The Journal of Physiology
A revised mechanism of action of hyperaldosteronism-linked mutations in cytosolic domains of GIRK4 (KCNJ5)
Boris Shalomov1 D, Reem Handklo-Jamal1 ID, Haritha P. Reddy1,2 (D, Neta Theodor1, Amal K. Bera2 (D and Nathan Dascal1 i İD
1 Department of Physiology and Pharmacology, School of Medicine, and Sagol School of Neuroscience, Tel Aviv University, Tel Aviv, Israel
2 Department of Biotechnology, Bhupat and Jyoti Mehta School of Biosciences, Indian Institute of Technology Madras, Chennai, India
Edited by: Peying Fong & Florian Lesage
Linked articles: This article is highlighted in a Perspectives article by Kubo. To read this article, visit https://doi.org/10.1113/JP282777.
The peer review history is available in the Supporting Information section of this article (https://doi.org/10.1113/JP282690#support-information-section).
KCNJ5 (GIRK4)
* Non-pore mutants/ Cytosolic domain mutants
* Pore mutants
-100
0
50
mV
-100
0
50
mV
- WT
- WT
Mutant
nA
Mutant
nA
Loss of function. (Low expression and/or poor activation). Normal selectivity. Depolarization or loss of repolarization
Excessive aldosterone secretion
Loss of selectivity, loss of rectification, Na* influx, depolarization
45
$5
*
$ *
#
38
$
#
Activators for non-pore mutant channels
Suggested treatment
Inhibitors for pore mutant channels
The Journal of Physiology
Boris Shalomov (MSc) is a graduate student in Dascal’s lab. Boris specializes in ion channel physiology and pharmacology, the role of ion channels, especially GIRKs, in the function of secretory cells and neurons. Nathan Dascal (PhD) is Professor of Physiology at the School of Medicine of Tel Aviv University. His long-standing interest is in the study of GPCR-G protein-ion channel signal transduction mechanisms and in the biophysics, biochemistry and cell biology of G proteins and ion channels.
This article was first published as a preprint: Shalomov B, Handklo-Jamal R, Reddy HP, Theodor N, Bera AK, Dascal N. 2019. A revised mechanism of action of hyperaldosteronism-linked mutations in cytosolic domains of GIRK4 (KCNJ5). bioRxiv. https://doi.org/10.1101/866202
Abstract G protein-gated, inwardly rectifying potassium channels (GIRK) mediate inhibitory transmission in brain and heart, and are present in the adrenal cortex. GIRK4 (KCNJ5) sub- units are abundant in the heart and adrenal cortex. Multiple mutations of KCNJ5 cause primary aldosteronism (PA). Mutations in the pore region of GIRK4 cause loss of K+ selectivity, Na+ influx and depolarization of zona glomerulosa cells followed by hypersecretion of aldosterone. The concept of selectivity loss has been extended to mutations in cytosolic domains of GIRK4 channels, remote from the pore. We expressed aldosteronism-linked GIRK4R52H, GIRK4E246K and GIRK4G247R mutants in Xenopus oocytes. Whole-cell currents of heterotetrameric GIRK1/4R52H and GIRK1/4E246K channels were greatly reduced compared with GIRK1/4wT. Nevertheless, all heterotetrameric mutants retained full K+ selectivity and inward rectification. When expressed as homotetramers, only GIRK4WT, but none of the mutants, produced whole-cell currents. Confocal imaging, single-channel and Förster Resonance Energy Transfer (FRET) analyses showed: (1) reduction of membrane abundance of all mutated channels, especially as homotetramers, (2) impaired interaction with GBy subunits, and (3) reduced open probability of GIRK1/4R52H. VU0529331, a GIRK4 opener, activated homotetrameric GIRK4G247R channels, but not GIRK4R52H or GIRK4E246K. In the human adrenocortical carcinoma cell line (HAC15), VU0529331 and over- expression of heterotetrameric GIRK1/4wT, but not overexpression of GIRK1/4 mutants, reduced aldosterone secretion. Our results suggest that, contrary to pore mutants of GIRK4, non-pore mutants R52H and E246K mutants are loss-of-function rather than gain-of-function/selectivity-loss mutants. Hence, GIRK4 openers may be a potential course of treatment for patients with cytosolic N- and C-terminal mutations.
(Received 5 December 2021; accepted after revision 21 December 2021; first published online 26 December 2021) Corresponding authors Nathan Dascal: Department of Physiology and Pharmacology, School of Medicine, and Sagol School of Neuroscience, Tel Aviv University, Tel Aviv 69978, Israel. Email: dascaln@tauex.tau.ac.il Boris Shalomov: Department of Physiology and Pharmacology, School of Medicine, and Sagol School of Neuroscience, Tel Aviv University, Tel Aviv 69978, Israel. Email: shalomov@mail.tau.ac.il
Abstract figure legend There are two mutation types in KCNJ5 (GIRK4) that can cause excessive secretion of aldosterone, leading to primary aldosteronism. Mutations of the first type render the channel non-selective to mono- valent cations and often constitutively active, thus depolarizing the zona granulosa cells. This previously described mechanism underlies the disease-causing effects of mutations of amino acid residues located in the pore region (red colour). Blockers of the channel may be useful as potential treatment to reduce aldosterone secretion. Here we show that mutations of the second type, located in the cytosolic domain remote from the pore, act by a different mechanism. They do not alter the channel’s ion selectivity or rectification but cause poor expression or poor activation by GBy, resulting in a reduction in the cell’s K+ conductance and depolarization. In this case, GIRK4 openers can potentially be useful to prevent the excessive aldosterone secretion.
Key points
· Mutations in GIRK4 (KCNJ5) G protein-gated channels cause primary aldosteronism, a major cause of secondary hypertension. The primary mechanism is believed to be loss of K+ selectivity.
· R52H and E246K, aldosteronism-causing mutations in cytosolic N- and C- termini of GIRK4, were reported to cause loss of K+ selectivity.
· We show that R52H, E246K and G247R mutations render homotetrameric GIRK channels non-functional. In heterotetrameric context with GIRK1, these mutations impair membrane expression, interaction with GBy and open probability, but do not alter K+ selectivity or inward rectification.
· In the human aldosterone-secreting cell line, a GIRK4 opener and overexpression of heterotetrameric GIRK1/4wT, but not overexpression of GIRK1/4 mutants, reduced aldosterone secretion.
· Aldosteronism-causing mutations in the cytosolic domain of GIRK4 are loss-of-function mutations rather than gain-of-function, selectivity-loss mutations. Deciphering of exact biophysical mechanism that impairs the channel is crucial for setting the course of treatment.
Introduction
G protein-gated, inwardly rectifying potassium channels (GIRK; Kir3) mediate inhibitory signalling via G protein-coupled receptors (GPCR) in the brain, heart and adrenal cortex. There are four GIRK sub- units (GIRK1-4) forming homo- (GIRK2, GIRK4) or heterotetrameric channels (GIRK1/2, GIRK1/3, GIRK1/4, GIRK2/3). GIRK channels are regulated by phosphotidylinositol-4,5-bisphosphate (PIP2) and Na+ (Sui et al. 1996; Huang et al. 1998; Sui et al. 1998) and activated by the GBy dimer, which is the main physio- logical gating factor of GIRK channels (Logothetis et al. 1987; Reuveny et al. 1994). Upon GPCR activation, GBy dissociates from Gai/0, binds to the channel and activates it (Dascal 2001). Like in all inward rectifiers, in GIRKs the outward current is smaller than the inward, due to the occlusion of the cytoplasmic pore by Mg2+ and polyamines (Lu 2004). GIRK4 subunits, encoded by the KCNJ5 gene, are mostly abundant in the heart and adrenal cortex, forming GIRK4 homo- and/or GIRK1/4 (KCNJ3/KCNJ5) heterotetrameric channels (Krapivinsky et al. 1995; Wickman et al. 1998; Choi et al. 2011). GIRK channels play critical roles in the regulation of heart rate and aldosterone secretion (Kurachi 1995; Choi et al. 2011). However, subunit composition of GIRK channels in human adrenal zona glomerulosa (AZG) cells of the adrenal cortex is yet to be elucidated. Notably, much lower mRNA levels of GIRK1 compared with GIRK4 have been reported in these cells (Choi et al. 2011), indicating a possible predominance of GIRK4 homotetramers.
Aldosterone, the steroid hormone secreted by the AZG cells, is essential for blood pressure regulation. In physiological conditions, aldosterone secretion is regulated by the interplay between angiotensin II and plasma K+ concentration, [K+] (Spat 2004). Angiotensin activates the angiotensin II type 1 receptor, leading to depolarization of aldosterone-secreting cells through inhibition of K+ channels and the Na+/K+ pump (Al-Salameh et al. 2014). Consequently, L- and T-type voltage-gated Ca2+ channels (VGCC) open, Ca2+ enters the cell and activates aldosterone secretion. Any change in the level of angiotensin II or [K+] (even 1 mM) affects aldosterone secretion (Spat 2004).
Primary aldosteronism (PA) is a disease characterized by hypersecretion of aldosterone. PA accounts for 90% of secondary hypertension cases, approximately 10% of hypertensive patients worldwide (Fernandes-Rosa
et al. 2017). The main causes of PA are somatic and germline mutations in the KCNJ5, CACNA1D, CACNA1H, ATP1A1, ATP2B3, CTNNB1 and ARMC5 genes (Fernandes-Rosa et al. 2017). Sporadic and familial mutations in KCNJ5 account for about half of PA cases, often accompanied by adrenal adenoma (Williams et al. 2015; Fernandes-Rosa et al. 2017). KCNJ5 germline mutations cause familial hyperaldosteronism type III, out of four types (Monticone et al. 2017). Choi et al. showed that PA-causing mutations in the GYG motif of the selectivity filter (G151R) or in close proximity to it (T158A and L168R) in GIRK4 cause a loss of selectivity for K+, yielding K+/Na+ non-selective GIRK4 and GIRK1/4 channels (Choi et al. 2011). It is widely accepted that this results in depolarization caused by the influx of Na+ and a consequent influx of Ca2+ through L- and T-type VGCCs followed by constitutive aldosterone secretion and hyper- tension (Al-Salameh et al. 2014; Hattangady et al. 2016; Funder 2019). Since 2011, additional PA-linked mutations were discovered in the KCNJ5 gene, in the pore region or in the cytosolic N- and C-terminal domains of GIRK4 (Murthy et al. 2012, 2014; Cheng et al. 2015; Monticone et al. 2017).
