INTRODUCTION

Top Abstract Introduction Procedures Results Discussion References

Glucocorticoid hormones exert profound effects on cellular excitability in endocrine and nerve cells through regulation of ion channel activity that requires the rapid induction of new proteins (1-3). Increasing evidence suggests that potassium channels are major targets for glucocorticoid action. Although glucocorticoids rapidly induce potassium channel subunits in some systems (4, 5), the mechanisms of channel regulation by glucocorticoids in endocrine cells and neurons are largely not understood (1-3).

Anterior pituitary corticotrope cells have been widely used as a physiologically relevant model system to explore the mechanisms of early glucocorticoid action (6, 7). In the mouse corticotrope cell line, AtT20 D16:16, the cAMP-mobilizing neuropeptide, corticotrophin-releasing factor, stimulates adrenocorticotropin secretion through the concerted action of protein kinase A (PKA)¹ to activate L-type calcium channels and inhibit BK channels and subsequent enhancement of calcium influx through L-type calcium channels (3, 8). In turn, glucocorticoids rapidly (within 2 h) inhibit corticotropin-releasing factor-stimulated secretion through the induction of new proteins (9, 10). We have previously demonstrated in metabolically intact AtT20 D16:16 corticotropes that glucocorticoid-induced proteins render BK channels insensitive to inhibition by protein kinase A and that the action of the steroid is central for the early inhibition of adrenocorticotropin hormone secretion in this system (3). Intriguingly glucocorticoids also block protein kinase A-mediated inhibition of calcium-activated potassium channels underlying the slow after-hyperpolarization in hippocampal neurons (1, 11), suggesting that calcium-activated potassium channels are common targets for reciprocal regulation of cellular excitability by glucocorticoid-induced proteins and cAMP-dependent phosphorylation.

In corticotropes, glucocorticoid-induced proteins may modulate other signaling pathways to regulate BK channel activity or may directly modulate BK channel function themselves. Increasing evidence suggests that the activity of BK channels are dynamically regulated by the interaction of protein kinases and phosphatases intimately associated with the channel complex (12-15). Because glucocorticoids specifically antagonize PKA-mediated regulation of BK channels but not L-type calcium channels in AtT20 D16:16 cells (3), we hypothesized that glucocorticoids may exert their effects through protein phosphatase activity at the level of the BK channel complex itself. Indeed increasing evidence from other systems suggests that glucocorticoids mediate some of their effects through regulation of serine/threonine as well as tyrosine-protein phosphatase activity (16-19).

Inhibition of protein phosphatases modulates cAMP accumulation and metabolism in intact AtT20 D16:16 cells (20) and glucocorticoid receptor function in many cell types (21) thus precluding definitive analysis of the mechanism of glucocorticoid action at the level of BK channels themselves in intact cells. Thus to directly address whether protein phosphatases are involved in the ability of glucocorticoids to block PKA-mediated inhibition of BK channel activity we have examined the regulation of BK channel activity in excised inside-out patches from control and glucocorticoid-pretreated AtT20 D16:16 corticotropes. The data in this report demonstrate that glucocorticoid regulation of BK channel activity requires protein phosphatase 2A activity closely associated with the BK channel complex.

	EXPERIMENTAL PROCEDURES

Top Abstract Introduction Procedures Results Discussion References

AtT20 D16:16 Cell Culture-- Clonal mouse anterior pituitary (AtT20 D16:16, passage 18-32) cells were maintained as described previously (3) and used 3-7 days post-plating on glass coverslips. Cells were treated with 1 µM of the synthetic glucocorticoid dexamethasone or vehicle (<0.01% Me₂SO) for 2 h at 37 °C in serum-free Dulbecco's modified Eagle's medium, pH 7.4, buffered with 25 mM HEPES and containing 0.25% bovine serum albumin. Cells were then transferred to the bath solution (dexamethasone-free) outlined below for electrophysiological recording. Regulation of single channel events in isolated inside-out patches from control or dexamethasone-treated cells was performed in parallel on the same passage of cells to avoid potential intra-passage variations in responsiveness.

