The Renin-Angiotensin-Aldosterone System and Calcium-Regulatory Hormones

Abstract

There is increasing evidence of a clinically relevant interplay between the renin-angiotensin-aldosterone system and calcium regulatory systems. Classically, the former is considered a key regulator of sodium and volume homeostasis while the latter is most often associated with skeletal health. However, emerging evidence suggests an overlap in regulatory control. Hyperaldosteronism and hyperparathyroidism represent pathophysiologic conditions that may contribute to or perpetuate each other; aldosterone regulates parathyroid hormone and associates with adverse skeletal complications, and parathyroid hormone regulates aldosterone and associates with adverse cardiovascular complications. As dysregulation in both systems is linked to poor cardiovascular and skeletal health, it is increasingly important to fully characterize how they interact in order to more precisely understand their impact on human health and potential therapies to modulate these interactions. This review describes the known clinical interactions between these two systems including observational and interventional studies. Specifically, we review studies describing the inhibition of renin activity by calcium and vitamin D, and a potentially bidirectional and stimulatory relationship between aldosterone and parathyroid hormone. Deciphering these relationships might clarify variability in outcomes research, inform the design of future intervention studies, and provide insight into the results of prior and on-going intervention studies. However, before these opportunities can be addressed, more effort must be placed on shifting observational data to the proof of concept phase. This will require reallocation of resources to conduct interventional studies and secure the necessary talent.

INTRODUCTION

The renin-angiotensin-aldosterone system (RAAS) plays a crucial role in the physiologic regulation of sodium and potassium balance, intravascular volume, and blood pressure(1). It is now also well established that excess RAAS activity increases cardiovascular disease risk that can be mitigated by inhibiting or blocking the RAAS(2, 3).

Parathyroid hormone (PTH) and vitamin D are calcium-regulatory hormones that play a crucial role in skeletal health(4-6). PTH plays several key roles, including: 1) raising circulating calcium by mobilizing calcium from skeletal reservoirs; 2) promoting the 1-alpha-hydroxylation of 25-hydroxyvitamin D (25[OH]D); 3) indirectly increasing intestinal calcium absorption (via vitamin D receptor activation); 4) and increasing renal calcium absorption. Elevations in circulating calcium, in turn, negatively regulate PTH and synthesis of 1,25-dihydroxyvitamin D (1,25[OH]₂D). In addition to these known physiologic roles of PTH and vitamin D, high PTH and low vitamin D have been repeatedly associated with cardiovascular disease and mortality(7-15), although consistent and conclusive evidence from intervention studies to support these observations have yet to be reported.

This review highlights emerging interactions between calcium and calcium-regulatory hormones with the RAAS that may describe novel endocrine relationships and/or may represent mechanistic explanations for the links between calcium-regulatory hormones and cardiovascular diseases.

THE RAAS AND CALCIUM

Calcium dysregulation has been implicated as a potential mechanism for negative cardiovascular outcomes(16-21). Acute hypercalcemia increases blood pressure in normal, healthy humans(22-25). Elevated serum calcium concentrations increase cardiovascular risk as demonstrated in multiple large epidemiologic studies(16, 18, 20, 21). Calcification of large arteries, coronary arteries, and the microvasculature have all been associated with poor cardiovascular outcomes(16, 18, 21). The role of calcium in regulating the RAAS may represent an important mechanistic contribution in these observations.

Acute renin secretion is under inhibitory control via several calcium-mediated processes

Calcium cation is universally relevant in signal transduction, enzyme activation, and membrane potential in virtually all mammalian tissues(26). As such, both intra- and extracellular calcium concentrations are under tight regulatory control. It is not surprising, therefore, that a dependent relationship must exist between calcium homeostasis and the RAAS. Perhaps the most widely and detailed descriptions are the interactions of calcium and renin secretion by the juxtaglomerular and renal arteriolar cells(27-31). In contrast to almost all other interactions involving calcium-mediated signaling, in the case of renin release increasing calcium concentrations has an inhibitory effect. Renin secretion is mainly dependent on cyclic AMP formation. Cyclic AMP availability is the net effect of positive adenylyl cyclase activity and competing degradative activity of calmodulin-activated phosphodiesterase(30, 32-39). Increasing intracellular calcium concentrations decrease net cyclic AMP formation by dampening adenylate cyclase and enhancing phosphodiesterase activities.

