In addition to rapid responses comprising increases in blood pressure, drinking, and stimulation of natriuresis, ANG II induces the expression of transcription factors (TF) in the central nervous system. The ANG II metabolite ANG III (ANG 2–8) has been demonstrated to exert physiological effects similar to those of ANG II. We aimed to determine 1) whether ANG III induces TF expression in the brain, 2) which ANG II (AT) receptor subtype is involved, and 3) whether the two peptides, ANG II and ANG III, differ in their efficacy to stimulate TF expression. ANG II (100 pmol), ANG III (100 pmol), or vehicle was injected into the lateral brain ventricle of conscious rats alone or in combination with the AT1 receptor antagonist losartan (10 nmol), the AT2 receptor antagonist PD-123319 (5 nmol), or the aminopeptidase inhibitor amastatin (10 nmol). Similar to ANG II, ANG III induced the expression of c-Fos, c-Jun, and Krox-24 in four brain regions, subfornical organ, median preoptic area, paraventricular nucleus, and supraoptic nucleus of the hypothalamus, with the same efficacy. This effect was AT1 receptor mediated. Pretreatment with amastatin reduced the expression of TF in response to ANG II, indicating that this expression is partly mediated by ANG III. Interestingly, the AT2 receptor antagonist PD-123319 alone slightly enhanced the expression of c-Fos, c-Jun, and Krox-24 in different populations of neurons of the paraventricular nucleus. These data indicate that different populations of neurons in the paraventricular nucleus are tonically inhibited by AT2 receptors under physiological conditions.
stimulation of brain ANG II type 1 (AT1) receptors by ANG II, the main effector peptide of the renin-angiotensin system (RAS), induces a number of immediate effects comprising a rise in blood pressure, the release of AVP from the pituitary terminals of magnocellular neurons originating in the hypothalamic supraoptic (SON) and paraventricular (PVN) nuclei, sympathetic modulation, and drinking (5). However, ANG II, acting through its central AT1 receptors, also induces a temporally and spatially highly differentiated expression of transcription factors of the AP-1 and Krox families that is restricted to the subfornical organ (SFO), the median preoptic nucleus (MnPO), and the PVN and SON (10, 12, 16, 25). The distribution of c-Fos-positive cells in the SFO depends on the route of ANG II administration. Whereas a stimulation of periventricular AT receptors increases c-Fos mainly in the cells adjacent to the ventricle, intravenous (iv) injection of ANG II, which reaches the brain via the circumventricular organs that lack the blood-brain barrier, results in stimulation of neurons and c-Fos induction in the central parts of the SFO (15, 16, 24). Besides the SFO, regions like the PVN, SON, or MnPO display the same patterns of c-Fos expression to ANG II injected iv or intracerebroventricularly (icv) (15, 16). When ANG II is injected icv, the expression of transcription factors in the SON and PVN can, at least in part, be ascribed to the activation of AT receptors in the SFO and MnPO. Lesions of the SFO or the anteroventricular region of the third ventricle were shown to inhibit Fos-like immunoreactivity in the SON and PVN (24, 31). The half-life of ANG II in brain tissue and cerebrospinal fluid is very short, only a few seconds, because the enzyme aminopeptidase A (APA; glutamyl aminopeptidase, EC 184.108.40.206) catalyzes the degradation of ANG II to ANG III [(des-Asp1)-ANG II] (9). Both ANG II and ANG III bind with similar affinities to the two ANG receptor subtypes, AT1 and AT2. When injected icv or directly into certain brain regions, ANG III produces a cardiovascular response and a vasopressin release almost identical to those produced by ANG II (30, 33). However, ANG III injected centrally is less effective (∼50%) than ANG II in promoting dipsogenic responses (6). Both ANG peptides are equally released in the hypothalamic PVN after hyperosmotic stimulation in conscious rats (18).
