Hypothalamic angiotensinergic fibre systems terminate in the neurohypophysis

Oxytocin response to controlled dietary sodium and angiotensin II among healthy individuals

Oxytocin, while classically known for its role in parturition, lactation, and social behavior, also has been implicated in the control of sodium homeostasis in animal models. To improve our understanding of oxytocin physiology in humans, we measured basal oxytocin levels under low- and liberal-dietary-sodium conditions and following a peripheral angiotensin II (ANG II) infusion. Ten healthy individuals underwent a 6-day standardized low-sodium diet and a 6-day liberal-sodium diet. Each diet was followed by a graded ANG II infusion for 30-min sequential intervals at doses of 0.3, 1.0, and 3.0 ng·kg^-1·min^-1. Fasting serum oxytocin was assessed before and after ANG II infusion. Basal oxytocin levels (1,498.5 ± 94.7 vs. 1,663.3 ± 213.9 pg/ml, P = 0.51) did not differ after the low- and liberal-sodium diets. Following the ANG II infusion, ANG II levels and mean arterial pressure significantly increased as expected. In contrast, the ANG II infusion significantly lowered oxytocin levels from 1,498.5 ± 94.7 vs. 1,151.7 ± 118.1 pg/ml ( P < 0.001) on the low-sodium diet and from 1,663.3 ± 213.9 vs. 1,095.2 ± 87.4 pg/ml ( P = 0.03) on the liberal-sodium diet. The percent change in oxytocin following the ANG II infusion did not differ by sodium diet (-25 ± 5% vs. -28 ± 7% low- vs. liberal-sodium conditions, P > 0.99). Dietary sodium intake did not affect circulating oxytocin levels among healthy individuals. Systemic oxytocin levels were significantly suppressed following a peripheral ANG II infusion independent of dietary sodium conditions.

Redox signaling in central neural regulation of cardiovascular function

One of the most prominent concepts to emerge in cardiovascular research over the past decade, especially in areas focused on angiotensin II (AngII), is that reactive oxygen species (ROS) are critical signaling molecules in a wide range of cellular processes. Many of the physiological effects of AngII are mediated by ROS, and alterations in AngII-mediated redox mechanisms are implicated in cardiovascular diseases such as hypertension and atherosclerosis. Although most investigations to date have focused on the vasculature as a key player, the nervous system has recently begun to gain attention in this field. Accumulating evidence suggests that ROS have important effects on central neural mechanisms involved in blood pressure regulation, volume homeostasis, and autonomic function, particularly those that involve AngII signaling. Furthermore, oxidant stress in the central nervous system is implicated in the neuro-dysregulation associated with some forms of hypertension and heart failure. The main objective of this review is to discuss the recent progress and prospects for this new field of central redox signaling in cardiovascular regulation, while also addressing the molecular tools that have spurred it forward.

Osmotic regulation of angiotensin AT1 receptor subtypes in mouse brain

Experiments were performed to study angiotensin (Ang) AT1a and AT1b mRNA expression in mice, including, examination of brain distribution and the effect of salt loading. In situ hybridization (ISH) methods showed that the pattern of mRNA expression was identical for AT1a and AT1b, with cellular labeling in rostral forebrain, hypothalamus and brainstem. Receptor mRNAs were concentrated in brain regions involved in the regulation of electrolyte and cardiovascular balance. Immunocytochemistry with AT1 specific antisera showed a pattern that was consistent with the ISH. Reverse transcriptase-polymerase chain reaction (RT-PCR) of hypothalamus and pituitary verified the presence of both AT1a and AT1b mRNA. Using quantitative ISH, we found that AT1a mRNA expression was significantly increased after 5 days of 2% NaCl consumption in anterior third ventricle (AV3V), paraventricular hypothalamus (PVN) and subfornical organ (SFO), but unchanged in anterior pituitary. There were no significant changes in AT1b mRNA. These results document the utility of ISH coupled with quantitative imaging techniques for the study of subtype specific expression. Using ISH and RT-PCR, we verified that AT1a and AT1b receptors are expressed in mouse brain and pituitary and show a similar pattern of distribution. Salt loading produced a specific increase in AT1a mRNA in osmosensitive regions, suggesting that this receptor subtype is regulated by sodium/osmolar input.