Murthy et al. (2014) identified PA-linked germline GIRK4 mutations R52H and E246K, located in GIRK4 cytosolic N- and C-terminal domains, respectively. Over- expression of GIRK4R52H and GIRK4E246K in human adrenocortical carcinoma cells, NCI-H295R, increased the production of aldosterone (Murthy et al. 2014). They further reported loss of K+ selectivity and inward rectification for R52H and E246K mutations expressed in Xenopus oocytes together with GIRK1 subunit, and proposed that the excessive secretion of aldosterone caused by these mutations shares the same biophysical mechanism with previously studied pore mutations (Murthy et al. 2014). However, unlike the well-established effect of mutations in the pore region (Hille 2002), it is more challenging to comprehend how mutations in cyto- solic domains, remote from the pore (Fig. 1), would affect selectivity. G247R is another mutation within the cyto- solic domain found to be causally related to PA, but the mechanism remains controversial. Both loss of function (Calloe et al. 2007) and no alteration in channel activity (Murthy et al. 2014) have been reported for the G247R mutation. Thus, the biophysical and cellular mechanisms of GIRK4 channel malfunction caused by PA-linked mutations in the cytosolic domain remain incompletely understood.
Here, we investigated GIRK4R52H, GIRK4E246K and GIRK4G247R mutants expressed in Xenopus oocytes and in the human adrenocortical carcinoma cell line (HAC15). We demonstrate that all heterotetrameric GIRK1/4 mutants examined remain selective to K+ and show an unimpaired inward rectification. In contrast, the activity of heterotetrameric GIRK1/4R52H and GIRK1/4E246K (but not GIRK1/4G247R) channels is severely impaired compared with wild-type (WT) GIRK1/4WT. Homo- tetrameric GIRK4R52H, GIRK4E246K and GIRK4G247R mutants are non-functional. Single-channel analysis, confocal imaging and Förster Resonance Energy Trans- fer (FRET) suggested both impaired gating and GBy interaction, and compromised plasma membrane (PM) expression in GIRK4R52H and GIRK4E246K. For GIRK4G247R, the only significant defect was a severe impairment of PM expression for its homotetrameric composition. A homotetrameric GIRK4 channel opener, VU0529331, enhanced GIRK4G247R currents in Xenopus oocytes and reduced aldosterone secretion in native HAC15 cells, suggesting a potential strategy for treatment in affected patients. Additionally, we found that compared with native HAC15 cells, transfection of GIRK1/4WT reduced aldosterone secretion. Transfection of GIRK1/4R52H induced significantly higher aldosterone secretion than GIRK1/4wT. Our results suggest that R52H and E246K mutations are loss-of-function mutations that impair channel gating and membrane abundance, but not K+ selectivity or inward rectification. Hence, we suggest that the mechanism previously proposed for R52H and E246K mutants should be revised.
Materials and methods
Ethical approval of Xenopus laevis, oocyte preparation and electrophysiology
Experiments were approved by Tel Aviv University Institutional Animal Care and Use Committee (permit #01-16-104). Maintenance and surgery of female frogs were as described previously (Dascal & Lotan 1992). All materials were from Sigma unless indicated otherwise. Oocytes were defolliculated by collagenase, injected with RNA and incubated for 3 days at 20-22℃ in ND96 solution (low K+) (in mM: 96 NaCl, 2 KCI, 1 MgCl2, 1 CaCl2, 5 Hepes; pH 7.5), supplemented with 2.5 mM sodium pyruvate, 100 µg ml-1 streptomycin and 62.75 µg ml-1 penicillin (or 50 µg ml-1 gentamycin) (Rubinstein et al. 2007). Whole-cell GIRK currents were measured using the standard two-electrode voltage clamp method at 20-22℃, in different K+ concentrations (8, 24, 48, 72 or 96 mM K+ (HK)) (Rubinstein et al. 2007). Different concentrations of K+ solutions were obtained by mixing ND96 with a 96 mM K+ solution containing, in mM: 96 KCI, 2 NaCl, 1 CaCl2, 1 MgCl2, 5 Hepes;
pH adjusted to 7.5 with KOH. Acetylcholine (ACh) was added to HK solution as agonist at 10 pM, Ba2+ (2.5 mM) was used to block the channel. Current-voltage (I-V) relationships were obtained using 2 s voltage ramps from -150 mV or -120 mV to +50 mV. Net GIRK I-V relationships were obtained by subtracting the current remaining after blocking all GIRK activity with 2.5 mM Ba2+. In one series of experiments (Fig. 2), we obtained the net GIRK I-V relationship by sub- tracting the averaged I-V relationship from native oocytes injected with only GBy. For small GIRK currents (in low [K+]out) this procedure is superior over the subtraction of Ba2+-blocked currents, avoiding possible inaccuracies caused by the different extent of a Ba2+ block of inward and outward GIRK currents (Rubinstein et al. 2007).
VU0529331 (Alomone Labs; V-155), an opener of homotetrameric GIRK2 and GIRK4 channels, was dissolved in 100% DMSO to a final concentration of 25 mM. To measure the GIRK4 response to VU0529331, the drug was diluted to a HK24 solution to get HK24 + 60 µM VU0529331 solution. Consecutively, this solution was diluted again to get 20 uM, and so on for all concentrations.
DNA constructs and RNA
cDNA constructs of rat GIRK1, N-terminally YFP-tagged GIRK1 (YFP-GIRK1), N-terminally CFP-labelled GY1 (CFP-Gy), GIRK4WT (wild-type), GIRK4R52H, GIRK4E246K and GIRK4G247R were inserted into pBS-MXT or pGEM-HJ vectors that contain 5’ and 3’ untranslated regions from Xenopus ß-globin (Berlin et al. 2011). DNA of human N-terminally YFP-tagged GIRK4WT (YFP-GIRK4WT) was a gift from Prof. Wolfgang Schreibmayer (Graz Medical University, Austria), and mutations were made to produce YFP-GIRK4R52H, YFP-GIRK4E246K and YFP-GIRK4G247R. mRNAs were prepared as described previously (Rubinstein et al. 2007).
The amounts of mRNA injected per oocyte were varied according to the experimental design, in ng/oocyte: 0.5-2 GIRK1, 0.5-2 YFP-GIRK1, 0.25-1 GIRK4, 0.25-5 YFP-GIRK4. For maximal channel activation by GBy, we injected 5 ng GB and 1-2 ng Gy mRNA. These weight ratios were chosen to keep approximately equal molar amounts of GB and Gy RNA. RNA of the muscarinic two receptor (M2R), if present, was 1 ng. Twenty-five nanograms of the anti-GIRK5 oligo nucleotide antisense XIR was routinely injected to prevent the formation of GIRK1/5 channels (Hedin et al. 1996).
HAC15 cells culture, transfection, biochemistry, electrophysiology and aldosterone measurements
Cells were acquired from ATCC (ATCC CRL3301) and cultured as described (Parmar et al. 2008). Cells
were grown in Dulbecco’s Modified Eagle’s/Ham’s F-12 medium (DMEM/F-12) (Gibco 11330-05) containing 10% cosmic calf serum (Hyclone SH30), 1% L-glutamine (Sigma-Aldrich; A7506), 1% ITS (Becton Dickenson - FAL354352) and penicillin-streptomycin (Sigma-Aldrich; P4333).
For western blot, cultured HAC15 cells were treated as described previously (Keren Raifman et al. 2017). GIRK1/4 RNA-injected oocytes were used as positive control. Oocytes were homogenized on ice in homo- genization buffer (20 mM Tris, pH 7.4, 5 mM EGTA, 5 mM EDTA and 100 mM NaCl) containing protease inhibitor mixture (Roche Applied Science). The homogenates were centrifuged at 3000 g for 5 min at 4°℃ and the pellet was discarded. Protein samples (35 µg) were electro- phoresed on 12% polyacrylamide-SDS gel and transferred to nitrocellulose membranes for western blotting with antibodies; anti-GIRK1 at 1:300 dilution (Alomone Labs, APC-005, Lot# APC005AN1102, RRID:AB_2040113) or anti-GIRK4 at 1:100 dilution (Alomone Labs, APC-027, Lot# APC027AN0702, RRID:AB_2039943) and GAPDH in 1:1000, 1:2000 or 1:4000 dilution (Abcam, ab37168, lo:GR303514-2, RRID:AB_732652). Goat anti-rabbit IgG antibody, (H+L) HRP conjugated secondary antibody at 1:40,000 dilution was applied (Jackson ImmunoResearch Labs 111-035-144, RRID:AB_2307391).
For electrophysiological experiments, cells were plated in a 6-well plate and were transfected with combinations of the following DNAs as designed for specific experiments: 0.5 ng GFP-GIRK1, 0.5 ng GIRK4WT, 0.5 ng GIRK4R52H, 0.5 ng GIRK4E246K, 0.5 ng GIRK4G247R, 0.5-1 ng GB1, 0.5-1 ng Gy2 (all in pcDNA3.1); 0.5 ng D2L receptor in pXOOM (the latter was kindly provided by Kristoffer Sahlholm, Umea University, Sweden). Cells were moved onto 13 mm cover slips 24-36 h after transfection, coated with fibronectin (F1141; Sigma-Aldrich) or Poly-L-lysine hydro-bromide (P2636; Sigma-Aldrich) and electro- physiological measurements were performed in the next 24-48 h.
Whole-cell currents were measured using Axopatch 200B (Molecular Devices, Sunnyvale, CA). Holding potential was -80 mV, and voltage ramps were from -120 to +20 mV; no correction for junction potential (~13 mV) was made. Currents were recorded in low-K+ solution (in mM): NaCl 136, KCI 4, CaCl2 2, MgCl2 2, Hepes 10, NaH2PO4 0.33, glucose 10; pH 7.4; or in high-K+ (HK48) solution (in mM): NaCl 92, KCI 48, CaCl2 2, MgCl2 2, Hepes 10, NaH2PO4 0.33, glucose 10; pH 7.4. GPCR response was measured using 100 µM dopamine. Electrode solution contained (in mM): NaCl 6, KCI 22, K-gluconate 110, MgCl2 2, Hepes 10, EGTA-HOH 1, ATP-K2 2, GTP-Tris 0.5; pH 7.2. VU0529331 was also applied in whole-cell recordings, VU0529931 was dissolved in DMSO and diluted in HK48 solution to final 0.1% DMSO. We added DMSO to the control solution and
saw that it had no effect on recordings. Hence, in later recordings we did not add it to the control solution.
