Primary aldosteronism: molecular medicine meets public health

Abstract

Primary aldosteronism is the most common single cause of hypertension and is potentially curable when only one adrenal gland is the culprit. The importance of primary aldosteronism to public health derives from its high prevalence but huge under-diagnosis (estimated to be <1% of all affected individuals), despite the consequences of poor blood pressure control by conventional therapy and enhanced cardiovascular risk. This state of affairs is attributable to the fact that the tools used for diagnosis or treatment are still those that originated in the 1970-1990s. Conversely, molecular discoveries have transformed our understanding of adrenal physiology and pathology. Many molecules and processes associated with constant adrenocortical renewal and interzonal metamorphosis also feature in aldosterone-producing adenomas and aldosterone-producing micronodules. The adrenal gland has one of the most significant rates of non-silent somatic mutations, with frequent selection of those driving autonomous aldosterone production, and distinct clinical presentations and outcomes for most genotypes. The disappearance of aldosterone synthesis and cells from most of the adult human zona glomerulosa is the likely driver of the mutational success that causes aldosterone-producing adenomas, but insights into the pathways that lead to constitutive aldosterone production and cell survival may open up opportunities for novel therapies.

Fig. 1. The cause, impact, and reversal with treatment, of sporadic primary aldosteronism.

a, In sporadic primary aldosteronism, somatic mutations result in autonomous aldosterone production, which leads to excess Na⁺retention by the kidneys. These in turn detect a high Na⁺flux through the Na⁺-K⁺−2Cl^- transporter in juxtaglomerular macula densa cells, which signal to the adjacent renin-producing cells of the afferent arteriole to switch off renin secretion. The net consequence is a sustained elevation of plasma aldosterone and suppression of plasma renin and angiotensin II (ANGII). b, Autonomous aldosterone production in the presence of salt is associated with continuous renal Na⁺ reabsorption despite the suppression of ANGII and renin. Such inappropriate sodium retention leads to renal damage, cardiac remodelling, and vascular remodelling^,. Patients with primary aldosteronism are at a higher risk than patients with essential hypertension of kidney and cardiovascular damage, detected as events such as atrial fibrillation and stroke. Relative risk data for atrial fibrillation and stroke are taken from Monticone et al.. c, The effects of aldosterone can be reversed with the use of competitive mineralocorticoid receptor antagonists (MRAs), such as spironolactone, which inhibit the reabsorption of Na⁺ in the distal convoluted tubule and collecting duct. The inhibition of Na⁺ retention can be detected as a rise in plasma renin. In patients whose renin is adequately de-suppressed with use of MRAs, tissue damage by aldosterone and salt is prevented or reversed, attenuating the increased risks of cardiovascular and renal events^,. Relative risk data for atrial fibrillation following MRA treatment or adrenalectomy are taken from Hundemer et al.. AT1R, ANGII type 1 receptor. Graphs showing the percentage of patients affected by microalbuminuria and cardiovascular events in part b are adapted with permission from ref. , Elsevier.

Fig. 2. Timeline of clinical and molecular advances in primary aldosteronism.

Seventy years after the first description of Conn’s syndrome, now known as primary aldosteronism, standard clinical practice for its management remains the removal of the entire adrenal, regardless of lesion size. Standard techniques include adrenal vein sampling, CT scanning and laparoscopic surgery. However, a number of key molecular scientific discoveries in the last 30 years have provided insights into the molecular pathogenesis of primary aldosteronism, and may identify new therapeutic approaches^{,,,,,,,,,–,,,,}. APA, aldosterone-producing adenoma; FH, familial hyperaldosteronism

Fig. 3. The contribution of KCNJ5 mutations to the pathogenesis of primary aldosteronism.

a Search for KCNJ5 (ENSG00000120457.11) in the GTEx portal shows that the adrenal gland expresses the highest levels of this gene. Data are shown as transcripts per million, calculated from a gene model with isoforms collapsed to a single gene. No other normalization steps have been applied. Box plots are shown as median and 25th and 75th percentiles; points are displayed as outliers if they are above or below 1.5 times the interquartile range. b, Structural analysis of the potassium channel Kir 3.4, which is encoded by KCNJ5, shows the originally described mutations (Gly151Arg and Leu168Arg) within or close to the selectivity pore of the channel. c, A typical 11C-metomidate PET-CT scan in a patient with a KCNJ5-mutant APA (arrow). Immunohistochemistry of tissue sections from mutant APAs showed them to more closely resemble cortisol-producing (zona fasciculata; ZF) than aldosterone-producing (zona glomerulosa; ZG) adrenal cells, with more striking staining for CYP11B1 than CYP11B2. d, Data from the MATCH study indicate that KCNJ5 mutations are the most common mutations found in APAs from self-identified white individuals. Data from MATCH also indicate that APA genotype is indicative of complete clinical success following adrenalectomy, with removal of KCNJ5-mutated APAs most likely to result in complete clinical success. The colours represent self-reported ethnicity. e, The presence of ZF- and ZG-specific enzymes (CYP17A1 and CYP11B2, respectively) in the same cell of a KCNJ5-mutant APA indicates that such cells can synthesize the ‘hybrid’ steroids 18-oxocortisol and 18-hydroxy-cortisol (18-OH-cortisol) that a normal ZG or ZF cell alone would not be able to produce. Data from the MATCH study have demonstrated the potential diagnostic utility of hybrid steroids as a metabolic marker of a KCNJ5-mutated APA. Part a reproduced from ref. , CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Part b reproduced from ref. , CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Parts d and e adapted from ref. , CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/).

