore reflect diminished secretion of CSF. More likely it might serve trophic functions, as growth factor characteristics promoting vascularization have been proposed for the octapeptide 4,m. This view is supported by the marked decrease in the density of ANGII target sites in the plexus already during late embryogenesis of the quickly maturing avian brain, whereas receptor downregulation could not be observed for the late maturing rat brain. Structures of the duck limbic system including the archistriatum, nucleus taeniae and the habenula were void of ANGII specific binding sites during early embryogenesis and presented moderate binding densities during all other ontogenetic stages investigated, thereby comprising the third group of neuronal structures with a consistent pattern of ANGII receptor development. Nuclei of the amygdala in the rodent brain, the structure homologuous to the avian archistriatum and nucleus taeniae, showed a similar development of ANGII receptor expression. In rat pups at 2 days of age, ANGII specific binding sites could not be detected 25, while they reached maximal expression during the early postnatal period. The density of the amygdala ANGII receptors characterized as AT 2 receptor subtypes subsequently declined to constant values 25'35. The present results obtained for autonomic and limbic structures in the duck central nervous system demonstrate, that functionally related structures exhibit a homogenuous ontogenetic development of their ANGII receptive system.

REFERENCES 1 Biirgisser, E., Methods for the determination of the specific radioactivity of radioligands, J. Recept. Res., 4 (1984) 357-369. 2 Butler, D.G., Siwanowicz, H. and Puskas, D., A reevaluation of experimental evidence for the hormonal control of avian salt glands. In M.R. Hughes and A. Chadwick (Eds.), Progress in Acian Osmoregulation, Leeds University Press, Leeds, 1989, pp. 127-141. 3 Evered, M.D. and Fitzsimons, J.T., Drinking and changes in blood pressure to angiotensin 11 in the pigeon Columba licia, J. Physiol., 310 (1981) 337-352. 4 Fernandez, L.A., Twickler, J. and Mead, A., Neovascularization produced by angiotensin II, J. Lab. Clin. Med., 105 (1985) 141148. 5 Gerstberger, R., Gray, D.A. and Simon, E., Circulatory and osmoregulatory effects of angiotensin II perfusion of the third ventricle in a bird with salt glands, J. Physiol., 349 (1984) 167-182. 6 Gerstberger, R., Healy, D.P., Hammel, H.T. and Simon, E., Autoradiographic localization and characterization of circumventricular angiotensin II receptors in duck brain, Brain Res., 4(10 (1987) 165-170. 7 Gerstberger, R., Miiller, A.M. and Simon-Oppermann, C., Functional hypothalamic angiotensin lI and catecholamine receptor

8

9

10

11

12

13 14

15

16

17

18

19 20

21

22

23

24

systems inside and outside the blood-brain barrier, I'rog. Brain Res., 92 (1992) 423-433. Gerstberger, R. and Gray, D.A., Fine structure, innervation and functional control of avian salt glands, Int. Ret'. Cytol., 144 (1993) 129-215. Gomez, R.A., Cassis, L., Lynch, K.R., Chevalier, R.L., Wilfong, N., Carey, R.M. and Peach, M.J., Fetal expression of the angiotensinogen gene, Endocrinology, 123 (1988) 2298-23112. Grady, E.F.. Secchi, L.A., Griffin, C.A., Schambelan, M. and Kalinyak, J.E., Expression of AT e receptors in the developing rat fetus, J. Clin. Ira,est., 88 (1991) 921-933. Gray, D.A. and Simon, E., Dehydration and arginine vasotocin and angiotensin 11 in CSF and plasma of Pekin ducks, Am. J. Physiol., 253 (1987) R285-R291. Gray, D.A., Gerstberger, R. and Simon, E., Role of angiotensin II in aldosterone regulation in the Pekin duck, J. Endocrinol., 123 (1989) 445-452. Hammel, H.T. and Maggert, J.E., Nasal salt gland secretion inhibited by angiotensin II, Physiologist, 26 (1983) A58. Imboden, J., Harding, J.W. and Felix, D., Hypothalamic angiotensinergic fibre systems terminate in the hypophysis, Neurosci. Lett., 96 (1989) 42-46. Klempt, M., Ellendorff, F. and Grossmann, R,, Functional maturation of arginine vasotocin secretory responses to osmotic stimulation in the chick embryo and the newborn chicken, Z Endocrinol., 133 (1992) 269-274. Kobayashi, H. and Takei, Y., Mechanism for induction of drinking with special reference to angiotensin II, Comp. Biochem. Physiol. A, 71 (1982) 485-494. Korf, H.-W., Neuronal organization of the avian paraventricular nucleus: intrinsic, afferent and efferent connections, J. Exp. Zool., 232 (1984) 387-395. Kuenzel, W.J. and Masson, M., A stereotaxic atlas of the brain of the chick (Gallus dornesticus), The Johns Hopkins University Press, Baltimore/London, 1988. Leshem, M. and Epstein, A.N., Ontogeny of renin-induced salt appetite in the rat pup, Dev. PsyehobioL, 22 (1989) 437-445. Lind, W.R., Swanson, W.L. and Ganten, D., Organization of angiotensin II immunoreactive cells and fibres in the rat nervous system, Neuroendocrinology, 40 (1985) 2-24. Maktabi, M.A., Heistad, D.D. and Faraci, F.M., Effects of angiotensin II on blood flow to choroid plexus, Am. J. Physiol., 258 (1990) H414-H418. Matsumura, K. and Simon E., Locations and properties of angiotensin II-responsive neurones in the circumventricular region of the duck brain, J. Physiol., 429 (1990) 281-296. Matsumura, K. and Simon, E,, Increase in basal firing rate and sensitivity to angiotensin I] in subfornical organ neurones of ducks adapted to salt water, J. Physiol., 429 (1990) 297-308. Milewski, N.. Ivell, R., Grossmann, R. and Ellendorff, F., Embry-