Immunocytochemical localization of angiotensin II receptor subtypes and angiotensin II with monoclonal antibodies in the rat adrenal gland

Angiotensin II (Ang II), a major regulator of cardiovascular function and body fluid homeostasis, mediates its biological actions via two subtypes of G protein-coupled receptors, termed AT(1) and AT(2). The primary goal of this study was to raise monoclonal anti-peptide antibodies specific to angiotensin AT(1)- and AT(2)-receptor subtypes and to Ang II itself and using these monoclonal antibodies to determine the intraadrenal localization of AT(1) and AT(2) receptors and Ang II in male adult rats. Immunocytochemistry unambiguously demonstrates a regional colocalization of Ang II and angiotensin II receptors in the adrenal gland. The novel antibodies localized Ang II and the AT(1) receptors to the zona glomerulosa of the cortex and to the medulla whereas AT(2) receptors were limited to the medulla. The specificity of immunostaining was documented by pre-adsorption of the antibody with the immunogenic peptide. Our data underscore that AT(1) appears to mediate most of the physiological actions of Ang II in adrenal. Western blot analysis of rat adrenal protein extracts using AT(1) antibody showed a predominant 73-kDa band and a weaker 97-kDa immunoreactive band corresponding to glycosylated forms of the AT(1) receptor. Immunostaining with anti-AT(2) yielded one major immunoreactive band of 73-kDa size and one additional fainter band of 120 kDa. These antibodies may prove of value in unraveling the subcellular localization and intracellular effector pathways of AT(1) and AT(2).

A new angiotensinergic system in the CNS of the rat

The use of two different polyclonal, affinity-purified, monospecific antibodies to ANG II (called BODE and BODE 1) revealed dissimilar distribution of ANG II immunoreactivity within the rat central nervous system (CNS). The ANG II-like material detected using BODE was concentrated in the neurosecretory hypothalamic nuclei, in the inner layer of the median eminence and in the posterior lobe of the pituitary. In contrast, the BODE 1 antibody did not stain the hypothalamic-neurohypophysial angiotensinergic system, and the staining pattern was much more broadly distributed throughout the CNS. BODE 1 is the first antibody that can be used to verify the locations of endogenous angiotensin and their receptor sites in the CNS. The diverse distribution of the ANG II-like material detected by the two antibodies provides strong evidence for the existence of at least two different angiotensinergic systems in the CNS.

Morphological and immunocytochemical characterization of electrophysiologically investigated neurons in the PVN of the rat

This study was carried out to characterize angiotensin II (ANG II) sensitive neurons in the hypothalamic paraventricular nucleus (PVN) of the rat. An approach was chosen in which a combination of an electrophysiological, a morphological, and an immunocytochemical method was focused on one single neuron. The cell's reaction to an application of ANG II and its specific antagonist Losartan (Dup753) was investigated using the technique of intracellular recording inside 450-microm-thick brain slices. A final injection of a fluorescent dye labelled the neurons. Optical sections were taken through the marked cells by a confocal laser-scanning microscope and made into a three-dimensional cell model on a computer. One-micrometer thin sections were cut from the thick slice at the level of the electrophysiologically characterized and marked cell body for immunocytochemical tests with different antibodies. Our results show an example of such a neuron inside the PVN excited by ANG II. It was possible to block this excitation with the specific ANG II receptor subtype 1 (AT1) antagonist Losartan. The result indicated that the ANG II reaction was mediated by the AT1 receptor subtype. Immunocytochemical studies show that this ANG II-sensitive neuron contains ANG II but no vasopressin. The combination of the results enables us to gain improved information on interactions of peptidergic systems.

Hypothalamic accessory nuclei and their relation to the angiotensinergic and vasopressinergic systems

The existence and colocalization of angiotensin II- and vasopressin-like immunoreactivity in individual magnocellular cell groups of the hypothalamus has been demonstrated by using immunocytochemical methods. These neurosecretory magnocellular groups consist of the paraventricular nucleus and the supraoptic nucleus, as well as different accessory cell groups. The fibers from the neurons of the accessory nuclei project directly to adjacent blood vessels and do not comigrate with the hypothalamo-neurohypophysial fiber pathway. On the basis of these findings it can be concluded that in the hypothalamus two different angiotensinergic and vasopressinergic neurosecretory systems exist: (1) an intrinsic hypothalamic and (2) a hypothalamo-neurohypophysial system. The distribution of the accessory cell groups in the hypothalamus is shown in a 3D reconstruction which includes the connection of these magnocellular nuclei with the vascular system in this area.

