Bovine adrenal glomerulosa (AZG) cells were shown to express bTREK-1 background K+ channels that set the resting membrane potential and couple angiotensin II (ANG II) receptor activation to membrane depolarization and aldosterone secretion. Northern blot and in situ hybridization studies demonstrated that bTREK-1 mRNA is uniformly distributed in the bovine adrenal cortex, including zona fasciculata and zona glomerulosa, but is absent from the medulla. TASK-3 mRNA, which codes for the predominant background K+ channel in rat AZG cells, is undetectable in the bovine adrenal cortex. In whole cell voltage clamp recordings, bovine AZG cells express a rapidly inactivating voltage-gated K+ current and a noninactivating background K+ current with properties that collectively identify it as bTREK-1. The outwardly rectifying K+ current was activated by intracellular acidification, ATP, and superfusion of bTREK-1 openers, including arachidonic acid (AA) and cinnamyl 1–3,4-dihydroxy-α-cyanocinnamate (CDC). Bovine chromaffin cells did not express this current. In voltage and current clamp recordings, ANG II (10 nM) selectively inhibited the noninactivating K+ current by 82.1 ± 6.1% and depolarized AZG cells by 31.6 ± 2.3 mV. CDC and AA overwhelmed ANG II-mediated inhibition of bTREK-1 and restored the resting membrane potential to its control value even in the continued presence of ANG II. Vasopressin (50 nM), which also physiologically stimulates aldosterone secretion, inhibited the background K+ current by 73.8 ± 9.4%. In contrast to its potent inhibition of bTREK-1, ANG II failed to alter the T-type Ca2+ current measured over a wide range of test potentials by using pipette solutions of identical nucleotide and Ca2+-buffering compositions. ANG II also failed to alter the voltage dependence of T channel activation under these same conditions. Overall, these results identify bTREK-1 K+ channels as a pivotal control point where ANG II receptor activation is transduced to depolarization-dependent Ca2+ entry and aldosterone secretion.
- patch clamp
- two-pore K+ channel
angiotensin ii (ANG II) is a principal physiological stimulus for aldosterone secretion by bovine adrenal glomerulosa (AZG) cells (3, 41). Although ANG II-stimulated aldosterone secretion is mediated through the activation of a losartan-sensitive AT1 receptor, the specific signaling pathways involved are only partially understood. In particular, the roles of specific ion channels and depolarization-dependent Ca2+ entry in the process have not been clarified. In this regard, both bovine and rat AZG cells maintain negative resting potentials and express both voltage-gated T- and L-type Ca2+ channels (26, 41, 46), as well as voltage-gated and background K+ channels (4, 10, 25, 29, 31, 43).
ANG II-stimulated aldosterone secretion depends, at least in part, on Ca2+ entry through voltage-gated T- and L-type Ca2+ channels (6, 24, 25, 46). Several studies indicate that ANG II enhances the activity of low voltage-activated T-type Ca2+ channels in AZG cells (6, 9, 26, 31). The enhanced activity of T-type Ca2+ channels was associated with an ∼10-mV negative shift in the voltage dependence of T channel activation (6, 31). These actions of ANG II may occur through activation of an AT1 receptor through a mechanism that involves calmodulin-dependent protein kinase II (1, 27).
In contrast to the above findings, other patch clamp studies on rat and bovine AZG cells reported that ANG II either has no effect or actually inhibits T-type Ca2+ channels in these cells (25, 45). Furthermore, ANG II does not directly activate high voltage-activated L-type Ca2+ channels in AZG cells (24, 25, 30). Overall, it is unlikely that ANG II enhances Ca2+ entry in AZG cells solely through a direct action on voltage-gated Ca2+ channels.
Other studies reported that ANG II depolarizes murine, feline, bovine, and human AZG cells by inhibiting unidentified background K+ channels, thereby suggesting a specific mechanism for the indirect activation of voltage-gated Ca2+ channels (4, 25, 41, 43). However, until recently, K+ channels that could set the resting potential of AZG cells and whose inhibition by ANG II would be coupled to depolarization-dependent Ca2+ entry have not been identified. In recent years, more than one dozen two-pore/four-transmembrane (2P/4TMS) background K+ channels have been identified. These background K+ channels exhibit little voltage dependence, remain open at negative membrane potentials, and set the resting potential of a wide range of cells (19, 39). Recently, rat AZG cells were shown to express the 2P/4TMS K+ channels TASK-1 and TASK-3 (10, 11). TASK-3 was reported to be the dominant background K+ channel in these cells (10). However, neither TASK-1 nor TASK-3 was shown to set the resting potential of AZG cells, nor was inhibition of either channel by ANG II shown to mediate membrane depolarization.