Extracellular concentrations of calcium affect intracellular concentrations via the calcium sensing receptor present on renal juxtaglomerular cells(39-42). Stimulation of the calcium sensing receptor with the calcimimetic cinacalcet results in a dramatic decrease in cyclic AMP formation and renin secretion(41). Mobilization of cytosolic calcium in the JG can occur via activation of L-type voltage-gated calcium channels or release from intracellular calcium stores via membrane action potentials(43). The exact signal transduction pathway in the juxtaglomerular apparatus is as yet unknown, but likely similar to that of calcium sensing receptor in parathyroid cells(44-46).

In vivo, acute activation of calcium sensing receptor inhibits renin release

In vivo studies of calcium-renin interaction are similar to those described in vitro. Acutely raising circulating plasma calcium concentration is mostly associated with inhibiting renin release with a signal that is more clearly visible under renin-stimulated conditions such as low dietary salt intake(41, 47). Pharmacologically stimulating the calcium sensing receptor decreases renin secretion and antagonizing it prevents the inhibitory effects of hypercalcemia on renin release(40, 41).

Clinically, chronic activation of calcium sensing receptor is associated with elevated plasma renin activity

Curiously, chronically stimulated calcium sensing receptor is most often associated with higher plasma renin activity. Primary hyperparathyroidism has been associated mostly with elevated plasma renin activity(48), although not always(49). Maximally stimulated calcium sensing receptor as seen in Type V Bartter’s syndrome results in hyperreninemia(50). Thus, apparently conflicting results may have more to do with the degree of calcium sensing receptor activation under pathophysiologic states, and the ability to detect plasma renin activity differences under normal dietary sodium conditions.

Overall, acute elevation of calcium inhibits renin release via several extra-, and in turn, intracellular mechanisms. States of chronic calcium elevation, or activation of the calcium sensing receptor, are associated with variable elevation in plasma renin activity. It is not known to what extent these clinical observations are due to secondary (indirect) activation of plasma renin activity.

THE RAAS AND PTH

Growing evidence points to a bi-directional and positive relationship between the RAAS and PTH(51, 52). Basic studies, observational studies, and a few intervention studies have now reported on this novel two-way interaction between the RAAS and PTH that may have clinical implications with respect to mechanisms of human cardiovascular and skeletal disease.

Influence of the RAAS on PTH

Studies in primary aldosteronism have repeatedly observed a link between excess aldosterone and hyperparathyroidism. Resnick and colleagues described high PTH levels and a negative calcium balance in a small cohort of subjects with primary aldosteronism(53); this finding was again observed by other investigators evaluating larger cohorts with primary aldosteronism(54-57), raising speculation that aldosterone could directly stimulate PTH secretion. Rossi et al. showed that PTH levels were significantly higher in patients with aldosterone producing adenomas when compared to bilateral adrenal hyperplasia, suggesting that the severity of aldosteronism correlated with the severity of hyperparathyroidism(57). Among observational studies that have evaluated the impact of primary aldosteronism on PTH levels following either surgical (adrenalectomy) or pharmacologic (mineralocorticoid receptor antagonism) therapy, both surgery and medical therapy have demonstrated reductions in PTH that parallel the treatment of primary aldosteronism(54-57).

The health implications of hyperparathyroidism in primary aldosteronism may be significant – beyond the known cardiovascular risks associated with primary aldosteronism, a concomitant state of hyperparathyroidism could contribute to skeletal diseases (such as osteoporosis and fracture) and compound cardiovascular risk since elevated PTH has also been associated with cardiovascular disease and mortality(7, 8). Animal studies(58, 59) and several human observational studies in primary aldosteronism have supported the concern that primary aldosteronism may result in declines in bone mineral density and an increased risk for fracture. Salcuni et al. demonstrated that patients with primary aldosteronism had higher PTH levels and lower bone mineral density when compared with hypertensive controls(60). In this study, primary aldosteronism was associated with a higher odds for osteoporosis (OR=15.4 [1.83-130]) and vertebral fractures (OR=30.4 [1.07-862]), although the sample sizes evaluated were small, as reflected by the wide 95% confidence intervals for the observed point estimates(60). Despite this limitation, among the subset treated with surgery or mineralocorticoid receptor antagonists, PTH levels were observed to decline and bone density observed to rise, suggesting that primary aldosteronism, hyperparathyroidism, decreased bone density, and vertebral fracture may all be potentially linked in a causal pathway. Similar findings were also independently seen by Ceccoli et al. and Petramala et al(61, 62). These human observations have implicated hyperaldosteronism as a potentially treatable cause of low bone density that may result in clinical fragility fractures and have further supported elevated PTH as a mediator of this effect. However, further studies are needed to determine whether aldosterone acts directly on bone(63), whether aldosterone induces a hyperparathyroidism that results in bone remodeling, or whether targeting aldosterone in a large clinical trial results in improved skeletal health .