Localization of angiotensin-converting enzyme, angiotensin II, angiotensin II receptor subtypes, and vasopressin in the mouse hypothalamus

The hypothalamic angiotensin II (Ang II) system plays an important role in pituitary hormone release. Little is known about this system in the mouse brain. We studied the distribution of angiotensin-converting-enzyme (ACE), Ang II, Ang II receptor subtypes, and vasopressin in the hypothalamus of adult male mice. Autoradiography of binding of the ACE inhibitor [125I]351A revealed low levels of ACE throughout the hypothalamus. Ang II- and vasopressin-immunoreactive neurons and fibers were detected in the paraventricular, accessory magnocellulary, and supraoptic nuclei, in the retrochiasmatic part of the supraoptic nucleus and in the median eminence. Autoradiography of Ang II receptors was performed using [125I]Sar1-Ang II binding. Ang II receptors were present in the paraventricular, suprachiasmatic, arcuate and dorsomedial nuclei, and in the median eminence. In all areas [125I]Sar1-Ang II binding was displaced by the AT1 receptor antagonist losartan, indicating the presence of AT1 receptors. In the paraventricular nucleus [125I]Sar1-Ang II binding was displaced by Ang II (Ki = 7.6 X 10(-9)) and losartan (Ki = 1.4 X 10(-7)) but also by the AT2 receptor ligand PD 123319 (Ki = 5.0 X 10(-7)). In addition, a low amount of AT2 receptor binding was detected in the paraventricular nucleus using [125I]CGP42112 as radioligand, and the binding was displaced by Ang II (Ki = 2.4 X 10(-9)), CGP42112 (Ki = 7.9 x 10(-10)), and PD123319 (Ki = 2.2 x 10(-7)). ACE, Ang II, and AT1 as well as AT2 receptor subtypes are present in the mouse hypothalamus. Our data are the basis for further studies on the mouse brain Ang II system.

Expression of AT1A and AT1B angiotensin II receptor messenger RNA in forebrain of 2-wk-old rats

The gene expression of angiotensin II receptor subtypes AT1A and AT1B was localized in the forebrain of 2-wk-old rats by in situ hybridization histochemistry and compared with [125I]Sar1-angiotensin II binding patterns. AT1A receptor mRNA was expressed in circumventricular organs, in hypothalamic nuclei like the paraventricular nucleus, in the lateral olfactory tract, in the basolateral amygdaloid and anterior olfactory nuclei, and in the piriform cortex. No AT1B receptor mRNA was detected in these areas. AT1A and AT1B receptor mRNA was detected in the hippocampus, cingulate cortex, and choroid plexus. No forebrain area studied expressed AT1B receptor mRNA exclusively. Most often, a good match for receptor mRNA and binding was found. In some areas, apparent mismatches suggested receptor formation elsewhere (median eminence) or receptor presence in local neuronal circuits (hippocampus, cingulate, and piriform cortex). Our results support the hypothesis that AT1A receptors are involved in the classical central functions of angiotensin II. Both AT1A and AT1B receptors may play roles in cortical and limbic system function, particularly early in development.

Oxytocin response to challenging stimuli in elderly men

The present study was carried out in order to establish possible alterations in oxytocin (OT) secretion with aging. Therefore, we evaluated the OT responses to insulin (0.15 U/kg)-induced hypoglycemia or to the administration of angiotensin II (i.v. infusion for 60 min of successively increasing doses of 4, 8 and 16 ng/kg min; each dose for 20 min) or apomorphine (60 micrograms/kg s.c.) in male subjects aged 22-80 yr and divided into 3 groups by age (group I (n = 9): 22-38 yr; group II (n = 9): 41-60 yr; group III (n = 9): 63-80 yr). Basal OT concentrations were similar in all groups. The OT response during the insulin tolerance test and the administration of ANG II had similar patterns and magnitudes in all groups. The OT response to apomorphine was similar in the two younger groups, with plasma OT levels increased 118% vs. baseline. In contrast, apomorphine was unable to induce a significant OT rise in the oldest group. During apomorphine test plasma OT concentrations were significantly lower in group III than in groups I and II. For the first time in elderly human subjects, these data show normal responsiveness of the OT secretory system to releasing stimuli such as hypoglycemia and ANG II. These findings indicate that in aged men production of OT and capability of responding to challenging stimuli is unchanged. On the other hand, the reduced OT responsiveness to apomorphine in group III might be an expression of the general dopaminergic dysfunction affecting the aging brain.