Electrophysiology-- Single BK channel events were recorded in the inside-out patch configuration of the patch clamp technique using physiological potassium gradients. The bath (intracellular face of patch) solution contained in (mM): 140 KCl, 1 1,2-bis-(o-aminophenoxy)ethane-N,N,N',N'-tetracetic acid, 10 HEPES, 30 glucose, 1 ATP, pH 7.35, and the respective concentrations of Mg²⁺ and Ca²⁺ as indicated in the figure legends. The patch pipette (extracellular face of patch) contained in (mM): 140 NaCl, 5 KCl, 5 MgCl₂, 0.1 CaCl₂, 10 HEPES, 20 glucose, pH 7.4, containing 0.002 tetrodotoxin. Single channel events were recorded for 20-30 s every 2-5 min at the voltages indicated in the figure legends. Preliminary stability plot experiments demonstrated that BK channel activity was stable for >1 h under the recording conditions used (data not shown). Data acquisition and voltage protocols were controlled by an Axopatch 200B amplifier and pCLAMP 6 software (Axon Instruments Inc., Foster City, CA). Pipettes were manufactured from Garner 7052 glass, sylgarded, with resistances of 1-3 M in physiological saline after fire polishing.

Western Blotting-- Crude membrane homogenates from AtT20 D16:16 cells were prepared by homogenizing ~10⁷ cells on ice in homogenization buffer (in mM): 50 Tris-HCl, pH 7.4, 140 KCl, 1 EGTA, 1 MgCl₂ containing 12 units/ml aprotinin, 5 µg/ml leupeptin, 6 mM 4-(2-aminoethyl)benzenesulfonylfluoride, and 4 mM pepstatin A followed by two freeze thaw cycles. After centrifugation for 5 min 1000 × g at 4 °C the resultant supernatant was pelleted at 20,000 × g to give the crude membrane fraction. Protein samples (15 µg) were separated on a 10% SDS gel and electroblotted to Immobilon polyvinylidene difluoride membranes. Membranes were blocked for 2 h at room temperature with PBS containing 0.1 mM EDTA, 0.1% Triton X-100, pH 7.4, (PBS-TE) and 5% (w/v) low fat milk (Marvel). Blots were incubated overnight at 4 °C with a 1:2000 dilution of the affinity purified antibody slo_(913-926) (directed toward residues 913-926 of the pore-forming -subunit of mouse brain BK channels; Ref. 22) in PBS-TE containing 1% (w/v) Marvel. Blots were washed five times with PBS-TE and incubated for 45 min at room temperature with horseradish peroxidase-labeled anti-rabbit IgG (Amersham Pharmacia Biotech, 1:5000 final dilution) in PBS-TE containing 5% (w/v) Marvel. After five washes in PBS-TE, blots were incubated with Amersham Pharmacia Biotech ECL reagents according to the manufacturer's protocol, and blots were exposed to ECL film in the linear response range (Amersham Pharmacia Biotech).

Protein Phosphatase Assays-- Protein phosphatase activity of crude cytosolic and membrane fractions were determined by using the molybdate:malachite green:phosphate complex assay using the synthetic phosphopeptide RRA(pT)VA as substrate essentially as described by the manufacturer (Promega Corporation, Madison, WI). Cytosolic and crude membrane fractions were prepared from control and dexamethasone-treated AtT20 D16:16 cells as for Western blotting in homogenization buffer (in mM): 50 Tris-HCl, pH 7.4, 140 KCl, 1 EGTA, 1 MgCl₂ containing 12 units/ml aprotinin, 5 µg/ml leupeptin, 6 mM 4-(2-aminoethyl)benzensulfonylfluoride, and 4 mM pepstatin A. To remove endogenous phosphate, cytosolic fractions were passed twice through a 10-ml bed volume of Sephadex G-25, resuspended membrane fractions were incubated for 20 min at 4 °C with 10 volumes of Sephadex G-25, and the 200 × g supernatant was washed and pelleted twice at 20,000 × g in Tris-HCl as above. Phosphatase assays were performed in a volume of 50 µl in imidazole buffer (in mM: 50 imidazole, pH 7.2, containing 0.2 EGTA, 0.02% (v/v) -mercaptoethanol, and 0.1 mg/ml bovine serum albumin) for 30 min at 30 °C using 100 µM of RRA(pT)VA as substrate. PP2A activity was determined as the difference in total phosphatase activity and phosphatase activity in the presence of 10 nM okadaic acid. Under the conditions used >80% of phosphatase activity was sensitive to 10 nM okadaic acid. Reaction was terminated by addition of the molybdate dye buffer and incubated for 30 min at room temperature, and absorbance was determined at 600 nm.