A number of findings derived from experiments in which new, more specific aminopeptidase inhibitors or an anticatalytic APA antiserum were used are in line with the previously formed hypothesis that ANG III is the main effector peptide of the RAS in the brain (26, 29, 33). In the present study, we investigated whether ANG III is also involved in the ANG II-induced expression of inducible transcription factors (ITF) in the brain. We compared the expression of c-Fos, c-Jun, and Krox-24 after stimulation of periventricular ANG receptors with ANG II and ANG III in conscious rats and investigated the contribution of ANG receptor subtypes to these effects. For this purpose, rats were pretreated with the selective ANG AT1 and AT2 receptor antagonists losartan and PD-123319, respectively, before ANG II or ANG III injections. The aminopeptidase inhibitor amastatin was used to determine which of the two angiotensin peptides is more effective as an inducer of transcription factor expression in the brain.
Male Wistar rats (280–300 g body wt) were obtained from Charles River (Sulzfeld, Germany). The animals were kept under controlled temperature, humidity, and light-dark period and had free access to food and water.
For icv injections, chronic cannulas were implanted into the lateral brain ventricle under chloral hydrate anesthesia [400 mg/kg body wt intraperitoneally (ip)]. The animals were housed individually and were allowed a 1-wk recovery period after surgery. During this time, rats were handled daily to avoid nonspecific, stress-induced expression of transcription factors on the day of the experiment.
All icv injections were made between 8 and 11 AM in conscious, freely moving rats to avoid the interference of circadian rhythms with transcription factor expression. The injection volume was 5 μl (1 μl substance followed by 4 μl isotonic saline or 5 μl isotonic saline alone in controls).
Rats were assigned to nine groups (n = 5 animals/group). Group 1 received isotonic saline. Group 2 received ANG II (100 pmol). Group 3 received ANG III (100 pmol). Groups 4, 5, 6, and 7 received the AT1 receptor antagonist losartan (10 nmol) or the AT2 receptor antagonist PD 123319 (5 nmol), respectively, alone or followed by ANG III (100 pmol). Groups 8 and 9 received the aminopeptidase inhibitor amastatin (10 nmol) alone or in combination with ANG II (100 pmol), respectively. The inhibitors were injected 10 min before treatment with ANG III or ANG II. The doses of the antagonists were chosen according to the literature (3, 11, 28).
Ninety minutes after the last injection, rats were deeply anesthetized with chloral hydrate (400 mg/kg body wt ip) and perfused intracardially with phosphate-buffered saline followed by 4% parafomaldehyde solution for fixation of the brain tissue. Brains were removed, postfixed overnight in 4% paraformaldehyde, and immersed in 30% sucrose for cryoprotection for 3 days.
Immunohistochemistry was performed on coronal, cryostat-cut, free-floating slices (50 μm). Incubation with the primary antiserum was followed by detection with the conventional avidin-biotin complex (ABC) peroxidase reaction with diaminobenzidine (DAB) as chromogen, as described previously (12). The dilutions of the antibodies were as follows: anti-c-Fos 1:20 000, anti-c-Jun 1:1,000, and anti-Krox-24 1:4,000. The secondary antibody, goat-anti-rabbit IgG (H+L), was used at a dilution of 1:400 according to the manufacturer’s instructions (Vectastain kit, Vector Laboratories, Burlingame, CA). For double immunohistochemistry, slices were first incubated with the anti-peptide antiserum, followed by visualization of the binding as described above with the Vectastain SG substrate as chromogen. This was followed by incubation with the specific antisera against the transcription factors, visualized with DAB as chromogen. The dilutions of the anti-peptide antisera were 1:5,000 for anti-oxytocin and anti-AVP and 1:100 for anti-CRF.
Drugs and antibodies used.
ANG II and amastatin were purchased from Sigma-Aldrich (Taufkirchen, Germany); ANG III was from Bachem (Bubendorf, Switzerland). Losartan was a generous gift from Dr. R. D. Smith (DuPont Merck Pharmaceutical, Wilmington, DE). PD-123319 was a generous gift from Dr. H. Heitsch (Aventis, Frankfurt, Germany). Both peptides and the antagonists were dissolved in physiological saline. The anti-c-Fos and anti-c-Jun rabbit antisera, as well as the anti-CRF antiserum, were from Oncogene (Cambridge, MA). The polyclonal anti-Krox-24 antibody was a generous gift from Rodrigo Bravo (BristolMyersSquibb, Princeton, NJ). The antibodies against AVP and oxytocin were from Chemicon (Hofheim, Germany). The Vectastain ABC kit and the blue Vectastain SG substrate were purchased from Biologo (Kiel, Germany).