Cortisol-secreting bovine adrenal zona fasciculata (AZF) cells express bTREK-1 background channels that are inhibited through activation of multiple native G protein-coupled receptors. These include receptors for the peptides ACTH and ANG II, as well as P2Y nucleotide and multiple P1 adenosine receptors (16, 32, 33, 51, 52). Inhibition of bTREK-1 through all of these receptors is coupled to AZF cell depolarization.
We now report that bovine AZG cells also robustly express bTREK-1 K+ channels and that, in these cells, they set the resting membrane potential. Inhibition of these channels by ANG II is tightly coupled to membrane depolarization. Specific activators of bTREK-1 channels reverse ANG II-mediated depolarization and suppress aldosterone secretion. Under conditions wherein ANG II produced nearly complete inhibition of bTREK-1, this peptide had no effect on the T-type Ca2+ current.
MATERIALS AND METHODS
Tissue culture media, antibiotics, fibronectin, and fetal bovine serum (FBS) were obtained from Invitrogen (Carlsbad, CA). Coverslips were from Bellco (Vineland, NJ). Phosphate-buffered saline (PBS), enzymes, 1,2,-bis(2-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid (BAPTA), ATP, arachidonic acid (AA), AMP-PNP, and ANG II were from Sigma (St. Louis, MO). Baicalein and cinnamyl 1–3,4-dihydroxy-α-cyanocinnamate (CDC) were obtained from Biomol (Plymouth Meeting, PA). rTASK-3 cDNA was the kind gift from both Dr. D. Kim (Department of Physiology and Biophysics, Finch University of Health Sciences/The Chicago Medical School) and Dr. R. Preisig-Müller (Institut für Normale und Pathologische Physiologie, Marburg University). Marathon-ready cDNA from normal human adrenals was obtained from Clontech (Palo Alto, CA).
Isolation and culture of AZG cells.
Bovine adrenal glands were obtained from steers (age range 2–3 yr) within 1 h of slaughter at a local slaughterhouse. Fatty tissue was removed immediately, and the glands were transported to the laboratory in ice-cold PBS containing 0.2% dextrose. Isolated AZG cells were obtained and prepared as previously described (40), with some modifications. Briefly, glomerulosa cells were isolated from adrenal capsular tissue and cells adherent to the capsule. Capsular tissue was cut into small (0.2–0.5 cm2) pieces. Tissue was digested for 1 h at 37°C in DMEM-F12 (1:1) with dispase (10 mg/ml), BSA (1%, wt/vol) and 50 μg/ml DNase. After digestion, the tissue suspension was strained through two layers of cheesecloth, and cells were either resuspended in DMEM-F12 (1:1) with 10% FBS, 100 U/ml penicillin, 0.1 mg/ml streptomycin, and the antioxidants 1 μM tocopherol, 20 nM selenite, and 100 μM ascorbic acid (DMEM-F12+) and plated for immediate use, or resuspended in FBS-5% DMSO, divided into aliquots, and stored in liquid nitrogen for future use. Cells were plated in 35-mm dishes for secretion experiments or 35-mm dishes containing 9-mm2 glass coverslips for electrophysiology experiments. Dishes or coverslips were treated with fibronectin (10 μg/ml) at 37°C for 30 min and then rinsed with warm, sterile PBS immediately before cells were added. Cells were plated in DMEM-F12+ and were maintained at 37°C in a humidified atmosphere of 95% air-5% CO2.
Patch clamp experiments.
Patch clamp recordings of K+ channel currents were made in the whole cell configuration. The standard pipette solution consisted of 120 mM KCl, 1 mM CaCl2, 2 mM MgCl2, 11 mM BAPTA, 10 mM HEPES, 1 mM ATP, and 200 μM GTP, with pH titrated to 7.1 with KOH. Pipette solution of this composition yielded a free Ca2+ concentration of 2.2 × 10−8 M, as determined by the Bound and Determined software program (5). In some experiments, MgATP in the pipette solution was raised to 5 mM and pH lowered to 6.4, as noted in the text. The external solution consisted of 140 mM NaCl, 5 mM KCl, 2 mM CaCl2, 2 mM MgCl2, 10 mM HEPES, and 5 mM glucose, with pH adjusted to 7.4 with NaOH. All solutions were filtered through 0.22-μm cellulose acetate filters.
Recording conditions and electronics.
AZG cells were used for patch clamp experiments 2–12 h after plating. Because AZG cells are significantly smaller than AZF cells, cells with capacitances of 7–12 pF were selected for recording. Coverslips were transferred from 35-mm culture dishes to the recording chamber (volume 1.5 ml), which was continuously perfused by gravity as a rate of 3–5 ml/min. Patch electrodes with resistances of 2–3 MΩ were fabricated from Corning 0010 glass (World Precision Instruments, Sarasota, FL). These electrodes routinely yielded access resistances of 1.5–5.0 MΩ and voltage clamp time constants of <100 μs. K+ currents were recorded at room temperature (22–25°C) according to the procedure of Hamill et al. (22) by use of a List EPC-7 patch clamp amplifier.
