| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Hypertension. 2006;48:127.)
© 2006 American Heart Association, Inc.
Original Articles |
From the Departments of Internal Medicine (Z.-H.Z., Y.-M.K., Y.Y., S.-G.W., R.B.F.), Physiology and Biophysics (T.J.S.), and Psychology (A.K.J.), Roy J. and Lucille A. Carver College of Medicine, University of Iowa, and Veterans Affairs Medical Center (R.B.F.), Iowa City, Iowa.
Correspondence to Robert B. Felder, MD, University of Iowa College of Medicine, E318-GH, 200 Hawkins Drive, Iowa City, IA 52242. E-mail robert-felder{at}uiowa.edu
| Abstract |
|---|
|
|
|---|
Key Words: aldosterone corticosterone brain
| Introduction |
|---|
|
|
|---|
The central nervous system sites of action of aldosterone and MR antagonists are still poorly defined. MR are widely distributed in the brain6 and are largely occupied by corticosterone,7 which is present in higher concentrations than aldosterone8 and binds to MR with equal affinity.7 The enzyme 11ß-hydroxysteroid dehydrogenase type 2 (11ß-HSD-2), which has a more restricted distribution in the brain,911 rapidly converts corticosterone to an inactive metabolite, increasing the likelihood of aldosterone binding to MR. The coexpression of 11ß-HSD-2 with MR may therefore identify brain regions that are particularly sensitive to aldosterone. Consistent with that hypothesis, a recent study found that the abundance of 11ß-HSD-2 protein in the nucleus tractus solitarius,11 a cardiovascular afferent integrative center in the hindbrain, correlates closely with the nuclear translocation of MR in response to systemically administered aldosterone.12
The paraventricular nucleus (PVN) of the hypothalamus is a well recognized cardiovascular integrative center of the forebrain that contributes to the augmented sympathetic drive in heart failure. PVN neurons express MR,13 but 11ß-HSD-2 activity has not previously been described in this region of the brain. Using sensitive molecular and physiological techniques, we tested the hypothesis that the PVN might also be a target for the central actions of aldosterone.
| Materials and Methods |
|---|
|
|
|---|
Drugs Administered
Glycyrrhizic acid (GA) and carbenoxolone (CBX) are both potent 11ß-HSD-2 inhibitors,15 but with important differences. CBX has the potential to block gap junctions; GA does not, and has been used to control for the gap junction blocking effect of CBX.16,17 From a practical standpoint, GA is stickier than CBX and so less amenable to microinjection via micropipettes.
CBX, GA, hexamethonium bromide, and SL were purchased from Sigma (St. Louis, Mo). CBX, GA, and SL were initially dissolved in absolute ethanol and then diluted with artificial cerebrospinal fluid to a desired concentration, with the final ethanol concentration <1%. The same volume of the vehicle ([VEH] <1% ethanol in artificial cerebrospinal fluid) was administered as a control for the ICV injections and tissue microinjections. Hexamethonium was dissolved in saline for intravenous injection. The doses of CBX and GA used in this study were derived from previous related studies18,19 and optimized in preliminary experiments.
Real-Time Polymerase Chain Reaction
Real-time polymerase chain reaction (PCR) was performed to determine whether messenger RNA for 11ß-HSD-2 is expressed in PVN. Eight rats were euthanized with an overdose of urethane. The brains were removed and cut into 450-µm coronal sections. A punch biopsy was obtained from right and left PVN and from right and left cortex (in the same section) using a 15-gauge needle stub (ID 1.5 mm),20 yielding 600 to 800 µg of each tissue. Total RNA was isolated from PVN and cortex using TRI reagent (Molecular Research Center Inc, Cincinnati, Ohio). cDNAs were synthesized using TaqMan reverse transcription reagents (Applied Biosystems). PCR primers and TaqMan probes for 11ß-HSD-2 were designed using the computer program Primer Express (Applied Biosystems). The sequences of primers and probe for 11ß-HSD-2 were the following: forward primer, 5'-TGGAACCTCCAAGGCAGCTA-3'; reverse primer, 5'-TGATACCCCAGGGAAGCAGTT-3'; probe, 5'-TGCACTGCTCATGGACACATTCAGCTG-3'. Real-time PCR was performed using a previously described method.21,22 To obtain a calibration curve, we amplified the amount of plasmid (PCR2.1-TOPO vector; Invitrogen) encompassing rat 11ß-HSD-2 cDNA. The standard curve for 11ß-HSD-2 was generated using serially diluted solutions (105, 106, 107, 108, 109, and 1010 copies) of plasmid. The target message in unknown samples was quantified by measuring the threshold cycle (Ct) and by using a calibration curve to determine the starting target message quantity. Quantification of 11ß-HSD-2 mRNA was expressed as copy number per nanogram of total RNA.
