Primary Aldosteronism: a Channelopathy?

Primary aldosteronism (PA) is the most common form of secondary hypertension accounting for approximately 10% of patients referred to hypertension clinics and as high as 20% of those with refractory hypertension ¹. The most important causes of PA are idiopathic hyperaldosteronism (IHA), also called bilateral zona glomerulosa hyperplasia, responsible for about 50–70% of PA and aldosterone-producing adenomas (APA) in 30–50% of PA, and less common causes in about 2% of patients with PA ¹. The latter include three familial types of PA or hyperaldosteronism. Familial hyperaldosteronism (FHA) I, or glucocorticoid-suppressible Aldosteronism, is due to gene duplication by an uneven crossover recombination of the 11β-hydroxylase (CYP11B1) and the aldosterone synthase (CYP11B2) genes resulting in the expression of an ACTH-regulated chimeric gene in the zona fasciculata which produces aldosterone. FHA II, the most common, is of yet unknown cause, though many cases are in linkage with chromosome 7p22. FHA III is described below ^2–4.

Until relatively recently research on PA concentrated in defining ways to diagnose and differentiate between the various types of PA and best way to treating them. Studies of the pathogenesis and the molecular basis of APAs had been limited to searches of aberrant expression of G-protein coupled receptors and altered expression of genes using microarray technology. Studies of the pathogenesis of IHA had focused on the futile search for unknown aldosterone stimulating factors in the 1970–80s. In 2011 use of the whole exome sequencing technique by the Lifton group led to the identification of somatic mutations of the G-protein activated potassium inward rectifying channel KCNJ5 (Kir3.4 channel) in adenomas of about 30% of patients with APA ² and initiated the use of this powerful technique to address the molecular pathogenesis of these adenomas by other laboratories. Mutations in the KCNJ5 were found in a highly conserved region corresponding to the selectivity filter of the channel; two mutations, G151R and L168R, were the most commonly found. Multiple studies have now shown that APA with KCNJ5 somatic mutations occur in 30–60% of APA patients and are more common in younger women ³. Germline mutations in the selectivity filter region of the KCNJ5 gene have been found in several families with FHA III ^2–4.

The Kir3.4 or KCNJ5 channel is expressed in the adrenal zona glomerulosa where it forms a homotetramer or heterotetramer with the Kir3.1 subunit ². Mutations in the selectivity filter result in a decreased selectivity for potassium and allow sodium to leak into the cell, depolarizing the membrane, resulting in the opening of calcium channels and an increase in aldosterone synthesis ². It was also postulated to result in adenoma cell proliferation ². Angiotensin II stimulation of the adrenal carcinoma cell line HAC15 downregulates KCNJ5 expression, resulting in a decrease in membrane potential, increase in intracellular calcium, and the expression of several enzymes in the biosynthetic pathway for aldosterone synthesis ⁵. Transduction of the mutant KCNJ5 T158A into the HAC15 cell, results in an increase in aldosterone biosynthesis, both under basal and stimulated conditions, due to the same sequence of events ⁶. While the G151R and the L168R mutations are by far the most commonly found in APAs to date, other somatic mutations in or close to the selectivity filter of the channel have been found ⁴. Seven families with germline mutations of the KCNJ5 gene have been described ^{3, 4}.

The study by Murthy et al ⁷ performed a comprehensive re-sequencing of genomic DNA from 251 Australian patients with florid sporadic PA. While they failed to identify germline mutations of the selectivity filter, they found 3 heterozygous missense mutations of the coding region of the KCNJ5 gene, two, R52H (SNP rs144062083) and G247R (SNP rs200170681), which are present in some dbSNP data bases, and one which had not yet been described, E246K. Twelve other subjects were carriers for the rare non-synonymous SNP, E282Q, with a population frequency of 2% in the 1000 genome cohort. Three of the four new mutations found, R52H, E246K and E282Q, produced a change in the function of the channel that substantially altered K⁺ selectivity of the channel. The G247R missense mutation had no discernable functional consequences. No signs of PA were identified in the families of the two individuals heterozygous for the functional missense mutations, R52H and E246K. This may be due to variable penetrance or an age-related penetrance of the mutations that later on might manifest with a PA phenotype.