Pulse generation and data acquisition were done using a personal computer and PCLAMP software with TL-1 interface (Axon Instruments, Burlingame, CA). Currents were digitized at 2–10 KHz after filtering with an eight-pole Bessel filter (Frequency Devices, Haverhill, MA). Linear leak and capacity currents were subtracted from current records by use of summed scaled hyperpolarizing steps of one-third to one-fourth pulse amplitude. Data were analyzed using PCLAMP (CLAMPFIT 6.04) and SigmaPlot (version 8.0) software. Drugs were applied by bath perfusion, controlled manually by a six-way rotary valve.
Measurement of bTREK-1 K+ currents.
The absence of time- and voltage-dependent inactivation of the bTREK-1 K+ current allowed it to be easily isolated for measurement in whole cell recordings from AZG cells, using either of two voltage clamp protocols. When voltage steps of 300-ms duration were applied from a holding potential of −80 mV to a test potential of +20 mV, bTREK-1 could be selectively measured near the end of the voltage step, where the rapidly inactivating bKv1.4 K+ current had completely inactivated. Alternatively, bTREK-1 was selectively activated with an identical voltage step, after a 10-s prepulse to −20 mV had fully inactivated bKv1.4 K+ current (see Fig. 2A).
Aldosterone secretion experiments.
AZG cells were cultured on fibronectin-coated 35-mm dishes at a density of 1.5 × 106 cells/dish in defined media [DMEM-F12 (1:1), 50 μg/ml BSA, 100 μM ascorbic acid, 1 μM tocopherol, 10 nM insulin, and 10 μg/ml transferrin]. After 1 h, the medium was aspirated and changed to defined medium (DMEM-F12 with 50 μg/ml BSA, 100 μM ascorbic acid, 1 μM tocopherol, 0.15 μg/ml insulin, and 10 μg/ml transferrin) either without (control), or with CDC or ANG II or these two in combination. Drugs were added directly to media in dishes from concentrated stock. Samples (200 μl) of media were collected at selected times and frozen at −20°C for later assay. Aldosterone concentration was determined using a solid-phase radioimmunoassay kit (Diagnostic Products, Los Angeles, CA). Experiments were performed in triplicate and assayed for aldosterone in duplicate.
Northern blot analysis.
RNeasy columns (Qiagen, Valencia, CA) that had been treated with RNase-free DNase (Qiagen) to remove genomic contamination were used to extract total RNA from AZG and AZF cells that had been cultured in DMEM-F12+ for 8 h.
Total RNA was separated on a denaturing 8% formaldehyde-1.0% agarose gel and transferred to a nylon membrane (Gene Screen Plus, NEN). RNA was fixed to the membrane by UV cross-linking using a Stratalinker (Stratagene, La Jolla, CA). Northern blot was prehybridized in a heat-sealable plastic bag for 2 h at 42°C in ULTRAhyb (Ambion, Austin, TX) and then hybridized with a random-primed [α-32P]dCTP radiolabeled 379-bp EcoRI fragment of bTREK-1 or a full-length TASK-3 cDNA (Prime-it-II, Stratagene) overnight in minimal volume of hybridization solution at 42°C. After 18–24 h, blots were washed twice at room temperature in 2× SSPE for 15 min, twice at 40°C in 1× SSPE, and 1% SDS for 30 min. For TREK-1 hybridization, a more stringent wash, twice at 65°C with 0.1× SSPE, and 1% SDS for 15 min was necessary. Autoradiograms were obtained by exposing blots to Kodak X-O-Mat AR film at −70°C for 1 h (bTREK-1) or 24 h (TASK-3).
In situ hybridization.
Bovine tissue, obtained as described above, was immersed in 4% paraformaldehyde-0.1 M sodium phosphate buffer, pH 7.4, at 4°C for 1–3 h to preserve morphology and then embedded in OCT (Tissue-Tek) embedding matrix for frozen sectioning in embedding molds. Frozen blocks were allowed to equilibrate with cryostat chamber at −7°C. Tissue was cut into 10-μm sections and then thaw-mounted onto charged slides (SuperFrost/Plus). Slides were stored at −80°C. Before hybridization, slides were allowed to equilibrate to room temperature and then fixed in a 4% paraformaldehyde solution. Slides were then subjected to a series of washes in 0.1 M PBS (pH 7.4) and acetic anhydride (0.25% I 0.1 M triethanolamine, 0.9% NaCl, pH 8.0) and subsequently dehydrated and delipidated in a series of ethanol washes. The hybridization reaction was carried out overnight at 37°C in the presence of 60 μl of hybridization buffer (Amresco) with 3′-terminal 35S-labeled oligo, 100 mM DTT, and 250 μg/ml yeast tRNA. Slides were washed in SSC of increasing stringency, dehydrated by a series of ethanol washes, and then exposed to Biomax film (Kodak) for 3 days to evaluate signal.
A bTREK-1 probe was designed using template region bp 1167–1203, where there is much less sequence similarity to other 2P/4TMS channels such as TREK-2, TASK, or TRAAK. A sense probe was used to assay nonspecific binding. A bovine 11β-hydroxylase (CYP11B) 32-nt oligo (5′-GTC CAG CTG GGA TGT GGT AGT TCT GCA GCA CC-3′) was used as a positive control to provide morphological identification of adrenal tissue regions. Specific bTREK-1 probe sequences were as follows: antisense: 5′-CTT GTC ATA AAT CTC CAC GCT CAG CCG CCT CCT GGT T-3′ (37 nt), and sense control: 5′-aac cag gag gcg gct gag cgt gga gat tta tga cca g-3′(37 nt). Oligos were synthesized and PAGE purified by IDT (Coralville, IA). Probe sequences were checked against sequences in GenBank to ensure no cross-reactivity with other two-pore K+ channel gene products or sequences in the database.
bTREK-1 mRNA Expression in Bovine Adrenal Cells
Northern blot analysis was used to characterize the relative expression of mRNA coding for TREK-1 and TASK-3 in bovine AZF and AZG cells. Figure 1A shows that AZF and AZG cells both express bTREK-1 mRNA, which is present in three transcripts of 4.9, 3.6, and 2.8 kb. Furthermore, as previously shown in AZF cells, bTREK-1 mRNA is also markedly induced in AZG cells by a 20-h treatment with forskolin (5 μM) (14).
In contrast to TREK-1, TASK-3 mRNA was poorly expressed in these same bovine AZF and AZG cells. TASK-3 was undetectable in the RNA from both control and forskolin-treated cells after the film was exposed for 24 h, whereas bTREK-1 mRNA was easily detected with a 1-h exposure.
In situ hybridization experiments confirmed the finding that bTREK-1 mRNA is robustly expressed throughout the bovine adrenal cortex, including the glomerulosa and fasciculata. These experiments also indicated that bTREK-1 is undetectable in the adrenal medulla. The experiment illustrated in Fig. 1B shows that mRNAs for bTREK-1 and CYP11B, a steroid hydroxylase in the pathways that convert cholesterol to cortisol and aldosterone (35), are both strongly expressed in bovine AZF and AZG tissue, but not in the adrenal medulla. In this regard, no distinction could be made between the level of bTREK-1 expression in the subcapsular glomerulosa and the adjacent fasciculata. The presence of bTREK-1 mRNA in AZF as well as AZG cells suggests that the corresponding TREK-1 K+ channels are expressed in glomerulosa cells.
bTREK-1 K+ Channels are Expressed in Bovine AZG Cells
AZG cells were isolated for patch clamp recording, and K+ currents from these cells were activated by two different voltage clamp protocols, as described in Methods. Whole cell patch clamp recordings showed that bovine AZG cells uniformly expressed two types of K+ currents that were similar to those previously described in bovine AZF cells (16, 17, 34). These included a rapidly inactivating A-type K+ current and a noninactivating K+ current with a large instantaneous and a smaller time-dependent component.
bTREK-1 is distinctive among 2P/4TMS channels in its activation by ATP and intracellular acidification (15, 16, 53). The noninactivating K+ current in AZG cells was activated by both ATP and low pH. In the experiment illustrated in Fig. 2, whole cell K+ currents were recorded with pipette solutions containing 5 mM MgATP at pH 6.4 or 1 mM MgATP at pH 7.1. When recordings were made with 5 mM MgATP at pH 6.4, the noninactivating K+ current spontaneously increased in amplitude for a period of minutes before it reached a stable maximum value (Fig. 2). In contrast, when the K+ current was recorded at pH 7.1 with 1 mM MgATP in the pipette, the noninactivating current was less prominent and failed to grow significantly beyond its initial amplitude. Overall, with acidified pipette solution containing 5 mM MgATP, the putative bTREK-1 reached a maximum current density of 27.9 ± 4.8 pA/pF (n = 24). By comparison with standard pipette solution, this current reached a maximum current density of only 11.1 ± 2.2 pA/pF (n = 12).
In situ hybridization experiments showed that bovine adrenal chromaffin cells express little or no bTREK-1 mRNA. Accordingly, bTREK-1 current was undetectable in whole cell patch clamp recordings from enzymatically dissociated chromaffin cells with acidified pipette solution containing 5 mM ATP. In recordings from eight cells, only voltage-gated K+ currents similar to those previously described were observed (28) (data not shown).
AA Activates the Background K+ Current in AZG Cells
Although the noninactivating current enhanced by acidified pipette solution containing 5 mM MgATP resembled that of bTREK-1, additional evidence was needed to establish its identity. Of the more than one dozen background K+ channels characterized thus far, only the mechanogated subgroup including TREK-1, TREK-2, and TRAAK is activated by AA and other polyunsaturated fatty acids (12, 18, 19, 39). In AZG cells, AA triggered a pronounced increase in the amplitude of a noninactivating current with voltage-dependent rectification indistinguishable from bTREK-1 (16).
In the experiment illustrated in Fig. 3A, whole cell K+ currents were activated with and without depolarizing prepulses, using pipettes containing standard internal solution (pH 7.1, 1 mM MgATP). After the noninactivating current reached a stable amplitude, the cell was superfused with 10 μM AA, which increased this K+ current more than 17-fold within 8 min, whereas the rapidly inactivating A-type current was completely inhibited.
The increase in amplitude of the noninactivating K+ current was rapidly reversed, whereas inhibition of the A-type K+ current was poorly reversible upon superfusion of control saline (Fig. 3, A and B). Overall, 10 μM AA increased the noninactivating current density of six AZG cells from 6.7 ± 2.0 to 162.4 ± 28.5 pA/pF (Fig. 3C).
AA also markedly increased the noninactivating K+ current in cells where the current had been preactivated with acidified pipette solution containing 5 mM MgATP. Although the noninactivating K+ current reached a maximum density of 22.0 ± 3.8 pA/pF (n = 6) in control saline, it grew to 222.0 ± 50.5 pA/pF (n = 6) in the presence of 10 μM AA (Fig. 3C).
In the presence of standard external solution, bTREK-1 appears as an outwardly rectifying K+ current (12, 16, 53). The voltage-dependent rectification of the background K+ current in AZG cells activated by ATP and acidified pipette solution and by 10 μM AA was characterized using voltage ramps. In the experiment illustrated in Fig. 3B, K+ currents were recorded with acidified pipette solution (pH 6.4) containing 5 mM MgATP before and after the cell was superfused with 10 μM AA. Under bath conditions, linear voltage ramps applied between +60 and −140 mV induced similar outwardly rectifying currents that reversed at potentials near the theoretical K+ equilibrium potential. Thus AZG cells express a background K+-selective current with properties indistinguishable from those of bTREK-1.
ANG II Inhibits bTREK-1 and Depolarizes AZG Cells
In situ hybridization, Northern blot, and patch clamp experiments indicate that bTREK-1 is the predominant background K+ channel expressed by bovine AZG cells. If bTREK-1 K+ channels set the resting membrane potential, then inhibition of these channels by ANG II should be coupled to depolarization.
In whole cell recordings, ANG II (10 nM) potently and selectively inhibited the noninactivating K+ current. Inhibition began within 90 s and typically reached a steady-state value in 5–7 min (Fig. 4A). Overall, ANG II (10 nM) inhibited bTREK-1 current by 82.1 ± 6.1% (n = 7; Fig. 4, A and C).
In current clamp recordings of AZG cell membrane potential, it was discovered that ANG II-mediated inhibition of bTREK-1 current was accompanied by membrane depolarization. In the experiment illustrated in Fig. 4B, ANG II (10 nM) depolarized this AZG cell by 30 mV from its resting value of −58 mV. Maximum depolarization occurred within 6 min. Overall, in current clamp recordings, AZG cells maintained a resting membrane potential of −63.6 ± 2.3 mV (n = 5). ANG II (10 nM) depolarized these cells by an average of 31.6 ± 2.3 mV with a temporal pattern that paralleled bTREK-1 inhibition.
Vasopressin Inhibits bTREK-1 Current in AZG Cells
If ANG II-stimulated secretion is mediated through bTREK-1 inhibition, then other agents that physiologically induce aldosterone secretion might also inhibit this background K+ current. Vasopressin stimulates aldosterone secretion through a phospholipase C-coupled receptor (48). Vasopressin (50 nM) inhibited bTREK-1 in AZG cells by an average of 73.8 ± 9.4% (n = 4; Fig. 4C).
AA Overwhelms ANG II-Mediated Inhibition of bTREK-1 and Hyperpolarizes AZG Cells
The correlation that exists between ANG II-mediated inhibition of bTREK-1 and membrane depolarization provides further evidence that these channels act pivotally in setting the membrane potential of AZG cells. If so, then activation of bTREK-1 channels by agents such as AA should oppose membrane depolarization by ANG II.
In whole cell patch clamp experiments, it was discovered that AA overwhelmed the inhibition of bTREK-1 by ANG II and completely reversed ANG II-mediated membrane depolarization. In the experiment illustrated in Fig. 5, ANG II (2 nM) produced nearly complete inhibition of bTREK-1 (Fig. 5A, trace 2) and depolarized the cell by 28.6 mV from its resting potential of −68.6 mV (Fig. 5B). Superfusion of the cell with saline containing 2 nM ANG II and 10 μM AA dramatically increased bTREK-1 to an amplitude nearly six times the control value (Fig. 5A, trace 3). This increase in bTREK-1 was accompanied by a rapid hyperpolarization from −40.6 to −72.7 mV within 2 min (Fig. 5B).
CDC Activates bTREK-1, Reverses ANG II-Stimulated Membrane Depolarization, and Inhibits Aldosterone Secretion
Recently, we demonstrated that CDC and other selected caffeic acid esters markedly enhance the activity of native AZF cell and cloned bTREK-1 channels (13). CDC also significantly increased the noninactivating K+ current in AZG cells. In the experiment illustrated in Fig. 6A, CDC (20 μM) increased the noninactivating K+ current more than 17-fold in 8 min before the gigohm seal was lost. Overall, CDC (10 or 20 μM) increased this current density in six AZG cells from a control value of 15.6 ± 3.1 to 192.6 ± 65.5 pA/pF (n = 6).
CDC resembles AA in effectively opening bTREK-1 K+ channels in AZG cells. Within the framework of our model, CDC would also be expected to reverse ANG II-stimulated depolarization. In the experiment illustrated in Fig. 6B, ANG II depolarized an AZG cell by 30.0 mV from its resting potential of −63.2 mV. Superfusing the cell with CDC (20 μM) repolarized the cell within 9 min from −33.2 to −58.0 mV. Similar results were obtained in each of three cells.
Because CDC reverses ANG II-stimulated membrane depolarization, it should also inhibit depolarization-dependent aldosterone secretion. In the experiment illustrated in Fig. 7, CDC (20 μM) inhibited ANG II-stimulated aldosterone secretion measured at 1.5 and 16 h by 83 and 95.4%, respectively. CDC also inhibited unstimulated aldosterone secretion at 1.5 and 16 h by 32 and 56.8%, respectively, in the same experiment. Overall, at 1.5 h in three separate experiments, CDC inhibited ANG II-stimulated and unstimulated aldosterone secretion by 70.5 ± 7.6 and 39.7 ± 4.4%. The inhibitory effects of CDC on aldosterone secretion were reversible, and this drug did not affect cell viability as determined by Trypan blue exclusion (data not shown).
It has been reported that ANG II-stimulated aldosterone secretion is mediated, in part, by 12-lipoxygenase products of AA (20, 36). Because CDC is a potent 12-lipoxygenase antagonist (IC50 = 0.06 μM), the possibility that CDC-mediated inhibition of aldosterone secretion occurs through inhibition of this enzyme, rather than bTREK-1 activation, had not been eliminated (8).
Baicalein inhibits 12-lipoxygenase with an IC50 of 0.015 μM but does not activate bTREK-1 (8, 13). At a concentration of 10 μM, baicalein failed to inhibit ANG II-stimulated aldosterone secretion at either 1.5 or 16 h (Fig. 7).
ANG II Has No Effect on IT-Ca in AZG Cells
ANG II inhibited bTREK-1 currents in AZG cells with pipette solutions containing 5 mM MgATP and intracellular Ca2+ concentration ([Ca2+]i) buffered to 22 nM using 11 mM BAPTA. Experiments were done to determine whether ANG II modulated voltage-gated Ca2+ currents through the same signaling pathway.
In whole cell recordings of Ca2+ current, the majority of AZG cells express only low voltage-activated T-type Ca2+ currents that are distinguished by their rapid inactivation and slow deactivation kinetics (Fig. 8A). The slow rate of T channel closing is observed in whole cell recordings as a prominent, decaying “tail” current upon repolarization after a brief depolarizing step (Fig. 8A, right traces).
The modulation of IT-Ca in AZG cells by ANG II was monitored in whole cell recordings with pipette solutions containing nucleotides and 11 mM BAPTA to buffer Ca2+, as described above for recording K+ currents. Under these conditions, ANG II (2 or 10 nM) failed to alter IT-Ca amplitudes measured in response to short (10 ms) or long (300 ms) voltage steps to −5 or −10 mV, from a holding potential of −80 mV (Fig. 8A). Overall, at concentrations of 2 and 10 nM, ANG II reduced IT-Ca insignificantly to 0.96 ± 0.02 (n = 6) and 0.97 ± 0.01 (n = 3) of its control amplitude (Fig. 8A).
ANG II also failed to alter the amplitude of IT-Ca measured over a wide range of test voltages. In the experiments illustrated in Fig. 8B, I T-Ca was activated by voltage steps between −60 and +50 mV before and after the cell was superfused with ANG II (2 nM) for 7–10 min. Plotting the averaged current densities from three cells against membrane voltage showed that ANG II did not significantly change IT-Ca at any of the 12 test potentials.
It was discovered that bovine AZG cells express bTREK-1 background K+ channels that set the resting membrane potential and couple ANG II receptor activation to membrane depolarization. These results suggest a model for aldosterone secretion wherein inhibition of bTREK-1 K+ channels by ANG II leads to depolarization and the activation of voltage-gated Ca2+ channels. Accordingly, TREK-1 K+ channel openers reverse ANG II-stimulated depolarization and inhibit aldosterone secretion. The pivotal role assigned to TREK-1 K+ channels in this model contrasts with previous studies that focused on a direct effect of ANG II on T-type Ca2+ channels. In our experiments, ANG II failed to produce a measurable change in the activity of T-type Ca2+ channels under conditions where bTREK-1 currents were nearly completely inhibited by this peptide hormone.
TREK-1 Is the Major Background K+ Channel of Bovine AZG Cells
The combination of Northern blot, in situ hybridization, and patch clamp studies revealed that TREK-1, rather than TASK-3, is the major K+ channel expressed by bovine AZG cells. Northern blot analysis showed that AZF and subcapsular AZG cells both expressed the same three bTREK-1 mRNA transcripts, each of which were similarly induced by forskolin. Thus bTREK-1 expression in AZG cells is likely regulated at the transcriptional level by ACTH through a cAMP-dependent mechanism as it is in AZF cells (14). In situ hybridization on bovine adrenal gland sections corroborated the Northern blot results and showed that bTREK-1 mRNA was uniformly distributed in AZF and AZG cells but was virtually undetectable in the adrenal medulla.
Results from patch clamp experiments were in agreement with those measuring TREK-1 mRNA distribution in the bovine adrenal gland. Bovine AZG cells expressed two K+ currents that were indistinguishable from those of AZF cells. Most importantly, these cells expressed hundreds of background K+ channels that are either dormant of have a low open probability and display a composite profile that matches that of bTREK-1. These outwardly rectifying channels were activated by intracellular acidification, ATP, AA, and CDC and inhibited by ANG II. Of the 2P/4TMS channels identified thus far, only bTREK-1 channels possess all of these properties (15, 16, 53). However, our results do not formally exclude the unlikely possibility that a background K+ channel in addition to bTREK-1 is expressed by bovine AZG cells.
Although we were extremely careful in our dissection to obtain only subcapsular glomerulosa tissue, it is possible that isolated glomerulosa cells were contaminated with a small fraction of AZF cells. By choosing smaller cells, we further reduced the possibility that patch clamp recordings mistakenly included AZF cells. In this regard, it is important to note that each of the more than 25 AZG cells exposed to the bTREK-1 openers responded with large increases in the noninactivating K+ current. Thus bTREK-1 appears to be uniformly expressed in both AZF and AZG cells.
Although TASK-3 may be the major background K+ current found in rat AZG cells (10), we found no evidence that this channel is expressed in bovine AZG. TASK-3 was undetectable in Northern blots of bovine AZG mRNA. When whole cell K+ currents were recorded with pipette solutions that minimized the expression of TREK-1, no background K+ current was activated. There is little doubt that bTREK-1 is the major background channel that sets the resting potential of bovine AZG cells.
In contrast to its expression in the adrenal cortex, no evidence of TREK-1 expression in neural crest-derived chromaffin cells was found in in situ hybridization or whole cell patch clamp experiments. When K+ currents were recorded from chromaffin cells with acidified pipette solution containing 5 mM MgATP, no bTREK-1 current was detected. Furthermore, neither CDC nor AA activated such a current in chromaffin cells (unpublished observations). Although these neural crest-derived cells express multiple K+ channel subtypes, including voltage- and Ca2+-activated K+ channels, the background K+ channel that sets their resting membrane potential remains to be identified (28).
ANG II Regulates AZG Membrane Potential through TREK-1 Inhibition: Model for Depolarization-Dependent Secretion
A requirement for Ca2+ in ANG II-stimulated aldosterone secretion is well established (7, 24, 25, 41, 47). In exploring the cellular mechanism, a number of studies have focused on ANG II modulation of T-type Ca2+ channels (6, 9, 26, 31). However, none of these has provided a satisfactory explanation whereby ANG II could trigger large increases in Ca2+ entry through voltage-gated channels.
The results of the present study suggest a specific model for ANG II-stimulated secretion that assigns a critical role to bTREK-1 K+ channels. In this model, ANG II-mediated inhibition of bTREK-1 is coupled to membrane depolarization, Ca2+ channel activation, and aldosterone secretion. This model allows for the indirect activation of T- or L-type Ca2+ channels through TREK-1 inhibition. Several possibilities exist that could produce efficient Ca2+ entry through either channel. If ANG II-mediated inhibition of bTREK-1 produces a sustained depolarization under physiological conditions, it would produce a continuous influx through noninactivating Ca2+ channels. The effects of ANG II on L-type Ca2+ channels in AZG cells are complex, and both enhancements and inhibition of Ca2+ entry have been reported (23–25, 30, 46). However, no L-type Ca2+ current was present in the majority of freshly cultured bovine AZG cells.
Alternatively, bTREK-1 may serve as a brake on the electrical activity of AZG cells. Inhibition of bTREK-1 by ANG II could remove this brake, triggering Ca2+-dependent action potentials driven by opposing T-type Ca2+ currents and A-type K+ currents. Regenerative Ca2+-dependent action potentials in AZG cells have been observed (42). It is unlikely that bovine AZG cells generate action potentials when membrane potential is recorded at 21–23°C with a whole cell patch electrode. Because membrane potential was sampled at 100-ms intervals in these experiments, fast action potentials would not have been readily detected. Action potentials would most likely be recorded from AZG cells in an adrenal slice with a sharp intracellular electrode at physiological temperatures. Regardless, the combined direct effects of ANG II on both T-type Ca2+ channels and bTREK-1 K+ channels produce the ionic effects that mediate aldosterone secretion.
Similar to ANG II, vasopressin stimulates aldosterone secretion through a PLC-coupled receptor requiring Ca2+ influx (21, 44, 50). The inhibition of bTREK-1 K+ channels in AZG cells by vasopressin suggests that it may also stimulate aldosterone secretion through depolarization-dependent Ca2+ entry.
Signaling Pathways for ANG II Modulation of Ion Channels in AZG Cells
The signaling pathways by which ANG II modulates the activity of ion channels in AZG cells are incompletely understood. In particular, the modulation of T-type Ca2+ channels by ANG II has produced conflicting results. In several studies, it has been reported that ANG II increased IT-Ca by a mechanism that involved a hyperpolarizing shift in the voltage dependence of channel activation (6, 9, 27, 31). These effects may occur through Ca2+-dependent activation of Ca2+-CaM kinase II (1, 6, 27).
However, ANG II has also been reported to inhibit T-type Ca2+currents in bovine AZG cells through activation of protein kinase C (45). Notably, the activity of protein kinase C by diacylglycerol is also enhanced by [Ca2+]i (37). Finally, in perforated patch whole cell recordings, ANG II failed to alter the activity of IT-Ca in rat AZG cells (25).
Although our results do not explain the conflicting findings described above, they do suggest that ANG II modulates ion channels in AZG cells by multiple signaling pathways, not all of which are Ca2+ dependent. In our experiments, including those measuring the activity of bTREK-1 and T-type Ca2+ channels, [Ca2+]i was strongly buffered to 22 nM using 11 mM BAPTA. If ANG II modulation of IT-Ca, including activation or inhibition, requires activation of Ca2+-dependent enzymes, this response would likely be blunted or eliminated in our experiments (2, 54).
Regardless, our results do show that ANG II potently and nearly completely inhibits bTREK-1 current and depolarizes AZG cells through a signaling pathway that does not modulate T-type Ca2+ channels. It appears that ANG II-mediated modulation of Ca2+ and K+ channels in bovine AZG cells occurs through multiple Ca2+-dependent and -independent signaling pathways.
TREK-1 Channel Activation and Inhibition of Aldosterone Secretion
The ability of AA and CDC to activate bTREK-1 channels and restore membrane potential to AZG cells depolarized by ANG II provides convincing evidence that these channels set the resting potential of AZG cells. The effectiveness of CDC in reversing ANG II-stimulated depolarization and inhibiting ANG II-stimulated aldosterone secretion is consistent with a model in which TREK-1 couples ANG II receptor activation to depolarization-dependent Ca2+ entry. Within this framework, CDC would negate the membrane depolarization that leads to Ca2+ channel activation.
These results also illustrate the utility of this new type of K+ channel activator in exploring the function of 2P/4TMS channels in cell physiology. However, additional studies will be needed to determine its specificity as an ion channel modulator. AA and other cis polyunsaturated fatty acids that activate the same subgroup of 2P/4TMS background channels also modulate several types of voltage-gated channels, limiting their value as pharmacological tools (38, 49).
In summary, the results of the present and previous studies clearly demonstrate that bTREK-1 is the predominant background K+ channel that sets the resting membrane potential of both bovine AZF and AZG cells and couples ACTH and ANG II receptors to depolarization-dependent Ca2+ entry and the secretion of cortisol and aldosterone. In contrast, TASK-type K+ channels may function similarly in the rat adrenal cortex, at least in the adrenal glomerulosa (10, 11).
Thus significant species differences that exist between rat and bovine adrenal cortical background K+ channels highlight the importance of identifying the background K+ channels in the human adrenal cortex. PCR using human cDNA as the template showed that TREK-1 is strongly expressed in the human adrenal (data not shown). Additional experiments will be required to determine whether TREK-1 functions in the human adrenal cortex as it does in bovine AZF and AZG. If so, then TREK-1 channel activators might serve as therapeutic agents in endocrine diseases marked by excessive cortisol or aldosterone secretion.
This work was supported in part by National Institute of Diabetes and Digestive and Kidney Diseases Grant R01-DK-47875 (to J. J. Enyeart), and in part by the National Science Foundation under Agreement 0112050 (S. Danthi).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 by American Physiological Society