For aldosterone secretion measurements, HAC15 cells were plated in 30 mm plates. Twenty-four hours later cells were transfected using lipofectamine 3000 (L3000008; Thermo Fisher Scientific, USA) with 1 µg GFP only (i.e. control group), and 1 µg GFP, 2.5 µg GFP-GIRK1 and 2.5 µg GIRK4 WT or mutants. The transfection efficiency was around 30-50%. The medium was replaced with new medium or medium with VU0529331 (25 [M) 72-96 h post-transfection, and 3 h later 1 ml of the medium were collected, frozen in liquid nitrogen and stored at -80°℃ for later analysis. After collection of medium the cells were scraped from plates in Dulbecco’s phosphate-buffered saline (02-023-1A; Biological Industries), centrifuged and used for western blotting as described above. A competitive enzyme immunoassay (KGE016; R&D Systems Inc., USA) was used to determine aldosterone levels. The bioluminescence signals were measured using SpectraMax I3 microplate reader (Molecular Devices). In the analysis we first tested the western blots to verify that the number of cells in all groups was similar by comparing GAPDH levels. Next, we normalized the aldosterone levels to the aldosterone level of the control group.
Single-channel recordings in Xenopus oocytes
Patch clamp experiments were done using Axopatch 200B as described (Yakubovich et al. 2015). Currents were recorded at -80 mV, filtered at 2 kHz and sampled at 20 kHz. Patch pipettes had resistances of 1.4-3.5 MS2. Pipette solution contained, in mM: 146 KCI, 2 NaCl, 1 MgCl2, 1 CaCl2, 1 GdCl3, 10 Hepes/KOH (pH 7.5). GdCl3 completely inhibited the stretch-activated channels. For cell-attached recordings, the bath solution contained, in mM: 146 KCI, 2 MgCl2, 1 EGTA, 10 Hepes/KOH (pH 7.5). For excised patch recordings, the bath solution was Na-free and contained, in mM: KCI 146, MgCl2 2, Mg-ATP 2, EGTA 1, Hepes 10 (pH 7.5). To obtain single-channel recordings, oocytes were injected with low doses of RNA of GIRK1 (10-100 pg/oocyte) and RNA of GIRK4 was half of that (5-50 pg); the amount of GIRK1/4WT RNA was 10-20/5-10 pg, respectively. The amounts of the GIRK1/4R52H and GIRK1/4E246K channels’ RNA were 20-40/10-20 pg and 40-100/20-50 pg, respectively. In addition, 25 ng of the antisense DNA oligonucleotide against the end- ogenous GIRK5 channel was injected together with the RNAs (Hedin et al. 1996). Single-channel current (isingle) was calculated from all-point histograms of the original records (Yakubovich et al. 2009), and open probability (Po) was obtained from event lists generated using an idealization procedure based on the 50% crossing criterion (Sakmann & Neher 1995). The number of
channels was estimated from overlaps of openings during the whole time of recording (at least 5 min). P. was calculated only from records that contained up to three channels. Thus, the probability of missing a channel was negligible. For channels activated by coexpressed GBy there was no decrease in Po over >4 min, and the Po was averaged from the first 4 min of the record. Single-channel conductance (g) was measured from I-V relationships, constructed from values of isingle measured at potentials ranging from -120 mV to -40 mV at 20 mV increments.
Confocal imaging
Confocal imaging of the oocytes and analysis were performed as described (Berlin et al. 2011), with a Zeiss 510 META confocal microscope, using a 20x objective. In whole oocytes, the image was focused on an oocyte’s animal hemisphere, at the equator. Images were acquired using spectral (2)-mode. YFP was excited with the 514 nm line of the argon laser and sampled at 534-546 nm. Fluorescence signals were averaged from three regions of interest using Zeiss LSM Image Browser, background and the average signal from uninjected oocytes were sub- tracted.
Förster resonance energy transfer (FRET)
FRET experiments were performed using the sensitized emission spectral method (Zheng et al. 2003) as described previously (Berlin et al. 2011). Two spectra were collected from the animal hemisphere of each oocyte, with 405 nm (CFP excitation) and 514 nm (YFP excitation) laser lines. Oocytes with low signal of one of the fluorophores (usually <100 arbitrary units (AU)), or with coefficient of variation of A parameter at 524, 535, 545, 566 nm of more than 0.25, were excluded from analysis. Net FRET signal was calculated in the YFP emission range (with the 405 nm excitation) by consecutive subtraction of a scaled CFP-only spectrum (giving the A ratio parameter) and then of the ratio A0, which reports the direct excitation of YFP by the 405 nm laser, as in eqns (1) and (2):
Ratio A F405 F514 F514 F514
FFRET F514 Fdirect FFRET (1)
(A - A0) = (2)
Because of the use of different experimental settings in independent experiments, the fluorescence intensities or A ratios cannot be used to construct FRET titration curves. We imaged a double-labelled protein, expressing both CFP and YFP at a 1:1 stoichiometry (YFP-GIRK2-CFP, DL-GIRK2) in each experiment to convert the fluorescence of CFP and YFP into their molar ratio (the average donor/acceptor ratio in AU corresponds
to a molar ratio of one) (Berlin et al. 2011). The molar ratio collected from many experiments can then be incorporated into one titration curve (Berlin et al. 2011).
The apparent FRET efficiency in an individual cell, Eapp, was calculated as in eqn (3):
Eapp = − ( Ratio A Ratio A0 1
)
EA ED (3)
where ED and EA are molar extinction coefficients for the donor and acceptor, respectively, at the donor excitation wavelength (Bykova et al. 2006; Berlin et al. 2011).
Data analysis
I-V curves were analysed using scripts written in-house and from Mathworks forum for analysing .abf files produced by Clampex (Molecular Devices), MATLAB R2015a (MathWorks, USA) and with Clampfit 10.7 software (Molecular Devices). The reversal potential (Vrev) was determined from the intercept of the net I-V curve with voltage axis (Rubinstein et al. 2007). The inward rectification index (Fir) was determined by dividing the current at 50 mV positive to Vrev by the current at 50 m V negative to Vrev (see Fig. 2B) (Rubinstein et al. 2007):
Fir r = IVrev +50 Ivrev-50 (4)
Estimated reversal potential (Vrev) was calculated using the Nernst equation:
Vrev=F In Out in ) (5)
RT zF
Permeability ratio of sodium potassium (pNa+/pK+) and was determined from Goldman-Hodgkin-Katz equation:
Erev =Flog ZF RT
PK+[K+]0 + PNa+ [Na+], PK+[K+]; + PNa+ [Na+]i ) (6)
Statistical analysis
To calculate the variance (Var) of the ratio I/EAU (I/E; for Fig. 6), we used the equation (Stuart & Ord 1998):
Var (!) = (MR)2 [Var (I) _2Cov (I, E) Var (E) =
(ps)2
(PI)2 μιμε (LE)2 I (7)
where u denotes sample mean, Cov stands for covariance.
Statistical analysis was performed using SigmaPlot 11 or 13 (Systat Software, Inc.) and GraphPad Prism version 8 for Windows (GraphPad Software, La Jolla California USA). In all analyses, if the data passed the
Shapiro-Wilk normality test and the equal variance test, two-group comparisons were performed using t tests, while a multiple group comparison was performed using one-way ANOVA. For normally distributed data, the data in the figures (bar graphs and line plots) are presented as means ± SD. If the data did not pass the normality test, a Mann-Whitney rank sum test or Kruskal-Wallis test were performed, respectively, and data in the figures are presented as box plots, with all data points shown. Boxes show the 25th and 75th percentiles and whiskers show the minimal and maximal values, black horizontal line shows the median and green line shows the mean.
Current amplitudes across several experiments (Fig. 1D) have been normalized as described previously (Kanevsky & Dascal 2006). For analysis across several experiments, Ibasal or IBy in each oocyte was normalized to the average current in the oocytes of the control group
of the same experiment. This procedure yields average normalized intensity as well statistical variability (e.g. SEM) in all treatment groups as well as in the control group.
Results
GIRK 1/4WT and GIRK1/4 mutants produce basal and GBy-activated currents in Xenopus laevis oocytes
The mutated amino acids (R52, E246, G247) are fully conserved in all GIRK subunits (Fig. 1A), suggesting their functional significance. They are located in the cytosolic domain far from the selectivity filter, as illustrated in Fig. 1B for the homologous amino acids in the crystal structure of GIRK2 (Whorton & Mackinnon 2013).
A
C
HK24
ND96
GIRK4 47 KKPRORYMEKTG 58 239 IKSROTKEGEFIPLNOTDINVGF 261
ND96
HK24 + 2.5 mM Ba2
GIRK1 40 KKKRQRFVDKNG 51 233 LKSROTPEGEFLPLDQLELDVGF 255
0+
no GBY
basal
GIRK2 52 KRKIQRYVRKDG 63 244 IKSKOTSEGEFIPLNQTDINVGY 266
IBY
GIRK3 17 RRGRORYVEKDG 28 210 IRSROTLEGEFIPLHQTDLSVGF 232
with GBY
B
Top view
4
Ramps
5
5 sec
Membrane
boundaries
D
71
26
23
40
73
33
32
40
4
(10)
(4)
(3)
(5)
2.5
(11) (5)
(4)
(5)
Normalized Ibasal
2.0
n.s.
3
Normalized IBy
R52
Bottom view
1.5
E246
2
G247
p=0.0002
1.0
%
1
p<0.0001
0.5
p<0.0001
0
#
0.0
WT
R52H
E246K
G247R
WT
R52H
E246K
G247R
E
GIRK1/4 (Ibasal)
GIRK1/4 + GBY (IBy)
F
no GBY (Ibasal)
with GBY (IBy)
V (mV)
1
V (mV)
5-
V (mV)
1-
V (mV)
5-
-120 -80
-40-1
-2
(μΑ)
40
-120
-80
40-5
40
-120
-80
-40
-1
40
120
-80.
-40 -5-
40
-10
(HA)
-15
-2
(NA)
-10-
(A)
-3
-3
-15-
-4
-20
-4
-20-
-5
-25
-25-
-1HK24
-30
-5
- WT
— R52H
-30-
-6
… |HK24+Ba -35
-6
-35-
Net IGIRK
E246K
… G247R
A, alignment of GIRK4 with GIRK1, GIRK2 and GIRK3 shows that R52, E246 and G247 are conserved amino acids. B, crystal structure of GIRK2 (Whorton & Mackinnon 2013) marked with highlighted positions of the mutations; R52 - red, E246 - blue and G247 - orange (numbering corresponds to GIRK4). C, example of the recording protocol in GIRK1/4wT-expressing oocytes without (grey) or with (black) coexpressed GBy. Horizontal bars show the corresponding external solutions. Voltage ramps were applied in HK24 and HK24 + 2.5 mM Ba2+ solutions. D, normalized /basal (left) and IBy (right) of GIRK1/4WT and mutants. Numbers above box plots denote number of cells and, in parentheses, number of experiments. Boxes show 25th and 75th percentiles and whiskers show minimal and maximal values, black horizontal line - median, green line -mean. E, representative I-V relationships of currents of GIRK1/4WT (/basal; left) and GIRK1/4WT + GBy (IBy; right). Net GIRK currents (/GIRK) were obtained by subtraction of the I-V curve in Ba2+ (‘Ba2+-subtraction’ procedure). F, net I-V relationships of GIRK 1/4WT and mutants; /basal (left) and IBy (right). [Colour figure can be viewed at wileyonlinelibrary.com]
We expressed heterotetrameric GIRK1/4 channels in Xenopus laevis oocytes and recorded whole-cell currents using a two-electrode voltage clamp. For GIRK1/4WT channels, switching from low-K+ (2 mM K+) ND96 solution to a high-K+ (24 mM K+) HK24 solution yielded an inward current reflecting the GIRK’s basal activity (Ibasal). GIRK1/4wT was further activated by GBy, which is evident from the strong increase (6.86 ± 0.42-fold) in the constitutive K+ current (IBy) in GBy-expressing oocytes (Fig. 1C). Ibasal of GIRK1/4R52H and GIRK1/4E246K was significantly smaller than Ibasal of GIRK1/4WT, and in some experiments it was indistinguishable from native oocyte’s currents, especially for GIRK1/4R52H (Fig. 1D). To determine GIRK I-V relationships, we used voltage ramp protocols (Fig. 1E and F). The I-V curves of basal and GBy-evoked currents of GIRK1/4wr showed the distinctive inward rectification pattern (Fig. 1E). Ibasal of GIRK1/4R52H and GIRK1/4E246K often showed seemingly non-rectifying I-V relationships (Fig. 1F, left); however, as we show below, this was an artefact arising from the very small GIRK current, comparable to oocyte’s endogenous currents. GIRK1/4G247R showed distinctive inward rectification, yet Ibasal was significantly lower compared with GIRK1/4WT (Fig. 1D and F). On the other hand, IBy of GIRK1/4G247R was similar to IBy of GIRK1/4WT, while IBy of GIRK1/4R52H and GIRK1/4E246K was significantly lower (Fig. 1D and F).
We also coexpressed GIRK1/4 with the muscarinic acetylcholine receptor M2 (M2R). Application of acetyl- choline (ACh) evoked additional GIRK current, Ievoked (Rubinstein et al. 2009). Incidentally, however, addition of ACh also activated oocyte’s endogenous Ca2+-dependent chloride channels, possibly via endogenous Gq-coupled muscarinic receptors occasionally present in the oocytes, or promiscuous coupling of M2R to Gq (Dascal 1987). Unlike GIRK currents, these Cl- currents were not blocked by Ba2+ but were blocked by the injection of the Ca2+ chelator BAPTA (data not shown). Ca2+-dependent chloride currents may be a major cause of artefact when estimating the reversal potential (Vrev) when the expressed GIRK currents are small. To avoid these artefacts, in the following we activated the GIRK channels by coexpressing GBy.
GIRK4 mutants are selective for K+ and retain inward rectification properties
To examine whether GIRK4 mutants lose selectivity or inward rectification, we coexpressed GIRK1/4 channels with GBy and measured whole-cell currents while applying voltage ramps in four concentrations of extracellular K+, [K+]. (Fig. 2A and C). Vrev was measured from the intercept of the net GIRK I-V curve with the voltage axis (Fig. 2B and C). Plots of Vrey as
a function of [K+]. (plotted on log10 scale, Fig. 2D) gave fairly straight lines. The slopes of the Vrev VS. log10[K+]o plots, in mV per tenfold change in [K.] (decade), were 53.33 ± 0.77, 58.49 ± 1.6, 58.9 ± 2.96 and 54.36 ± 0.42 for GIRK1/4WT, GIRK1/4R52H, GIRK1/4E246K and GIRK1/4G247R, respectively (Fig. 2D). These values are close to ~58 mV/decade predicted for a selective K+ channel (Hille 2002). In contrast, the slope for GIRK1/4Y152C, a known non-selective pore mutant (Monticone et al. 2013) used as control, was 13.45 ± 4 mV/decade (Fig. 2C and D). Permeability ratios (PNa/PK) for GIRK1/4wT and all mutants (R52H, E246K and G247R) were close to zero, as opposed to the pore mutant GIRK4Y152C (Fig. 2E). Thus, these cytosolic N- and C-terminal mutations do not affect the K+ selectivity. We also quantified the extent of inward rectification using the Fir index (Rubinstein et al. 2007; see Methods and Fig. 2B). Fir was similar for GIRK1/4wT and all mutants (Fig. 2F), usually <0.1, except for GIRK1/4Y152C where Fir was significantly higher, indicating partial loss of rectification in this channel, as reported (Monticone et al. 2013). We conclude that mutated channels are inwardly rectifying to the same extent as WT channels.
R52H, E246K and G247R mutations in GIRK4 impair channels’ surface expression and interaction with GBy
To assess the abundance of GIRK4 mutants in the PM, we expressed GIRK4 channels labelled with YFP at the N-terminus (YFP-GIRK4), with or without GB and CFP-Gy, as homo- and heterotetramers with GIRK1 (Figs 3 and 4, respectively). Since the abundance of GIRK4 vs. GIRK1/4 in aldosterone-secreting cells remains an open question (see Introduction), we first studied the GIRK4 homotetramers (Fig. 3). Each mutant was studied in a separate experiment and compared with YFP-GIRK4WT. Measurement of YFP fluorescence showed a significantly lower PM expression of YFP-GIRK4R52H and YFP-GIRK4E246K compared with YFP-GIRK4WT, with or without coexpression of GBy (Fig. 3A and B). PM expression of YFP-GIRK4G247R without GBy was reduced, but with GBy, it was similar to YFP-GIRK4WT (Fig. 3B). Electrophysiological measurements showed that Ibasal of GIRK4 homo- tetramers was very small, usually indistinguishable from native oocyte currents. Although I-V relationships occasionally showed inward rectification, the small size of GIRK currents did not allow a rigorous analysis (Fig. 3C and D). Conversely, GBy-evoked YFP-GIRK4WT currents were quite substantial, ~0.5 pA (at -80 mV) on the average (Fig. 3E and F) and showed typical GIRK current characteristics (Fig. 3E). In contrast, IBy of all homotetrameric mutants was almost undetectable
(Fig. 3E and F). This was true even for the homotetrameric YFP-GIRK4G247R, despite its comparable level of surface expression to YFP-GIRK4WT, in the presence of GBy. Thus, homotetramers of the three mutants under study show impaired surface expression and appear non-functional.
We next studied heterotetramers of GIRK1 with WT or mutated YFP-GIRK4. The surface expression of GIRK1/ YFP-GIRK4E246K was significantly lower than GIRK1/ YFP-GIRK4WT, with or without coexpression of GBy (Fig. 4A and B). The PM levels of GIRK1/YFP-GIRK4R52H and GIRK1/YFP-GIRK4G247R were not significantly different from GIRK1/YFP-GIRK4WT. However, a significant reduction for both mutants was observed when they were coexpressed with GBy. In the same
experiments, all mutants showed reduced currents compared with WT, with or without GBy (Fig. 4C).
To examine whether R52H, E246K and G247R mutations impair GIRK4 interaction with GBy, we used spectral FRET (Zheng et al. 2003). We coexpressed CFP-GBy (donor; usually at two concentrations) with YFP-GIRK4 (acceptor) and GIRK1 (Fig. 4D and E). YFP-labelled, G protein-insensitive inwardly rectifying K+ channel IRK1 (Kir2.1; KCNJ2) was used as negative control (Fig. 4D). To circumvent variability between oocyte batches, we calculated donor/acceptor molar ratio in each oocyte using the double-labelled YFP-GIRK2-CFP (coexpressed with GIRK1) as a molecular calliper (Berlin et al. 2011). In sensitized emission FRET, the FRET efficiency increases with the donor/acceptor
A
K+ (mM)
B
2
8
24
72
-50mV
0
+50mV
Vrev
10 µA
Native + GBY
Fir=
IVrev + 50 mV
20 sec
GIRK1/4WT + GBY
Vrev - 50 mV
C
WT
R52H
Y152C
E246K
G247R
V (mV)
10
V (mV)
1
1.
V (mV)
2
V (mV)
20
-120 -80 -40
40 -120 -80 -40_
0.5
40
V (mV)
-120 -80 -40-2-
40 -120 -80 -40
(HA)
-20
40 K+ (mM)
-10-
-20
-2
3
-120 80 40
40
-4
3
-40
-3
-0.5
2
-6
-60
2
-30
8
-4
-1.
-8
-80
24
-40
-5
-1.5
-10
-100
72
0.2
0.2
0.2
0.2
0.2
V (mV)
V (mV)
V (mV)
V (mV)
V (mV)
-120
-80
-40
40
-120 -80
-40
40
-120 -80 -40
40 -120 -80 -40
40
-120
80
40
40
10.2
-0 2-
0.2-
-0. 2
-0.2
D
0
N=2
n N=2
E
N=2
F
n N=2
-20
G247R
8
G247R
8
G247R
8
Vrev (mV)
-40
E246K
7
p<0.0001
E246K
7
p<0.0001
E246K
7 JH
p=0.0085
-60
Y152C
8
Y152C
7
Y152C
8
-80
WT
R52H
R52H
7
R52H
5
R52H
7
-100
Y152C
WT
14
WT
13
WT
14 3 …
-120
E246K
G247R
2
8
24
72
-20 0 20 40 60 80
-0.5 0.0 0.5 1.0 1.5
0.0
0 0.2 0.4 0.6
[K+]] (mM)
Slope (mV/decade)
PN&/PK
Fir at 24 mM K+
ratio (Bykova et al. 2006). The FRET titration curve reaches saturation at high donor/acceptor molar ratios (Bykova et al. 2006) (>5 for the GIRK-GBy FRET pair (Berlin et al. 2011)). Since in our experiments we have not achieved full saturation (Fig. 4D), we divided the data into bins with comparable donor/acceptor molar ratios, in the range where the representation of each mutant or WT was at least three experimental points (Fig. 4E). We found that FRET efficiency (Eapp), which reports interaction with GBy, was significantly impaired for GIRK1/YFP-GIRK4R52H and GIRK1/YFP-GIRK4E246K in all donor/acceptor molar ratio ranges. The interaction of GBy with GIRK1/YFP-GIRK4G247R was impaired at lower or mid donor/acceptor molar ratios compared with GIRK1/YFP-GIRK4WT (Fig. 4E).
Single-channel analysis: GIRK4 mutants are normally activated by Na+ but GIRK1/4R52H shows impaired activation by GBy
To directly address the possible defects in channel function on molecular level, we measured GBy-induced channel activity using cell-attached patch clamp (Fig. 5A). We used GIRK1/4 heterotetramers, since mutated GIRK4 homotetramers were found non-functional. GIRK1/4 channels were coexpressed at low surface density, with GBy using doses of RNA that usually produce maximal Po in GIRK1/4wT. Single-channel recordings were made at -80 mV. Figure 5A shows exemplary recordings of GIRK1/4WT, GIRK1/4R52H and GIRK1/4E246K. We found that the open probability (Po) of GIRK1/4R52H was
with GBy no GBy
WT
R52H
WT
E246K
WT
G247R
E
YFP-GIRK4WT+GBY
YFP-GIRK4R52H+GBY
V (mV)
500
200
-120 -80
-40
-500
40
V (mV)
100
-1500
I (nA)
-120 -80 -40
40
-100
I (nA)
B
-2500
-200
Expression (AU*1000)
3
YFP-GIRK4E246K+GBY
YFP-GIRK4G247R+GBY
2
150
300
A
100
200
·
·
1
V (mV)
50
V (mV)
100
¥
:
L
-120 -80 -40
40
-120
,80
-40
40
0
0
100
no GBY
GBY
no GBY
GBY
no GBY
GBY
-50
I (nA)
-200
(nA)
n= 12 10
12 12
12 12
12 14
12 16
12 12
-100
-300
IHK24
IHK24+Ba
Net |Ba subtracted
C
100
20
25
F
80
20
2.5
15
n.s. n.s.
WT
R52H
(nA)
60
15
2
10
E246K
40
10
IBy (HA)
1.5
G247R
20
5
₮
5
1
O
&
1
0
0
0
0.5
.
++
**
WT
R52H
uninjected
WT
E246K
uninjected
WT
G247R
uninjected
0
n=
6
6
10
6
7
6
D
Uninjected
YFP-GIRK4WT
YFP-GIRK4R52H
YFP-GIRK4E246K
YFP-GIRK4G247R
1500-
400-
300
150-
300
1000-
I (nA)
200
200
100
200
100-
V (mV)
50
500-
V (mV)
V (mV)
V (mV)
100
V (mV)
-120-80
-40
40
-120 .— 80
-40
40
-200-
I (nA)
-120-60
-40
-100.
I (nA)
40
50
-120 -80
-40
40
-120-80 -40
40
-100-
1
-100
I (nA)
-500-
-400
-200
-150-
-200-
HK24
HK24+Ba
Net I
Ba subtracted
significantly lower than Po of GIRK1/4WT, while Po of GIRK1/4E246K was not significantly different (Fig. 5B). Neither single-channel current (isingle) (Fig. 5C) nor single-channel conductance (g) (Fig. 5D) were affected by the mutations.
Intracellular Na+ activates GIRK channels (Sui et al. 1996; Ho & Murrell-Lagnado 1999), and this can contribute to basal (agonist-independent) activity of GIRK channels. To determine whether any of the mutations in our study altered channel activation by Na+, we measured channel activity in excised patches (Fig. 5E). After excision, basal activity was recorded in Na+-free bath solution for 3 min. Then, addition of 40 mM NaCl into the bath solution increased the activity of GIRK1/4WT and mutated channels (Fig. 5E). The extent of activation by Na+ was calculated by normalizing NP. measured during the first 2 min after NaCl addition, to NP. measured during the last minute before NaCl addition. Figure 5F
shows that the extent of activation by Na+ was similar for GIRK1/4wT and mutants.
Another way to assess mutation-induced changes in Po is from the whole-cell data of Figs 3 and 4. The whole-cell current, I, is given by the equation I = isingle × Po x N (Hille 2002), where N is the total number of channels in the PM. N is proportional to the fluorescence intensity of the labelled channels (measured in arbitrary units per unit area, EAU). Thus, whole-cell current normalized to PM level, I/EAU, is proportional to I/N, therefore I/EAU ~ isingle x Po. Since isingle is not affected by mutations, differences in I/EAU reflect differences in Po. We have calculated I/EAU for the averaged I and EAU in each experiment (because currents and channel expression were not measured in the same oocytes). All homo- tetrameric YFP-GIRK4 mutants exhibited a significantly reduced I/EAU (Fig. 6A), suggesting a great reduction in Po (impaired gating) in addition to the impaired expression.
A
D
WT
R52H
WT
E246K
WT
G247R
25
. WT
R52H
no GBY
20-+ E246K
G247R .
with GBY
Eapp, %
15
IRK1
10
5
B
2500
0
Expression (AU)
WT
2000
R52H
E246K
1
0
1
2
3
4
1500
G247R
donor/acceptor molar ratio
1000
n.s.
n.s.
E
15
500
A
A
WT
n.s.
**
#
T
R52H
A
18
3
4
I
0
T
E246K
no GBY
GBY
no
GBY
GØY
no
GBy GR
Eapp, %
10
G247R
40
n= 14 13 13 12
12 15 12 12
12 12 12 12
**
35
C
5
30
WT
22 ****
12 ***
R52H
9
17
8
*
E246K
5
0
4
5
T
20
G247R
·
(A)
0-0.5
0.5-1.5
1.5-2.5
**
·
10
-5
donor/acceptor molar ratio
E
**
F
0
no
GØY
GBY
1
no
GØY
GØY
no GBY GBY
n=
7
6
3
6
7
10 8
10
10 10 10 10
A, representative confocal images of GIRK1/YFP-GIRK4 expressing oocytes, with or without GBy. A separate experiment was performed for each mutant with GIRK1/YFP-GIRK4WT as control. B, GIRK1/YFP-GIRK4 expression in PM; mutated channels express less well than GIRK1/YFP-GIRK4WT. A separate experiment was performed for each mutant. C, basal and GBy-evoked currents of GIRK1/YFP-GIRK4WT and mutated channels. D, raw data from FRET experiments showing Eapp as a function of donor/acceptor molar ratio. Each symbol corresponds to Eapp from one oocyte. Data shown are from two to three experiments for each donor-acceptor FRET pair. F, binned analysis of Eapp shows that at all donor/acceptor molar ratios, the interaction of GB/CFP-Gy with GIRK1/YFP-GIRK4R52H and GIRK1/YFP-GIRK4E246K Was impaired, whereas the interaction of GB/CFP-Gy with GIRK1/YFP-GIRK4G247R Was more similar to GIRK1/YFP-GIRK4WT. Numbers of cells in bins are shown above the bars. n.s., not significant; ** P < 0.01; *** P < 0.001; **** P < 0.0001. [Colour figure can be viewed at wileyonlinelibrary.com]
For heterotetrameric GIRK1/YFP-GIRK4 mutants, the normalization procedure showed an impaired P. for GBy-activated GIRK1/YFP-GIRK4R52H but not for GIRK1/YFP-GIRK4E246K or GIRK1/YFP-GIRK4G247R, supporting the single-channel results (Fig. 6B).
The GIRK4 opener VU0529331 activates GIRK4 G247R homotetramers
Next, we aimed to test whether the activity of mutated channels can be rescued by the recently discovered GIRK2 and GIRK4 opener VU0529331, which activates
A
0
E
before Na+
after Na+
WT
2 pAL
10 s
n=2
C
WT
0
0
C
2 pAL
c
2 pA
50 ms
R52H
R52H
0
100 ms
0
n=1
C
C
0
G247R E246K
0
C
C
0
E246K
C
0
n=1
0
C
C
B
0.20
C
D
p=0.0051
4
60
F
20
0.15
3
55
Foldactivation
15
0. 0.10
A
single
g (pS)
50
2
10
45
1
0.05
1
40
5
1
0.00
·
I
0
35
0
WT
R52H
E246K
WT
R52H
E246K
WT
R52H
E246K
WT
R52H
E246K
G247R
A
YFP-GIRK4
B
GIRK1/YFP-GIRK4
6
WT
150
n.s.
R52H
p < 0.001
p < 0.001
4
E246K
IBV/EAU
G247R
I/EAU
100
n.s.
2
p = 0.003
50
p < 0.001
p < 0.001
n.s.
p < 0.001
0
0
GBY
GBY
GBY
no GBy
GØY
no GBY
GBY
no GBY
GY
Average whole-cell currents were normalized to average PM expression of YFP-GIRK4-containing channels (I/EAU), to assess changes in Po. The data for each mutant are from separate experiments. The differences in calculated I/EAU values of the WT channel result from the use of different settings of confocal imaging acquisition parameters in different experiments. Data are shown as means ± SD; statistical significance between WT and each mutant’s data was obtained using a t test. A, IBy/EAU of YFP-GIRK4 homotetramers indicates an impaired gating of the mutated channels. Only GBy-activated currents have been measured. B, I/EAU (in the absence and presence of GBy) of GIRK 1/YFP-GIRK4 heterotetramers indicates an impaired gating of GIRK1/YFP-GIRK4R52H heterotetramers; on the other hand, E246K and G247R mutations appear to have the same open probability, suggesting unimpaired gating. [Colour figure can be viewed at wileyonlinelibrary.com]
the channels in a GBy-independent manner (Kozek et al. 2018). We used oocytes expressing GIRK4 homo- tetramers, with or without GBy. VU0529331 activated GIRK4WT and GIRK4G247R in a dose-dependent manner, and the activated channels showed the typical inwardly rectifying I-V relationships (Fig. 7). No response could be seen in oocytes injected with GIRK4R52H and GIRK4E246K RNA.
Expression of GIRK in HAC15 cells and regulation of aldosterone secretion
The type of K+ channels in the AZG cells varies between species. GIRK1 and GIRK4 are expressed in human AZG (Velarde-Miranda et al. 2013; Enyeart & Enyeart 2021), but GIRK4 is not present in the adrenal cortex of rats, and probably also in mice (Aragao-Santiago et al. 2017). Thus, rodent models are inefficient to study the effect of GIRK4 mutations in vivo (Aragao-Santiago et al. 2017; Enyeart & Enyeart 2021). Hence, to corroborate that the mutation-induced changes in GIRK4 function observed in Xenopus oocytes also take place in a more relevant cellular environment, we used the human adrenocortical carcinoma cell line (HAC15).
We tested how VU0529331 affects the secretion of aldosterone by HAC15 cells. To measure aldosterone secretion, VU0529331 (25 [M) was added to the plating medium for 3 h. VU0529331 significantly reduced aldosterone secretion compared with untreated cells (Fig. 8A). In addition, we tested the effect of VU0529331 in whole-cell recordings of HAC15 cells. Net GIRK currents were revealed by subtracting the residual currents remaining after the addition of 400 nM TPNQ. In native cells, three out of seven cells showed significant VU0529331-activated inwardly rectifying GIRK currents, whereas no GIRK current activation was detected in the other four cells (Fig. 8B and C). In GIRK4WT transfected cells, basal current did not differ from the basal current of native cells (Fig. 8E); however, VU0529331 strongly activated GIRK currents (Fig. 8D and F).