25

26

27 28

29 30

31

32

33

34

35

36

37

38

39

onal development of arginine vasotocin/mesotociu gene expression in the chicken brain.. J. NeuroendocrinoL, l (1989) 473-484. Millan, M.A., Kiss, A. and Aguilera, G., Developmental changes in brain angiotensin I! receptors in the rat, Peptides. 12 (1991) 723-737. Miiller, A.R., Schiifer, F., Matsumura, K., Schmid, H. and Gerstberger, R., Circumventricular structures connected to the paraventricular nucleus in the duck brain: osmosensitivity and modulation by ANGII or VIP, (submitted). Phillips, M.I., Functions of angiotensin in the central nervous system, Annu. Rec. Physiol., 49 (1987) 413-435. Ramieri, G., An angiotensin-like immunoreactive neuronal system in the avian hypothalamus, Bol. Soc. It. Biol. Sper., 64 (1988) 835-839. Saavedra, J.M., Brain and pituitary angiotensin, Endocr. Ret., 13 (1992) 329-380. Schmid, H.A. and Simon E., Effect of angiotensin II and atrial natriuretic factor on neurons in the subfornical organ of ducks and rats in vitro, Brain Res., 588 (1992) 324-328. Simon, E., Gerstberger, R. and Gray, D.A., Central nervous angiotensin II responsiveness in birds, Prog. Neurobiol., 39 (1992) 179-207. Smeby, R.R. and Husain, A., Angiotensin I and lI lorming enzymes in the central nervous system. In J.P. Buckley and C.M. Ferrario (Eds.), Brain Peptides and Catecholamines in Cardiocascular Regulation, Raven Press, New York, 1987, pp. 301-311. Strittmatter, S.M., Lynch, D.R, and Snyder, S.H., Differential ontogeny of rat brain peptidases: prenatal expression of enkephalin convertase and postnatal development of angiotensinconverting enzyme, Dec. Brain Res., 29 (1986) 207-215. Takei, Y., The role of the subfornical organ in drinking induced by angiotensin in the Japanese quail (Coturnix coturnix japonica), Cell Tissue Res., 185 (1977) 175-181. Tsutsumi, K. and Saavedra, J.M., Characterization and development of angiotensin II receptor subtypes (AT l and AT e) in rat brain, Am. J. PhysioL, 261 (1991) R209-R216. Tsutsumi, K., Viswanathan, M., Str6mberg, C. and Saavedra, J.M., Type-1 and type-2 angiotensin II receptors in fetal rat brain, Eur. Z PharmacoL, 198 (1991) 89-92. Vogh, B.P. and Godman, D.R., Effects of inhibition of angiotensin converting enzyme and carbonic anhydrase on fluid production by ciliary process, choroid plexus and pancreas, J. Ocul. PharmacoL, 5 (1989) 303-311. Whiting, P., Nava, S., Mozley, L., Eatham, H. and Poat, J., Expression of angiotensin converting enzyme mRNA in rat brain, Mol. Brain Res., 11 (1991) 93-96. Wright, J.W. and Harding, J.W., Regulatory role of brain angiotensins in the control of physiological and behavioural responses, Brain Res. Rec., 17 (1992) 227-262,