Data are means ± SD. Corresponding sections of different brain regions were photographed with a video camera. For the MnPO and the SFO and for the SON and the PVN two and three sections per animal were used, respectively. In the case of the MnPO and the SFO, the whole area of the respective nucleus was counted. In case of the bilateral SON and PVN, each side was counted separately. Stained neurons were counted with a computer program (Leica Qwin, Leica, Bensheim, Germany), which allows direct selection of the area that should be counted and also calculates the size of the chosen area. Thus the obtained counts were comparable. Average numbers and SDs were calculated per area and per group. Statistical analysis was carried out with ANOVA followed by the Bonferroni test. A value of P < 0.05 was accepted as significant.
Stimulation of periventricular ANG receptors with ANG III resulted in a strong expression of c-Fos in neurons of the SFO, MnPO, PVN, and SON. The number of neurons in each of these four areas that expressed c-Fos was similar to the number of c-Fos-positive neurons after icv injection of ANG II (Table 1). ANG III-induced c-Fos expression was mediated by the AT1 receptor subtype, because pretreatment with the selective AT1 receptor antagonist losartan at a dose that had been shown previously to inhibit the expression of c-Fos to ANG II almost completely abolished the expression of c-Fos after icv ANG III (Table 1). Losartan given alone was without effect (Table 1). Pretreatment with the selective AT2 receptor antagonist PD-123319 did not affect ANG III-induced c-Fos expression, indicating that the AT2 receptor was not involved. An interesting finding of the present study is that the AT2 receptor antagonist PD-123319 alone induced c-Fos expression, which was almost exclusively restricted to the magnocellular part of the PVN, with single scattered neurons in the parvocellular part of the nucleus. The c-Fos protein was colocalized with AVP but not with oxytocin or CRF (Fig. 1). Inhibition of ANG II metabolism with the aminopeptidase inhibitor amastatin reduced the ANG II-induced expression of c-Fos by ∼50% in the SFO and the SON and by ∼30% in the MnPO and the PVN. Amastatin alone also slightly increased the expression of c-Fos, which was, however, much lower than c-Fos levels after a combined treatment with amastatin and ANG II but higher than control levels (Table 1).
c-Jun levels tended to be lower than those of c-Fos except in the PVN. The ANG II metabolite ANG III induced the expression of c-Jun in the MnPO, the SFO, and the PVN to a similar extent as ANG II (Table 1). In the SON, almost no expression of c-Jun was detected 90 min after stimulation. This finding is in line with previous results demonstrating that this transcription factor is induced at later time points in the SON than in the other nuclei. Again, the expression of c-Jun after ANG III could be inhibited with the selective AT1 receptor antagonist but not with the inhibitor of AT2 receptors (Table 1). Losartan alone had no effect on c-Jun expression, as reported previously. Similarly to c-Fos expression, PD-123319 alone stimulated expression of c-Jun in the magnocellular part of the PVN. When the aminopeptidase inhibitor amastatin was given before ANG II to inhibit the generation of ANG III from ANG II, the number of cells positively stained for c-Jun was reduced to ∼50% in the SFO, the MnPO, and the PVN. Amastatin alone also increased the expression of c-Jun slightly above basal levels of control animals treated with vehicle (Table 1).
