Bing S. Huang,* Sara Ahmadi,* Monir Ahmad,* Roselyn A. White, and Frans H. H. Leenen

Transcription

1 Am J Physiol Heart Circ Physiol 299: H422 H430, First published May 28, 2010; doi: /ajpheart Central neuronal activation and pressor responses induced by circulating ANG II: role of the brain aldosterone- ouabain pathway Bing S. Huang,* Sara Ahmadi,* Monir Ahmad,* Roselyn A. White, and Frans H. H. Leenen Hypertension Unit, University of Ottawa Heart Institute, Ottawa, Ontario, Canada Submitted 12 March 2010; accepted in final form 27 May 2010 Huang BS, Ahmadi S, Ahmad M, White RA, Leenen FH. Central neuronal activation and pressor responses induced by circulating angiotensin II: role of the brain aldosterone- ouabain pathway. Am J Physiol Heart Circ Physiol 299: H422 H430, First published May 28, 2010; doi: /ajpheart An increase in plasma ANG II causes neuronal activation in hypothalamic nuclei and a slow pressor response, presumably by increasing sympathetic drive. We evaluated whether the activation of a neuromodulatory pathway, involving aldosterone and ouabain, is involved in these responses. In Wistar rats, the subcutaneous infusion of ANG II at 150 and 500 ng kg min gradually increased blood pressure up to 60 mmhg at the highest dose. ANG II at 500 ng kg min increased plasma ANG II by 4-fold, plasma aldosterone by 25-fold, and hypothalamic aldosterone by 3-fold. The intracerebroventricular infusion of an aldosterone synthase (AS) inhibitor prevented the ANG II-induced increase in hypothalamic aldosterone without affecting the increase in plasma aldosterone. Neuronal activity, as assessed by Fra-like immunoreactivity, increased transiently in the subfornical organ (SFO) but progressively in the paraventricular nucleus (PVN) and supraoptic nucleus (SON). The central infusion of the AS inhibitor or a mineralocorticoid receptor blocker markedly attenuated the ANG II-induced neuronal activation in the PVN but not in the SON. Pressor responses to ANG II at 150 ng kg min were abolished by an intracerebroventricular infusion of the AS inhibitor. Pressor responses to ANG II at 500 ng kg min were attenuated by the central infusion of the AS inhibitor or the mineralocorticoid receptor blocker by 70 80% and by Digibind (to bind ouabain ) by 50%. These results suggest a novel central nervous system mechanism for the ANG II-induced slow pressor response, i.e., circulating ANG II activates the SFO, leading to the direct activation of the PVN and SON, and, in addition, via aldosterone-dependent amplifying mechanisms, causes sustained activation of the PVN and thereby hypertension. Fra-like immunoreactivity; aldosterone synthase inhibitor; eplerenone; telemetry; angiotensin II CHRONIC INCREASES in circulating ANG II lead to hypertension by direct arterial vasoconstriction and a gradually developing neurogenic response (5, 30), possibly related to sympathoexcitation. The subcutaneous infusion of ANG II for 2 wk significantly increased both blood pressure (BP) and resting splanchnic sympathetic nerve activity in rats (37). The chronic subcutaneous infusion of ANG II causes rapid and marked neuronal activation in circumventricular organs such as the subfornical organ (SFO) as well as the nucleus of the solitary tract (NTS) that diminishes over time, whereas a marked and sustained activation occurs in the paraventricular nucleus (PVN) and supraoptic nucleus (SON) (4). Davern and Head (4) * B. S. Huang, S. Ahmadi, and M. Ahmad contributed equally to this work. Address for reprint requests and other correspondence: F. H. H. Leenen, Hypertension Unit, Univ. of Ottawa Heart Institute, H3228, 40 Ruskin St., Ottawa, ON, Canada K1Y 4W7 ( fleenen@ottawaheart.ca). hypothesized that ANG II-induced neuronal input from the SFO to the PVN and SON diminishes over time and that an undetermined mechanism amplifies and maintains neuronal excitation in these nuclei, leading to sustained activation and sympathoexcitation. Plasma ANG II increases markedly after myocardial infarction (MI), followed by a modest increase during the chronic phase (33). After MI, neuronal activity shows a pattern similar to that caused by the chronic infusion of ANG II (4), i.e., transient increases in the SFO and NTS and progressive and persistent increases in the PVN and SON (26, 36, 44, 47). The oral administration of the mineralocorticoid receptor (MR) blocker eplerenone attenuates neuronal activation in the PVN by 50% in rats after MI (26). Central blockade of aldosterone synthase (AS) (22), MR (6, 20), or ouabain-like compounds ( ouabain ) (24, 32) prevents sympathetic hyperactivity after MI. We hypothesized that after MI, circulating ANG II causes neuronal activation in the SFO and thereby activates angiotensinergic pathways to the PVN and, in addition, causes the chronic activation of a neuromodulatory mechanism starting with aldosterone, which amplifies neuronal activation in the PVN and central sympathoexcitatory pathways (31). The present study was designed to examine the role of this putative central amplification pathway in persistent neuronal activation and the BP response to circulating ANG II in Wistar rats. For this, we first assessed the effects of a chronic subcutaneous infusion of ANG II at various rates on aldosterone levels in plasma and the hypothalamus and the pattern of neuronal activation in brain nuclei, as assessed by Fra-like immunoreactivity, as well as on BP and heart rate (HR), using telemetry. Second, we examined the effects of a chronic intracerebroventricular infusion of an AS inhibitor on hypothalamic and plasma aldosterone as well as the effects of a central infusion of an AS inhibitor, MR blocker, or Digibind (to bind ouabain ) on ANG II-induced neuronal activation and hypertension. METHODS Male Wistar rats weighing g were purchased from Charles River (Montreal, QC, Canada). All rats were housed in a climatized room on a 12:12-h light-dark cycle at constant room temperature and humidity and given standard laboratory chow (120 mol Na /g) and tap water ad libitum. This study was carried out in accordance with the guidelines of the Canadian Council on Animal Care, which conform to National Institutes of Health guidelines, and was approved by the University of Ottawa Animal Care Committee. All surgeries were performed under isoflurane inhalation. Several types of osmotic minipumps were used. Alzet models 2001D (8 l/h, 1 effective day), 2001 (1 l/h, 7 effective days), and 2002 (0.5 l/h, 14 effective days) were used for subcutaneous infusions over various periods, whereas Alzet model 2004 (0.25 l/h, 28 H /10 Copyright 2010 the American Physiological Society

