Aldosterone—A Hormone of Cardiovascular Adaptation and Maladaptation

Abstract

Aldosterone stimulates reabsorption of sodium, sustaining blood volume and pressure in the face of salt deprivation or extracellular fluid depletion. The steroid also stimulates excretion of potassium, protecting extracellular fluid from excessive levels of that ion. These two actions are relatively rapid and clearly adaptive when appropriately initiated and terminated, but maladaptive when prolonged or excessive, causing hypertension and electrolyte imbalance. Aldosterone and other mineralocorticoids exert slower, direct effects on cells in the heart, kidneys, and vessels, leading to hypertrophy, fibrosis, and dysfunction contributing to degenerative cardiovascular diseases. The maladaptive actions of aldosterone are exacerbated by sodium chloride, angiotensin, endothelin, and certain growth factors. Damage can be minimized by antagonists of aldosterone receptors, inhibitors of the renin system, depletion of salt, and repletion of potassium and magnesium. Specific inhibitors of fibrosis and hypertrophy, and more effective inhibitors of the renin system should be useful in the future.

Aldosterone enables animals and humans to live in environments where salt is scarce and potassium is plentiful and where dietary boluses of potassium from meat are intermittent. In other words, aldosterone allows humans and animals to live on dry land and eat each other. With such dramatic adaptive properties, aldosterone would appear to be an unambiguously beneficial hormone. However, like cortisol, thyroxine, other hormones and nutrients, and even oxygen, aldosterone in large amounts can be harmful. It can raise blood pressure above normal levels and deplete potassium within days or weeks. Adding to that well recognized set of potential liabilities, it now appears that prolonged exposure to barely elevated amounts of aldosterone exerts subtle damaging effects on the cardiovascular system, kidneys, and brain. This review emphasizes the pathologic effects of long‐term exposure to levels of aldosterone that may be only slightly higher than normal and the circumstances that aggravate or ameliorate those effects.

CLASSICAL EFFECTS OF ALDOSTERONE AND THEIR MECHANISM

Aldosterone stimulates reabsorption of sodium by the distal tubule of the nephron and a concomitant excretion of potassium, magnesium, and protons into the urine. Sodium reabsorption into the blood is also stimulated at three other sites where the ion may be excreted: the sweat glands, salivary glands, and colon. All of the affected cells are epithelial, and the actions of the hormone are lumped under the term epithelial effects. Stimulation of aldosterone secretion by angiotensin II and inhibition by atrial natriuretic peptides integrates the steroid into the body's regulation of fluid volume. Stimulation of secretion by extracellular potassium integrates the steroid into the body's defenses against potassium loads. Although there are other adrenal regulators that can affect aldosterone secretion, their homeostatic roles are largely unknown.

Aldosterone acts on renal tubular epithelium by binding to receptors, called mineralocorticoid receptors (Mrs), which are similar but not identical to glucocorticoid receptors. The complex of receptor and bound hormone affects the genetic (nuclear) apparatus of target cells, resulting in increased numbers of epithelial sodium channels. The sodium‐retaining effect of aldosterone is not only classical and traditional, it is epithelial, receptor‐mediated, nuclear, and genomic. Some of the other effects of aldosterone fall outside one or more of these categories. For example, aldosterone exerts a direct effect on pressor centers in the brain, elevating systemic blood pressure by an effect that is nonclassical and nonepithelial but is mediated by Mrs in the brain. Aldosterone also has a direct vasoconstrictor effect on vascular smooth muscle, another nonclassical and nonepithelial action that does not affect the genome of the muscle cell but contributes to elevated arterial pressure.

