The Promiscuous Mineralocorticoid Receptor
See related article, pp 897–905
The article by Briet et al1 in this issue, “Aldosterone-induced vascular remodeling and endothelial dysfunction require functional angiotensin type 1a receptors,” represents a prodigious amount of work by the authors: it is not surprising that the first 3 are credited as having equally contributed. The authors build on previous in vitro studies, generously cited, on the interactions between mineralocorticoid receptor (MR) activation by aldosterone and angiotensin type 1 receptor (AGTR-1) activation by angiotensin II. The hypothesis they tested in the present study was that the vascular remodeling and functional effects of MR activation would be blunted in AGTR-1a null (henceforth KO) mice.
In this they have clearly succeeded: as the opening sentence of the perspectives section puts it, “This is the first in vivo study to demonstrate that AGTR-1a is critical for aldosterone-induced vascular remodeling and endothelial dysfunction.” Like many ground-breaking studies, however, their new findings prompt new questions. Some of these refer to the lightly addressed cardiovascular differences in the basal state between wild-type (WT) and KO littermates; others the counterintuitive higher blood pressure elevation in response to aldosterone in KO rats; another, crucially important in terms of translation, is the possible roles of the physiological glucocorticoids (corticosterone in mice, cortisol in humans) in mimicking or blocking the effect obtained with aldosterone in the present experimental studies.
Before looking at each of these new questions, I will make several observations on the paper as a whole. It is clearly and linearly written, and the very impressive amount of work involved comes through. A minor criticism is the difference between the ≈30 (WT) and ≈50 (KO) mm Hg blood pressure rise cited in the abstract and the actual figures (23 and 44 mm Hg) in the Results section. A criticism not confined to the present article is the use of megadoses of agents—600 μg/kg per day of aldosterone, losartan at 10 mg/kg per day, and eplerenone at 100 mg/kg per day: <1% of this latter dose on a weight basis as monotherapy produced substantial falls in blood pressure in over 40% of essential hypertensives and a dose of <2% in 80% of such subjects.2 As previously wryly noted, such largesse makes extrapolation to physiology difficult.3
Finally, the section in the Discussion quoting Shibata et al4 on renal intercalated cells is not consistent with the data in that paper. Dephosphorylation of the remarkable S843-phosphorylated MR so that it can be activated is driven by angiotensin: it is thus unlikely to be a player in salt-loaded animals. What the Shibata study also showed is that the dephosphorylated MR was equivalently activated by aldosterone and cortisol (Figure [E and F]).
Back to the new questions. As the authors note, KO mice had a 35% greater aortic wall cross-sectional area at baseline, in response to 1% NaCl solution for 14 days, and similarly as shown in their Figure 3, baseline media collagen was >2-fold higher in KO and fibronectin ≈3-fold higher than in WT; the fluorescence, as a measure of ROS generation, was ≈4-fold higher in KO than in WT-absent aldosterone. These values in KO were unaltered by aldosterone; in WT, they were elevated to levels seen in KO with or without aldosterone. The authors acknowledge that this baseline elevation in KO might have blunted aldosterone-induced remodeling: true, but the question remains of how the baseline levels were increased, and an attempt to address the possible mechanisms involved would be welcome. Similarly, the doubling of daily sodium excretion (and presumably intake) by KO rats goes essentially unremarked.
The counterintuitive finding, of an ≈2-fold greater increase in blood pressure in KO than in WT mice in response to aldosterone, is one which to the authors credit they confirmed in a second study. A possible explanation for this finding is an effect via AGTR-1b, given no equivalent exaggerated elevation in losartan-treated rats, in which both AGTR-1 are blocked: this seems unlikely, in that the levels of aldosterone infused presumably reduce both renin and angiotensin to extraordinary low levels. Second, they propose that the observed difference in immediate (but not prolonged) response to a saline challenge between KO and WT—presumed to cause a sodium-loaded state in the KO, elevating blood pressure—is not really convincing, given the difference in baseline salt and water status. Finally, a 24% lower level of expression of Kcnmb1 is suggested as a possible contributor to increased myogenic tone, and thus of blood pressure levels. A more succinct presentation and discussion would have been welcome, and a conclusion that in all probability the question remains unanswered.
Now to the question crucial for physiology, pathophysiology, and translation. It has long been recognized that MR are highly promiscuous in that they bind corticosterone, cortisol, and aldosterone with the same high affinity. Under a variety of in vitro test conditions, the physiological glucocorticoids have been shown to have activity equal to aldosterone to 100-fold less potent. Epithelial tissues (and the vessel wall) express the enzyme 11βhydroxysteroid dehydrogenase type 2, which allows aldosterone selectively to activate MR: in such tissues, glucocorticoid-occupied MR are held inactive by as-yet-to-be adequately described redox-based mechanisms (Figure [A]). Where 11βhydroxysteroid dehydrogenase type 2 is not expressed, as in the myocardium, MR are overwhelmingly occupied (but normally not activated to mimic aldosterone) by corticosterone/cortisol (Figure [C]). In the context of tissue damage and ROS generation, MR–glucocorticoid complexes are activated, mimicking the effects of aldosterone (Figure [B and D]).
In vitro, this was shown by Mihaildou et al in studies in Langendorf ischemia/reperfusion preparation in rat hearts.5 Area at risk and infarct size were increased by aldosterone and decreased with spironolactone as previously shown6; importantly, they were equivalently increased by low nanomolar cortisol, blocked by spironolactone but not the glucocorticoid receptor/progesterone receptor antagonist RU486.
In vivo, that endogenous glucocorticoids are capable of sustaining ongoing vascular damage was shown by Young et al in 4- and 8-week studies on deoxycorticosterone acetate/salt rats.7 After 4 weeks, blood pressure was raised, and the rats divided into 4 groups—continued deoxycorticosterone acetate/salt, continued salt plus eplerenone, continued salt alone, and a control group killed at 4 weeks. In the group continuing on deoxycorticosterone acetate/salt for the full 8 weeks, blood pressure rose further, to plateau levels. In rats receiving salt and eplerenone, blood pressure (and other indices of cardiovascular damage) returned to baseline. In those continuing on salt but with no administered steroid, BP remained elevated over the following 5 to 8 weeks, and when killed, they were indistinguishable from control rats killed at week 4. The interpretation of these findings is that in the context of tissue damage, normal levels of glucocorticoids are sufficient to maintain elevated blood pressure and other cardiovascular sequelae.
In essential hypertensives, with normal levels of plasma electrolytes, renin, and aldosterone, administration of 50 to 200 mg/d of eplerenone as monotherapy substantially lowered blood pressure in 80% of subjects, as previously cited.2 This fall in blood pressure appeared unrelated to fluid and electrolyte effects; again, the inference is that what it reflects is an antagonist action of eplerenone on cortisol-activated MR in damaged resistance vessel walls.
As previously noted, there is no question that the authors have shown that experimentally aldosterone is effective in the measures tested. The question, however, is 2-fold. The first is whether an MR activated by corticosterone in the tissue-damaged mouse has the same effect as that shown with aldosterone: if this is the case, then given their relative concentrations, pathophysiologically aldosterone is not a player outside of experimental overloading (and possibly, but not probably, primary aldosteronism).
The second question is one of physiology, where angiotensin and aldosterone are coupled, and absent tissue damage vascular MR are physiological aldosterone receptors. Particularly, in the context of postural change, an interaction, additive rather than permissive, between aldosterone-induced MR activation and angiotensin acting via AGTR-1 might most probably—and perhaps most profitably—be sought.
Sources of Funding
This work was supported by the Victorian Government’s Operational Infrastructure Support Program.
The opinions expressed in this article are not necessarily those of the editors or of the American Heart Association.
- © 2016 American Heart Association, Inc.
- Briet M,
- Barhumi T,
- Mian MOR,
- Coelho SC,
- Ouerd S,
- Rautureau Y,
- Coffman TM,
- Paradis P,
- Schiffrin EL
- Shibata S,
- Rinehart J,
- Zhang J,
- Moeckel G,
- Castafieda-Bueno M,
- Stiegler AL,
- Boggon TJ,
- Gamba G,
- Lifton RP
- Mihailidou AS,
- Loan Le TY,
- Mardini M,
- Funder JW