Reagents-- Purified protein phosphatase 2A catalytic subunit and reagents for PP2A activity assay were from Promega Corporation (Southampton, UK). Calyculin A, okadaic acid, and norokadone were from LC Laboratories (Alexis Corporation Ltd., Nottingham, UK). The specific protein kinase A inhibitor peptide (PKI_(5-24)) was from Sigma or Calbiochem-Novabiochem Ltd. (Nottingham, UK). Tetrodotoxin was from Calbiochem-Novabiochem Ltd. (Nottingham, UK). Polyvinylidene difluoride membranes and reagents for SDS-polyacrylamide gel electrophoresis and Western blotting were from Bio-Rad Laboratories, Ltd. (Hertfordshire, UK). All other reagents were from Sigma or BDH-Merck (Poole, Dorset, UK). Dexamethasone was stored at 20 °C at 10 mM in Me₂S0.

Statistics-- Data are expressed as the means ± S.E. Statistical significance was determined by Student's t test for paired and unpaired data as appropriate. A p value of less than 0.05 was considered to be significant.

	RESULTS

Top Abstract Introduction Procedures Results Discussion References

In inside-out patches from control AtT20 D16:16 corticotropes single BK channel events were characterized by their slope conductance (125 ± 3 pS in physiological potassium gradients and 2 mM intracellular magnesium, reduced to 80 ± 4 pS with 10 mM internal magnesium) and sensitivity to voltage and calcium (Fig. 1 A-C). Over the physiological voltage range of AtT20 D16:16 cells, more than 75% (not shown) of single BK channels are half-maximally activated at positive (20-30 mV) potentials at "resting" 0.1 µM intracellular free calcium [Ca²⁺]_i levels (Fig. 1B). At levels of intracellular free calcium observed during secretagogue stimulation (1 µM) BK channels are maximally activated (Fig. 1B, half-maximal activation < 40 mV, not shown).

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Pretreatment of AtT20 D16:16 cells with dexamethasone does not modify single channel conductance or calcium and voltage sensitivity of BK channels in excised patches. A, plots of unitary current amplitude versus patch potential at room temperature in isolated inside-out patches from AtT20 D16:16 cells treated with vehicle (Control, solid symbols) or 1 µM DEX (open symbols) for 2 h at 37 °C. Unitary currents were recorded at various potentials in physiological potassium gradients (140 mM extracellular, 5 mM intracellular) in the presence of 2 mM ( and ) or 10 mM ( and ) intracellular Mg2+. The intracellular face of the patch was exposed to an intracellular free calcium concentration [Ca2+]i of 0.2 µM and 1 mM ATP. Means ± S.E., n = 3-4/group, unless otherwise indicated. Error bars are within the symbol size. B, representative plots of mean single channel open probability (Po) versus patch potential in patches containing a single BK channel from control and dexamethasone-treated cells as in A above at different [Ca2+]i (diamonds, 1 µM; squares, 0.1 µM; circles, Ca2+ free < 10 nM) and 2 mM Mg-ATP. Boltzmann functions were fitted to the data points in the presence of 0.1 µM intracellular free Ca2+, giving half-maximal activation (V1/2 max) of ~27 mV and ~25 mV for control and dexamethasone-treated cells, respectively. Steady-state Po was determined over a 30-s period under each condition. Plots are representative of >70% of all patches tested (see "Results"). C, representative single channel records from inside-out patches containing a single BK channel determined at 0.2 µM [Ca2+]i and 1 mM ATP as in B above.