PATHOLOGIC EFFECTS OF ALDOSTERONE

The best‐known deleterious effects of aldosterone can be predicted from its best‐known adaptive actions. Excessive retention of sodium (along with effects on the brain and vascular smooth muscle) leads to hypertension. Prolonged excretion of potassium, magnesium, and protons leads to hypokalemic alkalosis with hypomagnesemia. Aldosterone alone does not cause edema, but the hormone contributes to the edema and effusions that accompany cardiac and hepatic failure and some renal diseases. Some of the histopathologic effects described below are probably the combined result of direct effects of aldosterone on the affected target tissues and indirect results of the elevated pressure and perturbed electrolyte balance.

Evidence for direct pathologic effects of aldosterone comes from four types of experiments: 1) administration of aldosterone or other mineralocorticoids like deoxycorticosterone to animals; 2) administration of aldosterone antagonists to animals experiencing pathologic processes; 3) administration of aldosterone antagonists to human patients; and 4) observations of humans with excessive aldosterone production. The systems that have been studied most extensively are the kidneys, heart, arteries, and brain.

In various animal models of renal disease, aldosterone has been implicated as a cause or contributor to pathologic hypertrophy of the glomerular mesangium and fibrosis of the renal interstitium. Connective tissue overgrowth also affects arteries in the kidneys (as well as in the heart, brain, and periphery) of animals given exogenous aldosterone. In humans with preexisting renal disease, especially diabetic glomerulosclerosis, aldosterone antagonists reduce proteinuria. This strongly suggests that aldosterone contributes to pathologic lesions in the kidneys of humans as well as rats, even though the levels of aldosterone in the diabetic subjects were not high enough to qualify as “hyperaldosteronism.”

In animal models, administration of aldosterone or another mineralocorticoid in supraphysiologic amounts causes cardiac hypertrophy. The overgrowth affects cardiac muscle, but even more dramatically stimulates fibrosis. A similar effect apparently occurs in humans because patients with frankly elevated levels of aldosterone from adrenal tumors or hypertrophy have, on average, increased ventricular wall thickness on both the right and left sides of the heart. This hypertrophy is only partly attributable to the pressor effects of aldosterone, since arterial hypertension would not fully explain right ventricular overgrowth, and the degree of hypertrophy is greater than that seen in patients with comparable pressures from other causes.

Two large studies of aldosterone antagonists in humans with heart disease showed that the drugs can reduce cardiovascular mortality and morbidity. The clinical benefits could be taken as evidence that the drugs blocked aldosterone's direct effects on pathologic changes in human hearts. The survival curves in these studies showed a gradually widening difference between experimental and control groups, again suggesting that the drugs interfered with an ongoing process, like fibrosis. On the other hand, aldosterone antagonists contribute to repletion of intracellular potassium and magnesium and excretion of sodium, and such effects could produce clinical benefits that persist for years. As mentioned below, other data from those human trials offer additional circumstantial evidence that the drugs blocked a direct pathogenic effect of aldosterone.

In animals, aldosterone or related mineralocorticoids cause hypertrophic changes in cerebral vessels and, at the same time, worsen the effects of experimental cerebral vascular occlusions. The evidence that this occurs in humans is the observation that people with hyperaldosteronism have an unusually high incidence of stroke. As in the case of heart disease, the incidence of cerebrovascular events exceeds the statistical expectation deduced from the elevation of arterial pressure alone.

MECHANISMS OF PATHOLOGIC ACTIONS OF ALDOSTERONE

There is no evading the deleterious effects of hypertension caused by excessive aldosterone, especially on the left ventricle, arteries, and brain, but there are other ways the hormone can cause cardiovascular damage.

Aldosterone can cause total‐body potassium depletion if sodium intake is sufficient to place sodium in the distal tubule where its reabsorption wastes potassium. Potassium depletion impairs cellular protein synthesis, lowers insulin secretion, interferes with renal tubular function, affects muscle strength, and predisposes the heart to tachyarrhythmias. Potassium intake in the modern Western world is much lower than it was during prehistoric eras. At the same time, sodium intake is higher, so it is easier for aldosterone to deplete potassium today than it was in the remote past when aldosterone was necessary for survival in a salt‐poor environment.