Transfection experiments in the human adrenal cell carcinoma cell line H295R showed that the R52H and E282Q mutations did not increase the basal production of aldosterone, while the E246K mutant increased basal aldosterone synthesis to a similar degree that the delI157 KCNJ5, which was used as a positive control ⁷. Transfection of all three variants, R52H, E246K and E282Q enhanced the angiotensin II stimulation of aldosterone secretion. In previous studies with the KCNJ5 mutant T158A resulted in a significant increase in basal, as well as AII-, forskolin- and potassium-stimulated aldosterone, cortisol and 18-oxocortisol production by the HAC15 cells ⁶.

Aldosterone secretion in most patients with APA is very sensitive to stimulation with ACTH and less so with angiotensin-II, while in IHA aldosterone secretion is very sensitive to angiotensin-II stimulation. This finding has been used in the design of the postural maneuver to distinguish between the two conditions. Some APAs (10–30%) also are angiotensin-II responsive which behave as IHA in this maneuver.

An important finding not mentioned in the discussion of the paper by Murthy et al ⁷ is that the 3 new genomic mutations, R52H, E246K and G247R, occurred in IHA patients. In their cohort of 251 PA patients, 12 had the E282Q polymorphism and of these 9 had IHA. There were no clinical details on these patients nor on the histopathology of the adenomas of the 3 APA patients with the E282Q polymorphism ⁷. Many adrenals with an APA, including those with KCNJ5 mutations, also have zona glomerulosa hyperplasia with nodules and aldosterone-producing cell clusters surrounding the adenoma which also express stem/proliferative gene markers ⁸. The functional relevance of these areas is unknown. The pathophysiological characteristics of the zona glomerulosa surrounding the APA might depend on the functional consequences of the type of KCNJ5 mutation in the adenoma. Because IHA patients tend to respond to postural maneuvers that stimulate angiotensin-II, one may hypothesize that APA patients with the E282Q might have an angiotensin-II responsive adenoma.

A role in cell proliferation and resulting adenoma formation was suggested for KCNJ5 mutations found in APA ². Studies by Oki et al ⁶ using the H295R cells transduced with a lentivirus with the mutant KCNJ5-T158A showed a decrease proliferation of these cells using two different techniques. In cells transfected with the multiple mutants of the KCNJ5 described so far, electrophysiological studies demonstrate variation in sodium conductance and degree of loss of the K⁺ selectivity. The mutant KCNJ5-T158A has significantly greater sodium current than the wild type, but it is less than that of other mutants. G151E produces a channel with the most sodium conductance. While familial patients with the KCNJ5-T158A germ-line mutation exhibit a very severe PA phenotype, families with the KCNJ5-G151E have a mild phenotype. The lethal effect of the G151E mutation on glomerulosa cells appears to limit the number of aldosterone-producing cells while the functionally milder mutation T158A produces the most severe clinical pictures ^{2, 4}. The studies by Murthy et al also found that the novel functional mutations decrease cell proliferation in vitro ⁷. The depolarization of the membrane results in an increase in intracellular calcium by activating calcium channels. While a mild increase in intracellular calcium has a proliferative effect, excessive intracellular calcium is anti-proliferative and pro-apoptotic. KCNJ5 mutations are found in FHA III characterized by severe hyperplasia and in a large percentage of APA patients yet in vitro studies demonstrate an anti-proliferative consequence of the KCNJ5 mutations. One possibility to explain this discrepancy has been suggested is that adenomas also overexpress genes that have an anti-apoptotic effect, including Vsnl1 that binds calcium which would limit its toxic effect ⁹. Among current challenges is the discovery of mechanisms for the proliferation of zona glomerulosa cells harboring these mutations.

Somatic mutations of the calcium channel gene CACNA1D¹⁰ gene and the sodium/potassium ATPase alpha subunit gene (ATP1A1I) and the calcium ATPase gene (ATP2B3) have been reported¹⁰, but new mutations in somewhere between 40–50% of patients with APA remain to be uncovered.