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

The ontogenetic development of the central nervous angiotensin II (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 125I[1Sar,8Ile]ANGII 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[1Sar,8Ile]ANGII were 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 duck brain, functionally related structures exhibited a homogeneous ontogenetic development of their ANGII receptor system.

Immunohistochemical demonstration of angiotensin II receptors in rat brain by use of an anti-idiotypic antibody

In the present study we investigated the ability of an anti-idiotypic antibody which recognizes angiotensin II (AII) receptors to demonstrate the presence of such receptors under immunohistochemical conditions. The experiments revealed punctate immunoreactive granules on neurons of the nucleus supraopticus and of the nucleus paraventricularis of the hypothalamus. This localization of AII receptors is consistent with the findings obtained using other experimental approaches to the brain renin-angiotensin system. The conclusion of this study is that the applied anti-idiotypic antibody seems to be a reliable tool for mapping AII receptor distribution. The established experimental approaches to AII receptors are thus now supplemented with the possibility of immunohistochemical investigation. Moreover, the possible microscopic analysis of AII receptors on distinct cells will allow studies at an ultrastructural level.

Levels of angiotensin and molecular biology of the tissue renin angiotensin systems

The cloning of renin, angiotensinogen and angiotensin converting enzyme genes have established a widespread presence of these components of the renin-angiotensin system in multiple tissues. New sites of gene expression and peptide products in different tissues has provided strong evidence for the production of angiotensin independently of the endocrine blood borne system. In addition, the cloning of the angiotensin receptor (AT1) gene has confirmed the widespread distribution of angiotensin and suggested new functions for the peptide. This review of various tissues shows the variation in gene expression between tissues and angiotensin levels, and the fragmentary state of our knowledge in this area. As yet we cannot state that the gene expression of the substrates, enzymes and peptide products are involved in a single cell synthesis. This is not so much evidence against a paracrine function for tissue angiotensin, as lack of detailed, accurate intracellular information. The low abundance of renin in brain, spleen, lung and thymus compared to kidney, adrenal, heart, testes, and submandibular gland may suggest that there are both tissue renin-angiotensin systems (RAS) and nonrenin-angiotensin systems (NRAS). The NRAS could function through cleavage of angiotensinogen by serine proteinases such as tonin and cathepsin G to form Ang II directly. Although much angiotensinogen is extracellular and could therefore be a site of synthesis outside of the cell, intracellular angiotensinogen in a NRAS process could produce Ang II intracellularly without requiring extracellular conversion of Ang I to Ang II by ACE. In summary, renin mRNA is found in high concentrations in kidney, adrenal and testes and decreasing lower concentrations in ovary, liver, brain, spleen, lung and thymus. Angiotensinogen mRNA is found in the following tissues in descending order of abundance: liver, fat cells, brain (glial cells), kidney, ovary, adrenal gland, heart, lung, large intestine and stomach. It is debatable whether angiotensinogen and renin mRNA are expressed in blood vessels. The evidence that is lacking for a paracrine function of angiotensin is a complete description of the intracellular molecular synthesis and release of Ang II from single cells of promising tissues. Such tissues, SMG, ovary, testes, adrenal, pituitary and brain (neurons and glia) are potent sources of RAS components for future studies. Although the evidence for a paracrine function of angiotensin II is incomplete, it is an important concept for progressing toward the understanding of tissue peptide physiology and the significance of their gene regulation.