Accompanying renal excretion of potassium in hyperaldosteronism is excretion of magnesium. Magnesium depletion, like potassium depletion, predisposes the heart to tachyarrhythmias and also predisposes arteries to vasoconstriction. It is difficult to assess magnesium depletion, since the bulk of the electrolyte is intracellular. Measurements of red cell magnesium content are helpful but seldom performed; therefore, the contribution of magnesium depletion to the deleterious effects of aldosterone on the heart is unknown.

Oxidative stress is caused by production of reactive molecules that contain oxygen, such as hydrogen peroxide and superoxide ion, in amounts that exceed the capacity of protective enzymes like catalase and superoxide dismutase. The reactive oxygen species that evade degradation can damage cell proteins and nucleic acids, can oxidize lipids to form biologically active derivatives and alter cell membranes, and can disrupt physiologic regulation. Aldosterone in excess stimulates formation of reactive oxygen species. The immediate sources of the toxic molecules are oxidative enzymes like xanthine oxidase and nicotinamide adenine dinucleotide phosphate oxidase. The latter is found in blood vessels where there are receptors for aldosterone, completing a package that is potentially toxic to arteries.

Inflammation is protective and restorative in the face of trauma and infection but destructive when it arises inappropriately or lasts beyond its usefulness. The role of inflammation in atherosclerosis, especially in the so‐called unstable plaque, is well accepted. Aldosterone and other mineralocorticoids can exert proinflammatory effects in experimental animals that involve attraction and stimulation of blood‐borne mononuclear cells. In support of the hypothesis that aldosterone stimulates inflammation and fibrosis in humans, administration of aldosterone antagonists in one of the large heart failure trials mentioned above lowered circulating levels of collagen fragments and other markers of inflammation.

Aldosterone stimulates synthesis of an inhibitor of plasminogen activation (PAI‐1). Increased levels of that inhibitor favor persistence of blood clots, and that may be one way aldosterone promotes atherosclerosis.

Although the classical effects of aldosterone, those that can result in elevated blood pressure because of sodium retention, are mediated by mineralocorticoid receptors that affect gene expression, it has become apparent in recent years that aldosterone also activates membrane‐bound receptors that control intracellular signal cascades separate from the genome. Membrane‐bound receptors trigger vascular smooth muscle contraction in response to aldosterone, and that response is much more rapid than the increase in renal sodium reabsorption mediated through gene expression. Receptors in the cell membrane are not as easily blocked by standard MR antagonists as nuclear receptors, but they fulfill all the criteria of receptors, exhibiting structural specificity, high affinity, and coupling to classical signal cascades.

CHEMICAL COCONSPIRATORS

Both the classical and histopathologic effects of aldosterone are amplified and, in some cases, mediated by other molecules. Sodium chloride is the best known adjunct to mineralocorticoid action. Without ample salt, aldosterone has minimal effects on blood pressure or excretion of potassium and magnesium. In fact, some of the highest levels of aldosterone are found in the blood of primitive people who eat very little salt, yet their pressures and potassiums are normal and their cardiovascular structures pristine. By contrast, salt supplements can expose pressor, kaluretic, and pathologic effects of small amounts of mineralocorticoids in animals and, presumably, in humans. This synergy supports the hypothesis that apparently normal levels of aldosterone can be responsible for hypertension and related diseases when those levels are “inappropriately high” for the quantity of salt ingested. High‐salt intake by itself is epidemiologically associated with cardiovascular deterioration, but it is unknown if that effect depends on the presence of aldosterone. Clearly though, the combination is highly dangerous.

