The case against aldosterone: not proven

Abstract

In their paper, in the current issue [1], Fu et al. adduce three lines of evidence for aldosterone-induced osteopontin (OPN) expression in vascular smooth muscle cells (VSMCs). First, they show that spironolactone reduces vascular injury-induced neointimal hyperplasia in rat vessels in vivo. Second, aldosterone at 100 nM produces a doubling of OPN expression after 24 h (but not 6 or 12 h) exposure; OPN dose–response curves climb steadily (if disconcertingly) in a straight line for aldosterone concentrations from 1 to 100 nM. Finally, they show that MAPK, but not JUNK, pathways are involved in the acute effects of aldosterone (again at 100 nM) to elevate OPN levels. There are several minor points of detail in these findings that deserve comment, and one major issue of interpretation, as foreshadowed by the title of this commentary. The affinity of aldosterone for mineralocorticoid receptors (MR) is in the mid-to-high picomolar range: progressively increasing responses from 1 to 100 nM suggest mechanisms in addition to MR in producing the effect observed. Second, despite repeated assertions, neither spironolactone nor RU486 are specific antagonists for MR/glucocorticoid receptors (GR): spironolactone is an antagonist at androgen receptors (AR), and a progesterone receptor (PR) agonist, and RU486 is a PR antagonist. Finally, and very surprisingly, given its fivefold lower affinity for MR, spironolactone at 1 lM completely reversed the acute effect of aldosterone (100 nM) on OPN levels. The major assumption/inference is that aldosterone is the responsible pathophysiologic actor in vivo. The authors are not alone in this position: the opening sentence of the abstract reads ‘‘Osteopontin (OPN) is known to be one of the cytokines that is involved in the vascular inflammation caused by aldosterone.’’ The final sentence of paragraph 3 of the Discussion summarizes the findings thus: ‘‘These results suggest that aldosterone in its physiological and/or pathophysiological concentrations regulates OPN expression through MR in VSMCs.’’ To question both the opening sentence (... caused by aldosterone...) and the interpretation of the study findings requires a broader consideration of the promiscuity of MR, aldosterone selectivity-conferring mechanisms in epithelial cells (and VSMC), and the bivalent actions of the physiologic glucocorticoids at MR under normal conditions (antagonist) and in conditions of tissue damage (agonist). MR were the first of the MR/GR/AR/PR subfamily of nuclear receptors to branch off a common ancestor [2], are found in bony and cartilaginous fish, and antedate by millions of years the appearance of aldosterone synthase in terrestrial vertebrates. They are also spectacularly promiscuous, binding with equivalent high affinity aldosterone and deoxycorticosterone as agonists, progesterone as an antagonist, and cortisol and corticosterone as antagonist or agonist depending upon the circumstances. In epithelia, the vessel wall and the nucleus tractus solitarius MR are ‘‘protected’’ by the enzyme 11bhydroxysteroid dehydrogenase type 2 (11bHSD2), which converts cortisol and corticosterone to their receptor-inactive 11-keto analogs cortisone and 11-dehydrocorticosterone [3]. It is still commonly assumed that the enzyme protects MR from being occupied by glucocorticoids, which would require [999 of every 1,000 cortisol molecules being metabolized to cortisone in an organ (the kidney) which receives 20–25 % of the cardiac output. This was shown almost 20 years ago not to be the case [4]. What the enzyme does is debulk intracellular glucocorticoid levels of aldosterone from 100-fold to 10-fold, J. W. Funder (&) Prince Henry’s Institute, Clayton, VIC 3168, Australia e-mail: John.Funder@princehenrys.org