2 H423 effective days) was used for all intracerebroventricular infusions as well as for subcutaneous infusion of the AS inhibitor. Protocols Protocol I: dose-response experiment. A telemetry probe (model TA11PA-C40, DSI) was placed into the abdominal cavity and secured to the ventral abdominal wall with the catheter inserted into the abdominal aorta. The telemetry signal was obtained using an analog adapter and data-acquisition system, which was set to calculate and store the mean values of resting BP and HR with a 3-s interval over a 1-min period each hour. Continuous recordings of BP and HR during the day (7 AM 7 PM) and the night (7 PM 7 AM) were started 3 days after the probe implantation. After 3 days of control recordings of BP and HR, osmotic minipumps were implanted subcutaneously for a 2-wk infusion of vehicle (0.9% saline) or ANG II at 10, 50, 150, or 500 ng kg min at 0.5 l/h. Protocol II: neuronal activation by ANG II. ANG II- induced neuronal activation in various brain nuclei was assessed by Fra-like immunoreactivity (2). First, Fra expression was examined in rats infused with subcutaneous ANG II (150 or 500 ng kg min )or vehicle (0.9% saline) for 1, 4, or 14 days or without any treatment. In a second experiment, in four groups of rats, intracerebroventricular cannulae and osmotic minipumps were implanted for the chronic intracerebroventricular infusion of the AS inhibitor FAD-286 [100 g kg day dissolved in artificial cerebrospinal fluid (acsf); group 1], the MR antagonist spironolactone (10 g kg day dissolved in acsf with 2% ethanol; group 2), and the vehicle for groups 1 and 2, respectively at an infusion rate of 0.25 l/h (groups 3 and 4, respectively). Two days after the start of the intracerebroventricular infusions, a subcutaneous infusion of ANG II at 500 ng kg min in groups 1 3 and vehicle (0.9% NaCl) in group 4 was added for 14 days. FAD-286 hydrogen tartrate (Novartis Institutes for BioMedical Research) was used because it is soluble in acsf. Each 1.67 mg of FAD-286 hydrogen tartrate provides 1 mg of FAD-286 free base, and the amount of the drug in the pumps was adjusted accordingly. The infusion rates of FAD-286 and spironolactone were based on previous studies (6, 22). Neuronal activation was assessed by Fra expression using an antibody that recognizes all known members of the Fos family (Fos, Fos-B, Fra-1, and Fra-2), as described in previous studies (3, 38) and in the Supplemental Material. 1 Protocol III: effects of central blockades on ANG II-induced BP and aldosterone responses. Chronic intracerebroventricular infusions of various antagonists, or their vehicles as controls, followed by the subcutaneous infusion of ANG II were performed in rats at a time. In each set of rats, two to three rats with an intracerebroventricular infusion of acsf and a subcutaneous infusion of ANG II were included as positive controls. Generally, after 3 days of telemetry recording, an intracerebroventricular cannula and minipump were implanted for the intracerebroventricular infusion of FAD-286 (100 g kg day dissolved in acsf), MR blockers eplerenone (10 or 20 g kg day in acsf with 4% acetonitrile) or spironolactone (10 g kg day in acsf with 2% ethanol), or digoxin-specific Fab fragments (Digibind, 800 g kg day in acsf). In an additional group, FAD-286 at the same rate was infused subcutaneously. The doses for the intracerebroventricular infusion of eplerenone were extrapolated from studies using the oral administration of eplerenone (26) and oral versus intracerebroventricular administration of spironolactone (6, 29). Digibind binds not only to digoxin but also to ouabain and ouabain with high affinity (18, 40). For the infusion of Digibind, the content of one vial of Digibind (38 mg/vial in lyophilized form) was dissolved in ml acsf, and the osmotic minipump was filled with the solution. The dose for the intracerebroventricular 1 Supplemental Material for this article is available online at the American Journal of Physiology-Heart and Circulatory Physiology website. infusion of Digibind was based on our previous studies (19, 32) showing that an intracerebroventricular infusion of Digibind at the same dose as that used in the present study prevents hypertension and/or sympathetic hyperactivity in Dahl salt-sensitive rats on a high-salt diet or in Wistar rats after MI. Nonspecific sheep IgG Fab fragments (NS-Fab, Rockland, 800 g kg day in acsf) were infused intracerebroventricularly as a control for Digibind. Two days after the start of these intracerebroventricular or subcutaneous infusions, osmotic minipumps were implanted subcutaneously for a 2-wk infusion of ANG II at 150 or 500 ng kg min. BP and HR recordings continued for 14 days after the start of the ANG II infusion. Rats were then killed by decapitation early in the morning, and trunk blood was collected for measurements of ANGs and/or aldosterone and corticosterone. Plasma and the whole brain were frozen and stored at 80 C until analysis. The brain was later dissected for measurements of aldosterone and corticosterone content in the hypothalamus and hippocampus. Protocol IV: subcutaneous infusion of aldosterone and plasma and brain aldosterone. In three groups of Wistar rats, osmotic minipumps were implanted for the