Angiotensin peptides not only stimulate aldosterone secretion, and exert their own direct effects on vascular smooth muscle and renal tubules, they also amplify some of the nonclassical actions of aldosterone. Levels of PAI‐1 are increased by aldosterone or by angiotensin peptides alone, but the increase is markedly greater if the two hormones are administered together. Oxidative stress and inflammation are also induced more potently by experimental combinations of steroid and peptide than by either alone. The association between a hyperactive renin‐angiotensin‐aldosterone axis and cardiovascular disease cannot be attributed to any one component of that system—aldosterone and angiotensin are coconspirators.

The MR that acts on gene expression and increases epithelial sodium channels is avid for cortisol as well as aldosterone. Selectivity for aldosterone is conferred by a nearby enzyme that oxidizes cortisol but does not attack aldosterone. This enzyme is not located near all Mrs. In sites where the enzyme is less active, such as the brain, cortisol could induce the same MR‐mediated pathologic events as aldosterone. This may explain some of the cardiovascular accompaniments of stress.

An ingredient in licorice inhibits the enzyme that protects Mrs from cortisol, accounting for the hypertensive effect of that candy. Human plasma and urine contain endogenous enzyme inhibitors that mimic the action of licorice. Variations in endogenous enzyme inhibitors could explain why people with identical aldosterone and cortisol levels differ in their apparent mineralocorticoid effects.

Throughout this review, the term aldosterone has been used when the generic term mineralocorticoid is more appropriate. We should remember that the adrenal gland secretes other steroids apart from cortisol with actions like aldosterone, such as deoxycorticosterone. Those hormones have direct effects on epithelial and nonepithelial targets, but they are not usually measured in humans. If they were measured, the data might help to explain variations in target organ responses and pathology among humans with no significant elevations in plasma levels or urinary excretion of aldosterone itself.

Evidence from animal experiments and human observations indicate a gender difference in histopathologic responses to mineralocorticoids. In some ways, females are more sensitive and in other ways less so. Experimental ovariectomy converts female rats to the equivalent of males, so the gender differences must be caused by sex steroids.

All of the pathologic situations mentioned above, which aldosterone is known to cause or complicate, involve other hormones and modulators that surely work in concert with aldosterone to produce disease states. In the case of mineralocorticoid hypertension, other pressors play their customary roles. Elevated levels of mineralocorticoids sensitize vascular smooth muscle to catecholamines, so catecholamines themselves should be thought of as partners of aldosterone, amplifying its pressor and probably its pathologic effects.

In inflammation, atherosclerosis, oxidative stress, fibrosis, and hypertrophy, many known and unknown mediators and messengers participate; there is nothing about aldosterone that would suppress these other influences. Fibrosis and hypertrophy, for example, are regulated by a large number of growth factors and growth inhibitors, probably many more than we recognize today. Thickening of the walls of cerebral vessels in rats given mineralocorticoid, an effect inhibited by MR antagonists, is also inhibited by interruption of the actions of epidermal growth factor (EGF). In other models, endothelin, a polypeptide vasoconstrictor derived from endothelium, has been implicated as a mediator of the toxic vascular environment induced by mineralocorticoids. (Endothelin also mediates many of the vascular effects of angiotensin.) The promotion of inflammation by aldosterone involves participation by a number of cytokines that attract inflammatory cells, affect extracellular matrix, and modulate immune responses. It will be some time before experiments clarify the pathogenic sequence that begins with aldosterone binding to its receptors, proceeds by enlisting a variety of other signaling molecules, and results in histopathologic changes in the blood vessels, heart, and kidneys. It will be even longer before we know which of these steps is critical in human diseases, but that knowledge will provide many opportunities for pharmacologic interruption of the diseases.

We can safely assume that atherogenic molecules like LDL cholesterol, oxidized lipoprotein phospholipids, and homocysteine damage arteries during aldosterone excess, just as they do when hormone levels are normal.

PREDISPOSING DISEASES

Until the recent experiments with aldosterone antagonists were performed, it was assumed that aldosterone exerted pathologic effects only when its circulating levels were abnormally high. Adrenal hypertrophy or adenomas are not the only situations with increased levels of the steroid. The adrenal glomerulosa, the zone that produces aldosterone, is different from the zone that produces cortisol, for example, in its responsiveness to a wide variety of hormones and other molecules in vitro. It is not absolutely certain which of these potential regulators affect aldosterone secretion in living humans. There is excellent evidence for stimulation of the zona glomerulosa in humans by angiotensins, potassium, and adrenocorticotropic hormone, and inhibition by atrial natriuretic peptides and dopamine. Just considering those regulators, it can be assumed that aldosterone levels may be “inappropriately high” in patients with renal artery stenosis, renal insufficiency with hyperkalemia, and conditions that induce adrenocorticotropic hormone secretion.

If some of the other putative adrenal stimuli are active in humans, aldosterone could be expected to be elevated, for example, in vasculopathies where endothelin is produced and in those diseases where there is unusual production of serotonin, bradykinin, or oxidized fatty acids. A significant proportion of hypertensive humans, approximately 25%, have circulating levels of aldosterone that are normal or slightly elevated but higher than would be predicted from their renin activity (an indicator of angiotensin production) or serum potassium concentration. Those subjects are said to have low renin essential hypertension, and their hypertension looks very much like a manifestation of an unrecognized adrenal stimulus. It seems paradoxical that there is no evidence that subjects with low renin essential hypertension have an unusually high prevalence of cardiovascular disease of the type attributed to aldosterone.

MR antagonists have beneficial effects in animal models of hypertension and renal disease and in humans where circulating levels of aldosterone are normal or only slightly higher than normal. There are (at least) two explanations for this phenomenon: 1) the ill effects of the steroid come from prolonged exposure to modest amounts of the steroid; or 2) damage reflects unusual sensitivity to near‐normal or normal levels.

Humans undoubtedly differ in their sensitivity to the pathologic effects of aldosterone. In diseases marked by hyperaldosteronism, such as glucocorticoid‐remediable aldosteronism, blood pressures differ widely in subjects with very similar plasma levels and urinary excretion of steroid. The reasons for this heterogeneity are poorly understood. Some conditions, however, can be assumed to predispose humans to the damaging effects of mineralocorticoids. They include chronic kidney disease, essential hypertension, diabetes, obesity, sleep apnea, and the metabolic syndrome. Hyperlipidemia and smoking would fan any atherogenic fires ignited by aldosterone. Advanced age probably predisposes to pathologic changes induced by aldosterone, if for no other reason than it implies a prolonged exposure to the steroid as well as fragile cardiovascular, cerebrovascular, and renal structures.

Obesity presents an interesting compilation of factors that increase risks associated with aldosterone. It is frequently marked by hypertension that resists drug treatment, obstructive sleep apnea, dyslipidemia, diabetes, inflammation, and oxidative stress. Recent observations suggest that obesity is also marked by inappropriately high levels of aldosterone in the blood and urine. Aldosterone antagonists are surprisingly effective in lowering the blood pressure of obese patients with resistant hypertension. Weight loss ameliorates all of these risk factors, including overproduction of aldosterone, suggesting that the initiating cause is somehow related to excess fat.

ANTIDOTES AND PALLIATIVES

The obvious antidote to mineralocorticoid excess is a drug that blocks the mineralocorticoid receptor. Two MR antagonists are available: spironolactone and eplerenone. These drugs inhibit the classical effects of aldosterone on blood pressure and electrolytes. They also proved efficacious in reducing mortality in trials on select humans with heart failure. By virtue of being competitive inhibitors of aldosterone, spironolactone and eplerenone moderate, rather than ablate, the effects of this adaptive steroid. Depending on the amount of drug, high levels of aldosterone can overcome the inhibition. If the adrenal gland is working and responsive to angiotensin and potassium, any fall in renal perfusion or rise in serum potassium would stimulate aldosterone secretion and tend to overcome pharmacologic MR antagonism. That moderation helps to prevent excessive hypotension or hyperkalemia; it helps, but does not totally prevent those unwanted effects. Patients with impaired renal excretory function, for example, may become dangerously hyperkalemic when treated with aldosterone antagonists. The drugs have other side effects, unrelated to mineralocorticoid antagonism, that can limit their acceptability by patients.

Spironolactone and eplerenone have been the critical reagents in experiments with animals that impute aldosterone in the histopathologic changes described above. There is some evidence that the drugs do the same things in humans. In particular, spironolactone reduced proteinuria in diabetic hypertensive patients to a degree beyond that expected from its hypotensive effect.

Angiotensin‐converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs) reduce aldosterone secretion acutely. That reduction wanes in a matter of months, largely because inhibition of angiotensin generation (by ACEIs) or angiotensin action (by ARBs) is partially overcome by normal adaptive mechanisms, including increased secretion of renin. Although the drugs don't maintain very low levels of aldosterone, they lower levels or effectiveness of angiotensin (again, only partially). That reduces amplification of aldosterone by the renin‐angiotensin cascade, so the net effect is beneficial. When renin inhibitors become available, they will further the degree and duration of blockade of aldosterone production. With awareness of the danger of hypotension and hyperkalemia, clinicians will be able to orchestrate appropriate reductions in aldosterone levels using one, two, or three blockers of the renin‐angiotensin‐aldosterone system.

In view of the integral role of salt in the toxicity of aldosterone, limiting salt intake or increasing salt excretion should lessen damage from aldosterone. To increase sodium excretion, logic suggests that amiloride be the diuretic of choice, because it inhibits the same sodium channel (epithelial sodium channel) affected by aldosterone. On the other hand, amiloride is not a very potent depressor agent when used alone. Its diuretic effect raises circulating levels of angiotensins and aldosterone while it lowers blood pressure. Perhaps it is no wonder that amiloride has not been shown to prevent the histopathologic effects induced or aggravated by aldosterone. Whatever agent or combination is required, pressure reduction is central to preventing pathology that might be mediated or worsened by aldosterone. Drugs that block catecholamines or calcium channels would have the theoretic appeal that they interfere with chemical coconspirators of aldosterone while they lower blood pressure, but there is no clinical evidence that they are better at blocking histopathologic effects than other antihypertensives.

By increasing potassium excretion when necessary, aldosterone helps to maintain the appropriate gradients of potassium across cell membranes, gradients that sustain rhythmicity and responsivity of the heart. In the presence of adequate sodium, this otherwise adaptive property of aldosterone can be excessive, depleting intracellular potassium and magnesium. Therefore, another critical factor in treating or preventing aldosterone's cardiovascular damage is repletion of potassium and magnesium. MR antagonists and other potassium‐sparing diuretics are more likely to succeed than supplements alone.

Risk factor management is a cornerstone of prevention and therapy in cardiovascular disease, and the fact that aldosterone is a relatively newly recognized player does not lessen the importance of others. Smoking, excess body fat, sleep‐disordered breathing, hyperlipidemia, hyperhomocysteinemia, and diabetes should be addressed with as much vigor in patients with elevated aldosterone levels as in those with normal levels. There is no direct evidence that antioxidant or anti‐inflammatory modalities are beneficial in preventing cardiac hypertrophy, cardiovascular fibrosis, or renal disease associated with aldosterone, although there is logic to the idea.

In brief, aldosterone in the right amounts, under appropriate circumstances, and for reasonable periods of time can be adaptive and life‐saving, while higher levels under certain circumstances and for extended times can be pathogenic. We don't know exactly the optimal degrees and durations of mineralocorticoid effects under the many environments and comorbidities patients experience today. Still, we have plenty of ways to adjust these hormones and their actions, so we can help our patients live on dry land and enjoy potassium‐rich foods for much longer than our short‐lived ancestors.