Novel Insertion Mutation in KCNJ5 Channel Produces Constitutive Aldosterone Release From H295R Cells

Abstract

Primary aldosteronism accounts for 5%-10% of hypertension and in a third of cases is caused by autonomous aldosterone production by adenomas (APA). Somatic mutations in the potassium channel encoded by KCNJ5 have been detected in surgically removed APAs. To better understand the role of these mutations, we resequenced the KCNJ5 channel in a large Australian primary aldosteronism cohort. KCNJ5 mutations were detected in 37 APAs (45% of the cohort), including previously reported E145Q (n = 3), G151R (n = 20), and L168R (n = 13) mutations. In addition, we found a novel 12-bp in-frame insertion mutation (c.414-425dupGCTTTCCTGTTC, A139_F142dup) that duplicates the AFLF sequence in the pore helix upstream of the selectivity filter. Expressed in Xenopus oocytes, the A139_F142dup mutation depolarized the oocytes and produced a G-protein-sensitive Na(+) current with altered K(+) selectivity and loss of inward rectification but retained Ba(2+) sensitivity. Transfected into H295R cells, A139_F142dup increased basal aldosterone release 2.3-fold over the wild type. This was not increased further by incubation with angiotensin II. Although the A139_F142dup mutant trafficked to the plasma membrane of H295R cells, it showed reduced tetramer stability and surface expression compared with the wild-type channel. This study confirms the frequency of somatic KCNJ5 mutations in APAs and the novel mutation identified (A139_F142dup) extend the phenotypic range of the known KCNJ5 APA mutations. Being located in the pore helix, it is upstream of the previously reported mutations and shares some features in common with selectivity filter mutants but additionally demonstrates insensitivity to angiotensin II and decreased channel stability.

Figure 1.. A novel somatic 12-bp duplication mutation in KCNJ5 leads to the expansion of the highly conserved pore helix of GIRK4.

A, Chromatograms of partial KCNJ5 sequences from a patient's normal adjacent cortex and tumor genomic DNA. B, Protein sequence alignment and consensus sequence of Kir family members (using CLUSTAL omega), the duplicated region is highlighted by a red box. C, A homology model of GIRK4 embedded into the plasma membrane, highlighting the selectivity filter and downstream pore helix containing the mutation.

Figure 2.. Functional characterization of dupAFLF-GIRK4 shows CCh induced Na⁺ currents and reduced selectivity.

A, Continuous current recordings of oocytes held at −80 mV, expressing either wild type (WT) or dupAFLF-GIRK4 coinjected with M2R and GIRK1. Oocytes were exposed to bath solutions containing either 98 mM Na⁺ or 98 mM K⁺ with 6 μM CCh. B, Ratios of CCh evoked vs basal current for at least seven oocytes expressing wild type or dupAFLF-GIRK4 with M2R and GIRK1 in 98 mM K⁺ or 98 mM Na⁺ bath solutions. Error bars represent mean ± 95% confidence interval. *, P < .05 by unpaired t test. C and D, I-V plots of the mean of seven oocytes expressing wild type or dupAFLF-GIRK4 with M2R and GIRK1. Oocytes were clamped at voltages from −90 mV to +20 mV in 10-mV steps. Voltage families were recorded in 98 mM K⁺ (C) or 98 mM Na⁺ (D) bath solutions with or without the addition of 6 μM CCh. Currents were normalized to the maximal current for each oocyte and expressed relative to the maximal peak current (I/Imax). Error bars represent SEM of seven oocytes. *, P < .05, **, P < .01 by paired t test. E, E_rev in oocytes expressing wild type or dupAFLF-GIRK4 along with M2R and GIRK1, in three solutions containing 98 mM K⁺, 49 mM K⁺ with 49 mM Na⁺, and 98 mM Na⁺. Error bars represent mean ± SEM of at least 26 oocytes. F, Resting membrane potential of oocytes expressing wild type or dupAFLF-GIRK4. Error bars represent mean ± SEM of at least five oocytes. *, P < .005 by unpaired t test.

Figure 3.. Reduced expression and stability of dupAFLF-GIRK4 in mammalian cell lines.

A, Confocal microscopy of HeLa cells expressing wild type or dupAFLF-GIRK4-GFP. Plasma membrane (PM) stained with WGA 594, GIRK4 visualized by using a GFP-tag, and nuclei stained with 4′,6′-diamino-2-phenylindole (DAPI). B, Total and cell surface biotinylated proteins from HEK293T cells expressing GIRK4-GFP wild type or dupAFLF. Anti-GIRK4 antibody detected three bands with molecular weights that were consistent with tetramer, dimer, and monomer forms of GIRK4. As a loading control, β-actin was stained. C, Quantification of Western blot showing total GIRK4 expression in HEK293T cells expressing wild type or dupAFLF-GIRK4-GFP. Error bars represent mean ± SEM of three Western blots. *, P < .05 by unpaired t test. D, Relative fluorescent units of HEK293T cells expressing wild type or dupAFLF-GIRK4-GFP measured by flow cytometry. Error bars represent SEM of independent experiments, each with 10 000 events. *, P < .05 by unpaired t test.

Figure 4.. Angiotensin-independent increase in aldosterone secretion in cells expressing dupAFLF-GIRK4.

Aldosterone secretion from H295R cells expressing GIRK4-GFP wild type, delI157 (positive control), and dupAFLF over 24 hours, with or without 10 nM angiotensin II stimulation. Error bars represent mean ± SEM of at least five independent experiments *, P < .05 calculated by a two-way ANOVA.