The connection between hyperaldosteronism and skeletal disease is not restricted to primary aldosteronism and has been observed in conditions of secondary hyperaldosteronism such as heart failure and chronic kidney disease. In a case-control study consisting of 167 patients with heart failure and non-traumatic fractures and 668 age-matched control patients without fracture, Carbone et al. found that the use of spironolactone was associated with a reduced odds of fracture (OR=0.575 [0.346-0.955])(64). Hassan et al. reported that among a population of 950 patients with chronic kidney disease, those prescribed spironolactone had significantly lower PTH after taking the medication and a lower risk of hospitalization for heart failure when compared to patients with chronic kidney disease not prescribed spironolactone (HR=0.37 [0.19-0.74])(17). Koiwa and colleagues evaluated a cohort of patients with chronic kidney disease and observed that the use of any renin-angiotensin-aldosterone inhibitors was associated with significantly lower PTH levels(65).

Given that both primary aldosteronism and secondary forms of hyperaldosteronism have been linked with hyperparathyroidism and skeletal disease which appear to be mitigated with the treatment of the underlying hyperaldosteronism, what mechanisms might explain these observed phenomena? Some have speculated that aldosterone may exert a direct stimulatory influence on the parathyroid gland to induce PTH secretion, as the mineralocorticoid receptor has been found to be expressed in parathyroid glands(66, 67). A number of case reports have described the concomitant development of primary hyperparathyroidism with primary aldosteronism, suggesting that aldosterone may play a role in the pathogenesis of hyperparathyroidism(66, 68). Alternatively, or perhaps in addition, aldosterone may indirectly stimulate PTH secretion by acting on the nephron to induce hypercalciuria and subsequent hypocalcemia (secondary hyperparathyroidism)(54, 55, 61, 69, 70).

Mechanistic studies to evaluate the physiologic relationship between the RAAS and PTH have been reported in subjects without primary aldosteronism. Grant et al. initially demonstrated that a dose-dependent relationship between angiotensin II and PTH existed in studies where healthy humans were infused with exogenous angiotensin II(71). Brown et al. extended this experiment and confirmed that an acute infusion of exogenous angiotensin II to healthy humans increased PTH (and aldosterone) in a dose-dependent manner within 1-2 hours(67, 72). Further, treatment with the angiotensin-converting-enzyme inhibitor captopril acutely lowered PTH (by 10-12% from baseline), in addition to lowering angiotensin II and aldosterone concentrations. This was in contrast to infusion of aldosterone, which did not acutely affect PTH levels. However, in a randomized placebo-controlled trial, healthy subjects without primary aldosteronism treated with spironolactone 50mg daily for 6 weeks to block aldosterone effect displayed significant lowering of PTH levels with concomitant increases in serum calcium, whereas subjects assigned to placebo did not(67). These results suggest that in the acute setting, angiotensin II may exert a stimulatory effect on PTH that can be mitigated by lowering angiotensin II with an angiotensin converting enzyme inhibitor; however, acute increases in aldosterone do not appreciably alter PTH. In contrast, in the chronic setting aldosterone may increase PTH, and this effect may be mitigated by a mineralocorticoid antagonist--observations that were similar to those reported in the aforementioned cohorts of subjects with primary aldosteronism. Thus, this study by Brown and colleagues provides evidence implicating multiple RAAS components interacting with PTH, and a differential temporal effect dictating their relationships(67). In assessments of parathyroid pathology, both normal and adenomatous parathyroid tissue have been shown to express the angiotensin type1 receptor and the mineralocorticoid receptor, and the expression of these receptors was 3-4 fold greater among adenomatous tissue, giving further support to direct actions of angiotensin II and/or aldosterone on the parathyroid(67).

