Hypothesis on a relationship between hyperaldosteronism, inflammation, somatic mutations, and autoimmunity

After the description of the adaption syndrome, Selye hypothesized in 1949 that desoxycorticosterone (DOC), the first mineralocorticoid to be discovered, could induce an inflammatory effect. This hypothesis was based on the effect of DOC in worsening the symptoms of rheumatoid arthritis and inducing a strong inflammatory effect in animal models.1

Similar activity was also supposed for aldosterone, when it was discovered in 1953. Edith Glàz and Paul Vecsei in 1971 reported in their book on aldosterone that “Relying upon the fundamental concept of Selye, maybe it will be possible in the future to demonstrate that DOC, which has recently been shown to be secreted in significant amounts in the human adrenal vein, plays an important part in the pathogenesis of inflammatory diseases. No observations that might exclude this possibility have been made so far.”2

This concept was later abandoned for several decades and mineralocorticoids were only studied for their sodium and water‐retaining activity and related hypertension.

The inflammatory effect of aldosterone was later revisited in cardiovascular diseases, in particular, by Pitt and colleagues,3, 4 who pointed out the importance of aldosterone receptor blockers to prevent the relapse of cardiovascular or heart accidents even in patients with normal aldosterone values.

In 1985, we characterized mineralocorticoid receptors (MRs) in human mononuclear leukocytes,5 where aldosterone regulates intracellular electrolytes6 and volume.7 Subsequently, we demonstrated that incubation of mononuclear leukocytes with aldosterone increases the expression of two specific markers of inflammation, and this effect is blocked by coincubation with the MR antagonist canrenone.8 Nongenomic effects of aldosterone were also evaluated in mononuclear leukocytes and recently in erythrocytes.9 Both leukocytes and erythrocytes can interplay and contribute to the local inflammatory reaction, bringing their MRs in the site of inflammation, and this process could be modulated by canrenone, emphasizing a possible role of classical MRs in the genomic and nongenomic effects of aldosterone.

Inflammation and autoimmunity are strictly linked and recent studies are focusing on the relationship between aldosterone and autoimmune disorders. In particular, Herrada and coworkers10 demonstrated a direct action of aldosterone in promoting the induction of T‐helper 17 polarization of CD4+ T cells, which have been associated with the promotion of many organ‐specific autoimmune diseases. Recently, we also reported an increased association of Hashimoto thyroiditis in patients with primary aldosteronism (PA).11 From these studies it could be hypothesized that MR blockers could also modulate the onset and/or evolution of some autoimmune disorders, suggesting new therapeutic targets for MR antagonists.

THE ROLE OF ANTI–TYPE 1 ANGIOTENSIN RECEPTORS ANTIBODIE AND INFLAMMATION IN PA

Recent studies have shown an association between elevated levels of autoantibodies against the angiotensin type 1 receptor (AT1R‐AAs) and primary and secondary aldosteronism.

Wallukat and colleagues12 first reported an elevated titer of AT1R‐AAs in patients with preeclampsia, demonstrating a stimulatory effect of these autoantibodies. Subsequently, many studies showed that AT1R‐AAs activate AT1R on many different cells, such as cardiac myocytes, trophoblast cells, endothelial cells, and mesangial and vascular smooth muscle cells, causing biological responses that are relevant to the pathophysiology of preeclampsia. These findings support our previous study showing an increased effector mechanism of aldosterone in preeclampsia,13 probably mediated by AT1R‐AAs, but not in normal pregnancy where aldosterone levels are increased but AT1R‐AAs are normal.

Recently, AT1R‐AAs were identified even in patients with PA. In particular, Li and colleagues14 reported that approximately 60% of patients with PA harbor active autoantibodies against AT1R, whose action is markedly suppressed by AT1R blockers in vitro. The activity of AT1R‐AAs has been previously demonstrated to directly correlate with the ability of AT1R‐AAs to contract perfused cremaster‐resistant arteries and to stimulate aldosterone production in HAC15 cells in vitro, suggesting a possible pathological role in PA.

Subsequent studies have shown that AT1R‐AAs are able to activate NADPH‐oxidase, tumor necrosis factor‐α, and nuclear factor‐kB, contributing to the increased production of reactive oxygen species related to the genomic and nongenomic inflammatory effects of aldosterone.15 Moreover, these proinflammatory factors could induce not only autoimmune disorders but also somatic mutations in nearby cells.16 ^.

Recent advances in genome technology have allowed the discovery of recurrent somatic mutations in genes coding for ion channels and adenosine triphosphatases regulating intracellular ionic homeostasis and membrane potential, which could lead to increased aldosterone production and adrenal cortex cell proliferation.17 Currently, about 50% of sporadic aldosterone‐producing adenoma (APA) cases have been associated with some genetic defect.