Electrophysiology/Microinjection
The general methods have been described previously.23 Forty-two rats were anesthetized with urethane (1.5 g/kg IP). Supplemental urethane (0.1 to 0.3 g/kg IP or IV) was administered if spontaneous increases in arterial pressure (AP), heart rate (HR), or respiratory rate, or responses to a noxious stimulus, were observed during surgery or experimental recording. Recording sessions began at least an hour after completion of the surgical preparation.
At the end of each experiment, the net value of renal nerve activity was determined by subtracting the background noise recorded after injection of the ganglion blocker hexamethonium (30 mg/kg IV). The recordings of rectified and integrated RSNA, single unit PVN neuronal activity, mean arterial pressure (MAP), and HR were analyzed using methods described previously.23 Data were averaged over 1-minute intervals, and a 3-minute average of baseline activity was used as a control for responses to injection of CBX or GA. Absolute values were used to calculate the changes in MAP and HR, and a percent change from baseline was used to calculate changes in integrated RSNA. The relationship of PVN neuronal discharge to the cardiac cycle was determined by post-stimulus time histograms, triggered by the peak of the arterial pressure pulse across a 3-minute interval.
The last PVN recording or tissue microinjection site in each experiment was marked with pontamine sky blue, and the anatomic locations of other recording and microinjection sites were extrapolated with respect to this reference point. All sites were plotted on schematic tracings of the PVN, based on the rat atlas of Paxinos and Watson.24
Immunohistochemistry
CBX (15 µg in 2 µL, 5 rats) or VEH (2 µL, 4 rats) was administered ICV. Two hours later, rats were anesthetized and transcardially perfused with PBS and 4% paraformaldehyde. Brain tissues were removed and processed for c-fos expression using the avidinbiotinperoxidase complex technique, as previously described.25,26 A rabbit polyclonal anti-Fos antiserum (Santa Cruz Biotechnology, Santa Cruz, Calif) was used. The c-fos positive cells in posterior magnocellular, medial parvocellular, and dorsal parvocellular regions of PVN were counted manually in 2 adjacent coronal sections, one of which included the maximal expanse of PVN. An average value for the number of c-fos positive neurons counted in the 2 sections, reported as c-fos positive neurons/100 µm2, was used for data analysis.
Data Analysis
Statistical significance among multiple comparisons was determined by 1-way or 2-way repeated-measures ANOVA followed by post-hoc Tukey test. Paired t test was used for comparison between peak responses and baseline. Other group data were compared by t test. Values are presented as means±SEM. P<0.05 was considered to be statistically significant.
| Results |
|---|
|
|
|---|
|
Sympathoexcitatory Effects of 11ß-HSD-2 Inhibition
Microinjection of the 11ß-HSD-2 inhibitor CBX (2 µg, n=7) into PVN produced a sympathoexcitatory response (Figure 2A, 2B) characterized by significant (P<0.05) increases from baseline (Table) in RSNA (27.4±5.8%) MAP (8.4±2.5 mm Hg), and HR (26.2±6.2 bpm). The same volume of VEH (100 nL, n=5) elicited no significant response (Figure 2B). Microinjection of CBX into sites dorsal, lateral, and anterior to PVN did not significantly alter MAP (1.3±0.2 mm Hg), HR (4.3±3.8 bpm), or RSNA.(3.2±1.8%). Microinjection of CBX into the dorsomedial hypothalamic nucleus, 1.0 mm caudal to PVN, did induce a mild increase in MAP (5.7±0.2 mm Hg), HR (16.4±2.9 bpm), and RSNA (16.8±2.6%) in 2 of the 4 rats tested.
|
|
ICV CBX (Figure 3) elicited significant (P<0.05) increases from baseline (Table) in PVN neuronal firing (from 2.1±0.3 to 7.2±1.3 spikes/s), RSNA (by 84±21%), MAP (9.8±2.5 mm Hg), and HR (89.8±14.7 bpm). The onset latency of these responses was 23 minutes, with peak responses at
40 to 50 minutes and durations
90 minutes. The same volume of VEH (ICV, n=6) had no effects. Pretreatment with SL (20 µg ICV, n=6), prevented the CBX-induced increases in PVN activity (from 2.4±0.7 to 2.1±1.1 spikes/s), RSNA (4.5±7.0%), MAP (0.9±2.7 mm Hg), or HR (22.0±9.4 bpm). SL alone had no effect on baseline values (Table) of MAP, HR, or RSNA.