subcutaneous infusion of D-aldosterone (Sigma, St. Louis, MO) at 0.5 or 2.5 g/h or vehicle (0.9% NaCl with 1% ethanol) for 2 wk. At the end of the infusion period, rats were decapitated early in the morning, and trunk blood and the whole brain collected for aldosterone and corticosterone assays. Biochemical Assays The whole hypothalamus and hippocampus were dissected according to Glowinski and Iversen (10). Plasma and brain aldosterone were measured by RIA as previously described (28). For the standard curve of the aldosterone assay, the lowest detectable value was 0.5 pg/tube. Since the average weight of hypothalamic tissue was mg and the amount in each RIA tube was 4 mg, the sensitivity for hypothalamic aldosterone was 0.5/ pg/mg. Observed levels of hypothalamic aldosterone ( 0.10 pg/mg) were higher than this level. Plasma and tissue corticosterone were determined using a corticosterone 125 I RIA kit (product , MP Biomedicals). Plasma ANG I and II concentrations were measured by RIA after extraction on C18 Sep-Pak cartridges and separation by HPLC, as previously described (34). Statistical Analysis Average mean arterial pressure (MAP) and HR over 24 h as well as MAP differences between the night and day were used for comparisons. Two-way repeated-measures ANOVA was performed. When the F-values were significant for main effect, Duncan s method was used for multiple comparisons. For comparisons of other parameters, one-way ANOVA was performed followed by a Duncan s test as a post hoc procedure. For data with a failed normality test, one-way ANOVA on ranks was performed followed by Duncans s method for multiple comparisons. Statistical significance for all tests was defined as P RESULTS In all groups of rats, no behavioral changes were observed during the intracerebroventricular and subcutaneous infusions. Body weight gain was similar in rats with different treatments (not shown). BP and HR Responses to the Subcutaneous Infusion of ANG II The baseline MAP and HR were not significantly different among groups (Supplemental Table S1 and Supplemental Fig S1). The subcutaneous infusion of ANG II at 10 ng kg min did not change MAP and at 50 ng kg min increased MAP by 5

3 H424 mmhg. ANG II at 150 and 500 ng kg min significantly increased MAP by 8 and 20 mmhg, respectively, the first few days and then caused further increases with peaks of approximately 20 and 60 mmhg on days (P 0.05 for all vs. the corresponding baselines). HR was significantly decreased by 0 beats/min by ANG II at 500 ng kg min only on days 1 3 (P 0.05 vs. baseline). ANG II at lower rates did not change HR. Plasma ANG II and Plasma and Brain Aldosterone ANG II at 10 and 50 ng kg min did not change plasma ANG II levels (data not shown), nor did infusion of ANG II at 150 ng kg min (Fig. 1). In contrast, the infusion of ANG II at 500 ng kg min caused a fourfold increase in plasma ANG II levels (Fig. 1). The intracerebroventricular infusion of FAD-286 or Digibind had no effect on the increase in plasma ANG II. The intracerebroventricular infusion of ANG II at 500 but not 150 ng kg min significantly decreased plasma ANG I levels. ANG II at 150 ng kg min combined with FAD-286 significantly decreased plasma ANG I levels, whereas the decrease in ANG I levels by 500 ng kg min tended (P 0.07 and 0.08, respectively) to be enhanced by the concomitant intracerebroventricular infusion of FAD-286 or Digibind. Plasma aldosterone was 10 pg/ml in the vehicle groups and was markedly increased by ANG II at 500 but not at 150 ng kg min (Fig. 1). This increase was not affected by the intracerebroventricular infusion of FAD-286 or Digibind. Plasma corticosterone levels were similar for all groups (Fig. 1). In rats with intracerebroventricular and subcutaneous infusions of vehicles, aldosterone content was higher (P 0.05) in the hypothalamus than in the hippocampus [ vs pg/g (Fig. 2) and vs pg/g (Table 1)]. The subcutaneous infusion of ANG II at 150 ng kg min tended (P 0.1) to increase and at 500 ng kg min significantly increased hypothalamic aldosterone content (Fig. 2). This increase was significantly attenuated by intracerebroventricular FAD-286 but not by Digibind. The subcutaneous infusion of ANG II at 150 ng kg min had no effect but at 500 ng kg min significantly increased aldosterone content in the hippocampus (Fig. 2). Intracerebroventricular FAD or Digibind did not significantly attenuate this increase. None of the treatments affected corticosterone content in the hypothalamus or the hippocampus (Fig. 2). Plasma and Brain Aldosterone and Corticosterone After Subcutaneous Aldosterone Compared with vehicle, the subcutaneous infusion of aldosterone at 0.5 and 2.5 g/h increased plasma aldosterone concentrations significantly by 400 and 600% (Table 1). In contrast, subcutaneous aldosterone at both rates caused only minimal increases in hypothalamic aldosterone content. Subcutaneous aldosterone at 2.5 g/h increased aldosterone content in the hippocampus by 50%. Aldosterone infusion at both rates had no effect on corticosterone levels in the plasma, hypothalamus, and hippocampus. Neuronal Activation by the Subcutaneous Infusion of ANG II The subcutaneous infusion of vehicle caused mild increases in Fra expression, which were significant in the magnocellular PVN (mpvn) and parvocellular PVN (ppvn) on day 1 and in the median preoptic nucleus (MnPO) on day 14 compared with control rats without any intervention (Table 2). The subcutaneous infusion of ANG II increased Fra expression in a Fig. 1. Effects of an intracerebroventricular (icv) infusion of articial cerebrospinal fluid [acsf; vehicle (Veh)], aldosterone synthase inhibitor FAD-286 (FAD), or ouabain -binding Digibind on plasma ANG II, ANG I, aldosterone, and corticosterone concentrations in Wistar rats after a 14-day subcutaneous (sc) infusion of ANG II at 150 or 500 ng kg day or 0.9% saline (Veh). Data are means SE; see Fig. 5 for n values. *P 0.05 vs. the Veh Veh group.