Most recently, large-scale epidemiologic studies have added the greatest support for the aforementioned findings from small observational and intervention studies. In a cross-sectional analysis of more than 3000 participants, Fischer et al. showed that participants with an aldosterone-to-renin ratio that was greater than the 90^th percentile (suggestive of a “primary aldosteronism-like” phenotype) had higher PTH levels than those with an aldosterone-to-renin ratio that was less than the 90^th percentile(73). Similarly, among >1500 community-based participants without known hyperparathyroidism or primary aldosteronism from the Multi-Ethnic Study of Atherosclerosis, Brown et al. showed that higher serum aldosterone levels were significantly associated with higher serum PTH levels (+4.5 pg/mL per +1 ng/dL of aldosterone after multivariable adjustments), and that the use of RAAS inhibitor medications was associated with lower PTH levels when compared to the use of non-RAAS inhibitor anti-hypertensive medications (−2.0 pg/mL after multivariable adjustments)(74). In fact, of all potential anti-hypertensive therapies, the use of RAAS inhibitors associated with the lowest PTH. This study also observed that patients with the highest PTH values were those who had high serum aldosterone levels and low renin activity (a “primary aldosteronism-like” phenotype). In contrast, those with a secondary aldosteronism-like phenotype (high renin activity and high aldosterone) displayed PTH levels that were no different from participants with low or normal aldosterone levels, suggesting that the phenotype of hyperaldosteronism and possibly the chronicity of elevated aldosterone exposure may influence PTH elevations(74).

Influence of PTH on the RAAS

Increased PTH has been associated with vascular dysfunction and cardiovascular outcomes(7, 8, 75, 76); however, the mechanisms underlying this association remain unclear. In addition to the influence of the RAAS on PTH, several lines of evidence support a direct effect of PTH on components of the RAAS, which in turn may explain the link between PTH and cardiovascular disease(51, 77).

Basic laboratory studies have shown that PTH enhances aldosterone secretion from human adrenocortical cells in vitro by acting on PTH receptors present on zona glomerulosa cells, and that antagonists of the PTH-receptor block this effect(78, 79). In small intervention studies, infusions of exogenous PTH-like peptides have resulted in increases in adrenal aldosterone secretion and urinary tetrahydroaldosterone excretion(80), but also increases in plasma renin activity(71). In vitro studies describe similar stimulation of renin by PTH (81, 82), suggesting that PTH may enhance the activity of the RAAS via multiple methods. Patients with primary hyperparathyroidism (P-HPT) have been shown to demonstrate a heightened aldosterone secretory response to an infusion of angiotensin II(83), and in at least three well-documented reports, patients with P-HPT have developed concomitant primary aldosteronism(66, 68, 84). In some of these cases surgical parathyroidectomy resulted in cure of the aldosteronism, providing tantalizing evidence as to cause and effect(84). In contrast, some observational studies have been unable to find any difference in RAAS components between those with P-HPT and controls without P-HPT(85), underscoring the need for more dedicated and prospective study designs to evaluate this relationship.

THE RAAS AND VITAMIN D

The conversion of 25(OH)D to the active metabolite and vitamin D receptor (VDR) agonist 1,25(OH)₂D is a process governed by PTH(4), yet vitamin D may have independent influences on the RAAS. Low vitamin D status has been associated with clinical outcomes that are also traditionally associated with excess RAAS activity, including hypertension, inflammation, and cardiovascular disease(13-15, 86). Animal studies and human genetic association studies have provided mechanistic support for these observations; however, conflicting data exist, and there is a strong need for large-scale randomized studies to confirm the influence of vitamin D therapy on the RAAS and RAAS-mediated clinical outcomes.

Animal studies (mice) have shown that the 1,25(OH)₂D-VDR complex negatively regulates renin expression, and that this vitamin D-induced reduction in RAAS activity can prevent adverse vascular outcomes to a similar extent induced by pharmacologic angiotensin-receptor antagonism(15, 87-93). Human studies have supported this theory, demonstrating that low circulating vitamin D concentrations are associated with higher plasma renin activity and angiotensin II concentrations(86, 94, 95), and that vitamin D deficiency is associated with higher RAAS activity that can be lowered following intervention with vitamin D₃ therapy(72, 94, 96). Extrapolation of these results suggests that vitamin D therapy might serve to lower RAAS activity and improve complications associated with excess RAAS activity such as hypertension, nephropathy, and insulin resisitance(15).

To date, there have been mostly unimpressive or negative results from human interventional studies. Many human vitamin D intervention studies have focused on blood pressure or albuminuria as primary outcomes, and since both of these outcomes are related to excessive RAAS activity, these studies are particularly interesting to the topic of this review. One recent randomized study showed that reasonably high doses of vitamin D₃ therapy over 3 months modestly lowered systolic blood pressure in blacks(9). In addition, morbidly obese individuals with vitamin D deficiency have been reported to have modest reductions in mean arterial pressure following 1 month of very high-dose vitamin D₃ therapy, and this phenomenon was associated with reductions in the local vascular-tissue renin-angiotensin system(72). De Zeeuw et al., Larsen et al., and Joergensen et al. have shown that paracalcitol therapy, a VDR agonist, can lower albuminuria in advanced chronic kidney disease with or without diabetes; however, whether this effect is attributable to RAAS activity reduction remains to be proved(14, 97-99).

In contrast, a number of well-conducted intervention studies have found no effect of vitamin D therapy on blood pressure. Recently, Scragg et al. and Witham et al., in combination with several other small and short-term vitamin D₃ intervention studies(13, 14), have mostly reported no blood pressure lowering effect associated with vitamin D₃ therapy for time durations of 1 year and longer(100, 101). Perhaps the most notable vitamin D₃ intervention study to examine blood pressure lowering effects to date is the DAYLIGHT study. In this double-blinded, randomized, multi-center trial, including 534 individuals with stage I hypertension, 4,000 IU of daily vitamin D₃ for 6 months did not affect the mean 24-hour systolic blood pressure (primary endpoint) when compared to 400 IU of daily vitamin D₃₍₁₀₂₎.

These contradictory human intervention studies raise doubt about the consistency of observations, or at a minimum, the relevance of vitamin D therapy related to blood pressure lowering and cardiovascular outcomes. At the very least, it can be concluded that vitamin D3 therapy in durations of 3-12 months does not dramatically lower blood pressure akin to pharmacologic anti-hypertensives; however, long-term studies are lacking. It should be noted that except for the DAYLIGHT study, none of these studies was particularly large, and on the same token none of these studies (including the DAYLIGHT study) were particularly long in duration; therefore, larger and longer population-based interventional studies with more refined effect detection will be needed to satisfactorily determine whether vitamin D therapies can definitively modulate clinical outcomes related to excessive RAAS activity. Perhaps the best candidate for such a large study is the currently on-going VITAL study(103).

Beyond large and longer trials, it is important to remember and consider that “vitamin D” is not a single metabolite, rather is colloquially used to refer to a complex hormone system that includes several important players, each requiring metabolism and regulation at different steps. One factor complicating this physiologic relationship is the fact that 25(OH)D has low affinity for the VDR when compared to 1,25(OH)₂D, yet most vitamin D interventions in research and clinical settings use vitamin D₃ to raise 25(OH)D levels. However, the continued conversion to the active VDR agonist 1,25(OH)₂D by sheer substrate abundance may be down regulated depending on other pertinent variables. For example, with high dose vitamin D₃ therapy that increases serum calcium and 1,25(OH)₂D, parathyroid hormone is physiologically down-regulated and therefore the synthesis of 1,25(OH)₂D and dietary calcium absorption are reduced. In this scenario, normal physiology “buffers” the effect of sustained and chronic vitamin D₃ supplementation until the elevations in 25(OH)D exceed this buffering capacity and create a substrate-driven PTH-independent augmentation of 1,25(OH)₂D. Although circulating concentrations of these vitamin D metabolites may correlate with or dictate downstream biologic effect, it is well recognized that binding protein concentrations and genetic polymorphisms in the infrastructure governing this pathway play crucial roles in inter-individual responses to vitamin D therapy. Powe et al. demonstrated that black Americans had significantly lower 25(OH)D and vitamin D-binding protein concentrations when compared to white Americans; however, they had similar bioavailable 25(OH)D concentrations as a result(104, 105). This study points out a major weakness in the reliance on a circulating blood level of 25(OH)D to determine vitamin D status and correlations with downstream biologic activity. In most research studies that assess vitamin D interventions, the assessment of binding protein concentrations, bioavailable 25(OH)D, or 1,25(OH)₂D to assess active metabolite levels, is not performed or considered.