|
ICV injection of GA (10 µg, n=6; Figure 4) also significantly (P<0.01) increased PVN neuronal activity (1.8±0.3 to 7.7±1.6 spikes/s), RSNA (106.3±28.9%), MAP (29.3±7.8 mm Hg), and HR (73.0±11.2 bpm) from baseline (Table).
|
In ADX rats, neither CBX (15 µg ICV, n=5 rats) nor GA (10 µg ICV, n=5 rats) elicited a significant change in MAP, HR, RSNA, or PVN neuronal activity (data not shown). Consistent with earlier reports,2729 the ADX rats had lower baseline blood pressures, but HR, RSNA, and PVN neuronal activity were not significantly different from VEH-treated intact rats (Table).
The PVN microinjection and PVN recording sites are shown in Figure 5. The PVN neurons recorded were distributed throughout the nucleus. The majority of recorded PVN neurons (21/30 tested) had a pulse-related discharge pattern (see insets in Figures 3A, 3B, 4
A).
|
Locations of PVN Neurons Excited By ICV Injection of CBX
Immunohistochemistry of brain tissue harvested 2 hours after ICV administration of CBX (5 rats) revealed increased c-fos expression in neurons throughout the PVN, compared with VEH (4 rats). Similar increases were observed in all 3 regions in which the neurons were counted (Fig 6). Areas outside PVN but at the same stereotaxic level had scattered c-fos expression. However, there was no difference between CBX- and VEH-treated rats.
|
| Discussion |
|---|
|
|
|---|
Other forebrain regions express 11ß-HSD-2 activity,9,10 but, to the best of our knowledge, this report is the first to suggest that aldosterone might act directly on PVN neurons endowed with 11ß-HSD-2protected MR. The lack of evidence for 11ß-HSD-2 expression in PVN in previous studies911 likely reflects the sensitivity of the methods used. We used real-time PCR to demonstrate the presence of mRNA for 11ß-HSD-2 in PVN, and chemical inhibition of 11ß-HSD-2 activity with CBX to demonstrate the presence of functionally significant levels of 11ß-HSD-2 in PVN. These 2 lines of evidence leave little room to doubt that 11ß-HSD-2 is expressed and enzymatically active in PVN.
The cardiovascular and autonomic responses to ICV injections of CBX or GA are more dramatic than the responses elicited by microinjecting CBX directly into PVN. This is the expected result, since other brain regions expressing 11ß-HSD-2911 are likely recruited by the ICV injection. This interpretation is consistent with the observation that ICV injection of CBX elicits a widespread increase in c-fos expression throughout PVN. The recently identified sodium- and aldosterone-sensitive solitary tract nucleus neurons,12 which were noted to have ascending projections, may be one source of extrinsic input to PVN.
The differences in the magnitude and timing of the sympathetic responses to ICV CBX and GA, as compared with the responses to PVN microinjection of CBX, probably reflect differences in experimental conditions (eg, drug solubility, tissue penetration, or numbers of neurons activated). They have no bearing on the relative influence of 11ß-HSD-2 activity in any specific brain region on sympathetic regulation under normal or pathophysiological conditions.
