Aldosterone breakthrough is usually defined as an elevation of plasma aldosterone levels (PALs) during chronic administration of angiotensin-converting enzyme inhibitors (ACEis) and/or angiotensin type 1 (AT1) receptor blockers (ARBs) [1]. This phenomenon, which has been observed in conditions such as hypertension, heart failure and diabetic nephropathy, is thought to be involved in the gradual attenuation of cardiovascular and renal protection that is observed over time in patients receiving these treatments [1]. In support of this assumption, aldosterone is known to exert direct deleterious effects on heart, vessels and renal tissues, and mineralocorticoid receptor antagonists have shown high efficacy in decreasing mortality and preserving renal function in patients with cardiovascular and renal diseases [2, 3]. The mechanisms involved in aldosterone breakthrough are still a matter of debate. Incomplete blockade of the renin-angiotensin system (RAS) has been proposed as a potential explanation for this phenomenon in some patients showing an increase in plasma angiotensin II concentration during treatments with ACEis. However, PAL is not correlated with angiotensin II levels in this situation, and aldosterone breakthrough is also observed in patients in whom angiotensin II production remains suppressed [4]. In addition, no correlation was found between PALs and plasma potassium levels during aldosterone rebound [1]. Thus, it is likely that aldosterone breakthrough may result from activation of a compensatory mechanism that aims to overpass RAS inhibition.

In the April issue of the Journal, Masaki Mogi has provided an up-to-date and well-sourced review of the putative mechanisms involved in aldosterone breakthrough with a special focus on the role of the angiotensin type 1 (AT1) receptor [5]. The potential role of RAS-independent regulatory factors such as endothelin, serotonin (5-hydroxytryptamine, 5-HT), leptin and neuropeptides is also mentioned. Notably, all of these bioactive signals can be released within the adrenocortical tissue and are therefore capable of stimulating aldosterone secretion through a paracrine mode of action (Fig. 1). For instance, the adrenal cortex is richly vascularized such that each adrenocortical cell is in direct contact with endothelial cells, and endothelin has been shown to activate aldosterone production by cultured human adrenal cells. 5-HT, which can be released by subcapsular mast cells, is able to stimulate in vitro mineralocorticoid secretion via activation of 5-HT4 receptors expressed by zona glomerulosa cells. Consistently, oral administration of 5-HT4 receptor agonists such as cisapride to healthy volunteers induces a significant renin-independent elevation of plasma aldosterone levels [6]. In this regard, it seems likely that the well-documented in vivo stimulatory effect of metoclopramide on aldosterone release is mediated through its 5-HT4 receptor agonistic properties rather than its dopaminergic antagonistic action. In addition, aldosterone secretion can be activated by leptin produced by adipocytes. Circulating leptin may thus act as an endocrine factor to regulate mineralocorticoid synthesis. However, leptin can also be produced by intra-adrenal islets of adipocytes, raising the possibility that the peptide may be an important paracrine stimulator of aldosterone secretion. Finally, the zona glomerulosa contains a dense network of nerve fibers that can release numerous neuropeptides and conventional neurotransmitters. Most of them are able to modulate in vitro aldosterone secretion by adrenocortical cells [7]. Establishing the hierarchy of these diverse mechanisms is not easy. This is in part because in vivo evaluation of the role of paracrine factors in a human organ is still challenging at present. Nevertheless, it has been postulated that the strongest evidence for the physiological function of an intra-adrenal regulatory factor is the demonstration that its inhibition actually modifies corticosteroid production. Very few adrenal signals meet this criterion. Endothelin receptor antagonists have been shown to reduce PALs in hypertensive patients, but the effect of the compounds was mainly mediated by the RAS, since the decrease in PALs was associated with a significant lowering of plasma renin activity [8]. Conversely, in vivo investigation of the role of intra-adrenal 5-HT in the regulation of aldosterone has been hampered by the lack of specific 5-HT4 receptor antagonists available for clinical studies. Recently, we have shown that substance P, a neuropeptide belonging to the tachykinin family and released by adrenocortical nerve fibers, stimulates in vitro aldosterone production by adrenocortical cells [9]. The action of substance P is mediated by the neurokinin 1 receptor, which can be detected in zona glomerulosa cells expressing aldosterone synthase. Interestingly, whereas the AT1 receptor potently activates the intracellular calcium pathway to stimulate aldosterone synthesis, the NK1 receptor enhances mineralocorticoid production via activation of the ERK/pERK pathway in adrenocortical cells. In vitro evaluation of the role of substance P in the regulation of aldosterone synthesis has been complemented by a placebo-controlled clinical trial designed to investigate the impact of the NK1 receptor antagonist aprepitant on mineralocorticoid secretion in healthy volunteers. Aprepitant, which is currently used as an antiemetic in patients receiving anticancer chemotherapy, was found to reduce global aldosterone production by approximately 30% as assessed by the measurement of 24-hr urinary aldosterone excretion. Interestingly, aprepitant also induced a significant decrease (approximately 25%) in PALs without any effect on plasma renin concentrations. These data indicate that the autonomic nervous system directly stimulates aldosterone production and thus regulates hydromineral homeostasis through intra-adrenal release of substance P and activation of the NK1 receptor. The degree of short-term mineralocorticoid synthesis inhibition induced by aprepitant appeared to be on the same order of magnitude as that previously reported with RAS blockers, suggesting that substance P may be an important regulator of the activity of adrenocortical zona glomerulosa cells. In this regard, it is likely that the action of substance P on aldosterone-producing cells also encompasses indirect mechanisms involving intra-adrenal mast cells and 5-HT. In fact, it is well established that in numerous organs, substance P–positive nerves establish close connections to mast cells, which are known to express the NK1 receptor. As a result, substance P is able to trigger the release of various bioactive substances by mast cells, which thus appear to act as local amplifiers of the autonomic nervous system. In the adrenal gland, an increase in sympathetic tone may induce dual stimulation of zona glomerulosa cells by both substance P released by nerve fibers and 5-HT produced by mast cells. Another finding deserves emphasis because it helps clarify the global regulation of aldosterone secretion. Aprepitant was able to decrease the PAL in recumbency but had no effect on the PAL in upright position, a situation that is associated with RAS activation. Thus, it appears that substance P and RAS exert complementary actions on aldosterone synthesis: substance P controls basal aldosterone secretion, while RAS induces an increase in mineralocorticoid production in physiological and/or pathological conditions that require a rapid and strong stimulation of zona glomerulosa cells.