We also verified the presence of endogenous GIRK1 and GIRK4 using western blot. Both GIRK1 and GIRK4 proteins were present in HAC15 cells (Fig. 9A). Whole-cell recordings of untransfected or transfected HAC15 cells with D2 dopamine receptor did not show any dopamine (DA)-evoked GIRK current (Fig. 9B, left). Next, we trans- fected HAC15 cells with GFP-GIRK1 and GIRK4 (WT or mutants), with GBy subunits and the D2 receptor. Ievoked was elicited by application of 100 µM DA. The GBy-evoked currents of GIRK1/4R52H and GIRK1/4E246K mutants were significantly reduced compared with GIRK1/4WT, while GIRK1/4G247R did not significantly differ from GIRK1/4WT (Fig. 9B), corroborating the results from the Xenopus oocyte expression model.
A
no GBY
B
with GBY
1.5
20
· GIRK4WT
· GIRK4WT
GIRK4G247R
Ibasal (HA)
1-
15
a GIRK4R52H
& GIRK4E246K
10
v GIRK4G247R
0.5
2
5
0
0
0
VU0529331 (μM)
1
5
20
60
0
1
VU0529331 (μM)
5
20
60
C
150
no GBY
D
100
with GBy
100
80
I/ control
50
control
60
40
0
20
-50
0
0
m
VU0529331 log(µM)
1
10
100
VU0529331 log(µM)
1
10
100
no GBY
with GBY
E
V (mV)
200
F
5
-120 80
-40
40
V (mV)
-400
-120 -80
40
40
WT
(nA)
-5
-800
-10
-15
-1200
-20
-1600
-25-
V (mV)
200
V (mV)
0.5
-120 -80 4
40
40
-120 -80
40
-0.5
40
G247R
-200
VU (μM)
-400
(nA)
0
-1
-600
1
-1.5
-2
-800
5
-2.5
-1000
-20
-60
-3-
RNAs of GIRK4 mutants and WT were injected into oocytes with or without GBy. Cells were exposed to four concentrations of VU0529331 (1, 5, 20 and 60 µM). Only the GIRK4G247R mutant responded to VU0529331. A, Ibasal of GIRK4WT (n = 7) and GIRK4G247R (n = 6). GIRK4WT and GIRK4G247R react similarly to VU0529331. In A and B, the central horizontal line shows the mean and the whiskers show SD. B, IBy of GIRK4WT (n = 8), GIRK4R52H (n = 4), GIRK4E246K (n = 3), and GIRK4G247R (n = 7). GIRK4WT had larger current at 0 pM VU0529931, while GIRK4G247R did not show any response to concentrations lower than 20 µM VU0529331. C, normalized /basal (///control) of /basal. GIRK4WT and GIRK4G247R Were activated similarly by VU0529331. D, I/Icontrol of IBy shows that GIRK4G247R Was activated by VU0529331 stronger than GIRK4WT, in relative terms. E, I-V relationship of GIRK4WT (top) and GIRK4G247R (bottom) without GBy. F, I-V relationship of GIRK4WT (top) and GIRK4G247R (bottom) with GBy. [Colour figure can be viewed at wileyonlinelibrary.com]
Additionally, we wanted to assess how GIRK4 mutations affect aldosterone secretion. We have trans- fected HAC15 cells with GFP, GFP-GIRK1 and either GIRK4WT, GIRK4R52H, GIRK4E246K or GIRK4G247R-GFP. After 72-96 h we washed the cells, exposed them to fresh medium and collected medium for aldosterone level measurement after 3 h. In parallel, we collected cells from all wells and measured GAPDH (internal control) and GIRK4 levels using western blot. The number of cells in all plates was similar, as verified by similar GAPDH levels in all groups (Fig. 9C). Cells expressing GIRK1/4WT showed significantly higher levels of GIRK4 than control (GFP-transfected) cells. GIRK4 levels in cells transfected with GIRK1/4E246K showed the same level of GIRK4 as the control, suggesting negligible expression of the mutant protein (Fig. 9D and E). GIRK1/4R52H and GIRK1/4G247R showed a tendency of lower expression compared with GIRK1/4wr but it did not reach statistical significance. Next, we found that transfection of HAC15 cells with GIRK1/4wT reduced the secreted aldosterone by ~35% compared with control. Aldosterone secretion was significantly higher in GIRK1/4R52H compared with GIRK1/4WT. GIRK1/4E246K and GIRK1/4G247R had no effect on aldosterone secretion (Fig. 9F). To note, in the case of GIRK1/4E246K this could reflect the negligible expression of the channel.
Discussion
Loss of selectivity of GIRK4 (KCNJ5) channels is believed to be the leading cause of PA, by producing constitutive depolarization of AZG cells (Choi et al. 2011; Monticone et al. 2013; Al-Salameh et al. 2014; Funder 2019). This concept was extended from mutations in the GIRK4 pore region (Choi et al. 2011; Monticone et al. 2013) to additional mutations in the transmembrane domain of GIRK4 (Murthy et al. 2012; Kuppusamy et al. 2014) and, eventually, to mutations in the cytosolic N- and C-termini (Al-Salameh et al. 2014; Murthy et al. 2014), despite the identification of a cytosolic domain mutation, E246G, which impaired membrane abundance of GIRK4 rather than selectivity (Cheng et al. 2015). We studied PA-linked GIRK4 mutations of three conserved amino acids located in the cytosolic domain (Fig. 1): R52 and E246, which were reported to cause loss of selectivity of GIRK4 channels (Murthy et al. 2014), and G247. Here, we show that R52H, E246K and G247R mutations do not impart selectivity loss; they cause a complete or partial loss of function and/or channel abundance in the PM, but do not alter channel selectivity or inward rectification. We also show that, following GIRK1/4 transfection into HAC15 cells, only GIRK1/4wT reduces aldosterone secretion, whereas all mutants have lost this ability.
A
B
1000-
0.0362
Native
GIRK4WT transfected
V (mV)
20-
Aldosterone, pg/ml
V (mV)
800
-125 -100 -75 -50 -25 25 450 -125 -100 -75 -50 -25
25
50
-20
600
-100
-40
I (pA)
400
-60
-200
(pA)
200
-80
-300
I
0
-100
Basal
-400
no
VU
VU
-120
VU0529331
-500
C
D
E
F
Native
WI
Basal
VU0529331
Current density (pA/pF)
8
Current density (pA/pF)
25
0.0345
Current density (pA/pF)
6.
Current density (pA/pF)
25
0.0157
.
6
20
20
15
4
15
4
10
10
2
2
5
5
C
0
0
0
0
basal VU
basal
VU
Native WT
Native WT
Many of the KCNJ5 mutant studies (including our work) were done in Xenopus laevis oocytes used as a heterologous expression system (Calloe et al. 2007; Murthy et al. 2012, 2014; Hardege et al. 2015). We first thoroughly investigated possible sources of artefact in this system. Like other cells, Xenopus oocytes possess a number of endogenous ion channels that yield small currents, usually in the range of a few tens of nA, but they may give rise to artefactual interpretations if the expressed exogenous ion channels show poor expression or function. For instance, activation of M2 receptors by ACh often elicited outwardly rectifying endogenous Ca2+-dependent Cl- currents (Dascal 1987) that were abolished by the Ca2+ chelator BAPTA. Even the remaining leak currents may be mistaken for currents of GIRK mutants if the expressed currents are small. To avoid these artefacts, we activated GIRK channels by coexpressing GBy subunits
at doses that maximally activate the WT GIRKs; yet, even under these conditions the homotetrameric GIRK4R52H and GIRK4E246K did not produce detectable currents.
In this study we focused on cytosolic domain mutations in GIRK4 - R52H, E246K and G247R. Three lines of evidence strongly link these mutations to PA. First, these mutations are accompanied by PA and high aldosterone levels in humans (Murthy et al. 2014). Second, over- expression of R52H and E246K in the human adreno- cortical NCI-H295R cell line (the precursor of the HAC15 cells, Wang & Rainey 2012) were reported to cause excessive aldosterone secretion (Murthy et al. 2014). Third, a similar cytosolic mutation, E246G, is also PA-linked; it was reported to cause loss of function but not loss of K+ selectivity (Cheng et al. 2015), similarly to what we find for E246K. In all, there is little doubt that these cytosolic mutations are causally linked to PA. However,
A
oocyte
× oocyte
HAC15
, oocyte
× oocyte
, oocyte
, oocyte
HAC15
B
LK
HK48
HK48+100 µM DA
Currentd ensity (pA/pF)
400
DA
GIRK1
GIRK4
βγ
GIRK4
YFP-GIRK4
300
+
+
75
60
75
← YFP-GIRK4
60
200
45
GIRK1
35
45
GAPDH
← GIRK4
35
GAPDH
25
25
1 nA
100
**
WT
E246K
15
13
5 s
G247R
0
15
R52H
WT
R52H
E246K
10
G247R
WB: GIRK1
WB: GIRK4
Untransfected
D2R only
C
D
20
E
1.5
F
3
E246K
R52H
G247R
GIRK4 net signal (x103)
.
WT
aldosterone secretion
15
normalized GIRK4 expression
*
**
GIRK4
GFP-GIRK1
+
Y
1.0
Normalized
2
**
+
x
+
*
**
75
10
*
**
**
60
← GIRK4G247R-GFP
0.5
1
45
← GIRK4
5
35
← GAPDH
0
GFP
WT
R52H
E246K
G247R
0.0
GFP
R52H
E246K
G247R
0
GFP
WT
R52H
E246K
G247R
we were not convinced by the proposed biophysical mechanism (Murthy et al. 2014), whereby mutations remote from the pore region have been suggested to impair selectivity and rectification.
Most heterologous expression studies have been done with GIRK1/4 heterotetramers (Charmandari et al. 2012; Monticone et al. 2013; Kuppusamy et al. 2014; Cheng et al. 2015; Hardege et al. 2015). However, in AZG a low abundance of GIRK1 RNA (Choi et al. 2011) and protein (Enyeart & Enyeart 2021) has been reported. Thus, despite the presence of GIRK1 subunit in HAC15 (Fig. 9A), GIRK4 homotetramers may be the predominant GIRK channel form in aldosterone-secreting cells of the adrenal cortex (Enyeart & Enyeart 2021). Among the cytosolic domain mutations, only GIRK4G247R has been tested in a homotetrameric context. However, it was tested on the background of an additional mutation, S143T (Calloe et al. 2007). The homotetrameric GIRK4S143T is a pore mutant that expresses better than GIRK4wT and preserves inward rectification (Vivaudou et al. 1997). The GIRK4S143T homotetramers showed large Ibasal (Calloe et al. 2007), which is different from GIRK4wr homotetramers that have small Ibasal and large IBy, as we have demonstrated both in oocytes (Fig. 3) and HAC15 cells (Figs 8 and 9). To address this conundrum, we examined both homo- tetrameric GIRK4 and heterotetrameric GIRK1/4, WT and mutants.