4 H425 Fig. 2. Effects of an intracerebroventricular infusion of acsf (Veh), aldosterone synthase inhibitor FAD, or ouabain -binding Digibind on aldosterone and corticosterone contents in the hypothalamus and hippocampus of Wistar rats after a 14-day subcutaneous infusion of ANG II at 150 or 500 ng kg day or 0.9% saline (Veh). Data are means SE; see Fig. 5 for n values. *P 0.05 vs. the Veh Veh group; a P 0.05 vs. the Veh 500 ng kg day ANG II group. dose-related manner (Table 2 and Supplemental Fig S2). ANG II infused for 1 day increased Fra expression in all nuclei except the MnPO and after 4-days in all nuclei. After 14 days of infusion, ANG II at both rates caused sustained increases in Fra expression in the ppvn, mpvn, and SON but no longer in the MnPO. In the SFO after 14 days, ANG II at 150 ng kg min no longer had an effect and at 500 ng kg min still increased Fra expression, but less than after 1 and 4 days of infusion. In contrast, activation of the ppvn, mpvn, and SON was significantly greater after 14 versus 1 day of infusion of ANG II at 500 ng kg min. In the second experiment, compared with rats with both intracerebroventricular and subcutaneous infusions of vehicles, rats with the intracerebroventricular infusion of vehicle and subcutaneous infusion of ANG II at 500 ng kg min for 14 days showed significant increases in Fra expression in the SFO, mpvn, ppvn, and SON (Fig. 3). These increases were significantly attenuated in the ppvn and mpvn by the intracerebroventricular infusion of either FAD-286 or spironolactone. In contrast, intracerebroventricular FAD-286 or spironolactone did not affect neuronal activation in the SON and further increased Fra expression in the SFO compared with rats with intracerebroventricular vehicle and subcutaneous ANG II. Effects of Central Blockades on Pressor Responses to Subcutaneous ANG II The intracerebroventricular infusion of blockers or their corresponding vehicles without the subcutaneous infusion of ANG II did not affect 24-h MAP and HR (data not shown). ANG II at 150 or 500 ng kg min combined with intracerebroventricular acsf or other vehicles caused similar increases in MAP as in protocol I (Fig. 4). The intracerebroventricular infusion of FAD-286 fully prevented the increase in MAP induced by ANG II at 150 ng kg min and 70 80% of the increase by ANG II at 500 ng kg min. The subcutaneous infusion of FAD-286 (n 4) at the same rate had no effect on the increase in MAP by ANG II at 500 ng kg min, i.e., 28 and 62 mmhg after 4 and 10 days. The intracerebroventricular infusion of eplerenone at 10 and 20 g kg day significantly attenuated the increases in MAP by ANG II at 500 ng kg min by 70% and 80%, respectively (Fig. 4 and Supplemental Fig S3). The extent of attenuation by the intracerebroventricular infusion of eplerenone at the two rates was not statistically different. The intracerebroventricular infusion of spironolactone attenuated the increase Table 1. Plasma and brain aldosterone and corticosterone concentrations in Wistar rats after a 2-wk subcutaneous infusion of vehicle or aldosterone at 0.5 or 2.5 g/h Plasma Hypothalamus Hippocampus Aldosterone, pg/ml Corticosterone, ng/ml Aldosterone, pg/g Corticosterone, ng/g Aldosterone, pg/g Corticosterone, ng/g Vehicle Aldosterone 0.5 g/h * g/h * * Values are means SE; n 4 5 rats/group. *P 0.05 vs. the vehicle group.

5 H426 Table 2. Fra expression in various brain nuclei in intact (untreated) rats and rats after 1, 4, or 14 days of subcutaneous infusion of vehicle or ANG II at 150 or 500 ng kg min Subfornical Organ Median Preoptic Nucleus Parvocellular Paraventricular Nucleus Magnocellular Paraventricular Nucleus Supraoptic Nucleus Day 1 Intact Vehicle ANG II 150 ng kg min 28 5* * 46 7* 71 6* 500 ng kg min 51 2* * 56 5* 83 8* Day 4 Vehicle ANG II 150 ng kg min 52 8* 51 8* * 49 6* 93 4* 500 ng kg min 70 5* 80 6* 150 8* 70 7* * Day 14 Intact Vehicle ANG II 150 ng kg min * 82 4* 97 10* 500 ng kg min 26 5* * 85 5* 134 9* Values are means SE (expressed as numbers of Fra-positive cells as an average of 3 4 sections/nucleus from each rat); n 4 rats/group. *P 0.05 vs. vehicle-treated or intact animals; P 0.05 vs. 150 ng kg min ANG-treated animals; P 0.05 vs. intact animals. in MAP by 50%, which was somewhat less than that by eplerenone (Supplemental Fig S3). The intracerebroventricular infusion of Digibind attenuated the pressor responses to ANG II at 500 ng kg min by 50% (Fig. 4), whereas the intracerebroventricular infusion of nonspecific Fab fragments at the same rate did not affect the increase in BP (Supplemental Fig S3). MAP was significantly higher by 7 8 mmhg at night versus in the day in the different groups before the subcutaneous infusion of ANG II (Table 3). During the subcutaneous infusion of ANG II, this difference between the night and day MAP persisted and was not affected by any of the treatments. The small decrease in HR by ANG II at 500 ng kg min was absent in rats with the intracerebroventricular infusion of a blocker (Supplemental Fig S4). DISCUSSION The mechanisms mediating the central, slowly developing pressor response to circulating ANG II have so far not been elucidated. The present study provides evidence for the concept that in the central nervous system (CNS) activation of a local aldosterone - ouabain system acts as a neuromodulatory pathway to induce the slow pressor response to circulating ANG II. The subcutaneous infusion of ANG II at a modest rate induced minimal increases in plasma ANG II and aldosterone but caused persistent neuronal activation in the PVN and SON and a gradual pressor response of 20 mmhg. ANG II at a higher rate caused clear increases in plasma ANG II and aldosterone, an increase in hypothalamic aldosterone, further neuronal activation (especially in the PVN and SON), and a marked pressor response of mmhg. The central infu- Fig. 3. Effects of an intracerebroventricular infusion of aldosterone synthase inhibitor FAD, mineralocorticoid receptor (MR) blocker spironolactone (Spiro), or spironolactone Veh on Fra expression in the magnocellular paraventricular nucleus (mpvn), parvocellular paraventricular nucleus (ppvn), subfornical organ (SFO), and supraoptic nucleus (SON) of Wistar rats treated with a subcutaneous infusion of ANG II at 500 ng kg day or 0.9% saline (Veh) for 14 days. Data are means SE; n 5 6. *P 0.05 vs. the Veh Veh group; a P 0.05 vs. the Veh ANG II group.

6 Fig. 4. Effects of an intracerebroventricular infusion of aldosterone synthase inhibitor FAD, MR blocker eplerenone (Epler) at 20 g kg day, ouabain - binding Digibind, or acsf (Veh) on changes in 24-h mean arterial pressure (MAP) as assessed with telemetry in Wistar rats during a 14-day subcutaneous infusion of 0.9% saline (Veh) or ANG II at 150 (A)or500(B) ng kg day. Data are means SE. *P 0.05 vs. the Veh Veh group; **P 0.05 vs. the other groups. sion of an AS inhibitor or MR blocker prevented most of the ANG II-induced neuronal activation in the PVN and the increase in BP. The AS inhibitor also prevented the ANG II-induced increase in hypothalamic aldosterone without affecting the increase in plasma aldosterone. Consistent with this dissociation of hypothalamic and plasma aldosterone, the chronic subcutaneous infusion of aldosterone caused clear dose-related increases in plasma aldosterone with only minimal increases in hypothalamic aldosterone content. In the present study, a 2-wk subcutaneous infusion of ANG II at 150 ng kg min did not change plasma ANG II and aldosterone levels but at 500 ng kg min increased plasma ANG II by 4-fold and plasma aldosterone by 25-fold. In Sprague-Dawley rats, an infusion of ANG II at 200 ng kg min intraperitoneally increased plasma aldosterone by 2.5-fold (16) and at 300 ng kg min infused subcutaneous increased urinary aldosterone excretion by 7-fold (37). Plasma ANG II was not measured in these studies. An intravenous infusion of ANG II at 60 and 200 ng kg min for 5 days increased plasma ANG II in a dose-related manner, but only ANG II at 200 ng kg min increased plasma aldosterone significantly (42). Together, these data suggest that the chronic infusion of ANG II only at higher rates causing clear increases in plasma ANG II leads to sustained and marked adrenal stimulation. As expected, the infusion of ANG II decreased plasma ANG I levels. This decrease tended to be clearer in the presence of central blockades, possibly reflecting the prevention of an increase in renal sympathetic nerve activity and the resulting effect on renin release. Infusion of aldosterone at 0.5 and 2.5 g/h caused doserelated but nonproportional increases in plasma aldosterone levels, i.e., the lower infusion rate increased levels from to pg/ml and the fivefold higher rate increased levels further, but only to pg/ml. A similar pattern of aldosterone levels in response to the subcutaneous infusion of aldosterone has been previously reported by Garwitz and Jones (8). Whether this pattern reflects nonlinear kinetics of aldosterone or other mechanisms cannot be assessed from the present study. The subcutaneous infusion of aldosterone increased plasma aldosterone but did not change hypothalamic aldosterone. In contrast, studies in adrenalectomized rats with the subcutaneous infusion of aldosterone at low rates of g/h for 1 2 wk showed aldosterone levels in the hypothalamus (49) or in the whole brain (13) proportional to plasma aldosterone levels over the range of pg/g and pg/ml (49). Limited penetration of the blood-brain barrier by aldosterone (9) and substantial competition from corticosterone for both uptake and binding to MR in the cell (9, 12, 48) may explain these different findings in adrenalectomized rats without corticosterone replacement compared with the findings in intact rats. Uptake mechanisms may also vary by brain regions, since the subcutaneous infusion of aldosterone clearly increased aldosterone levels in the hippocampus but not in the hypothalamus of intact rats, consistent with different regulation of uptake. Enzymes involved in steroid biosynthesis are present in the CNS (38), and aldosterone can be synthesized in the brain (11). In the present study, the chronic subcutaneous infusion of ANG II increased aldosterone content in the whole hypothalamus and hippocampus. The subcutaneous infusion of ANG II Table 3. Effects of an intracerebroventricular infusion of FAD-286, eplerenone, Digibind, or acsf on changes in day and night MAP as well as diurnal differences in MAP as assessed with telemetry in Wistar rats after 1, 5, 10, and 14 days of a subcutaneous infusion of 0.9% saline or ANG II at 500 ng kg min After ANG II H427 Treatment Baseline (Before ANG II) Night-Day Difference Day 14 Intracerebroventricular Infusion Subcutaneous Infusion Day Night Night-Day Difference Day 1 Day 5 Day 10 Day 14 Day Night acsf (vehicle) 0.9% Saline (vehicle) * * acsf (vehicle) ANG II * * FAD-286 ANG II * * Digibind ANG II * * Eplerenone ANG II * * Data are means SE (in mmhg); see Fig. 4 for n values. FAD-286 is an inhibitor of aldersterone synthase, Digibind binds ouabain and ouabain-like compounds; and eplerenone is an inhibitor of mineralocorticoid receptors. acsf, artifical cerebrospinal fluid. *P 0.05 vs. the corresponding day; P 0.05 vs. all other groups on day 14; P 0.05 vs. the vehicle vehicle-treated group on day 14.

7 H428 at 500 ng kg min increased plasma aldosterone by 25- fold but increased hypothalamic aldosterone only by 3.5-fold. The intracerebroventricular infusion of the AS inhibitor prevented most of this increase in hypothalamic aldosterone but had no effect on the marked increase in plasma aldosterone, consistent with an increase in local aldosterone production in the hypothalamus. The intracerebroventricular infusion of the AS inhibitor had only a modest, nonsignificant effect on the increase in aldosterone in the hippocampus induced by ANG II. Considering also that AS mrna levels are lower in the hippocampus versus in the hypothalamus (14), and that the subcutaneous infusion of aldosterone increases aldosterone levels in the hippocampus but not in the hypothalamus, most of the increase in aldosterone in the hippocampus by the subcutaneous infusion of ANG II appears to reflect uptake rather than local production. Immunohistochemical detection of the Fos family of proteins is extensively being used to identify chronically activated neuronal populations in the CNS, including CNS regions involved in cardiovascular homeostasis (4, 36, 44). Central effects of circulating ANG II are generally considered to be initiated by the activation of neurons in circumventricular organs such as the SFO (4, 52). In the present study, the subcutaneous infusion of ANG II caused transient neuronal activation in the SFO and progressive neuronal activation in the PVN and SON. These findings are consistent with a previous study (4) in rabbits. The intracerebroventricular infusion of an AS inhibitor or MR blocker markedly attenuated neuronal activation in both the mpvn and ppvn. Neither blocker attenuated the neuronal activation in the SON. One may speculate that circulating ANG II activates descending pathways from the SFO to activate magnocellular neurons in the SON (43), leading to an increase in aldosterone synthesis, and aldosterone via MR causes further activation of the mpvn and ppvn. Neuronal activation in the PVN and SON has also been observed in other rat models showing an increase in brain aldosterone content with or without an increase in plasma aldosterone, such as in rats with the chronic intracerebroventricular infusion of aldosterone (51), Dahl salt-sensitive rats on a high-salt diet (1), and rats after MI (26, 36, 44, 47). This activation in the PVN can be prevented by the intracerebroventricular infusion of an MR blocker (51) or attenuated by 50% by the oral administration of an MR blocker (26). Interestingly, the intracerebroventricular infusion of the AS inhibitor or MR blocker prevented the decrease in neuronal excitation in the SFO after 14 days of ANG II infusion. It appears that sustained activation of the PVN decreases neuronal activation of the SFO through a negative feedback mechanism, perhaps indirectly by the increase in BP (35), which is attenuated by the blockers. The subcutaneous infusion of ANG II caused dose-related increases in BP. These findings are fairly similar to previous studies (17, 27, 30). The subcutaneous infusion of ANG II at the low rate of 150 ng kg min after a few days increased BP by 20 mmhg but did not significantly increase plasma ANG II and aldosterone. Previous studies using a subcutaneous infusion of ANG II at 200 ng kg min (15) or an intravenous infusion of ANG II at 30 ng kg min (25) have also reported significant increases in BP without detectable increases in plasma ANG II or aldosterone levels. These findings indicate that pressor responses can be induced by the long-term infusion of ANG II at low rates without detectable increases in plasma ANG II and aldosterone. Pressor responses to ANG II depend on both peripheral (arterial and renal) and neurogenic mechanisms (5, 37). Our results suggest that the infusion of ANG II at low rates mainly increases BP by activation of CNS pressor pathways, since the intracerebroventricular infusion of the AS inhibitor fully prevented the pressor response to ANG II at 150 ng kg min. The subcutaneous infusion of ANG II at 500 ng kg min during the first few days increased BP by 30 mmhg and after 8 10 days by 60 mmhg. Most of these pressor responses to ANG II were prevented by the AS inhibitor and MR blockers, both during the first few days and the more chronic phase. It appears that modest increases in circulating ANG II cause the activation of CNS pathways, which substantially contributes to both the initial and long-term increase in BP. The results on neuronal activation, hypothalamic aldosterone, and BP together suggest that an ANG II-induced increase in aldosterone locally produced in the brain plays a major role in the progressive activation of the PVN and thereby contributes to the progressive increase in BP. An intracerebroventricular infusion of the AS inhibitor also prevents hypertension and/or sympathetic hyperactivity in rats with an intracerebroventricular infusion of Na -rich acsf (21), Dahl saltsensitive rats on a high-salt diet (14, 23), and rats after MI (22). In these models, plasma aldosterone may or may not be elevated, further demonstrating the role of brain aldosterone per se. The intracerebroventricular infusion of eplerenone or spironolactone also largely attenuated the ANG II-induced hypertension. The effects of the AS inhibitor and MR blockers were rather similar, and the results from the two different blockades suggest that locally produced aldosterone can access free MR in, e.g., the PVN (50) to cause neuronal activation and hypertension. Using an AS inhibitor as well as an MR blocker provides specific evidence for the role of aldosterone locally produced in the brain in the activation of MR and BP regulation. As expected, BP measured by telemetry showed higher values during the night compared with the day. This diurnal rhythm was not affected by the subcutaneous infusion of ANG II or any of the central blockades. This finding supports the concept that ANG II via aldosterone and MR resets CNS pathways at a higher level but does not interfere with their physiological responses. The central infusion of aldosterone increases hypothalamic ouabain levels (46), and both MR blockers and Digibind prevent the sympathoexcitation and pressor responses to the central infusion of aldosterone (46, 51), indicating that MR and ouabain mediate central responses to exogenous aldosterone. Since the intracerebroventricular infusion of Digibind did not affect the ANG II-induced increase in hypothalamic aldosterone but did inhibit the pressor response, it appears that ouabain is involved in the central pathways downstream to aldosterone synthesis. Aldosterone via MR possibly activates magnocellular neurons in the PVN, increasing ouabain synthesis/release. Ouabain may inhibit Na -K -ATPase and thereby lower the neuronal membrane potential or increase intracellular Ca 2 (39), which can enhance the activity of angiotensinergic sympathoexcitatory pathways leading to hypertension (7). Ouabain blockade was somehow less effective than AS inhibition in preventing ANG II-induced hyper-

8 tension, suggesting that aldosterone s effects in the CNS may depend not only on ouabain. It appears less likely that higher doses are needed, since Digibind at the same rate is effective in rats after MI (32) and Dahl salt-sensitive rats on high salt intake (19). The intracerebroventricular infusion of eplerenone was somewhat more effective in attenuating the pressor effects of ANG II than the intracerebroventricular infusion of spironolactone. Eplerenone has very low affinities for other steroid hormone receptors, such as androgen receptors, and does not have an antiandrogenic effect in vivo (41). Spironolactone may block androgen receptors in the brain in favor of, e.g., AVP release (45), offsetting the BP-lowering effects induced by the MR blockade. Limitations of the Study FAD-286 hydrogen tartrate was used as the AS inhibitor. The central infusion of FAD-286 free base (14) and FAD hydrogen tartrate (23) similarly prevent salt-induced hypertension in Dahl salt-sensitive rats, but possible effects of tartrate per se cannot be excluded, and the intracerebroventricular infusion of L-tartaric acid is an appropriate vehicle control to consider. Neither the aldosterone content of the whole hypothalamus nor intracerebroventricular infusion of blockers provide insight into the actual hypothalamic nuclei possibly responsible for local production nor into the areas mediating the actions of aldosterone. Further studies on gene expression and local inhibition may provide more specific information in this regard. Perspectives The present study shows a novel CNS mechanism for ANG II-induced hypertension, i.e., a chronic increase in circulating ANG II appears to activate an aldosterone-dependent neuromodulatory pathway, which, via MR activation, causes ouabain release and sustained neuronal activation in the PVN. Our findings suggest that this CNS aldosterone-dependent neurogenic component is a major mechanism for hypertension induced by circulating ANG II. Peripheral arterial and renal effects (3) of circulating ANG II have also been well documented. Our findings suggest that these effects may be facilitated by this central mechanism. After MI, circulating ANG II rapidly increases, and neurons in, e.g., the SFO, SON, and PVN are activated in a similar pattern as caused by the subcutaneous infusion of ANG II. Central blockade of AS, MR, or ouabain largely prevents sympathetic hyperactivity, left ventricular dysfunction, and remodeling (6, 20, 22, 24, 32), indicating an activation of this central aldosterone- ouabain pathway also in rats after MI. Further studies are needed to assess the relative role of circulating ANG II in activating this pathway after MI. ACKNOWLEDGMENTS F. H. H. Leenen holds the Pfizer Chair in Hypertension Research, an endowed chair supported by Pfizer Canada, University of Ottawa Heart Institute Foundation, and Canadian Institutes of Health Research. FAD-286 was a generous gift from Novartis Institutes for BioMedical Research, and eplerenone was kindly provided by Pfizer Canada. GRANTS This work was supported by Canadian Institutes of Health Research Operating Grant FRN:MOP DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the author(s). REFERENCES H Budzikowski AS, Vahid-Ansari F, Leenen FH. Chronic activation of brain areas by high-sodium diet in Dahl salt-sensitive rats. Am J Physiol Heart Circ Physiol 274: H2046 H2052, Budzikowski AS, Vahid-Ansari F, Robertson GS, Leenen FH. Patterns of neuronal activation during development of sodium sensitive hypertension in SHR. Hypertension 30: , Crowley SD, Gurley SB, Herrera MJ, Ruiz P, Griffiths R, Kumar AP, Kim HS, Smithies O, Le TH, Coffman TM. Angiotensin II causes hypertension and cardiac hypertrophy through its receptors in the kidney. Proc Natl Acad Sci USA 103: , Davern PJ, Head GA. Fos-related antigen immunoreactivity after acute and chronic angiotensin II-induced hypertension in the rabbit brain. Hypertension 49: , Fink GD. Long-term sympatho-excitatory effect of angiotensin II: a mechanism of spontaneous and renovascular hypertension. Clin Exp Pharmacol Physiol 24: 91 95, Francis J, Weiss RM, Wei SG, Johnson AK, Beltz TG, Zimmerman K, Felder RB. Central mineralocorticoid receptor blockade improves volume regulation and reduces sympathetic drive in heart failure. Am J Physiol Heart Circ Physiol 281: H2241 H2251, Gabor A, Leenen FH. Mechanisms in the PVN mediating local and central sodium-induced hypertension in Wistar rats. Am J Physiol Regul Integr Comp Physiol 296: R618 R630, Garwitz ET, Jones AW. Aldosterone infusion into the rat and dosedependent changes in blood pressure and arterial ionic transport. Hypertension 4: , Geerling JC, Loewy AD. Aldosterone in the brain. Am J Physiol Renal Physiol 297: F559 F576, Glowinski J, Iversen LL. Regional studies of catecholamines in the rat brain. The disposition of [ 3 H]norepinephrine, [ 3 H]dopamine and [ 3 H]dopa in various regions of the brain. J Neurochem 13: , Gomez-Sanchez CE, Zhou MY, Cozza EN, Morita H, Foecking MF, Gomez-Sanchez EP. Aldosterone biosynthesis in the rat brain. Endocrinology 138: , Gomez-Sanchez EP. The mammalian mineralocorticoid receptor: tying down a promiscuous receptor. Exp Physiol 95: 13 18, Gomez-Sanchez EP, Ahmad N, Romero DG, Gomez-Sanchez CE. Is aldosterone synthesized within the rat brain? Am J Physiol Endocrinol Metab 288: E342 E346, Gomez-Sanchez EP, Gomez-Sanchez CM, Plonczynski M, Gomez- Sanchez CE. Aldosterone synthesis in the brain contributes to Dahl salt-sensitive rat hypertension. Exp Physiol 95: , Griffin SA, Brown WC, MacPherson F, McGrath JC, Wilson VG, Korsgaard N, Mulvany MJ, Lever AF. Angiotensin II causes vascular hypertrophy in part by a non-pressor mechanism. Hypertension 17: , Hauger RL, Aguilera G, Catt KJ. Angiotensin II regulates its receptor sites in the adrenal glomerulosa zone. Nature 271: , Hendel MD, Collister JP. Contribution of the subfornical organ to angiotensin II-induced hypertension. Am J Physiol Heart Circ Physiol 288: H680 H685, Huang BS, Harmsen E, Yu H, Leenen FH. Brain ouabain-like activity and the sympathoexcitatory and pressor effects of central sodium in rats. Circ Res 71: , Huang BS, Leenen FH. Both brain angiotensin II and ouabain contribute to sympathoexcitation and hypertension in Dahl S rats on high salt intake. Hypertension 32: , Huang BS, Leenen FH. Blockade of brain mineralocorticoid receptors or Na channels prevents sympathetic hyperactivity and improves cardiac function in rats post-mi. Am J Physiol Heart Circ Physiol 288: H2491 H2497, Huang BS, White RA, Ahmad M, Jeng AY, Leenen FH. Central infusion of aldosterone synthase inhibitor prevents sympathetic hyperactivity and hypertension by central Na in Wistar rats. Am J Physiol Regul Integr Comp Physiol 295: R166 R172, 2008.