Fig. 1
figure 1

Paracrine regulation of aldosterone production in the human adrenal gland. The secretory activity of aldosterone-producing adrenocortical cells is subject to complex regulatory mechanisms involving several bioactive signals released by the diverse cell types present in the cortex. Notable components of these mechanisms include endothelin, serotonin (5-HT), leptin and substance P produced by endothelial cells, mast cells, adipocytes and nerve fibers, respectively. Substance P may stimulate aldosterone secretion through both a direct effect on zona glomerulosa cells and an indirect effect involving mast cells and serotonin release

To date, there is no clinical evidence showing that these regulatory systems are actually involved in aldosterone breakthrough. However, some data suggest that they could play a significant role in the pathophysiology of aldosterone excess associated with various cardiovascular diseases. In particular, endothelin levels are increased in chronic heart failure resulting from diverse cardiac pathologies, including diabetic cardiomyopathy and ischemic heart disease [8]. In these conditions, elevation of plasma endothelin levels is regarded as a consequence of inflammation and endothelial dysfunction. Given that aldosterone breakthrough is observed in the context of heart failure, even in patients undergoing double blockade of the RAS through combined administration of angiotensin-converting enzyme inhibitors and AT1 receptor antagonists, it is tempting to speculate that endothelin may be one of the promoters of aldosterone breakthrough through both endocrine and paracrine mechanisms. However, clinical trials with endothelin receptor antagonists have produced disappointing results. In particular, long-term administration of endothelin receptor antagonists failed to improve clinical outcomes in patients with heart failure [8]. Stimulation of the neural command for aldosterone secretion would represent another conceivable explanation for aldosterone breakthrough. Indeed, the sympathetic nervous system is known to be activated in both heart failure and hypertension associated with metabolic syndrome. In the latter, the enhancement of sympathetic nervous system activity is considered to result from insulin resistance–associated hyperinsulinism. It is also noteworthy that obesity/metabolic syndrome is frequently associated with idiopathic primary (i.e., RAS-independent) aldosteronism. Interestingly, analysis of the database of the Japan Primary Aldosteronism Study suggests that obesity-related factors may contribute to the pathogenesis of hyperaldosteronism [10]. These mechanisms, which remain to be elucidated, may involve leptin and the sympathetic nervous system. The implication of the autonomic nervous system in the pathophysiology of aldosterone breakthrough is also supported by other considerations. In particular, as previously pointed out by Sato and Saruta [1], RAS blockers are frequently combined with diuretics, which promote an increase in sympathetic tone via reduced blood volume and arterial blood pressure.

In conclusion, recent advances in the comprehension of the physiological regulation of aldosterone secretion indicate that, in humans, aldosterone production is controlled by intra-adrenal bioactive signals that are independent of the circulating RAS. These intra-adrenal regulatory systems are likely to play a pivotal role in the increase in PAL observed during prolonged treatment with RAS blockers. However, further clinical investigations are required to test this hypothesis and decipher the hierarchy of the diverse paracrine factors involved in these complex processes.