Regulatory role of brain angiotensins in the control of physiological and behavioral responses

Considerable evidence now indicates that a separate and distinct renin-angiotensin system (RAS) is present within the brain. The necessary precursors and enzymes required for the formation and degradation of the biologically active forms of angiotensins have been identified in brain tissues as have angiotensin binding sites. Although this brain RAS appears to be regulated independently from the peripheral RAS, circulating angiotensins do exert a portion of their actions via stimulation of brain angiotensin receptors located in circumventricular organs. These circumventricular organs are located in the proximity of brain ventricles, are richly vascularized and possess a reduced blood-brain barrier thus permitting accessibility by peptides. In this way the brain RAS interacts with other neurotransmitter and neuromodulator systems and contributes to the regulation of blood pressure, body fluid homeostasis, cyclicity of reproductive hormones and sexual behavior, and perhaps plays a role in other functions such as memory acquisition and recall, sensory acuity including pain perception and exploratory behavior. An overactive brain RAS has been identified as one of the factors contributing to the pathogenesis and maintenance of hypertension in the spontaneously hypertensive rat (SHR) model of human essential hypertension. Oral treatment with angiotensin-converting enzyme inhibitors, which interfere with the formation of angiotensin II, prevents the development of hypertension in young SHR by acting, at least in part, upon the brain RAS. Delivery of converting enzyme inhibitors or specific angiotensin receptor antagonists into the brain significantly reduces blood pressure in adult SHR. Thus, if the SHR is an appropriate model of human essential hypertension (there is controversy concerning its usefulness), the potential contribution of the brain RAS to this dysfunction must be considered during the development of future antihypertensive compounds.

Angiotensin peptides in brain and pituitary of rat and sheep

Several lines of evidence indicate that angiotensin peptides may be formed in the brain, where angiotensin II (Ang II) and angiotensin-(1-7) (Ang-(1-7)) may function as neurotransmitters. However, there is considerable controversy concerning the identity and levels of angiotensin peptides in the brain. We have used a novel high performance liquid chromatography-based radioimmunoassay to measure Ang-(1-7), Ang II, Ang-(1-9) and Ang I in various brain regions and in the pituitary of the rat and sheep. We also studied the effect of different methods of tissue extraction, and the effect of the converting enzyme inhibitor ramipril, on angiotensin peptide levels in the rat hypothalamus. The levels of Ang-(1-7), Ang II, Ang-(1-9) and Ang I were low (<25 fmol/g) in all brain regions examined, except for the sheep median eminence and cerebellar cortex where Ang II levels were 385±116 and 193±37 fmol/g (mean ± SEM, n = 6), respectively. Pituitary Ang II levels were 103±13 fmol/g in the rat and 63±18 fmol/g in the sheep. The levels of Ang-(1-7), Ang-(1-9) and Ang I were much lower than those of Ang II in brain and pituitary. Ang-(1-7) levels in the rat hypothalamus were low (<6 fmol/g) but methods of extraction which involved freezing and thawing of the tissue resulted in substantially higher levels of this peptide. Ang II levels in the rat hypothalamus (18±3 fmol/g) were reduced to undetectable levels (<6 fmol/g) by ramipril administration. The low levels of angiotensin peptides in the hypothalamus and brainstem indicate that if these peptides function as neurotransmitters in these regions, then they are of particularly low abundance. Moreover, our results indicate that the high levels of Ang-(1-7) reported previously for rat hypothalamus may be artefactual, due to the method of tissue extraction.

Mutant vasopressin precursor producing cells of the homozygous Brattleboro rat as a model for co-expression of neuropeptides

The homozygous Brattleboro rat (di/di) synthesizes a VP precursor with an abnormal C terminus, which is not transported from the rough endoplasmic reticulum to the Golgi apparatus. In addition, the phenotypic expression of co-existing peptides is differentially disturbed. Ang II and 7B2 are two of the peptides which are not detectable, whereas other peptides (e.g. galanin) are clearly expressed in mutant VP cells. During postnatal life a small but increasing number of solitary post-mitotic VP neurons of the di/di rat undergoes a switch to a heterozygous phenotype. At the same time Ang II and 7B2 show up again in these heterozygous cells, which suggests that for the expression of 7B2, but not for that of other peptides (e.g. galanin), a normal VP precursor is required. A possible underlying mechanism (i.e. the existence of several domains on the endoplasmic reticulum involved in the translocation of sets of neuropeptides) for this differential phenotypic expression of co-existing peptides is discussed.