9 H Huang BS, White RA, Ahmad M, Tan J, Jeng AY. Central infusion of aldosterone synthase inhibitor attenuates left ventricular dysfunction and remodeling in rats after myocardial infarction. Cardiovasc Res 81: , Huang BS, White RA, Jeng AY, Leenen FH. Role of central nervous system aldosterone synthase and mineralocorticoid receptors in salt-induced hypertension in Dahl salt-sensitive rats. Am J Physiol Regul Integr Comp Physiol 296: R994 R1000, Huang BS, Yuan B, Leenen FH. Chronic blockade of brain ouabain prevents sympathetic hyper-reactivity and impairment of acute baroreflex resetting in rats with congestive heart failure. Can J Physiol Pharmacol 78: 45 53, Kanagy NL, Pawloski CM, Fink GD. Role of aldosterone in angiotensin II-induced hypertension in rats. Am J Physiol Regul Integr Comp Physiol 259: R102 R109, Kang YM, Zhang ZH, Johnson RF, Yu Y, Beltz T, Johnson AK, Weiss RM, Felder RB. Novel effect of mineralocorticoid receptor antagonism to reduce proinflammatory cytokines and hypothalamic activation in rats with ischemia-induced heart failure. Circ Res 99: , Kimura S, Zhang GX, Abe Y. Malfunction of vascular control in lifestyle-related diseases: oxidative stress of angiotensin II-induced hypertension: mitogen-activated protein kinases, and blood pressure regulation. J Pharmacol Sci 96: , Lal A, Veinot JP, Ganten D, Leenen FH. Prevention of cardiac remodeling after myocardial infarction in transgenic rats deficient in brain angiotensinogen. J Mol Cell Cardiol 39: , Lal A, Veinot JP, Leenen FH. Critical role of CNS effects of aldosterone in cardiac remodeling post myocardial infarction in rats. Cardiovasc Res 64: , Li Q, Dale WE, Hasser EM, Blaine EH. Acute and chronic angiotensin hypertension: neural and nonneural components, time course, and dose dependency. Am J Physiol Regul Integr Comp Physiol 271: R200 R207, Leenen FH. Brain mechanisms contributing to sympathetic hyperactivity and heart failure. Circ Res 101: , Leenen FH, Huang BS, Yu H, Yuan B. Brain ouabain mediates sympathetic hyperactivity in congestive heart failure. Circ Res 77: , Leenen FH, Skarda V, Yuan B, White R. Changes in cardiac ANG II postmyocardial infarction in rats: effects of nephrectomy and ACE inhibitors. Am J Physiol Heart Circ Physiol 276: H317 H325, Leenen FH, White R, Yuan B. Isoproterenol-induced cardiac hypertrophy: role of circulatory versus cardiac renin-angiotensin system. Am J Physiol Heart Circ Physiol 281: H2410 H2416, Li YW, Dampney RA. Expression of Fos-like protein in brain following sustained hypertension and hypotension in conscious rabbits. Neuroscience 61: , Lindley TE, Doobay MF, Sharma RV, Davisson RL. Superoxide is involved in the central nervous system activation and sympathoexcitation of myocardial infarction-induced heart failure. Circ Res 94: , Luft FC, Wilcox CS, Unger T, Kühn R, Demmert G, Rohmeiss P, Ganten D, Sterzel RB. Angiotensin-induced hypertension in the rat. Sympathetic nerve activity and prostaglandins. Hypertension 14: , Mellon SH, Griffin LD, Compagnone NA. Biosynthesis and action of neurosteroids. Brain Res Rev 37: 3 12, Pulina MV, Zulian A, Berra-Romani R, Beskina O, Mazzocco-Spezzia A, Baryshnikov SG, Papparella I, Hamlyn JM, Blaustein MP, Golovina VA. Upregulation of Na and Ca 2 transporters in arterial smooth muscle from ouabain-induced hypertensive rats. Am J Physiol Heart Circ Physiol 298: H263 H274, Pullen MA, Brooks DP, Edwards RM. Characterization of the neutralizing activity of digoxin-specific Fab toward ouabain-like steroids. J Pharmacol Exp Ther 310: , Rogerson FM, Yao Y, Smith BJ, Fuller PJ. Differences in the determinants of eplerenone, spironolactone and aldosterone binding to the mineralocorticoid receptor. Clin Exp Pharmacol Physiol 31: , Schiffrin EL, Gutkowska J, Genest J. Effect of angiotensin II and deoxycorticosterone infusion on vascular angiotensin II receptors in rats. Am J Physiol Heart Circ Physiol 246: H608 H614, Sgro S, Ferguson AV, Renaud LP. Subfornical organ supraoptic nucleus connections: an electrophysiologic study in the rat. Brain Res 303: 7 13, Vahid-Ansari F, Leenen FH. Pattern of neuronal activation in rats with CHF after myocardial infarction. Am J Physiol Heart Circ Physiol 275: H2140 H2146, Viau V, Meaney MJ. The inhibitory effect of testosterone on hypothalamic-pituitary-adrenal responses to stress is mediated by the medial preoptic area. J Neurosci 16: , Wang H, Huang BS, Leenen FH. Brain sodium channels and ouabainlike compounds mediate central aldosterone-induced hypertension Am J Physiol Heart Circ Physiol 285: H2516 H2523, Wei SG, Yu Y, Zhang ZH, Weiss RM, Felder RB. Angiotensin IItriggered p44/42 mitogen-activated protein kinase mediates sympathetic excitation in heart failure rats. Hypertension 52: , Yongue BG, Roy EJ. Endogenous aldosterone and corticosterone in brain cell nuclei of adrenal-intact rats: regional distribution and effects of physiological variations in serum steroids. Brain Res 436: 49 61, Yu Y, Wei SG, Zhang ZH, Gomez-Sanchez E, Weiss RM, Felder RB. Does aldosterone upregulate the brain renin-angiotensin system in rats with heart failure? Hypertension 51: , Zhang ZH, Kang YM, Yu Y, Wei SG, Schmidt TJ, Johnson AK, Felder RB. 11 -Hydrosteroid dehydrogenase type 2 activity in hypothalamic paraventricular nucleus modulates sympathetic excitation. Hypertension 48: , Zhang ZH, Yu Y, Kang YM, Wei SG, Felder RB. Aldosterone acts centrally to increase brain renin-angiotensin system activity and oxidative stress in normal rats. Am J Physiol Heart Circ Physiol 294: H1067 H1074, Zimmerman MC, Lazartigues E, Sharma RV, Davisson RL. Hypertension caused by angiotensin II infusion involves increased superoxide production in the central nervous system. Circ Res 95: , 2004.