ANG II and ANG III were equipotent in stimulation of Krox-24 expression (Table 1). Compared with c-Fos and c-Jun, Krox- 24 expression was especially high in the SON. Other nuclei showed no differences between the expression of Krox-24 and the AP-1 transcription factors (Table 1). The ANG III-induced Krox-24 expression was mediated by AT1 receptors, because pretreatment with losartan almost completely abolished the expression of this transcription factor in all four brain areas, whereas pretreatment with PD-123319 had no effect (Table 1). Again, the AT2 receptor antagonist alone increased the number of Krox-24-positive neurons. In contrast to the expression patterns of c-Fos and c-Jun, Krox-24 protein was localized mainly in the parvocellular part of the PVN. The colocalization experiments showed that neither vasopressinergic nor oxytocinergic neurons expressed this transcription factor (Fig. 2). Pretreatment with amastatin diminished ANG II-induced Krox-24 by ∼60%, indicating that the increase in Krox-24 expression after ANG II was partly mediated by ANG III (Table 1). Animals treated with amastatin alone displayed slightly enhanced Krox-24 protein compared with vehicle-treated controls.
The present study shows that ANG III is able to stimulate the expression of transcription factors of the AP-1 and Krox families. ANG III-induced transcription factor expression was restricted to the same brain regions as transcription factor expression induced by the main effector of the RAS, ANG II, namely, the SFO, MnPO, PVN, and SON. Our results clearly demonstrate that ANG III increased transcription factor expression on binding to the AT1 receptor subtype. The APA inhibitor amastatin, applied before ANG II, reduced the expression of transcription factors in response to the peptide, indicating that at least part of ANG II-induced transcription factor expression in vivo is mediated by the ANG II metabolite ANG III.
ANG III is generated from ANG II by an APA-catalyzed removal of an NH2-terminal amino acid. Both peptides bind to the AT1 and AT2 receptor subtypes with similar affinities (8). As early as 1971, Blair-West et al. (2) demonstrated that ANG III increases the secretion of aldosterone in a manner equipotent to ANG II. Since then, a number of studies have reported on the effects of ANG III in peripheral tissues and organs. Most of these effects, which are similar to those produced by ANG II, are mediated by the AT1 receptor subtype. Besides its effects in the periphery, ANG III is also generated in the brain and, similarly to ANG II, may act as a neurotransmitter/neuromodulator. Thus ANG III has been demonstrated to be as potent as ANG II in the central regulation of blood pressure and renal functions, as evidenced by an increase in renal blood and urine flows, glomerular filtration rate, or sodium and potassium excretion (4). As mentioned above, stimulation of brain AT1 receptors with ANG III increases the release of vasopressin from pituitary terminals of SON and PVN magnocellular neurons and stimulates drinking (19). Brain ANG III also seems to exert a tonic stimulatory control over arterial blood pressure (20). Our results, along with the findings of others (26, 33), suggest that ANG III and not ANG II is the main effector peptide of the RAS in the brain.
In the present study, ANG II also induced the expression of ITF when its conversion to ANG III was inhibited. We used a commercially available APA inhibitor, amastatin, which has been used in a number of previous studies (28). This inhibitor inhibits APA and aminopeptidase N with equal potencies. The latter enzyme catalyzes the formation of ANG IV from ANG III (22). The ANG IV binding site was recently identified as the enzyme insulin-regulated membrane aminopeptidase (1). ANG IV can also stimulate the expression of ITF in the brain. The brain regions that respond to ANG IV with increased transcription factor expression also contain ANG IV binding sites and are not identical with the areas in which ANG II or ANG III induces ITF expression (23). Therefore, the observed reduction of ITF expression to ANG II after pretreatment with amastatin is caused by lack of ANG III resulting from its decreased generation from ANG II rather than by a perturbation of ANG IV levels.