Similarly, the influence of genetic polymorphisms in the metabolism and activity of the vitamin D system has been shown to play an important role, particularly in mediating vitamin D-RAAS interactions; almost every step of vitamin D metabolism has been shown to be modified by genetic variation(106, 107). The FokI polymorphism within the VDR has been shown to be a functional variant that alters the length and activity of the 1,25(OH)₂D-VDR complex(108, 109). Genetic variation at FokI has been associated with hypertension(110-112) and has been shown to predict plasma renin activity in both normotensives and hypertensives(113). Further, the combination of 25(OH)D concentrations and FokI genotype enhances the prediction of renin activity in hypertension suggesting important pharmacogenetic considerations when evaluating the impact of vitamin D interventions to raise 25(OH)D or influence a clinical RAAS-mediated outcome(113). This is best exemplified in the large prospective study evaluating the influence of vitamin D status on composite cardiovascular, skeletal, and cancer outcomes in the Cardiovascular Health Study(114). In this study, more than 1500 participants were followed for greater than 10 years, and analyses to evaluate whether 25(OH)D deficiency in combination with genetic variation predicted the composite outcomes. The authors observed that when they stratified their clinical outcomes by VDR genotype, they observed significant risk modification that was dependent on genotype, further supporting the important pharmocogenetic interplay that underlies the relationship between vitamin D and clinical outcomes(114). Whether the RAAS plays a mediating role in these relationships was not assessed in this study, but should be strongly considered when evaluating future study designs given the supportive pre-clinical data.

CONCLUSIONS

In summary, growing evidence suggests interactions between calcium-regulatory hormones and the renin-angiotensin-aldosterone system with potentially important clinical implications. Calcium and vitamin D have been shown to inhibit renin secretion and expression, whereas a bi-directional and stimulatory relationship between parathyroid hormone and aldosterone (and possibly angiotensin II) has been observed. With this mounting evidence, new questions are raised that may have notable implications for future clinical outcomes research. Does inhibition of the renin-angiotensin-aldosterone system or antagonism of the mineralocorticoid receptor induce clinically relevant parathyroid hormone reductions? And if so, are skeletal outcomes such as bone density and fragility fracture influenced by pharmacotherapies that target the RAAS? Does the administration of vitamin D₃ or direct agonism of the vitamin D receptor with 1,25(OH)₂D result in clinically relevant reductions in the activity of the renin-angiotensin-aldosterone system? And if so, are cardiovascular outcomes modifiable with vitamin D therapies? Contemplating the interactive complexity of two highly evolved regulatory systems is daunting, yet deciphering these relationships might clarify variability in outcomes research and inform the design of future intervention studies. This in turn may provide the basis for endocrine pleiotropy in human disease management. However, interventional, prospective, well-designed human studies are severely lacking. A realignment of resources and talents will be required to move the over-abundance of epidemiological association observations to the critical proof of concept stage and beyond.

Interactions between the renin-angiotensin-aldosterone system and calcium-regulatory hormones.

The 1,25(OH)₂D-VDR complex, and calcium ion, can decrease plasma renin activity either by decreases renin expression or decreasing renin secretion, respectively. Some evidence has shown that PTH may directly increase renin secretion in the acute setting. Aldosterone and angiotensin II stimulate the secretion of PTH via the MR and AT1R; the former with chronic exposure whereas the latter in the acute setting. PTH, in turn, has been implicated as an aldosterone secretagogue via interactions with the PTH-R in the adrenal zona glomerulosa.

25(OH)D=25-hydroxyvitamin D

1,25(OH)₂D=1,25-dihydroxyvitamin D

VDR=vitamin D receptor

PTH=parathyroid hormone

Ca²⁺=calcium cation

CaSR=calcium sensing receptor

MR=mineralocorticoid receptor

AT1R=angiotensin receptor type 1

ACE=angiotensin converting enzyme