An immunocytochemical comparison of the angiotensin and vasopressin hypothalamo-neurohypophysial systems in normotensive rats

In the present study we investigated the possibility that angiotensin II/III and vasopressin coexist in the hypothalamo-neurohypophysial pathway. For our experiments 8-week-old male rats not treated with colchicine were used. The anatomical orientation of the entire pathway for angiotensin and vasopressin was facilitated by examining a series of subsequent coronal, horizontal and sagittal sections. Arching fibre tracts are formed mainly by projections emanating from cell bodies in the paraventricular nucleus, the accessory magnocellular nuclei, the supraoptic nucleus and the retrochiasmatic part of the supraoptic nucleus. The majority extend as far as the median eminence and the neurohypophysis, where major terminal fields exist. However, there is a difference between the staining pattern within the suprachiasmatic nucleus and the hypophysis. The results clearly show the colocalization of angiotensin and vasopressin in neurones as well as in fibres of the hypothalamo-neurohypophysial system.

Vasopressin and angiotensin II are absent but spontaneously reappear in solitary hypothalamic neurons of the homozygous Brattleboro rat

The homozygous Brattleboro rat (di/di) synthesizes a vasopressin (VP) precursor with a different C-terminus, which is not packaged in granules. In addition, the expression of a coexisting peptide, angiotensin II (Ang II), is disturbed. During postnatal life a small but increasing number of solitary post-mitotic hypothalamic neurons of the di/di rat undergoes a switch to a genuine heterozygous phenotype. Here we report the reappearance of Ang II in these heterozygous cells, which suggests that for the expression of Ang II a normal VP precursor is required. Based upon the present study and literature data it is proposed that at the level of the endoplasmic reticulum a compartmentalization of the synthesis of various peptide precursor occurs.

Neurophysiological responses to angiotensin-(1-7)

The aim of this study was to investigate the action of the heptapeptide angiotensin-(1-7) on the spontaneous activity of paraventricular neurons using microiontophoresis. Recent immunocytochemical investigations have shown that this product of angiotensin I is predominantly located in cells and fibers of the forebrain and brain stem. Our results show that most neurons in the paraventricular nucleus are excited by angiotensin-(1-7) at a dose of 50-80 nA. In comparison with angiotensin II or angiotensin III, the onset of response and the occurrence of the maximal effect were significantly delayed. With higher doses of angiotensin-(1-7), there was a decrease in latency and a dose-dependent increase in firing frequency. Of all the angiotensin compounds tested, angiotensin III was the most potent. Preliminary results obtained with an angiotensin antagonist show that the action of angiotensin II, angiotensin III, and angiotensin-(1-7) is blocked by the angiotensin receptor subtype 2 antagonist CGP 42112A. Because the angiotensin-(1-7) system in the brain is associated with central vasopressinergic pathways, vasopressin was tested in a similar way. Neurons in the paraventricular nucleus that were excited by iontophoretically applied angiotensins showed a weak response to vasopressin. Occasionally, a small excitatory action was observed. Our results support the hypothesis that the heptapeptide angiotensin-(1-7) is a biologically active neuropeptide. The data also suggest that amino terminal fragments of angiotensin II are not inactive degradation products.

Immunocytochemical Localization of Angiotensinogen and Angiotensin II in the Rat Pituitary

Abstract The presence and distribution of angiotensinogen and angiotensin II (All) were demonstrated in rat pituitary by immunocytochemical staining with the avidin-biotin-peroxidase method, using primary polyclonal antibodies specific for angiotensinogen and All. Silver enhancement of the reaction product was used to intensify lightly stained areas. Attempts were made to identify immunopositive cells by colocalization studies with antisera against luteinizing hormone, prolactin and S-100, a glial cell protein. In the anterior pituitary, angiotensinogen-immunoreactivity was observed in cells lining follicle-like structures. These cells, which were irregularly shaped and had processes extending between the glandular cells, did not colocalize with any of the reference antisera and are therefore of unknown cell type. The follicular endothelium was also immunopositive for angiotensinogen. After silver intensification, dispersed immunoreactive glandular cells were consistently observed in the anterior lobe. A proportion of these costained for luteinizing hormone, but not prolactin or S-100, indicating their identity as gonadotrophs. In the posterior pituitary, angiotensinogen immunostaining was associated only with the vasculature, while groups of immunopositive cells were observed in the medial region of the intermediate lobe after silver enhancement. All-immunoreactivity was observed in large cells preferentially located at the poles of the anterior pituitary which also costained for luteinizing hormone. No staining was observed in either the posterior or intermediate lobes. The presence of immunoreactive angiotensinogen in all three lobes of the pituitary suggests that there are sites, in addition to gonadotrophs, at which the intracellular production of All could occur.