We report in the present study that, compared with controls, treatment with the AT2 receptor antagonist PD-123319 alone increased c-Fos, c-Jun, and Krox-24 in the PVN. The AT2 receptor is G protein coupled like the AT1 receptor and is predominantly expressed in the brain during fetal development. In most regions of the adult rat brain this receptor subtype is found at very low levels. Although the AT2 receptor appears to be suppressed in the adult brain, a few areas, such as the cerebellum, the superior and inferior olive, and the locus coeruleus display remarkable levels of AT2 receptors (13). The AT2 receptor has been implicated in such diverse processes as wound healing (e.g., it is upregulated in the cortex after brain ischemia; Refs. 14, 32), the regeneration and differentiation of neuronal tissue, as well as apoptosis (7, 17, 21, 27). The results of the present study demonstrate that AT2 receptors exert a tonic inhibitory control on neurons of the hypothalamic PVN. Once this tonic inhibition is removed by blockade of the AT2 receptor, the neurons respond with an expression of transcription factors. It appears that this reaction can only be observed within a very limited range of doses of the AT2 receptor antagonist. In previous experiments, when we used a higher dose of PD-123319 we did not observe any effect of this compound on the expression of transcription factors in the PVN. This was possibly due to the fact that PD-123319 at higher concentrations can also interact with the AT1 receptor subtype. At lower doses, however, presumably only AT2 receptors are selectively blocked and ANG peptides can increasingly interact with AT1 receptors, because the AT2 receptor antagonist leaves the AT1 receptor unopposed and rather exposes it to elevated ANG II levels.
We showed previously (11) that AT1 receptor-mediated vasopressin release and drinking response are potentiated after pretreatment with PD-123319 at a dose similar to that used in the present study. Interestingly, the transcription factors whose expression is upregulated in both the magnocellular and parvocellular parts of the PVN after stimulation of periventricular AT receptors show distinct expression patterns after AT2 receptor blockade. The members of the AP-1 family of transcription factors were detected mainly in magnocellular vasopressinergic neurons, although single scattered c-Fos-positive neurons could also be found in the lower parvocellular part of the PVN close to the third ventricle. Colocalization studies revealed no c-Fos expression in oxytocinergic neurons or CRF-positive neurons. On the other hand, Krox-24 was localized mainly in the upper parvocellular part of the PVN. Again, a few scattered neurons were detected close to the third ventricle in the lower PVN. Krox-24 was not colocalized with AVP or with oxytocin.
The number of AT2 receptors in the PVN is very low. Nevertheless, the AT2 receptor antagonist selectively increased transcription factors in the PVN. Other brain areas, the SFO, the MnPO, and the SON, did not respond to PD-123319 with increased expression of transcription factor, although 1) they also contain AT2 receptors; 2) the ratios of AT1 and AT2 receptors in the PVN, SFO, SON, and MnPO are almost identical; and 3) all brain areas, except for the SON, are localized in close vicinity to the third ventricle and can therefore be equally targeted with the antagonist. Obviously, ANG peptides acting on AT2 receptors selectively inhibit neurosecretory neurons in the PVN but not other types of peptidergic neurons, e.g., in the MnPO.
Oxytocin- and vasopressin-synthesizing neurons are also localized in the SON. However, the SON is localized far from the third ventricle, and PD-123319 applied as a single icv injection could hardly reach these neurons to produce an effective inhibition of AT2 receptors. Therefore, our data do not allow any conclusion regarding the role of AT2 receptors in this brain region. It appears that ANG peptides on activation of AT1 and AT2 receptors can stimulate, but also inhibit, neurosecretory neurons in the PVN to ensure a fine tuning of both rapid responses, including hormonal regulations and central control of blood pressure and volume homeostasis (11) and delayed effects comprising the activation of transcription factors and the regulation of the expression patterns of various inducible genes.
We conclude that the ANG peptides ANG II and ANG III, on binding to AT1 receptors, are able to stimulate the expression of transcription factors in the same brain regions. Our results also indicate that, under physiological conditions, ANG III considerably contributes to the increased activation of transcription factors in response to ANG II. Besides stimulatory effects, which are mediated by AT1 receptors, ANG peptides exert, by binding to AT2 receptors, a tonic inhibitory control on the expression of transcription factors of the AP-1 and Krox families.
Thus it seems that AT2 receptors exert their tonic inhibition on different populations of neurons, and these neurons respond with distinct signals to the discontinuation of the inhibitory control. This finding makes the expression of transcription factors a powerful tool to study tonic inhibitory signaling systems.
This study was supported by a grant of the Deutsche Forschungsgemeinschaft (DFG, SFB 415, project C2) to A. Blume.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2005 the American Physiological Society