Our central finding was that, in the GIRK1/4 heterotetrameric context, all mutations strongly reduced the whole-cell basal GIRK current in Xenopus oocytes, and R52H and E246K (but not G247R) mutations greatly reduced the GBy-induced currents, both in Xenopus oocytes and in HAC15 cells (Figs 1 and 9). The homo- tetrameric mutant channels gave practically no detectable currents (Fig. 3). These results indicate that all three mutations cause loss of function, which is more severe in homotetrameric than in heterotetrameric channels.
In contrast, we found no changes in K+ selectivity or inward rectification (in the heterotetrameric context; Fig. 2). The transmembrane pore of K+ channels is a conserved structure that enables K+ ions to permeate rapidly through the channel, and mutations in the pore region shift the permeability ratio towards Na+ (Doyle et al. 1998; Lu 2004). The cytosolic domain of inwardly rectifying K+ channels forms an additional, cytoplasmic pore that is a continuation of the transmembrane pore but does not contain a selectivity filter (Nishida & Mackinnon 2002; Pegan et al. 2005; Inanobe et al. 2007). Several amino acids of the cytosolic domain lining the pore control channel’s inward rectification by altering Mg2+ and polyamine block, but not the ion selectivity (Kubo & Murata 2001; Lu 2004; Pegan et al. 2005). We measured the reversal potential of GBy-activated GIRK1/4 current in four external K+ concentrations and found a linear dependence between log[K+] and Vrev with a slope close
to that predicted by the Nernst equation for a pure K+ pore, and identical to that of the WT channel (Fig. 2). Analysis using the Goldman-Hodgkin-Katz equation confirmed that the mutated channels conducted K+ but not Na+. Evaluation of the degree of inward rectification showed that inward rectification of mutated channels was similar to WT (Fig. 2) (Rubinstein et al. 2007). Our results clearly demonstrate that all mutants studied here are selective to K+ over Na+, and rectify, to a similar extent.
To understand the mechanisms underlying loss of function of the mutants, we employed several approaches. Confocal imaging in Xenopus oocytes revealed that PM abundance of mutated homo- or heterotetrameric channels was significantly reduced compared with GIRK4WT (Figs 3 and 4). FRET experiments suggested an impaired interaction of the R52H and E246K mutant channels with the main gating factor, GBy (Fig. 4D and E). Analysis of macroscopic whole-cell currents normalized to PM expression indicated impaired gating (in addition to reduced PM abundance) of R52H, and impaired PM abundance of E246K and G247R heterotetramers (Fig. 6). Single-channel recordings (Fig. 5) confirmed the reduction in open channel probability (Po), which indicates impaired gating by GBy, of the heterotetrameric GIRK1/4R52H. On the other hand, while the heterotetrameric GIRK1/4E246K had an impaired interaction with GBy (FRET data), we did not see a significant reduction in Po. Hence, we suggest that R52H mutation impairs both gating by GBy, and PM abundance of homo- and heterotetrameric channels. E246K mutation impairs mainly channel expression, and to a lesser extent channel activity (in the heterotetrameric context), mainly because of a reduced GBy affinity. G247R mutation had the smallest impact on the heterotetrameric channel, and showed a mild impairment in activity (Ibasal, but not IBy) and in PM abundance. None of the mutations impaired the direct activation of GIRK1/4 by intracellular Na+ (Fig. 5), which contributes to the basal activity of GIRK channels (Sui et al. 1996); however, in AZG cells, a decrease in GIRK’s basal activity may be expected because of the lower abundance of the mutant channels in the PM.
In our study, none of the mutated homotetramers produced detectable whole-cell currents in oocytes, with or without coexpression of GBy. Homotetrameric GIRK4G247R showed PM expression similar to GIRK4WT, but it was not activated by GBy. Remarkably, however, it was activated by the GIRK4 opener, VU0529331 (Fig. 7). On the other hand, GIRK4R52H and GIRK4E246K showed very low expression in the PM compared with GIRK4WT, and this could be a major factor underlying the absence of current. These two mutants could not be activated by VU0529331 or GBy. However, since we see some expression, it is possible that other openers will be able to activate these channels.
We next characterized native HAC15 cells. About half of the cells contained endogenous inwardly rectifying channels activatable by VU0529331 (Fig. 8B and C). We hypothesize that these were mostly homotetrameric GIRK4, in view of their high sensitivity to VU0529331 (GIRK1/4 show much lower activation by this opener (Kozek et al. 2018)). The basal GIRK current in native HAC15 cells was small and similar to that in GIRK4-transfected cells (Fig. 8C-E), resembling the low Ibasal of GIRK4 in oocytes (Fig. 3). These observations also support the notion that native cells contain mainly homo- tetrameric GIRK4, although the presence of GIRK1/4 heterotetramers cannot be ruled out. Furthermore, we show that activation of endogenous GIRK channels in HAC15 cells dramatically regulates aldosterone secretion: VU0529331 significantly attenuated aldosterone secretion (Fig. 8A), even though only a subpopulation of cells contained the VU0529331-activated K+ channels.
The expression profile and electrophysiology of GIRK1/4 heterotetramers expressed in HAC15 cells supported the results obtained in Xenopus oocytes. All GIRK4 mutants showed a tendency of reduced whole-cell expression in HAC15 cells; GIRK1/4E246K did not express at all and showed very small GBy-activated currents (Fig. 9C-E). GIRK1/4R52H Was expressed, but also showed negligible currents, most probably owing to the impaired gating of GIRK1/4R52H, as revealed in oocyte experiments. The GIRK1/4G247R was similar to WT in almost all parameters that we measured. However, we note a slight tendency of reduced protein expression and whole-cell current, though neither were statistically significant. A greater functional deficiency would be expected for this and other mutants in native AZG cells, if homotetramers are the major GIRK4 form, as discussed above.
The changes in aldosterone secretion caused by the coexpression of GIRK1/4 are generally in line with the functional data. Only GIRK1/4WT significantly reduced aldosterone secretion, whereas GIRK1/4R52H, which showed the greatest functional impairment, also showed the greatest difference from GIRK1/4WT (Fig. 7F). We note that, because the transfection efficiency was only 30-50%, our measurements of the effects of GIRK expression on aldosterone secretion probably underestimate the actual differences between the WT and mutants. In all, the inability of the mutants to reduce basal aldosterone secretion on HAC15 cells (in contrast to GIRK1/4WT) supports the hypothesis that loss of function of GIRK4 may lead to deregulation of aldosterone secretion.
Notably, unlike HAC15 cells, native human AZG cells possess sizeable GIRK basal currents probably carried by homotetrameric GIRK4, which are inhibited by angiotensin II (Enyeart & Enyeart 2021). We therefore hypothesize that in human AZG cells there is a yet unidentified endogenous activator, not present in cultured HAC15 cells, that augments the basal activity
of this channel (which is intrinsically low). The effect of the loss-of-function mutations will be much more pronounced, in terms of regulation of aldosterone synthesis, than in HAC15 cells.
Conclusions and prospects
A well-established molecular mechanism of KCNJ5 (GIRK4)-linked hyperaldosteronism is gain-of-function and loss of selectivity of GIRK4 channels, caused by mutations in their pore region. Our results suggest that R52H and E246K mutations, located in the cytosolic domains, impair channel activity by a completely different mechanism. We demonstrate that N- and C-terminal domain mutations studied here reduce channel activity, PM abundance and lose the ability to regulate aldosterone secretion. Knowing the biophysical mechanism that impairs the channel function is crucial for setting the course of treatment. While patients with pore mutations, which yield gain-of-function non-selective Na+/K+ channels, may benefit from treatment with channel blockers, patients with cytosolic domain mutations may potentially benefit from treatment with GIRK4 channel openers to achieve the same effect: prevention of constitutive depolarization of zona glomerulosa secretory cells.
References
Al-Salameh A, Cohen R & Desailloud R (2014). Overview of the genetic determinants of primary aldosteronism. Appl Clin Genet 7, 67-79.
Aragao-Santiago L, Gomez-Sanchez CE, Mulatero P, Spyroglou A, Reincke M & Williams TA (2017). Mouse models of primary aldosteronism: from physiology to pathophysiology. Endocrinology 158, 4129-4138.
Berlin S, Tsemakhovich VA, Castel R, Ivanina T, Dessauer CW, Keren-Raifman T & Dascal N (2011). Two distinct aspects of coupling between Gai protein and G protein-activated K+ channel (GIRK) revealed by fluorescently labelled G@13 protein subunits. J Biol Chem 286, 33223-33235.
Bykova EA, Zhang XD, Chen TY & Zheng J (2006). Large movement in the C terminus of CLC-0 chloride channel during slow gating. Nat Struct Mol Biol 13, 1115-1119.
Calloe K, Ravn LS, Schmitt N, Sui JL, Duno M, Haunso S, Grunnet M, Svendsen JH & Olesen SP (2007). Characterizations of a loss-of-function mutation in the Kir3.4 channel subunit. Biochem Biophys Res Commun 364, 889-895.
Charmandari E, Sertedaki A, Kino T, Merakou C, Hoffman DA, Hatch MM, Hurt DE, Lin L, Xekouki P, Stratakis CA & Chrousos GP (2012). A novel point mutation in the KCNJ5 gene causing primary hyperaldosteronism and early-onset autosomal dominant hypertension. J Clin Endocrinol Metab 97, E1532-E1539.
Cheng CJ, Sung CC, Wu ST, Lin YC, Sytwu HK, Huang CL & Lin SH (2015). Novel KCNJ5 mutations in sporadic aldosterone-producing adenoma reduce Kir3.4 membrane abundance. J Clin Endocrinol Metab 100, E155-E163. Choi M, Scholl UI, Yue P, Bjorklund P, Zhao B,
Nelson-Williams C, Ji W, Cho Y, Patel A, Men CJ, Lolis E, Wisgerhof MV, Geller DS, Mane S, Hellman P, Westin G, Akerstrom G, Wang W, Carling T & Lifton RP (2011). K+ channel mutations in adrenal aldosterone-producing adenomas and hereditary hypertension. Science 331, 768-772.
Dascal N (1987). The use of Xenopus oocytes for the study of ion channels. CRC Crit Rev Biochem 22, 317-387.
Dascal N (2001). Ion-channel regulation by G proteins. Trends Endocrinol Metab 12, 391-398.
Dascal N & Lotan I (1992). Expression of exogenous ion channels and neurotransmitter receptors in RNA-injected Xenopus oocytes. vol. 13. In Protocols in Molecular Neuro- biology, ed. Longstaff A & Revest P, pp. 205-225. Humana Press, Totowa, NJ.
Doyle DA, Morais Cabral J, Pfuetzner RA, Kuo A, Gulbis JM, Cohen SL, Chait BT & Mackinnon R (1998). The structure of the potassium channel: molecular basis of K+ conduction and selectivity. Science 280, 69-77.
Enyeart JJ & Enyeart JA (2021). Human adrenal glomerulosa cells express K2P and GIRK potassium channels that are inhibited by AngII and ACTH. Am J Physiol Cell Physiol 321, C158-C175.
Fernandes-Rosa FL, Boulkroun S & Zennaro MC (2017). Somatic and inherited mutations in primary aldosteronism. J Mol Endocrinol 59, R47-R63.
Funder JW (2019). Primary aldosteronism. Hypertension 74, 458-466.
Hardege I, Xu S, Gordon RD, Thompson AJ, Figg N, Stowasser M, Murrell-Lagnado R & O’Shaughnessy KM (2015). Novel insertion mutation in KCNJ5 channel produces constitutive aldosterone release from H295R cells. Mol Endocrinol 29, 1522-1530.
Hattangady NG, Karashima S, Yuan L, Ponce-Balbuena D, Jalife J, Gomez-Sanchez CE, Auchus RJ, Rainey WE & Else T (2016). Mutated KCNJ5 activates the acute and chronic regulatory steps in aldosterone production. J Mol Endocrinol 57,1-11.
Hedin KE, Lim NF & Clapham DE (1996). Cloning of a Xenopus laevis inwardly rectifying K+ channel subunit that permits GIRK1 expression of IKACh currents in oocytes. Neuron 16, 423-429.
Hille B (2002). Ion Channels of Excitable Membranes. Sinauer, Sunderland.
Ho IH & Murrell-Lagnado RD (1999). Molecular determinants for sodium-dependent activation of G protein-gated K+ channels. J Biol Chem 274, 8639-8648. Huang CL, Feng SY & Hilgemann DW (1998). Direct activation of inward rectifier potassium channels by PIP2 and its stabilization by GBy. Nature 391, 803-806.
Inanobe A, Matsuura T, Nakagawa A & Kurachi Y (2007). Structural diversity in the cytoplasmic region of G protein-gated inward rectifier K+ channels. Channels (Austin) 1, 40-46.
Kanevsky N & Dascal N (2006). Regulation of maximal open probability is a separable function of Cavß subunit in L-type Ca2+ channel, dependent on NH2 terminus of @ 1C (Cay1.2x). J Gen Physiol 128, 15-36.
Keren Raifman T, Kumar P, Haase H, Klussmann E, Dascal N & Weiss S (2017). Protein kinase C enhances plasma membrane expression of cardiac L-type calcium channel, Cay1.2. Channels (Austin) 11, 604-615.
Kozek KA, Du Y, Sharma S, Prael FJ, Spitznagel BD, Kharade SV, Denton JS, Hopkins CR & Weaver CD (2018).
Discovery and characterization of VU0529331, a synthetic small-molecule activator of homomeric G protein-gated, inwardly-rectifying, potassium (GIRK) channels. ACS Chem Neurosci 10, 358-370.
Krapivinsky G, Gordon EA, Wickman K, Velimirovic B, Krapivinsky L & Clapham DE (1995). The G-protein-gated atrial K+ channel IKACh is a heteromultimer of two inwardly rectifying K+-channel proteins. Nature 374, 135-141.
Kubo Y & Murata Y (2001). Control of rectification and permeation by two distinct sites after the second trans- membrane region in Kir2.1 K+ channel. J Physiol 531, 645-660.
Kuppusamy M, Caroccia B, Stindl J, Bandulik S, Lenzini L, Gioco F, Fishman V, Zanotti G, Gomez-Sanchez C, Bader M, Warth R & Rossi GP (2014). A novel KCNJ5-insT149 somatic mutation close to, but outside, the selectivity filter causes resistant hypertension by loss of selectivity for potassium. J Clin Endocrinol Metab 99, E1765-E1773. Kurachi Y (1995). G protein regulation of cardiac muscarinic potassium channel. Am J Physiol 269, C821-C830.
Logothetis DE, Kurachi Y, Galper J, Neer EJ & Clapham DE (1987). The By subunits of GTP-binding proteins activate the muscarinic K+ channel in heart. Nature 325, 321-326.
Lu Z (2004). Mechanism of rectification in inward-rectifier K+ channels. Annu Rev Physiol 66, 103-129.
Monticone S, Hattangady NG, Penton D, Isales CM, Edwards MA, Williams TA, Sterner C, Warth R, Mulatero P & Rainey WE (2013). A novel Y152C KCNJ5 mutation responsible for familial hyperaldosteronism type III. J Clin Endocrinol Metab 98, E1861-E1865.
Monticone S, Tetti M, Burrello J, Buffolo F, De Giovanni R, Veglio F, Williams TA & Mulatero P (2017). Familial hyper- aldosteronism type III. J Hum Hypertens 31, 776-781. Murthy M, Azizan EA, Brown MJ & O’Shaughnessy KM (2012). Characterization of a novel somatic KCNJ5 mutation delI157 in an aldosterone-producing adenoma. J Hypertens 30, 1827-1833.
Murthy M, Xu S, Massimo G, Wolley M, Gordon RD, Stowasser M & O’Shaughnessy KM (2014). Role for germline mutations and a rare coding single nucleotide polymorphism within the KCNJ5 potassium channel in a large cohort of sporadic cases of primary aldosteronism. Hypertension 63, 783-789.
Nishida M & Mackinnon R (2002). Structural basis of inward rectification. Cytoplasmic pore of the G protein-gated inward rectifier GIRK1 at 1.8 A resolution. Cell 111, 957-965.
Parmar J, Key RE & Rainey WE (2008). Development of an adrenocorticotropin-responsive human adreno- cortical carcinoma cell line. J Clin Endocrinol Metab 93, 4542-4546.
Pegan S, Arrabit C, Zhou W, Kwiatkowski W, Collins A, Slesinger PA & Choe S (2005). Cytoplasmic domain structures of Kir2.1 and Kir3.1 show sites for modulating gating and rectification. Nat Neurosci 8, 279-287.
Reuveny E, Slesinger PA, Inglese J, Morales JM, Iniguez-Lluhi JA, Lefkowitz RJ, Bourne HR, Jan YN & Jan LY (1994). Activation of the cloned muscarinic potassium channel by G protein By subunits. Nature 370, 143-146.
Rubinstein M, Peleg S, Berlin S, Brass D & Dascal N (2007). Ga13 primes the G protein-activated K+ channels for activation by coexpressed GBy in intact Xenopus oocytes. J Physiol 581, 17-32.
Rubinstein M, Peleg S, Berlin S, Brass D, Keren-Raifman T, Dessauer CW, Ivanina T & Dascal N (2009). Divergent regulation of GIRK1 and GIRK2 subunits of the neuronal G protein gated K+ channel by Ga; GDP and GBy. J Physiol 587, 3473-3491.
Sakmann B & Neher E (1995). Single-Channel Recording. Plenum Press, New York.
Spat A (2004). Glomerulosa cell-a unique sensor of extracellular K+ concentration. Mol Cell Endocrinol 217, 23-26.
Stuart A & Ord JK (1998). Kendall’s Advanced Theory of Statistics, 6th edition, vol. 1, pp. 351. Arnold, London.
Sui JL, Chan KW & Logothetis DE (1996). Na+ activation of the muscarinic K+ channel by a G-protein-independent mechanism. J Gen Physiol 108, 381-391.
Sui JL, Petit-Jacques J & Logothetis DE (1998). Activation of the atrial KACh channel by the By subunits of G proteins or intracellular Na+ ions depends on the presence of
phosphatidylinositol phosphates. Proc Natl Acad Sci U S A 95, 1307-1312.
Velarde-Miranda C, Gomez-Sanchez EP & Gomez-Sanchez CE (2013). Regulation of aldosterone biosynthesis by the Kir3.4 (KCNJ5) potassium channel. Clin Exp Pharmacol Physiol 40, 895-901.
Vivaudou M, Chan KW, Sui JL, Jan LY, Reuveny E & Logothetis DE (1997). Probing the G-protein regulation of GIRK1 and GIRK4, the two subunits of the KACh channel, using functional homomeric mutants. J Biol Chem 272, 31553-31560.
Wang T & Rainey WE (2012). Human adrenocortical carcinoma cell lines. Mol Cell Endocrinol 351, 58-65.
Whorton MR & Mackinnon R (2013). X-ray structure of the mammalian GIRK2-By G-protein complex. Nature 498, 190-197.
Wickman K, Nemec J, Gendler SJ & Clapham DE (1998). Abnormal heart rate regulation in GIRK4 knockout mice. Neuron 20, 103-114.
Williams TA, Monticone S & Mulatero P (2015). KCNJ5 mutations are the most frequent genetic alteration in primary aldosteronism. Hypertension 65, 507-509.
Yakubovich D, Berlin S, Kahanovitch U, Rubinstein M, Farhy-Tselnicker I, Styr B, Keren-Raifman T, Dessauer CW & Dascal N (2015). A quantitative model of the GIRK1/2 channel reveals that its basal and evoked activities are controlled by unequal stoichiometry of Ga and GBy. PLOS Comput Biol 11, e1004598.
Yakubovich D, Rishal I, Dessauer CW & Dascal N (2009). Amplitude histogram-based method of analysis of patch clamp recordings that involve extreme changes in channel activity levels. J Mol Neurosci 37, 201-211.
Zheng J, Varnum MD & Zagotta WN (2003). Disruption of an intersubunit interaction underlies Ca2+-calmodulin modulation of cyclic nucleotide-gated channels. J Neurosci 23, 8167-8175.
Additional information
Data availability statement
All our data, including the relevant statistics, are presented in the Figures. We report no new code, sequences or similar data that could be deposited separately. All data are also summarized in the Statistical Summary Document.
Competing interests
The authors report no conflicting interests.
Author contributions
Conceptualization, B.S., N.D. and A.K.B; Methodology, B.S., R.H .- J. and N.D .; Investigation and analysis, B.S., R.H .- J., H.P.R. and N.T .; Writing - original draft, B.S., H.P.R. and N.D .; Writing - review & editing, B.S., R.H .- J. and N.D .; Funding acquisition and supervision, N.D. and A.K.B.
Funding
This work was supported by the joint Israel-India grant ISF # 2255/15 and the Israel Science Foundation grant # 1282/18.
Acknowledgements
We are grateful to Alomone Labs for the generous gift of VU0529331 and to Mariam Ashkar for assistance in some of the experiments.
Keywords
electrophysiology, K+ channel opener, Kir3, primary aldosteronism, selectivity
Supporting information
Additional supporting information can be found online in the Supporting Information section at the end of the HTML view of the article. Supporting information files available: