ELSEVIER
Molecular and Cellular Endocrinology
journal homepage: www.elsevier.com/locate/mce
Molde har and Cellular Endocrinokgy
NCI-H295R cell line as in vitro model of hyperaldosteronism lacks functional KCNJ5 (GIRK4; Kir3.4) channels
CrossMark
Marie-Cécile Kienitz a,*, Evanthia Mergia b, Lutz Pott ª
a Institute of Physiology, Ruhr-University Bochum, D-44780 Bochum, Germany
b Department of Pharmacology and Toxicology, Ruhr-University Bochum, D-44780 Bochum, Germany
ARTICLE INFO
Article history: Received 10 February 2015 Received in revised form 22 April 2015 Accepted 11 May 2015 Available online 18 May 2015
Keywords: GIRK4 KCNJ5 Primary aldosteronism Adrenocortical cell line Electrophysiology
ABSTRACT
As a major cause of aldosterone producing adenomas, numerous gain-of-function mutations in the KCNJ5 gene (encoding the K+ channel subunit GIRK4) have been identified. The human adrenocortical carcino- ma cell line NCI-H295R is the most frequently used cellular model for in vitro studies related to regulation of aldosterone-synthesis. Because of the undefined role of KCNJ5 (GIRK4) in regulating synthesis of al- dosterone, we aimed at identifying basal and G protein-activated GIRK4 currents in this paradigmatic cell line. The GIRK-specific blocker Tertiapin-Q did not affect basal current. Neither loading of the cells with GTP-Y-S via the patch-clamp pipette nor agonist stimulation of an infected A1-adenosine receptor resulted in activation of GIRK current. In cells co-infected with KCNJ5, robust activation of basal and adenosine-activated inward-rectifying current was observed. Although GIRK4 protein can be detected in Western blots of H295R homogenates, we suggest that GIRK4 in aldosterone-producing cells does not form functional GBy-activated channels.
@ 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction
Aldosterone, via the mineralocorticoid receptor (NR3C2), regu- lates plasma volume by modulating absorption of Na+ in the colon and kidney. Aldosterone is produced by glomerulosa cells of the adrenal cortex. Its synthesis is regulated via the expression level of aldosterone synthase, encoded by CYP11B2 (Tauber et al., 2014). Ex- pression of this key enzyme is regulated in a Ca2+-dependent manner. Ca2+ entry in glomerulosa cells is promoted by opening of voltage- activated Ca2+ channels (L-Type, T-type) upon membrane depolarization (see Hattangady et al., 2012 for review).
Primary aldosteronism, a major cause of secondary (non- essential) hypertension, is characterized by increased constitutive secretion of aldosterone. It is either caused by adrenal hyperpla- sia, which is treated using aldosterone receptor antagonists, or by aldosterone-producing adenoma (APA) which requires surgical treat- ment (Funder et al., 2008).
Recently, somatic and germline mutations in the KCNJ5 gene have been identified as a major common cause of APA (Choi et al., 2011; Moraitis et al., 2013; Mulatero et al., 2013; Scholl et al., 2012). KCNJ5 encodes for an inward-rectifying K+ channel subunit (Kir3.4; GIRK4). In the heart, where this type of channel has first been identified,
and where its physiological role is well defined (Sakmann et al., 1983; Yamada, 2002), the functional channel is a tetrameric complex of GIRK4 (KCNJ5) and GIRK1 (KCNJ3) in a 2:2 stoichiometry (Krapivinsky et al., 1995). GIRK channel complexes are activated (opened) by direct, membrane-delimited interaction with By subunits of heterotrimeric G proteins of the pertussis toxin-sensitive class (Gio) upon agonist activation of appropriate GPCRs (Hibino et al., 2010; Lüscher and Slesinger, 2010). Apart from agonist-dependent opening activity, basal GIRK currents have been described in native and ex- pression systems. In their native environment, homotetrameric GIRK4 channels can be activated, independent of Gøy subunits, by high con- centrations of intracellular Na+ ions (Mintert et al., 2007; Rishal et al., 2003). Opening of GIRK channels results in membrane hyperpo- larization and/or reduced excitability. In addition, basal (receptor- independent) activity of GIRK channels contributes to maintaining resting potential in a variety of cells (Ito et al., 1994; Wiser et al., 2006). Thus far, the majority of identified mutations in the KCNJ5 gene related to APA affect the selectivity filter, a highly conserved feature common to all K+ channel species (Doyle et al., 1998) and result in a loss in ion channel selectivity of the homotetrameric GIRK4 complex (Choi et al., 2011; Murthy et al., 2012). The increase in Na+- conductance is supposed to result in depolarized resting membrane potential and increased Ca2+-entry via voltage-gated Ca2+ channels and reduced Ca2+-extrusion via the Na+/Ca2+-exchanger (Maturana et al., 1999; Tauber et al., 2014).
In previous electrophysiological studies on rodent glomerulosa cells, two-pore K+ channels (TASK-1/TASK-3) had been identified as dominating K+ background conductance pathway and important
* Corresponding author. Institute of Physiology, Ruhr-University Bochum, Universitätsstrasse 150, D-44780 Bochum, Germany. Tel .: +49 234 3229200; fax: +49 234 3214449.
E-mail address: cecile.kienitz@rub.de (M .- C. Kienitz).
target of regulation of aldosterone synthesis (Czirják and Enyedi, 2002; Czirják et al., 2000; Davies et al., 2008; Hu et al., 2009; Penton et al., 2012). Furthermore, expression of TASK-3 and TREK-1 was confirmed in the H295R cell line (Brenner and O’Shaughnessy, 2008). Inhibition of these channels was accompanied by membrane de- polarization and increased aldosterone release, suggesting that two- pore K+ channels are important regulators of the resting membrane potential.
Though the discovery of APA-causing GIRK4 mutations sug- gests an important role of this channel in regulating membrane potential in glomerulosa cells, thus far no information on its con- tribution to maintaining resting potential in primary glomerulosa cells is available. GIRK4 channel subunits assemble into heteromeric channels together with GIRK1 subunits in atrial myocytes (Corey et al., 1998) or GIRK2 in the cerebellum (Aguado et al., 2008). However, a large fraction of GIRK4 subunits is processed to the cell membrane to form homomultimers (Kennedy et al., 1999). Upon heterologous expression in Xenopus oocytes, CHO or HEK293 cells, GIRK4 channel subunits are able to form functional channels (Bender et al., 2001; Treiber et al., 2013) which are blocked by the GIRK- specific inhibitor T-Q with higher sensitivity as compared to GIRK1/ GIRK4 heteromeric channels (Mintert et al., 2007).
In H295R cells a membrane depolarization upon exposure to the GIRK channel inhibitory peptide T-Q has been described recently (Cheng et al., 2015). Identification of a corresponding conduc- tance pathway, however, is missing. The NCI H295R cell line, originally derived from a human adrenocortical carcinoma, was modified to increase cell growth rate and attachment to the culture dishes and was renamed H295R to differentiate it from the origi- nal NCI-H295 cells (Rainey et al., 2004). Due to its responsiveness to angiotensin II, elevated external K+ and ACTH, the NCI-H295R cell line is the most frequently used cellular model for in vitro studies related to synthesis of aldosterone and other adrenal steroids (Bird et al., 1993; Lichtenauer et al., 2013). In previous studies related to KCNJ5 mutations, endogenous expression of Kir3.4 protein had been demonstrated in these cells by Western blot analysis (Murthy et al., 2012). Because of the emerging but physiologically undefined role of KCNJ5 (GIRK4) in regulating synthesis of aldosterone, we aimed at identifying basal (agonist-independent) and G protein-activated macroscopic current carried by GIRK4 in this paradigmatic cell line.
2. Methods
2.1. Cell culture
H295R cells (NCI-H295R, CRL-2128; American Type Culture Col- lection, Manassas, VA) were maintained in DMEM/F12, containing 5% fetal calf serum (FCS), 400 IU/ml penicillin and 0.4 mg/ml strep- tomycin (Invitrogen, Carlsbad, CA). Cells were seeded in 30 mm culture dishes at about 105 cells/ml. Electrophysiological measure- ments were performed 3-5 days after seeding. For confocal microscopy, cells were grown on ibidi u-dishes (Ø 35 mm, ibidi GmbH, Martinsried, Germany). Adult rat atrial myocytes were iso- lated and cultured as described previously (Bünemann et al., 1996).
2.2. Molecular biology and adenovirus constructs
CDNAs of mouse KCNJ3 (GIRK1, GenBank: D45022.1), KCNJ5 (Kir3.4, GIRK4) kindly provided by Dr. Y. Kurachi, Osaka, Japan, and rat A1R (ADORA1) obtained from Dr. A. Karschin, Göttingen,Germany, were cloned into the pAd-Track-CMV vector using standard meth- odology, to generate pAd-Kir3.4 and pAd-A1R.
The pAd-Easy1 plasmid encoding for the adenovirus type 5, pAd- Track and pAd-Track-CMV were kindly provided by Dr B. Vogelstein (Johns Hopkins University, Baltimore, MD, USA). Production and pu- rification of the recombinant virus were performed as described in
detail previously (Luo et al., 2007). In brief, the cDNAs of rat AR (ADORA1) and Kir3.4 (KCNJ5) were subcloned into the pAd-Track- CMV shuttle vector to generate pAd-A1R and pAd-Kir3.4, respectively. Recombinant adenovirus was generated by homologous recombi- nation between the shuttle vectors and pAd-Easy-1 in Escherichia coli to produce Ad-A1R and Ad-Kir3.4 (Ad-GIRK4). Recombinant viruses were propagated in HEK 293 cells and recovered after several freezing-thawing cycles. For infection of H295R cells, virus titers were adjusted to yield about 50% GFP-positive cells after 3 days. Patch clamp experiments were performed at days 2-4 post-infection. As a rule, time matched GFP-positive cells infected with the empty GFP- encoding virus (pAd-Track-CMV) served as controls.
2.3. Current measurements
Membrane currents were recorded using whole cell voltage clamp. The DC resistance of the filled pipettes ranged from 4 to 6 MQ. Current measurements were performed using a patch clamp amplifier (List LM/EPC 7). Signals were analog-filtered (1 kHz), digitally sampled, and stored on a computer equipped with a hardware/software package (ISO2, MFK) for voltage control, data acquisition, and anal- ysis. Experiments were performed at ambient temperature (22-24 ℃). Cells were voltage-clamped at a holding potential of -90 mV, i.e. negative to EK, resulting in inward K+ currents. As a rule, every 10 s, voltage ramps (duration 500 ms) from -120 to-60 mV were applied to assess stability of the recording conditions and to generate current- voltage relations (membrane currents in response to depolarizing voltage ramps are shown as rapid deflections). Rapid exposure to agonist-containing solutions was performed by means of a custom- made solenoid-operated flow system.
2.4. Solutions and chemicals
For whole cell measurements of membrane currents, an extra- cellular solution of the following composition was used (mM): NaCl 120; KCl 20; CaCl2 0.5; MgCl2 1.0; Hepes/NaOH 10.0; pH 7.4. The pipette solution contained (mM): K-aspartate 100; KCl 40; NaCl 5.0; MgCl2 2.0; Na2ATP 5.0; EGTA 2.0; GTP 0.025; Hepes/KOH 20.0; pH 7.4. The K+ reversal potential under these conditions was calcu- lated as -48 mV. Standard chemicals were from Merck (Darmstadt, Germany). EGTA, Hepes, Na2ATP, GTP, GTP-Y-S and adenosine were from Sigma Aldrich (München, Germany). Tertiapin-Q (T-Q) was from Tocris (Ellisville MI,USA).
2.5. Preparation of homogenates
To obtain homogenates of H295R cells, isolated rat atrial myocytes, mouse whole brain, and mouse adrenal gland, cells or tissues were homogenized in buffer (in mM: triethanolamine (TEA)/ HCI 50, NaCl 50, EDTA 1, DTT 2, benzamidine 0.2, PMSF 0.5 and 1 µM pepstatin A, pH 7.4, 4 ℃) using a glass/glass homogenizer (900 rpm). After centrifugation (800 x g, 5 min, 4 ℃), supernatants were sub- jected to Western blot analysis. The protein concentration was determined in triplicate and repeated (Bradford, Bio-Rad).
2.6. Western blots
For Western blot analysis, protein components (homogenates 20 µg/lane) were separated by 12% SDS-PAGE and transferred to ni- trocellulose (Protran BA-85, Schleicher & Schuell). The membranes were blocked with Roti-Block (Roth) and incubated at 4 ℃ over- night with the rabbit polyclonal C-terminal GIRK4 antibody (dilution 1: 1000; SAB4501633, Sigma-Aldrich) or the rabbit polyclonal anti- GIRK1 (dilution 1:1000; APC-005, alomone) or the rabbit polyclonal anti-GIRK2 (dilution 1:1000, APC-006, alomone) in 3% ovalbumin in Tris-buffered saline-Tween. After washing, the membranes were
incubated with a peroxidase labeled anti-rabbit-IgG antibody (1:10,000; Sigma-Aldrich) in 3% ovalbumin for 1 h at ambient tem- perature. Chemiluminiscence detection was performed with SuperSignal® West Dura chemiluminiscence substrate (Thermo Sci- entific) and a 16-bit cooled CCD camera system (GDS 8000, UVP). All lanes were loaded with equal amounts of solubilized protein.
2.7. Immunocytochemistry and confocal microscopy
H295R cells and isolated atrial myocytes cells were washed with PBS and fixed with 4% paraformaldehyde (PFA) (4 ℃, 20 min). Cells were permeabilized (20 min) with 0.2% Triton and 5% goat serum in PBS (pH7.4), incubated (90 min) with the primary antibody (rabbit polyclonal C-terminal GIRK4 antibody (SAB4501633, Sigma- Aldrich, diluted 1:500) in PBS containing 0.2% Triton and 5% goat serum (to minimise non-specific staining), washed three times and incubated with the secondary antibody (Alexa 488 goat, anti- rabbit (Life Technologies, Darmstadt, Germany), diluted 1:200) for 60 min. Cells were washed three times with PBS and afterwards DAPI-stained (diluted 1: 1000 in PBS) for 20 min. As a control (data not shown), time-matched sister cultures were incubated with the secondary antibody only.
Confocal images were aquired using the LSM 510 Meta system (Carl Zeiss MicroImaging GmbH,), equipped with argon and HeNe lasers and a 63x (NA 1.4) oil objective. Data were analyzed with the LSM 510 META software as described previously (Zoidl et al., 2007).
3. Results
3.1. GIRK4 subunits fail to form functional channels in H295R cells
Intrinsic expression of GIRK4 protein in H295R cells has been demonstrated previously by Western blot analysis (Cheng et al., 2015; Murthy et al., 2012). Using an C-terminal GIRK4 antibody, this was confimed in the present study as shown in Fig. 1C. Overexpression of GIRK4 in atrial myocytes (Mintert et al., 2007) or expression of GIRK4 in CHO cells (Bender et al., 2001) results in formation of func- tional homomeric GIRK4 channels. These channels have distinctive properties as compared to endogenous GIRK1/4 channels. GIRK4 homomeric channels can be distinguished from endogenous GIRK channels by differences in inward rectification, activation by intra- cellular [Na+] and sensitivity to the selective GIRK-inhibitor tertiapin-Q (T-Q) (Mintert et al., 2007). Since endogenous cardiac or neuronal GIRK channels are heteromers comprised of GIRK4 and GIRK1 or GIRK2, we investigated the expression of these GIRK sub- units in H295R cells and mice adrenal gland homogenates. As shown in Fig. 1D, we were able to detect GIRK1 in both adrenal gland homogenates and H295R cells. GIRK2 subunits are expressed in neu- ronal cells (homogenates from mouse brain served as control), but could not be detected in H295R cells and mice adrenal glands.
The obvious interpretation of how APA-associated KCNJ5 mu- tations cause a Ca2+-dependent increase in CYP11B expression is based on the assumption that the resting potential is constitutively reduced
A
B
Į[pA]
TP-Q
+1000
+600
-100
-50
+200
+50
-200
Vm [mV]
-600
a
b
8
c
1000
a
b
C
500 pA
20 s
C
D
1.
2.
3.
M
1. 2. 3.
M 4. 5. 6.
50 kDa
97 kDa
40 kDa
66 kDa
*
*
35 kDa
45 kDa
*
25 kDa
1. rat atria
1. adrenal gland
M molecular weight markers
2. H295R cells
2. whole heart
4. adrenal gland
3. H295R cells + Ad-GIRK4
3. H295R cells
5. whole brain
M molecular weight markers
6. H295R cells
in glomerulosa cells of affected individuals, resulting in activation of voltage-activated Ca2+-channels (Mulatero et al., 2013). This implies that the GIRK conductance pathway contributes to basal K+- conductance, as e.g. in atrial myocytes or certain neurons (Ito et al., 1994; Wiser et al., 2006). To identify a putative contribution of homomeric GIRK4 or heteromeric GIRK1/4 channels to constitu- tive (background) conductance of H295R cells, we used the inhibitory peptide T-Q. It has been shown previously that homotetrameric rat GIRK4 channels are inhibited by this compound with an EC50 of 0.6 nM as compared to heteromeric GIRK1/4 channels (EC50: 12 nM; Mintert et al., 2007). As illustrated by the representative current recording in Fig. 1A, superfusion of a cell held at -90 mV in whole-cell voltage clamp with T-Q at a highly saturating concentration (200 nM) had no effect on holding current. The typical current-voltage relations (I/V-curves) generated by voltage ramps from -120 to +60 mV were linear with a reversal potential around -10 mV and appear undistinguishable in the absence and in the presence of the block- ing peptide (Fig. 1B). This behavior was found in each of 26 cells tested by means of this experimental protocol. Thus, in H295R cells cultured under standard conditions, a contribution of GIRK chan- nels to basal K+-conductance can be excluded.
3.2. Dialyzing native H295R cells with GTP-Y-S failed to activate inward-rectifying GIRK currents
Apart from basal opening activity of GIRK channels, which appears to be independent of canonical activation by direct binding of GBy (Kienitz et al., 2014), GIRK channels can be canonically opened by exposure of the inner face of the plasma membrane to the stable GTP analogue GTP-Y-S, e.g. by supplementing the pipette filling so- lution with this compound in a whole-cell voltage-clamp setting (Mintert et al., 2007). Figure 2 shows a representative current trace recorded from a native cell using a pipette filling solution contain- ing 500 µM GTP-Y-S. After rupture of the membrane patch, the holding current was slowly shifted in the inward direction, con- comitant with an increase of the voltage-ramp evoked current peaks. Exposure of the cell to T-Q (200 nM) did not affect this current com- ponent. The I/V-curve of this current component, obtained by subtraction, as described in detail in the legend, showed an outward- rectifying behavior and a reversal potential between -20 and -30 mV (Fig. 2B, n = 13). This is reminiscent of a Ca2+-activated chloride current previously identified in adrenal Zona fasciculata cells upon stimulation with angiotensin II (Chorvatova et al., 1998). Since its voltage-dependence safely excluded a contamination by GIRK chan- nels, its identity was not further analyzed. Its gradual development serves as an indicator of intracellular GTP-Y-S loading by the pipette solution. The fact that we did not observe GIRK current activation under these experimental conditions supports our hypothesis that GIRK channels do not contribute to basal membrane conductance.
3.3. Tertiapin-Q sensitive background currents in GIRK4-infected cells
In previous studies, we have confirmed that key properties of the pathway under study were not different in native and GFP- expressing cells (Beckmann et al., 2008; Kienitz et al., 2014). As shown in the representative current recording in Fig. 3A, baseline currents in mock-infected H295R cells were insensitive to Tertiapin- Q (n =6), excluding a non-specific effect of viral infection or GFP- expression on basal K+ conductance. Although both GIRK1 and GIRK4 subunits are expressed in H295R cells, formation of functional heteromeric GIRK1/GIRK4 in H295R cells is unlikely to occur since no T-Q sensitive GIRK currents were recorded in native H295R cells (see Fig. 1). GIRK1 subunits alone are not able to form functional channels, but are processed to the cell membrane by interaction with
GTP-Y-S
A
TP-Q
a
100 pA
b
20 s
B
I[pA]
1200-
1000
800
600
400
-100
-50
200
Vm [mV]
-200
+50
-400
b - a
-600
GIRK4 subunits that confer translocation of heteromeric GIRK1/ GIRK4 channels to the cell surface (Kennedy et al., 1999). To investigate whether increasing the expression level of endog- enous GIRK1 subunits results in formation of functional heteromeric GIRK1/GIRK4 channels, we infected H295R cells with an GIRK1- encoding vector. As shown in Fig. 3B, application of T-Q had no effect on the holding current, indicating that overexpression of GIRK1 sub- units did not result in expression of heteromeric GIRK1/GIRK4 channels at the cell surface. However, as expected, cotransfection of H295R cells with GIRK1- and GIRK4-encoding vectors resulted in basal, T-Q-sensitive GIRK conductance (Fig. 3C) which is charac- terized by a strong inward rectifying I/V curve (Fig. 3D).
We next tested if infection of H295R cultures with a GIRK4- encoding vector would result in a T-Q-sensitive component of basal current. The trace recorded from a representative infection- positive cell presented in Fig. 3E reveals a T-Q-sensitive component of baseline current at -90 mV. The I/V curve of the T-Q-sensitive current, obtained by subtraction is characterized by strong inward rectification (Fig. 3F) and a reversal potential (-38 ± 1.8 mV; n = 11) only slightly positive close to the calculated Nernst potential in the
A
mock-infected
B
Ad-GIRK1
TP-Q
TP-Q
500 pA
200 pA
20 s
20 s
C
Ad-GIRK1 + Ad-GIRK4
D
TP-Q
I[pA]
+200
-100
-50
Vm [mV]
+50
-200
a -b
b
-400
500 pA
-600
a
20 s
-800
E
Ad-GIRK4
F
TP-Q
l[pA]
+100
+50
-100
-50
Vm [mV]
50
+50
a -b
-100
-150
200
400 pA
250
a
b
-300
20 s
present conditions (see Methods). Thus, vector-induced expres- sion or overexpression, respectively, of GIRK4 - either alone or together with GIRK1 - results in a constitutive (agonist-independent) T-Q-sensitive inward-rectifying K+-selective current pathway.
3.4. Raising the intracellular Na+-concentration activates GIRK currents in GIRK4-infected cells, but not in native H295R cells
Raising the cytosolic Na+-concentration to levels ≥ 15 mM results in opening of homotetrameric GIRK4 (GIRK2) channels by binding of Na+ to a structurally defined binding site at D223 (GIRK2 D228) (Mintert et al., 2007; Sui et al., 1998; Whorton and Mackinnon, 2011).
Na+-dependent GIRK channel current is independent of the canon- ical GBy-associated pathway (Kienitz et al., 2014). In order to identify functional expression of homomeric GIRK4 channels in H295R cells, pipette filling solutions were supplemented with 60 mM Na+, which results in an increase in cytosolic Na+ paralleled by activation of homomeric GIRK4 channels (Mintert et al., 2007). As shown in Fig. 4A in a representative native H295R cell, transition from cell-attached to whole-cell mode caused an inward step in the current trace fol- lowed by a decrease of current within 40 s to a small steady state level of 70 pA. The nature of the instantaneous current, which de- creased upon dialysis of the cell with the pipette filling solution, was not further analyzed. In a representative infection-positive cell,
A
B
Ad-A,R Ad-GIRK4
Ad-A,R
60mM Na
A
B
Ad-GIRK4
TP-Q
TP-Q
Ado
Ado
500 pA
20 s
1 nA
b
500 pA
20 s
200 pA
a
20 s
C
C
I[pA]
20 s
I[pA]
200
+200
100
-100
-50
-100
-50
Vm [mV]
Vm [mV]
+50
+50
-200
-100
a -b
-200
a -b
-400
-300
-600
400
-800
inward holding current upon rupture of the membrane patch in- creased in a time-dependent fashion (Fig. 4B). Both the inward- rectifying I/V-curve and sensitivity to T-Q clearly identify GIRK4 channels as charge-carrying mechanism underlying this Na+- dependent current. In this series of experiments in none of 5 native cells tested, an inward rectifying current component upon Na+- loading for up to 10 minutes was identified, whereas each of 8/8 GIRK-4 infection-positive cells exhibited a Na+-dependent inward- rectifying component of current of 9.2 + 2.5 pA/pF on average. This behavior is analogous to previous experiments using this experi- mental protocol on GIRK4 channels in atrial myocytes and infected CHO-cells (Bender et al., 2001; Mintert et al., 2007).
3.5. Adenosine-induced stimulation of A1R failed to activate GIRK currents in native H295R cells
It is conceivable that, for unknown reasons, endogenous GIRK channels are resistant to raising intracellular Na+, e.g. by associa- tion with endogenous GIRK1 subunits. As shown previously, the prototypic atrial GIRK4/GIRK1 channel lacks any sensitivity to changes in intracellular Na+ between nominally zero and 60 mM (Mintert et al., 2007).
We therefore investigated if endogenous GIRK channels in H295R cells can be canonically opened by activation of an expressed re- ceptor coupled to pertussis-toxin sensitive G-proteins (Gi/o). We used the A1 adenosine receptor (A1R), which either endogenously, such as in atrial myocytes and hippocampal neurons, or upon heterolo- gous expression, efficiently couples to GIRK channels via intrinsic
Fig. 5. Agonist-stimulation of A1R failed to activate endogenous GIRK currents in H295R cells. A. Representative recording of membrane current from an Ad-A1R- infected H295R cell showing no current responses to the application of adenosine (10 uM). B. Analogous current trace from a H295R cell coexpressing A1R and GIRK4. C. I/V curve of the T-Q-sensitive current.
Gijo (Kienitz et al., 2011; Kurachi et al., 1986; Wellner-Kienitz et al., 2000; Wetherington and Lambert, 2002).
In H295R cells, infection-positive for the A1R, exposure to ad- enosine (Ado, 10 uM) had no measurable effect on membrane holding current and the I/V curve between -120 and +60 mV (Fig. 5A). In contrast, cells double-infected with Ad-A1R and Ad-GIRK4 responded to application of Ado with robust activation of an inward-rectifying T-Q-sensitive current (Fig. 5B). The requirement for co-expression of GIRK4 with the Gifo-coupled GPCR clearly provides additional ev- idence that the cell line under study does not express functional GIRK channels susceptible to standard maneuvers of activation.
As illustrated in Fig. 1, GIRK 4 protein is detected in immune blots of H293R lysates in line with a study by Murthy et al. (2012). Since this is in conflict with the negative results of the current measure- ments, we studied the subcellular distribution of GIRK4 by means of confocal immunocytochemistry. As shown in Fig. 6, immunore- activity in native H295R cells is weak and appears to be diffuse with some preference to a perinuclear compartment. No association with the plasma membrane could be detected. In contrast, GIRK4- infected cells showed a distinct localization of immunoreactivity to the plasma membrane. This was analogous to localization of this protein in native adult atrial myocytes, the paradigmatic cell type with robust functional endogenous expression of GIRK channel com- plexes (Fig. 6C).
4. Discussion
Membrane potential is an important factor contributing to reg- ulation of secretion of aldosterone by adrenal glomerulosa cells
A
H 10 μm
B
110 pm1
C
1
10 um
(Lotshaw, 1997). As predominant background K+conductance path- ways channels of the two-pore (K2P) family (TASK-3, TREK-1) had been identified in previous studies (Czirják and Enyedi, 2002; Enyeart et al., 2002, 2010) (see Gomez-Sanchez and Oki, 2014 for review).
Expression of these channels has been confirmed in the H295R model cell line (Brenner and O’Shaughnessy, 2008), and various modes of regulation by cellular signals have been functionally ana- lyzed (Brenner and O’Shaughnessy, 2008; Enyeart et al., 2011).
Since the discovery of somatic and inherited KCNJ5 mutations in APAs and early onset aldosteronism (Choi et al., 2011; Murthy et al., 2014), GIRK4 channels are being considered as a regulatory target important for controlling synthesis of aldosterone via Ca2+- dependent transcription of CYP11B2. Paradoxically, to date demonstration of functional GIRK channels in terms of a distinct membrane current activated via a Gi/o-coupled GPCR or intracellu- lar Na+ is lacking, both in glomerulosa cells and standard models such as the H295R cell line. The latter has been used as a cellular vehicle to compare biophysical properties of expressed wild-type and mutant subunits (Murthy et al., 2014; Tauber et al., 2014). In this regard, there is no particular advantage as compared to other standard cell lines as expression systems (Monticone et al., 2013). Evidence for endogenous expression of functional channels had not been vig- orously tested in these studies. Interestingly, in line with the present data, Western blots of H295R homogenates in a previous study were positive for GIRK4 (Murthy et al., 2012).
Previous studies have established inward rectifying channels com- posed of KCNJ (GIRK) subunits in various combinations as an online detector of outstanding dynamic sensitivity of activation of heterotrimeric G proteins of the PTx-sensitive Gilo class (Hatcher-Solis et al., 2014; Kahanovitch et al., 2014; Rubinstein et al., 2009). This applies to endogenous and expressed channels activated by either endogenous or expressed GPCRs. Consistently, in mammalian or amphibian-oocyte expression systems, availability of endogenous heterotrimeric G proteins is not limiting sensitivity of this assay (e.g. Aryal et al., 2009; Kahanovitch et al., 2014; Rubinstein et al., 2009). Expression levels of GPCRs or GPCR-associated proteins might affect kinetics of responses to a GPCR agonist (Wellner-Kienitz et al., 2000). For example, overexpression of Gai-coupled A1-receptors in cardiomyocytes increased agonist-induced GIRK current ampli- tude and accelerated activation kinetics (Wellner-Kienitz et al., 2000) as compared to cells with endogenous A1 receptor expression levels.
Receptor-independent (basal) activity of endogenous GIRK chan- nels is sensitive to Tertiapin-Q (Kienitz et al., 2011), a bee venom peptide, which blocks GIRK channels with high affinity, though not in a completely specific manner (Felix et al., 2006; Jin and Lu, 1999). As demonstrated in the present study, T-Q did not affect basal current, indicating that GIRK channels do not contribute to maintaining resting potential of H295R cells. Dialyzing the cells with GTP-Y-S failed to activate endogenous GIRK channels (Fig. 2) which are known to be canonically opened by tonic activation of Gio by this stable GTP-analog. This supports the notion that functional GIRK chan- nels are not expressed in H295R cells. In view of the fact that GIRK4 protein had been unambiguously detected in Western blots previ- ously and in the present study, this is a surprising finding. (Cheng et al., 2015; Murthy et al., 2012). Furthermore, we provide evi- dence for the expression of GIRK1 protein in H295R cells (see Fig. 1). GIRK1 subunits alone are not able to constitute functional chan- nels (Kennedy et al., 1999), but interact with GIRK4 to form heteromeric GIRK channels. To exclude that formation of heteromeric GIRK channels is impeded by minor GIRK1 expression as com- pared to GIRK4, we increased endogenous GIRK1 expression levels. However, adenoviral-induced overexpression of GIRK1 protein does not promote formation of heteromeric channels as baseline cur- rents remain insensitive to Tertiapin-Q (Fig. 3).
GIRK4 homotetramers were first purified from bovine heart atria (Corey and Clapham, 1998) and have been shown to be localized to the cell membrane of GIRK4-transfected COS-7 cells (Kennedy et al., 1999). Electrophysiological experiments confirmed the func- tionality of GIRK4 homotetrameric channels in GIRK4-transfected atrial myocytes and CHO cells (Bender et al., 2001) as well as in Xenopus oocytes (Treiber et al., 2013).
The contribution of native GIRK4 channels to the electrical ac- tivity of human adrenocortical cells and aldosterone synthesis is still under debate. In a previous study, shRNA-induced knockdown of
GIRK4 in H295R cells did not affect membrane voltage and intra- cellular calcium concentration (Oki et al., 2012). On the other hand it was shown that T-Q-application resulted in membrane depolar- ization and increased expression of aldosterone synthase (CYP11B2), suggesting a role of native GIRK4 channels in maintaining resting potential and regulating aldosterone synthesis (Cheng et al., 2015) in H295R cells. Up to date, current recordings of native GIRK4 chan- nels, which confirm the functionality of expressed GIRK4 subunits unequivocally, are not available. Homotetrameric GIRK4 channels are activated via G protein-coupled receptors or by raising the cy- tosolic Na+-concentration to levels ≥15 mM (Mintert et al., 2007). In the present study, neither increasing intracellular Na+ to 60 mM via the patch pipette (Fig. 4) nor agonist stimulation of a co- expressed Gi/o-coupled A1-adenosine receptor (Fig. 5) activated endogenous GIRK4 channels. In contrast, these maneuvers in cells (co-)infected with KCNJ5 resulted in robust activation of T-Q- sensitive inward-rectifying currents.
Using confocal microscopy, GIRK4 immunoreactivity in native cells was detected exclusively in an intracellular compartment, whereas in GIRK4-infected cells additional immunoreactivity as- sociated with the plasma membrane was observed. Interestingly, a recent study showed that a combination of GIRK4 and Dab2 (disabled-2 protein) immunostaining is suitable to distinguish between aldosterone-producing adenomas (APA) and non- functioning adrenal incidentalomas (NFAI; Fernandes-Rosa et al., 2015). Positive GIRK4 and Dab2 membrane staining in that study was found in two thirds of APA, whereas the majority of NFAI was characterized by the absence of GIRK4 membrane staining but strong cytoplasmic GIRK4 accumulation.
We conclude that H295R cells express GIRK4 protein, which, however, is not exported and assembled to functional channels in the plasma membrane at endogenous expression levels. It is con- ceivable that a signal yet to be identified is required for translocation of GIRK4 channels to the plasma membrane. Alternatively, GIRK4 in these cells serves a hitherto unknown function, unrelated to forming ion channel complexes in the plasma membrane.
Acknowledgement
This work was supported by FoRum (F735-2011).
References
Aguado, C., Colón, J., Ciruela, F., Schlaudraff, F., Cabañero, M.J., Perry, C., et al., 2008. Cell type-specific subunit composition of G protein-gated potassium channels in the cerebellum. J. Neurochem. 105 (2), 497-511.
Aryal, P., Dvir, H., Choe, S., Slesinger, P.A., 2009. A discrete alcohol pocket involved in GIRK channel activation. Nat. Neurosci. 12 (8), 988-995.
Beckmann, C., Rinne, A., Littwitz, C., Mintert, E., Bösche, L.I., Kienitz, M .- C., et al., 2008. G protein-activated (GIRK) current in rat ventricular myocytes is masked by constitutive inward rectifier current (IK1). Cell. Physiol. Biochem. 21 (4), 259-268.
Bender, K., Wellner-Kienitz, M.C., Inanobe, A., Meyer, T., Kurachi, Y., Pott, L., 2001. Overexpression of monomeric and multimeric GIRK4 subunits in rat atrial myocytes removes fast desensitization and reduces inward rectification of muscarinic K+ current (IK(ACh)). Evidence for functional homomeric GIRK4 channels. J. Biol. Chem. 276 (31), 28873-28880.
Bird, I.M., Hanley, N.A., Word, R.A., Mathis, J.M., McCarthy, J.L., Mason, J.I., et al., 1993. Human NCI-H295 adrenocortical carcinoma cells: a model for angiotensin-II- responsive aldosterone secretion. Endocrinology 133 (4), 1555-1561.
Brenner, T., O’Shaughnessy, K.M., 2008. Both TASK-3 and TREK-1 two-pore loop K channels are expressed in H295R cells and modulate their membrane potential and aldosterone secretion. Am. J. Physiol. Endocrinol. Metab. 295 (6), E1480- E1486.
Bünemann, M., Brandts, B., Pott, L., 1996. Downregulation of muscarinic M2 receptors linked to K+ current in cultured guinea-pig atrial myocytes. J. Physiol. 494 (Pt 2), 351-362.
Cheng, C .- J., Sung, C .- C., Wu, S .- T., Lin, Y .- C., Sytwu, H .- K., Huang, C .- L., et al., 2015. Novel KCNJ5 mutations in sporadic aldosterone-producing adenoma reduce Kir3.4 membrane abundance. J. Clin. Endocrinol. Metab. 100 (1), E155-E163.
Choi, M., Scholl, U.I., Yue, P., Björklund, P., Zhao, B., Nelson-Williams, C., et al., 2011. K+ channel mutations in adrenal aldosterone-producing adenomas and hereditary hypertension. Science 331 (6018), 768-772.
Chorvatova, A., Guyot, A., Ojeda, C., Rougier, O., Bilbaut, A., 1998. Activation by angiotensin II of Ca2+-dependent K+ and Cl- currents in zona fasciculata cells of bovine adrenal gland. J. Membr. Biol. 162 (1), 39-50.
Corey, S., Clapham, D.E., 1998. Identification of native atrial G-protein-regulated inwardly rectifying K+ (GIRK4) channel homomultimers. J. Biol. Chem. 273 (42), 27499-27504.
Corey, S., Krapivinsky, G., Krapivinsky, L., Clapham, D.E., 1998. Number and stoichiometry of subunits in the native atrial G-protein-gated K+ channel, IKACh- J. Biol. Chem. 273 (9), 5271-5278.
Czirják, G., Enyedi, P., 2002. TASK-3 dominates the background potassium conductance in rat adrenal glomerulosa cells. Mol. Endocrinol. 16 (3), 621-629.
Czirják, G., Fischer, T., Spät, A., Lesage, F., Enyedi, P., 2000. TASK (TWIK-related acid-sensitive K+ channel) is expressed in glomerulosa cells of rat adrenal cortex and inhibited by angiotensin II. Mol. Endocrinol. 14 (6), 863-874.
Davies, L.A., Hu, C., Guagliardo, N.A., Sen, N., Chen, X., Talley, E.M., et al., 2008. TASK channel deletion in mice causes primary hyperaldosteronism. Proc. Natl. Acad. Sci. U.S.A. 105 (6), 2203-2208.
Doyle, D.A., Morais Cabral, J., Pfuetzner, R.A., Kuo, A., Gulbis, J.M., Cohen, S.L., et al., 1998. The structure of the potassium channel: molecular basis of K+ conduction and selectivity. Science 280 (5360), 69-77.
Enyeart, J.A., Liu, H., Enyeart, J.J., 2010. cAMP analogs and their metabolites enhance TREK-1 mRNA and K+ current expression in adrenocortical cells. Mol. Pharmacol. 77 (3), 469-482.
Enyeart, J.J., Xu, L., Danthi, S., Enyeart, J.A., 2002. An ACTH- and ATP-regulated background K+ channel in adrenocortical cells is TREK-1. J. Biol. Chem. 277 (51), 49186-49199.
Enyeart, J.J., Liu, H., Enyeart, J.A., 2011. Calcium-dependent inhibition of adrenal TREK-1 channels by angiotensin II and ionomycin. Am. J. Physiol. Cell Physiol. 301 (3), C619-C629.
Felix, J.P., Liu, J., Schmalhofer, W.A., Bailey, T., Bednarek, M.A., Kinkel, S., et al., 2006. Characterization of Kir1.1 channels with the use of a radiolabeled derivative of tertiapin. Biochemistry 45 (33), 10129-10139.
Fernandes-Rosa, F.L., Amar, L., Tissier, F., Bertherat, J., Meatchi, T., Zennaro, M .- C., et al., 2015. Functional histopathological markers of aldosterone producing adenoma and somatic KCNJ5 mutations. Mol. Cell. Endocrinol. 408, 220-226.
Funder, J.W., Carey, R.M., Fardella, C., Gomez-Sanchez, C.E., Mantero, F., Stowasser, M., et al., 2008. Case detection, diagnosis, and treatment of patients with primary aldosteronism: an endocrine society clinical practice guideline. J. Clin. Endocrinol. Metab. 93 (9), 3266-3281.
Gomez-Sanchez, C.E., Oki, K., 2014. Minireview: potassium channels and aldosterone dysregulation: is primary aldosteronism a potassium channelopathy? Endocrinology 155 (1), 47-55.
Hatcher-Solis, C., Fribourg, M., Spyridaki, K., Younkin, J., Ellaithy, A., Xiang, G., et al., 2014. G protein-coupled receptor signaling to Kir channels in Xenopus oocytes. Curr. Pharm. Biotechnol. 15 (10), 987-995.
Hattangady, N.G., Olala, L.O., Bollag, W.B., Rainey, W.E., 2012. Acute and chronic regulation of aldosterone production. Mol. Cell. Endocrinol. 350 (2), 151-162.
Hibino, H., Inanobe, A., Furutani, K., Murakami, S., Findlay, I., Kurachi, Y., 2010. Inwardly rectifying potassium channels: their structure, function, and physiological roles. Physiol. Rev. 90 (1), 291-366.
Hu, T., Fu, Q., Chen, P., Zhang, K., Guo, D., 2009. Generation of a stable mammalian cell line for simultaneous expression of multiple genes by using 2A peptide-based lentiviral vector. Biotechnol. Lett. 31 (3), 353-359.
Ito, H., Ono, K., Noma, A., 1994. Background conductance attributable to spontaneous opening of muscarinic K+ channels in rabbit sino-atrial node cells. J. Physiol. 476 (1), 55-68.
Jin, W., Lu, Z., 1999. Synthesis of a stable form of tertiapin: a high-affinity inhibitor for inward-rectifier K+ channels. Biochemistry 38 (43), 14286-14293.
Kahanovitch, U., Tsemakhovich, V., Berlin, S., Rubinstein, M., Styr, B., Castel, R., et al., 2014. Recruitment of GBy controls the basal activity of G-protein coupled inwardly rectifying potassium (GIRK) channels: crucial role of distal C terminus of GIRK1. J. Physiol. 592 (Pt 24), 5373-5390.
Kennedy, M.E., Nemec, J., Corey, S., Wickman, K., Clapham, D.E., 1999. GIRK4 confers appropriate processing and cell surface localization to G-protein-gated potassium channels. J. Biol. Chem. 274 (4), 2571-2582.
Kienitz, M .- C., Littwitz, C., Bender, K., Pott, L., 2011. Remodeling of inward rectifying K+ currents in rat atrial myocytes by overexpression of A1-adenosine receptors. Basic Res. Cardiol. 106 (6), 953-966.
Kienitz, M .- C., Mintert-Jancke, E., Hertel, F., Pott, L., 2014. Differential effects of genetically-encoded GBy scavengers on receptor-activated and basal Kir3.1/Kir3.4 channel current in rat atrial myocytes. Cell. Signal. 26 (6), 1182-1192.
Krapivinsky, G., Gordon, E.A., Wickman, K., Velimirović, B., Krapivinsky, L., Clapham, D.E., 1995. The G-protein-gated atrial K+ channel IKAch is a heteromultimer of two inwardly rectifying K+-channel proteins. Nature 374 (6518), 135-141.
Kurachi, Y., Nakajima, T., Sugimoto, T., 1986. On the mechanism of activation of muscarinic K+ channels by adenosine in isolated atrial cells: involvement of GTP-binding proteins. Pflügers Arch. 407 (3), 264-274.
Lichtenauer, U.D., Shapiro, I., Osswald, A., Meurer, S., Kulle, A., Reincke, M., et al., 2013. Characterization of NCI-H295R cells as an in vitro model of hyperaldosteronism. Horm. Metab. Res. 45 (2), 124-129.
Lotshaw, D.P., 1997. Effects of K+ channel blockers on K+ channels, membrane potential, and aldosterone secretion in rat adrenal zona glomerulosa cells. Endocrinology 138 (10), 4167-4175.
Luo, J., Deng, Z .- L., Luo, X., Tang, N., Song, W .- X., Chen, J., et al., 2007. A protocol for rapid generation of recombinant adenoviruses using the AdEasy system. Nat. Protoc. 2 (5), 1236-1247.
Lüscher, C., Slesinger, P.A., 2010. Emerging roles for G protein-gated inwardly rectifying potassium (GIRK) channels in health and disease. Nat. Rev. Neurosci. 11 (5), 301-315.
Maturana, A.D., Casal, A.J., Demaurex, N., Vallotton, M.B., Capponi, A.M., Rossier, M.F., 1999. Angiotensin II negatively modulates L-type calcium channels through a pertussis toxin-sensitive G protein in adrenal glomerulosa cells. J. Biol. Chem. 274 (28), 19943-19948.
Mintert, E., Bosche, L.I., Rinne, A., Timpert, M., Kienitz, M .- C., Pott, L., et al., 2007. Generation of a constitutive Na+-dependent inward-rectifier current in rat adult atrial myocytes by overexpression of Kir3.4. J. Physiol. 585 (Pt 1), 3-13.
Monticone, S., Hattangady, N.G., Penton, D., Isales, C.M., Edwards, M.A., Williams, T.A., et al., 2013. a Novel Y152C KCNJ5 mutation responsible for familial hyperaldosteronism type III. J. Clin. Endocrinol. Metab. 98 (11), E1861- E1865.
Moraitis, A.G., Rainey, W.E., Auchus, R.J., 2013. Gene mutations that promote adrenal aldosterone production, sodium retention, and hypertension. Appl. Clin. Genet. 7, 1-13.
Mulatero, P., Monticone, S., Rainey, W.E., Veglio, F., Williams, T.A., 2013. Role of KCNJ5 in familial and sporadic primary aldosteronism. Nat. Rev. Endocrinol. 9 (2), 104-112.
Murthy, M., Azizan, E.A., Brown, M.J., O’Shaughnessy, K.M., 2012. Characterization of a novel somatic KCNJ5 mutation delI157 in an aldosterone-producing adenoma. J. Hypertens. 30 (9), 1827-1833.
Murthy, M., Xu, S., Massimo, G., Wolley, M., Gordon, R.D., Stowasser, M., et al., 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 (4), 783-789.
Oki, K., Plonczynski, M.W., Lam, M.L., Gomez-Sanchez, E.P., Gomez-Sanchez, C.E., 2012. The potassium channel, Kir3.4 participates in angiotensin II-stimulated aldosterone production by a human adrenocortical cell line. Endocrinology 153 (9), 4328-4335.
Penton, D., Bandulik, S., Schweda, F., Haubs, S., Tauber, P., Reichold, M., et al., 2012. Task3 potassium channel gene invalidation causes low renin and salt-sensitive arterial hypertension. Endocrinology 153 (10), 4740-4748.
Rainey, W.E., Saner, K., Schimmer, B.P., 2004. Adrenocortical cell lines. Mol. Cell. Endocrinol. 228 (1-2), 23-38.
Rishal, I., Keren-Raifman, T., Yakubovich, D., Ivanina, T., Dessauer, C.W., Slepak, V.Z., et al., 2003. Na+ promotes the dissociation between GOGDP and GBy, activating G protein-gated K+ channels. J. Biol. Chem. 278 (6), 3840-3845.
Rubinstein, M., Peleg, S., Berlin, S., Brass, D., Keren-Raifman, T., Dessauer, C.W., et al., 2009. Divergent regulation of GIRK1 and GIRK2 subunits of the neuronal G protein gated K+ channel by GoiGDP and GBy. J. Physiol. 587 (Pt 14), 3473-3491.
Sakmann, B., Noma, A., Trautwein, W., 1983. Acetylcholine activation of single muscarinic K+ channels in isolated pacemaker cells of the mammalian heart. Nature 303 (5914), 250-253.
Scholl, U.I., Nelson-Williams, C., Yue, P., Grekin, R., Wyatt, R.J., Dillon, M.J., et al., 2012. Hypertension with or without adrenal hyperplasia due to different inherited mutations in the potassium channel KCNJ5. Proc. Natl. Acad. Sci. U.S.A. 109 (7), 2533-2538.
Sui, J.L., Petit-Jacques, J., Logothetis, D.E., 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 (3), 1307- 1312.
Tauber, P., Penton, D., Stindl, J., Humberg, E., Tegtmeier, I., Sterner, C., et al., 2014. Pharmacology and pathophysiology of mutated KCNJ5 found in adrenal aldosterone-producing adenomas. Endocrinology 155 (4), 1353-1362.
Treiber, F., Rosker, C., Keren-Raifman, T., Steinecker, B., Gorischek, A., Dascal, N., et al., 2013. Molecular basis of the facilitation of the heterooligomeric GIRK1/GIRK4 complex by cAMP dependent protein kinase. Biochim. Biophys. Acta 1828 (4), 1214-1221.
Wellner-Kienitz, M.C., Bender, K., Meyer, T., Bünemann, M., Pott, L., 2000. Overexpressed A1 adenosine receptors reduce activation of acetylcholine-sensitive K+ current by native muscarinic M2 receptors in rat atrial myocytes. Circ. Res. 86 (6), 643-648.
Wetherington, J.P., Lambert, N.A., 2002. GABAB receptor activation desensitizes postsynaptic GABAB and A1 adenosine responses in rat hippocampal neurones. J. Physiol. 544 (Pt 2), 459-467.
Whorton, M.R., Mackinnon, R., 2011. Crystal structure of the mammalian GIRK2 K+ channel and gating regulation by G proteins, PIP2, and sodium. Cell 147 (1), 199-208.
Wiser, O., Qian, X., Ehlers, M., Ja, W.W., Roberts, R.W., Reuveny, E., et al., 2006. Modulation of basal and receptor-induced GIRK potassium channel activity and neuronal excitability by the mammalian PINS homolog LGN. Neuron 50 (4), 561-573.
Yamada, M., 2002. The role of muscarinic K+ channels in the negative chronotropic effect of a muscarinic agonist. J. Pharmacol. Exp. Ther. 300 (2), 681-687.
Zoidl, G., Petrasch-Parwez, E., Ray, A., Meier, C., Bunse, S., Habbes, H .- W., et al., 2007. Localization of the pannexin1 protein at postsynaptic sites in the cerebral cortex and hippocampus. Neuroscience 146 (1), 9-16.