Dexamethasone Does Not Modify Single Channel BK Channel Properties or Expression Levels-- Pretreatment of AtT20 D16:16 cells with a maximally effective concentration (1 µM) of the synthetic glucocorticoid agonist, dexamethasone (3, 10), had no significant effect on single channel slope conductance (125 ± 2 and 80 ± 3 pS with 2 and 10 mM internal magnesium, respectively, Fig. 1A) or sensitivity to voltage or calcium in inside-out patches (Fig. 1, A-C). Half-maximal activation of BK channels in 0.1 µM [Ca²⁺]_i in greater than 70% of patches was observed between 20-30 mV (Fig. 1B).

View larger version (28K): [in this window] [in a new window]   Fig. 2.   Dexamethasone does not modify expression levels of BK channel -subunit. Proteins from membrane fractions (15 µg) of AtT20 D16:16 cells treated with vehicle alone (Control) or 1 µM DEX for 2 h at 37 °C were separated by SDS-polyacrylamide gel electrophoresis and blotted to Immobilon polyvinylidene difluoride membranes. Immunoblots were probed with a 1:2000 dilution of the affinity purified rabbit antibody (slo(913-926)) directed against residues 913-926 of the pore forming -subunit of mouse brain BK channels and detected by ECL as described under "Experimental Procedures." The right-hand panel shows the mean (± S.E.) expression level determined by densitometric scanning of the ~125-kDa immunoreactive bands from three separate paired AtT20 D16:16 cell extracts.

Single BK Channels Are Inhibited by Activation of Closely Associated Protein Kinase A-- In eight of eight control patches application of cAMP (0.1 mM) to the intracellular face of the patch in the presence of 1 mM Mg-ATP and 0.5 µM [Ca²⁺]_i resulted in a significant inhibition of mean channel open probability, Po (expressed as the percentage of change of pretreatment Po, 72.9 ± 9.5% p < 0.01 t test, determined 10 min after cAMP application compared with pretreatment Po, n = 8, Fig. 3, A-C) that was maximal within 10 min and was maintained for more than 30 min. On washout of cAMP mean channel open probability gradually returned toward pretreatment levels; this reversal was accelerated by removal of ATP from the intracellular face of the channel (not shown). The inhibitory action of cAMP was mediated through protein kinase A-like activity closely associated with the channel in the patch because no significant inhibition of Po was observed on application of cAMP in the presence of the specific protein kinase A inhibitor peptide, PKI_(5-24) (percentage of change in Po, 0.9 ± 5.3%; n = 4) or in the absence of ATP (percentage of change in Po, 7.0 ± 10.1%; n = 4) (Fig. 3C).

View larger version (23K): [in this window] [in a new window]   Fig. 3.   PKA-mediated inhibition of BK channels is blunted in isolated patches from dexamethasone-treated cells. A, representative single channel traces from an isolated inside-out patch from a control cell before (Control) and 10 min after (+ cAMP) application of 0.1 mM cAMP to the intracellular face of the patch. The intracellular face of the patch containing a single BK channel was exposed to 0.5 µM [Ca2+]i and 1 mM ATP at + 40 mV. B, representative plots of mean single channel open probability (Po) versus time after perfusion of cAMP (0.1 mM, application started at time = 0 min) to the intracellular face of isolated inside-out patches from control and DEX-treated (1 µM pretreatment for 2 h) cells. Po was determined over 30 s at + 40 mV in the presence of 0.5 µM [Ca2+]i and 1 mM ATP and is expressed as a percentage of the Po determined 5 min before application of cAMP. Mean Po at 5 min was 0.67 and 0.81 for the isolated patch from the control and dexamethasone-treated cell, respectively. C, control, application of 0.1 mM cAMP to the intracellular face of the patch from vehicle-treated AtT20 D16:16 cells results in significant inhibition of Po (cAMP, n = 8). No significant inhibition of Po was observed in the absence of ATP (n = 4) or in the presence of 0.45 µM of the specific PKA inhibitor peptide PKI(5-24) (n = 4). DEX, cAMP-mediated inhibition of BK channel activity was severely blunted in patches from dexamethasone-pretreated (1 µM, 2 h) cells compared with inhibition in patches from control cells (n = 8). Full inhibition was restored using 100 µM ATPS as the phosphate donor (n = 5). All data are expressed as the percentage of change in pretreatment Po measured at + 40 mV in the presence of 0.5 µM [Ca2+]i and 1 mM ATP ( indicates inhibition). Mean Po before application of cAMP in each group was as follows. Control: cAMP, 0.67 ± 0.05; cAMP/no ATP, 0.69 ± 0.07; cAMP + PKI(5-24), 0.69 ± 0.09. DEX: cAMP, 0.68 ± 0.08; cAMP + ATPS, 0.69 ± 0.08. The ranges of inhibition of Po by cAMP alone in patches from control and DEX-treated cells were 35-99% and 3-54%, respectively. The means ± S.E. are shown. **, p < 0.01 compared with control cAMP group; #, p < 0.01 compared with cAMP + ATPS group (Student's t test).