Autoptic local lymphocytic infiltration has been reported in the adrenal cortex of 63% of patients older than 60 years, while it was observed in only 7.4% of younger patients.18 Moreover, a considerable proportion of CD4+ T cells was activated in adrenal of elderly as judged by interleukin 2 receptor expression, and may represent a preclinical manifestation of inflammatory disorders leading to somatic mutations and/or organ‐specific autoimmunity. In fact, a lymphoid adrenalitis has also been reported in patients with APA19 and those with other adrenal adenomas.20 In addition, patients with APA with KJNJ5 mutations have relatively high amounts of stromal and inflammatory cells that could be positive for Ki67 proliferation markers.

THE ONSET OF PA AND THE THEORY OF EVOLUTION

The pathogenesis of PA remains unknown. It is difficult to understand why a patient with essential hypertension or normotension at one moment in life develops PA in its different forms. Surely, an epigenetic cause or other somatic mutations can be proposed in these sporadic cases.

A possible link between bilateral hyperplasia and unilateral adenoma has been hypothesized in the theory of a continuous progression from low‐renin or normal‐renin essential hypertension to bilateral adrenal hyperplasia, unilateral hyperplasia, and finally unilateral adenoma.21, 22 A gradual increase in autonomous aldosterone synthesis and a parallel decrease in sensitivity of glomerulosa to angiotensin II has been hypothesized.

The formation of APA could be related to somatic mutations caused by the inflammatory reaction induced by cross‐talk between increased aldosterone levels and MRs of mononuclear leukocytes infiltrated in the glomerulosa of predisposed patients. This hypothesis is supported by the observation of a possible reversal of PA in some patients after long‐term treatment with spironolactone or its derivatives.21, 22 A reversal of the DNA methylation or of somatic mutations can be proposed to explain the recovery and normal aldosterone secretion in PA.23

The recent finding of increased titers of AT1R‐AAs in PA suggests two other possible pathological hypotheses: (1) hyperaldosteronism induces AT1R‐AAs, or (2) AT1R‐AAs are responsible for PA. Considering the first hypothesis, hyperaldosteronism or its increased activity (as occurs in preeclampsia) could cause a proinflammatory systemic state that induces the formation of AT1R‐AAs. These autoantibodies may play a role in the pathogenesis of some hypertensive disorders, such as low‐renin essential hypertension, and could enhance the inflammatory status that leads to the onset of somatic mutations in the adrenal cortex, responsible for PA in predisposed patients. On the other hand, the presence of AT1R‐AAs could be the expression of an autoimmune disorder involving adrenal cortex cells, which are infiltrated by activated lymphocytes and can be directly stimulated by these autoantibodies to produce more aldosterone. The finding of multiple nodules in the tissue surrounding the APA and in the controlateral adrenal24 supports this hypothesis, suggesting the presence of a trigger that stimulates cell proliferation and aldosterone synthesis. However, any possible relationship between AT1R‐AAs and the mutational changes observed in APA is still not known.

DISCUSSION

Excessive synthesis of aldosterone or an increased effector mechanism in situations of normal aldosterone secretion can induce hypertension and cardiovascular diseases by generating an inflammatory state and promoting T‐cell immunity. This consequent proinflammatory state could also induce somatic mutations, which may induce autoimmune disorders or even PA.

The onset and progression of PA could be related to a combination of all these mechanisms, and a possible link between the various forms of PA is suggestive but requires confirmation with specific studies.

In predisposed patients, the presence of T lymphocytes with their MRs in the adrenal could allow local high aldosterone levels to induce inflammation or autoimmunity and a progressive autonomization of adrenals, leading to AT1R‐AAs or somatic mutations, low‐renin hypertension, bilateral hyperplasia, and later APA. This pathogenic model could explain the reported reversal of both APA and bilateral hyperplasia after long‐term treatment with MR blockers, and the reversal of the inflammatory reaction at the level of adrenals could be responsible for this change. The description of spironolactone bodies in the APA is an indirect demonstration of a direct effect of the drug at the level of adrenals.25

A genetic predisposition to an increased effector mechanism of aldosterone can also be proven from the usefulness of MR blockers to prevent the relapse of cardiovascular or cerebral accidents in all forms of hypertension and, in particular, resistant hypertension. Finally, some variants of PA may also be the expression of a new autoimmune disease, characterized by the presence of AT1R‐AAs and lymphoid adrenalitis.

CONCLUSIONS

All of these hypotheses should be investigated, and the notion that PA is a syndrome with different mechanisms involved in the individual case should not be excluded.

DISCLOSURES

The authors have no conflicts of interest to disclose.