Fig. 4. Germline and somatic mutations associated with primary aldosteronism.

Pathological autonomous aldosterone production in a cell can occur by a number of different mechanisms. (1) In familial hyperaldosteronism (FH)-I, the CYP11B1/CYP11B2 hybrid gene is activated by cAMP signalling, which occurs when adrenocorticotropic hormone (ACTH) binds to the melanocortin 2 receptor (MC2R) in response to hypothalamic–pituitary–adrenal axis (HPAA) activation. (2) GNA11 and GNAQ mutations in aldosterone-producing adenomas (APAs) prevent termination of G protein signalling downstream of the renin–angiotensin–aldosterone system, which is activated when angiotensin II (ANGII) binds to the angiotensin 1 receptor (AT1R)), leading to increased calcium release from intracellular stores via the inositol trisphosphate (IP₃) signalling pathway. These mutations co-occur with CTNNB1 mutations, which are found in APAs and adrenocarcinomas, and prevent β-catenin degradation, leading to increased transcription of target genes such as LHCGR. (3) CLCN2 mutations in FH-II and APAs, KCNJ5 mutations in FH-III and APAs, and ATP1A1 and ATP2B3 mutations in APAs lead to abnormal permeabilities for Cl^- Na⁺, H⁺, and Ca²⁺. These abnormalities cause cell depolarization, which increases intracellular calcium concentrations, thereby stimulating CYP11B2 expression and aldosterone production. Acidification and an impaired pump function may also have a pathological role (not shown). (4) CACNA1H mutations in FH-IV and APAs and CACNA1D in primary aldosteronism with seizures and neurological abnormalities (PASNA) and APAs increase calcium permeability by stimulating the calcium signalling pathway directly. High-probability pathogenic somatic mutations have also been found in DPYSL2 and VAPA. DPYSL2 is highly abundant in the zona glomerulosa and has previously been linked to ion-channel trafficking. Based on sequence prediction, we hypothesize that it may bind to the β-subunit binding site of Cav1.3 (refs. 63,171). VAPA is also linked to protein trafficking and is thought to be expressed on the endoplasmic reticulum where IP3 signalling acts on the calcium channel to release calcium from intracellular stores.

Fig. 5. The plethora of CACNA1D mutations associated with aldosterone-producing adenomas and aldosterone-producing micronodules.

a, CACNA1D, which encodes the pore-forming α1 subunit of a voltage-gated calcium channel (Cav1.3), is the most commonly mutated gene in aldosterone-producing adenomas (APAs) and aldosterone-producing micronodules (APMs) from patients with primary aldosteronism who self-identify as Black or who have bilateral adrenal hyperplasia. The Ca_v1.3 α1-subunit comprises four homologous repeats (I–IV), each comprising six transmembrane segments (S1–S6). Germline mutations (red circles) and somatic mutations (blue circles) have been identified throughout regions I–IV. More than 70 mutations have been described, mostly in small APAs or APMs. The majority of mutations in CACNA1D with evidence of pathological function are in conserved sites within functional domains, such as the voltage-sensing domain, the channel-activation gate and the cytoplasmic S4-S5 linker, which couples the voltage-sensing domain to the pore. b, The identification of APMs (and small zona glomerulosa (ZG)-like APAs) was facilitated by the development of a specific monoclonal antibody for CYP11B2, which discriminated it from the highly abundant and homologous CYP11B1 (ref. 19). These APMs and small ZG-like APAs commonly harbour a CACNA1D somatic mutation^,.

Fig. 6. A proposal for the evolutionary selection of aldosterone-producing adenomas and aldosterone-producing micronodules.

a, Confocal analysis of adrenal tissue from mice with heterozygous (A^+/−) and homozygous (AS^−/−) knockout of the gene that encodes aldosterone synthase (CYP11B2) shows that homozygous CYP11B2 deletion induces centripetal migration of outer zona glomerulosa (ZG; green) cells towards the inner zona fasciculata (ZF) layer by 5 weeks of age. At 6 months, ZG cells that have migrated to the zona reticularis (ZR)–medulla boundary in AS^−/− mice expresses markers for apoptosis, terminal deoxynucleotidyl transferase dUTP nick end labelling (TUNEL) staining (yellow). b, The neuronal gene, SLC35F1 is highly expressed both in APM cells (encapsulated by the dashed brown lines) and in nerve-like structures adjacent to these (yellow circle). c, We propose that salt-induced suppression of aldosterone synthesis in the ZG is a scenario analogous to that induced by genetic CYP11B2 deletion. We also propose that the presence of mutations in aldosterone-driver genes such as CACNA1D may facilitate the evolutionary selection of APMs, by imparting a survival advantage. The presence of gap junctions between normal cells of the ZG may enable propagation of neuronal signals and thereby also contribute to the inhibition of aldosterone production in this region. By contrast, the main gapjunction proteins are largely absent from APMs. We also propose that neuronal excitation may over-ride negative feedback regulation in APMs and contribute to autonomous CYP11B2 production. Note that this model in which APMs arise from clonal selection is unproven. Part a adapted from ref. , Oxford Academic Press, and ref. , Elsevier. Part b is from The Human Protein Atlas, image available at https://www.proteinatlas.org/ENSG00000196376-SLC35F1/tissue/adrenal+gland#img.