Most of the spontaneously discharging neurons we recorded in PVN had cardiac-related activity. Pulse-related activity was recently described in a small group of pre-autonomic neurons in dorsal, lateral, and ventrolateral PVN that were identified by their response to antidromic stimulation of the spinal cord.30 In that study, neurons in other regions of PVN were not tested. Thus, it remains unknown whether pulse-related activity is an exclusive property of presympathetic neurons. If so, our data would argue that presympathetic neurons are scattered diffusely throughout the PVN. An intermingling of preautonomic and neuroendocrine neurons is consistent with the integrative role of the PVN and with retrograde tracing studies that have demonstrated a broad distribution of presympathetic PVN neurons.31,32
Finally, our results are consistent with the emerging concept33 that nongenomic as well as genomic effects may be elicited by activation of classical MR. The cardiovascular and sympathetic responses to the ICV or PVN injection of 11ß-HSD-2 inhibitors appear to be too rapid in onset to be mediated by genomic actions of MR, and yet are blocked by SL. Early effects of MR stimulation that are responsive to classical MR antagonists have also been described in kidney, colon, and vasculature.33 Corticosterone activation of aldosterone-sensitive MR is an appealing explanation for the early onset of cardiovascular and sympathetic responses after ICV administration of CBX or GA. In human arteries, application of CBX permits cortisol to induce early nongenomic MR effects.34
That said, the precise mechanism by which corticosterone (or cortisol) binding to MR can induce an aldosterone-like response is the subject of continued investigation and speculation.35,36 The complexity of this issue is illustrated by the well known ability of glucocorticoids to inhibit aldosterone effects, presumably by competitively occupying the MR. For example, the ICV infusion of corticosterone blocks the hypertension induced by ICV infusion of aldosterone.37 It has recently been suggested that another effect of 11ß-HSD-2the conversion of NAD to NADHmay be important in determining the outcome of glucocorticoids binding to MR.35,38 By mechanisms not fully understood, an altered redox state induced by inhibition of 11ß-HSD-2 activity may enable corticosterone to function as a MR agonist.35,38
Perspectives
It is clear from the present experiments that 11ß-HSD-2 activity in the PVN and other brain regions effectively prevents corticosterone from inappropriately activating cardiovascular presympathetic neurons. It remains to be determined whether altered 11ß-HSD-2 activity in PVN contributes to the increased sympathetic drive that occurs in cardiovascular disease states. It is noteworthy, however, that the same physiological stressors that downregulate 11ß-HSD-2 activity in peripheral tissues (eg, angiotensin II and tumor necrosis factor
3941) are upregulated in the PVN in heart failure.21,42 In a preliminary study,43 we found that 11ß-HSD-2 mRNA is reduced in PVN of rats with experimentally induced heart failure. Thus, it is conceivable that the beneficial effects of a MR antagonist in experimental heart failure3,4,44 may be ascribed, at least in part, to blocking the effects of cortiocosterone on brain MR that are normally reserved for aldosterone by 11ß-HSD-2 activity.
| Acknowledgments |
|---|
This work was supported by National Institutes of Health RO1 HL-063915 to R.B.F., and by the Department of Veterans Affairs.
Disclosures
None.
Received January 22, 2006; first decision February 8, 2006; accepted April 19, 2006.
| References |
|---|
|
|
|---|
after coronary artery ligation in rats. Am J Physiol Regul Integr Comp Physiol. 2003; 284: R328R335.
and phorbol 12-myristate-13-acetate down-regulate human 11ß-hydroxysteroid dehydrogenase type 2 through p50/p50 NF-
B homodimers and Egr-1. Faseb J. 2005; 19: 650652.
on 11ß-hydroxysteroid dehydrogenase type 2 gene expression in heart failure rats. Circulation. 2005; 112: II191.Abstract.
in normal rats. Am J Physiol Regul Integr Comp Physiol. 2003; 285: R1402R1409.This article has been cited by other articles:
![]() |
Y. Yu, S.-G. Wei, Z.-H. Zhang, E. Gomez-Sanchez, R. M. Weiss, and R. B. Felder Does Aldosterone Upregulate the Brain Renin-Angiotensin System in Rats With Heart Failure? Hypertension, March 1, 2008; 51(3): 727 - 733. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. H. Yang, L. H. Li, S. Y. Shin, S. Lee, S. Y. Lee, S. K. Han, and P. D. Ryu Adrenalectomy Potentiates Noradrenergic Suppression of GABAergic Transmission in Parvocellular Neurosecretory Neurons of Hypothalamic Paraventricular Nucleus J Neurophysiol, February 1, 2008; 99(2): 514 - 523. [Abstract] [Full Text] [PDF] |
||||
![]() |
Z.-H. Zhang, Y. Yu, Y.-M. Kang, S.-G. Wei, and R. B. Felder Aldosterone acts centrally to increase brain renin-angiotensin system activity and oxidative stress in normal rats Am J Physiol Heart Circ Physiol, February 1, 2008; 294(2): H1067 - H1074. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Marissal-Arvy, A. Gaumont, A. Langlois, F. Dabertrand, M. Bouchecareilh, C. Tridon, and P. Mormede Strain differences in hypothalamic pituitary adrenocortical axis function and adipogenic effects of corticosterone in rats J. Endocrinol., December 1, 2007; 195(3): 473 - 484. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Fujita, K. Ando, A. Nagae, and T. Fujita Sympathoexcitation by Oxidative Stress in the Brain Mediates Arterial Pressure Elevation in Salt-Sensitive Hypertension Hypertension, August 1, 2007; 50(2): 360 - 367. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Naray-Fejes-Toth and G. Fejes-Toth Reply to Geerling and Loewy Am J Physiol Renal Physiol, July 1, 2007; 293(1): F442 - F443. [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2006 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |