Classics in Cardiovascular Endocrinology: Aldosterone Action Beyond Electrolytes

In 1856, Charles-Edouard Brown-Sequard demonstrated that adrenalectomy was fatal in laboratory animals (1). He interpreted these experiments as evidence that the adrenals produced substances that are essential for life. The early crude adrenal extracts contained several properties: stimulating hepatic glycogenolysis, increasing blood pressure (BP), increasing sodium retention, and even some androgenic activity. When techniques to isolate organic compounds became available around the turn of the century, Takamine identified epinephrine as the hypertensinogenic substance from the medulla (2). Kendall and coworkers (3) and Reichstein (4) independently isolated cortisone from the cortex, which was the primary “cortin” controlling carbohydrate metabolism. Selye, however, had introduced the terms “glucocorticoid” and “mineralocorticoid” to distinguish these properties and to suggest that different compounds possessed these activities (5). Kendall and others noted that the sodium-retaining hormone or “electrocortin” activity was concentrated in an “amorphous polar fraction” generated during cortisone preparation from adrenal extracts. In the early 1950s, Simpson, Tait, Reichstein, and their coworkers isolated aldosterone from this fraction (6, 7), and its secretion from the zona glomerulosa under angiotensin II (AngII) stimulation was later documented (8). Thus, the renin-angiotensin-aldosterone system (RAAS) was identified as a closed-loop system, in which renin derived from the kidney during volume depletion catalyzed AngII generation, which stimulated aldosterone production, which then acted on the kidney to increase sodium retention and volume expansion, thus attenuating renin secretion. When Conn described hypertension and hypokalemia as a syndrome of autonomous electrocortin secretion from an adrenal tumor about the same time as aldosterone was isolated (8), the entire physiology of aldosterone appeared to be explained by the RAAS, invoking the adrenal and kidney alone.

The 2 papers I have chosen as classics in cardiovascular endocrinology for Endocrinology are landmark papers that demonstrate aldosterone actions outside the RAAS. The first paper (Gomez-Sanchez [9]), explored the action of aldosterone on the brain. Models of mineralocorticoid-induced hypertension such as metyrapone treatment of dogs (10) and deoxycorticosterone acetate implants in pigs (11) sometimes found increases in cardiac output and peripheral vascular resistance, which were opposite the predicted changes from volume expansion alone. Studies with radiolabeled steroids found aldosterone binding sites not only in the kidney and colon but also in vascular smooth muscle cells (12) and several brain regions (13). Could aldosterone directly act on the brain to increase BP?

To answer this question, Gomez-Sanchez (9) treated Sprague-Dawley rats after single nephrectomy with intracerebroventricular (ICV) low-dose (5 ng/h) aldosterone infusion or vehicle and compared the BP changes induced in these animals to those with peripheral infusion of low or high dose (5 or 500 ng/h) aldosterone for 4 weeks. All animals received 0.9% sodium chloride instead of drinking water and standard chow, and systolic BP (SBP) rose from 103 to 121 or 123 mmHg in the vehicle-treated control animals or animals receiving 5-ng/h aldosterone peripherally, respectively. In contrast, the rise in SBP was significantly higher with ICV infusion of 5-ng/h aldosterone to 146 mmHg, even compared with peripheral infusion of 500-ng/h aldosterone (136 mmHg). To demonstrate the specificity of these results, a second experiment was performed in which the SBP rise in response to ICV infusion of aldosterone was blocked by coadministration of the mineralocorticoid receptor antagonist prorenone, and prorenone alone gave similar SBP changes as vehicle controls.

Simple yet elegant, these data validated the effect of mineralocorticoids on the brain and a mechanism of pressor action independent of electrolyte handling. Subsequent studies have shown that activation of mineralocorticoid receptors in the brainstem increases sympathetic nerve activity (14). The clinical implications of this paper are evident today. Patients with primary aldosteronism have increased sympathetic nerve activity similar to those with essential hypertension, and surgical cure reduces sympathetic trafficking to normal (15). Furthermore, mineralocorticoid receptor blockade with spironolactone lowers BP in anephric hemodialysis patients with refractory hypertension (16).

The second paper (Rocha et al [17]) provided compelling evidence for direct actions of aldosterone on the vasculature in the heart and kidney. In the spontaneously hypertensive stroke-prone rat model, spironolactone greatly reduced damage to the vasculature in the brain (18) and kidney (19), effects that exceeded that anticipated from BP reduction alone. Conversely, chronic aldosterone administration caused myocardial fibrosis in rats, disproportionate to the hypertension burden (20). Was the vascular damage a direct effect of aldosterone beyond an indirect effect of BP?

To explore this paradox, the Rocha et al (17) gave male Wistar rats 1% sodium chloride for drinking alone as controls and to the others also administered the nitric oxide synthase inhibitor N^ω-nitro-L-arginine methyl ester (L-NAME) for 2 weeks with AngII the last 3 days to induce hypertension. These rats were divided into 4 treatment groups: L-NAME + AngII alone, L-NAME + AngII + eplerenone, L-NAME + AngII + adrenalectomy, and L-NAME + AngII + adrenalectomy + aldosterone. BP rose in all groups compared with saline-only controls. Renin was suppressed in all the L-NAME + AngII groups except those animals who also underwent adrenalectomy without aldosterone replacement. Despite similar BPs as the other 2 groups, rats treated with either adrenalectomy or mineralocorticoid receptor blockade using eplerenone showed reduced cardiac hypertrophy normalized to body weight and greatly reduced areas of myocardial necrosis. In the kidney, the L-NAME + AngII treatment with salt caused the development of prominent microvascular lesions, which included myointimal proliferation, thrombosis, and fibrinoid necrosis. Adrenalectomy or coadministration of eplerenone greatly ameliorated this pathology. The conclusion of these experiments is that aldosterone, or more precisely mineralocorticoid receptor agonists, act directly on the vasculature in states of sodium-dependent hypertension to cause inflammation and fibrosis. I should be cautious here, as a many studies suggest that cortisol and corticosterone, acting through both mineralocorticoid and glucocorticoid receptors, also contribute to cardiovascular damage in certain pathologic situations (21).

Today, we recognize the profound implications of direct aldosterone action on the vasculature in the heart and kidney. Patients with primary aldosteronism have disproportionate cardiac fibrosis (22, 23), nephropathy (24), and markedly elevated risk of atrial fibrillation (25, 26) for the degree of hypertension. Fortunately, this damage is substantially reversed with surgical cure or with mineralocorticoid receptor antagonist therapy (27–29). Large clinical trials, including RALES (30) and EPHASUS (31), have confirmed the benefit of mineralocorticoid receptor antagonists in the treatment of heart failure, and these drugs are now standard of care. Indeed, even populations such as hemodialysis patients who are presumably less likely to benefit from spironolactone show reduced cardiovascular events and improved survival with treatment (32). Electrocortin might have been a clever name for the “polar substance” known to alter electrolyte balance, but aldosterone has proven to be so much more.