Pretreatment of Cells with Dexamethasone Attenuates PKA-mediated Inhibition of BK Channels in Isolated Inside-Out Patches-- In parallel experiments, application of cAMP to the intracellular face of inside-out patches from dexamethasone-treated cells resulted in a significantly attenuated inhibition of mean channel open probability compared with inhibition observed in patches from control cells (Fig. 3, B and C). In patches from dexamethasone-treated cells the percentage of change in Po was 22.4 ± 7.1%, n = 9 (compared with 72.9 ± 9.5%, n = 8, in control patches, p < 0.01 t test, Fig. 3C). The blockade of PKA-mediated inhibition was not a result of delayed responsiveness to cAMP (Fig. 3B); in addition the effect of cAMP was mediated through activation of endogenous PKA (Fig. 4B).

View larger version (24K): [in this window] [in a new window]   Fig. 4.   Inhibition of an okadaic acid- and/or a calyculin A-sensitive protein phosphatase in isolated patches from dexamethasone-treated cells restores PKA-mediated inhibition of BK channels to that seen in patches from control cells. A, representative single channel traces from an isolated inside-out patch from a control cell before (Control) and 10 min after (+ Calyculin A + cAMP) application of 10 nM calyculin A and 0.1 mM cAMP to the intracellular face of the patch. Recordings were determined at + 40 mV with the intracellular face of the channel exposed to 0.5 µM [Ca2+]i and 1 mM ATP. B, control, in the presence of 10 nM okadaic acid (n = 8) or 10 nM calyculin A (n = 5) application of 0.1 mM cAMP to the intracellular face of the patch from vehicle-treated AtT20 D16:16 cells resulted in inhibition of Po to a similar extent as cAMP alone. The effect of cAMP was completely blocked in the presence of 0.45 µM of the specific PKA inhibitor peptide PKI(5-24) (n = 4). Application of the catalytic subunit of PP2A (1 unit/ml) to control patches completely blocked cAMP-mediated inhibition of BK channel Po (n = 5). DEX, application of 10 nM okadaic acid (n = 6) or 10 nM calyculin A (n = 4) to the intracellular face of patches from dexamethasone-treated (1 µM, 2 h) cells restored cAMP-mediated inhibition of BK channel Po to that observed in patches from control cells. The effect of cAMP was completely blocked in the presence of 0.45 µM of the specific PKA inhibitor peptide PKI(5-24) (n = 6). All data are expressed as percentages of change of pretreatment control Po measured at + 40 mV in the presence of 0.5 µM [Ca2+]i and 1 mM ATP as in Fig. 3. Mean pretreatment Po for each group was as follows. Control: cAMP, 0.67 ± 0.05; cAMP + Okadaic acid, 0.69 ± 0.09; cAMP + Calyculin A, 0.78 ± 0.08; cAMP + PKI(5-24) + Okadaic acid, 0.70 ± 0.08; cAMP + PP2A, 0.73 ± 0.07. DEX: cAMP, 0.68 ± 0.08; cAMP + Okadaic acid, 0.73 ± 0.07; cAMP + calyculin A, 0.79 ± 0.09; cAMP + PKI(5-24)> + Okadaic acid, 0.80 ± 0.05. The means ± S.E. are shown (n = 4-9). **, p < 0.01 compared with control cAMP group (Student's t test).

An Okadaic Acid-sensitive Phosphatase, Closely Associated with BK Channels, Is Required for Dexamethasone Action-- The thiophosphate of ATPS can be used by protein kinases to phosphorylate target proteins, but the resultant phosphoprotein is largely resistant to dephosphorylation. Using ATPS as the phosphate donor in place of ATP in patches from dexamethasone-treated cells, cAMP inhibited Po to the same extent as that observed in control cells (percentage of change in Po, 60.8 ± 12.3%, Fig. 3C). These data suggest that a closely associated protein phosphatase is responsible for the attenuation of PKA-mediated inhibition of single BK channels in isolated patches from dexamethasone-treated cells.

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Dexamethasone does not enhance global PP2A activity in AtT20 D16:16 cell extracts. Protein phosphatase activity from cytosolic (S, open bars) and membrane fractions (P, shaded bars) from AtT20 D16:16 cells treated with vehicle alone (Control) or 1 µM DEX for 2 h at 37 °C and expressed as pmol of phosphate released/µg of protein/min using the synthetic phosphopeptide RRA(pT)VA as substrate at 30 °C. Phosphatase activity was determined as the phosphatase activity sensitive to 10 nM okadaic acid, which under the conditions used was >80% of total phosphatase activity measured. The means ± S.E. are shown (n = 3/group).

	DISCUSSION

Top Abstract Introduction Procedures Results Discussion References

This study demonstrates that (i) BK channels in AtT20 D16:16 corticotropes are dynamically regulated by protein kinase A and protein phosphatase 2A-like activity closely associated with the BK channel complex and (ii) glucocorticoid regulation of BK channels is dependent upon protein phosphatase 2A activity at the level of the BK channel complex. Importantly, this action of glucocorticoids is context-sensitive because glucocorticoids do not modify the intrinsic properties (calcium or voltage sensitivity) of the BK channel; rather they block PKA-mediated inhibition of BK channel activity. These data support our previous electrophysiological and secretion studies in metabolically intact AtT20 D16:16 cells (3), suggesting that glucocorticoid-induced proteins render BK channels insensitive to inhibition by PKA and that this action of steroids is central to the mechanism of early inhibition of adrenocorticotropin hormone secretion in this system. Furthermore, these data support a growing body of evidence that suggests that reversible phosphorylation of ion channels acts as a dynamic process to finely tune ion channel behavior (12-15, 23).

Mechanism of Glucocorticoid Regulation of BK Channels?-- In order to directly examine the effects of PKA activation and protein phosphatases on BK channel behavior in this paper, we examined regulation in isolated patches of membrane from cells that had been pretreated with a maximally effective dose of glucocorticoid so that the full effects of steroid-induced proteins could be exerted on the channel complex. Because PKA-mediated inhibition of BK channel activity was significantly attenuated in isolated patches of membrane, as we previously observed in whole cell current recordings (3), these data strongly suggest that glucocorticoid-induced proteins exert their effect through pathways that are tightly associated with the BK channel complex. Thus it is reasonable to exclude effects of steroid that require the maintained presence of a diffusible mediator. For example, arachidonic acid metabolites and cGMP exert powerful activation of BK channels in pituitary cells through activation of protein phosphatases (14, 15); however, glucocorticoids inhibit arachidonic acid release (24) and have no effect on cGMP levels in AtT20 D16:16 cells.²

Conclusions and Perspectives-- The data in this report demonstrate that glucocorticoid inhibition of BK channel activity is dependent upon protein phosphatase 2A activity closely associated with the BK channel complex. Such a mechanism may be a common determinant to allow the reciprocal regulation of calcium-activated potassium channels by glucocorticoid-induced proteins and cAMP-dependent protein phosphorylation in excitable cells. Identification of the BK channel complex as a target for glucocorticoid action should allow us to characterize glucocorticoid-induced proteins involved in ion channel regulation and provide further insights into the mechanism and role of rapid glucocorticoid regulation of excitability in neuroendocrine and neuronal cells.