Adenosine inhibits a non-inactivating K⁺ current in bovine adrenal cortical cells by activation of multiple P₁ receptors

Abstract

1. Bovine adrenal zona fasciculata (AZF) cells express a non-inactivating K⁺ current (I_AC) that sets the resting potential while it is activated by intracellular ATP. In whole-cell patch clamp recordings from bovine AZF cells, we found that adenosine selectively inhibited I_AC by a maximum of 78·4 ± 4·6% (n = 8) with an IC₅₀ of 71 nM. The non-selective adenosine receptor agonist NECA effectively inhibited I_AC by 79·3 ± 2·9% (n = 24) at a concentration of 100 nM.

2. Inhibition of I_AC was mediated through multiple P₁ adenosine receptor subtypes. The A₁-selective agonist CCPA (10 nM), the A_2A-selective agonist CGS 21680 (100 nM) and the A₃-selective agonist IB-MECA (10 nM) inhibited I_AC by 64·8 ± 8·4, 78·4 ± 4·6 and 69·3 ± 6·9%, respectively.

3. Specific adenosine receptor subtype antagonists including DPCPX (A₁), ZM 241385 (A_2A) and MRS 1191 (A₃) effectively blocked inhibition of I_AC by adenosine receptor-selective agonists.

4. A mixture of the three adenosine receptor antagonists completely suppressed inhibition of I_AC by adenosine, but failed to alter inhibition by external ATP which acts through a separate P₂ nucleotide receptor.

5. Inhibition of I_AC by adenosine or NECA was eliminated by substituting GDP-β-S for GTP in the pipette, or by replacing ATP with AMP-PNP or UTP.

6. In addition to inhibiting I_AC, adenosine (10 μM) depolarized AZF cells by 46·2 ± 5·8 mV (n = 6).

7. These results show that bovine AZF cells express at least three adenosine receptor subtypes (A₁, A_2A, A₃), each of which is coupled to the inhibition of I_AC K⁺ channels through a G-protein-dependent mechanism requiring ATP hydrolysis. Adenosine-mediated inhibition of I_AC is associated with membrane depolarization. Adenosine and other purines may co-ordinate the stress-induced secretion of corticosteroids and catecholamines from the adrenal gland.

Adenosine is an extracellular signalling molecule that acts as a neuromodulator or local hormone at a variety of sites. When released from myocytes, neurons, endocrine cells and immune system cells, adenosine modulates a range of cell functions, including neural activity, secretion of neurotransmitters and hormones, and contraction of cardiac and smooth muscle (Williams & Burnstock, 1997; Ralevic & Burnstock, 1998).

In neurons and muscle cells, adenosine acts as a protective agent during hypoxia and ischaemia. Activation of adenosine receptors in neurons reduces neuronal activity through membrane hyperpolarization (Pan et al. 1995). In heart, activation of adenosine receptors reduces heart rate and the force of myocardial contraction (Olsson & Pearson, 1990). In the vascular system, activation of adenosine receptors leads to vasodilatation and increased blood flow (Conti et al. 1993). Many of these effects of adenosine are coupled to ionic conductance changes, particularly the activation of K⁺ channels (reviewed in Ralevic & Burnstock, 1998). Overall, the protective effects of adenosine in the brain and cardiovascular system improve the balance between oxygen supply and demand.

Adenosine has been reported to regulate the secretion of glucose-regulatory hormones including glucagon, insulin and corticosteroids from endocrine cells (Wolff & Cook, 1977; Chapal et al. 1985; Shima, 1986; Hillaire-Buys et al. 1989). The modulation of peptide and steroid hormone secretion by adenosine could also contribute to the optimization of energy balance by tissues under metabolically stressful conditions.

The cellular effects of adenosine are all mediated through P₁ purinergic receptors, which are expressed by a wide-range of tissues throughout the body. Four distinct G-protein-coupled P₁ purinergic receptor subtypes have been identified (A₁, A_2A, A_2B, A₃) which differ in their molecular structure, pharmacological profile, signalling pathways and tissue distribution (Williams & Burnstock, 1997; Ralevic & Burnstock, 1998).

Several lines of evidence suggest that adenosine and other purines may act as modulators in the adrenal cortex. Adenosine is released along with ATP, catecholamines and other nucleotides from adrenal medullary chromaffin cells (Nussdorfer, 1996). Clusters of chromaffin cells are distributed throughout the three zones of the adrenal cortex, including the adrenal zona fasciculata (AZF) where cortisol is synthesized (Nussdorfer, 1996). ATP and adrenaline (epinephrine) stimulate cortisol secretion from AZF cells (Walker et al. 1988; Hoey et al. 1994).

Bovine AZF cells express a novel K⁺ current (I_AC) that sets the resting potential, while it is inhibited by ACTH, angiotensin II (AII) and external ATP at concentrations that depolarize these cells and stimulate cortisol secretion (Mlinar et al. 1993; Enyeart et al. 1996b; Xu & Enyeart, 1999). To determine whether bovine AZF cells might express adenosine receptors linked to I_AC K⁺ channels, we have studied the effects of adenosine and specific P₁ receptor agonists and antagonists on I_AC in whole-cell patch clamp recordings from bovine AZF cells.

METHODS

Tissue culture media, antibiotics, fibronectin and fetal bovine sera were obtained from Gibco Laboratories. Coverslips were purchased from Bellco Glass, Inc. (Vineland, NJ, USA). Enzymes, MgATP, Na₂ATP, Na₂UTP, 5-adenylyl-imidodiphosphate (AMP-PNP, lithium salt), NaGTP, GDP-β-S and BAPTA were obtained from Sigma. 5-N-Ethylcarboxamidoadenosine (NECA); 8-cyclopentyl-1,2-dipropylxanthine (DPCPX), 3-ethyl-5-benzyl-2-methyl-4-phenylethylnyl-6-phenyl-1,4-(±)-dihydropyridine-3,5-dicarboxylate (MRS 1191) and 2-chloro-N⁶-cyclopentyl-adenosine (CCPA) were obtained from Research Biochemicals International (Natick, MA, USA). 4-[2-[[6-Amino-9-(N-ethyl-β-D-ribofuranuronamidosyl)-9H-purin-2-yl] amino]ethyl]benzenepropanoic acid(CGS 21680), 1-deoxy-1-[6-[[(3-iodo phenyl) methyl] amino] - 9H-purin-9-yl] -N-methyl-β-D-ribofuranuron amide (IB-MECA) and 4-((2-[7-amino-2-(2-furyl)[1,2,4]-triazolo[2,3-α][1,3,5]triazin-5-yl-amino]ethyl)phenol) (ZM 241385) were obtained from Tocris (Ballwin, MO, USA).

Isolation and culture of AZF cells

Bovine adrenal glands were obtained from steers (age range, 1–3 years) within 30 min 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 AZF cells were prepared as previously described (Enyeart et al. 1997). Cells were plated in Dulbecco's modified Eagle's medium/F12 (1:1) containing 10% fetal bovine serum, 100 u ml⁻¹ penicillin, 0·1 mg ml⁻¹ streptomycin, and the antioxidants 1 μM tocopherol, 20 nM selenite and 100 μM ascorbic acid, in 35 mm dishes containing 9 mm² glass coverslips that had been treated with fibronectin (10 μg ml⁻¹) at 37°C for 30 min then rinsed with warm, sterile PBS immediately before adding cells. Dishes were maintained at 37°C in a humidified atmosphere of 95% air and 5% CO₂.

Patch clamp experiments

Patch clamp recordings of K⁺ channel currents were made in the whole-cell configuration. The standard pipette solution was 120 mM KCl, 2 mM MgCl₂, 1 mM CaCl₂, 20 mM Hepes, 11 mM BAPTA, 200 μM GTP and 5 mM MgATP with pH buffered to 7·2 using KOH. Deviations from the standard solution are described 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, pH 7·4 using NaOH. All solutions were filtered through 0·22 μm cellulose acetate filters. Drugs were applied externally by bath perfusion controlled manually by a six-way rotary valve.

AZF cells were used for patch clamp experiments 2–12 h after plating. Typically, cells with diameters of < 15 μm and capacitances of 8–15 pF were selected. Coverslips were transferred from 35 mm culture dishes to the recording chamber (volume, 1·5 ml) which was continuously perfused by gravity at a rate of 3–5 ml min⁻¹. Patch electrodes with resistances of 1·0-2·0 MΩ were fabricated from Corning 0010 glass (Garner Glass Co., Claremont, CA, USA). These routinely yielded access resistances of 1·5–4 MΩ and voltage clamp time constants of less than 100 μs. K⁺ currents were recorded at room temperature (22–25°C) following the procedure of Hamill et al. (1981), using an Axopatch-1D patch clamp amplifier (Axon Instruments).

Pulse generation and data acquisition were done using a personal computer and pCLAMP software with a TL-1 interface (Axon Instruments). Currents were digitized at 5–20 kHz after filtering with an 8-pole Bessel filter (Frequency Devices, Haverhill, MA, USA). Linear leak and capacity currents were subtracted from current records using scaled hyperpolarizing steps of 1/3 to 1/4 amplitude. Data were analysed and plotted using pCLAMP 5.5 and 6.04 (Clampan and Clampfit) and SigmaPlot (version 3.0). The mean amplitude of I_AC in AZF cells was less than 750 pA. A current of this size in combination with a 4 MΩ access resistance produces a voltage error of only 3 mV, which was not corrected.

RESULTS

Bovine AZF cells express two types of K⁺ current, a rapidly inactivating, voltage-gated current and a non-inactivating, weakly voltage-dependent current activated by intracellular ATP (I_AC) (Mlinar & Enyeart, 1993; Mlinar et al. 1993; Enyeart et al. 1997). Although present initially in whole-cell recordings, I_AC grows dramatically over a period of minutes, provided that ATP or other nucleotides are present at millimolar concentrations in the recording pipette (Enyeart et al. 1996b; Enyeart et al. 1997).

The absence of time- or voltage-dependent inactivation of the I_AC K⁺ current allows it to be easily isolated and measured in whole-cell recordings, using either of two voltage clamp protocols. When voltage steps of 300 ms duration are applied from a holding potential of −80 mV to a test potential of +20 mV, I_AC can be measured near the end of a voltage step where the transient A-type current (I_A) has inactivated (Fig. 1A, left traces). Alternatively, I_AC can be selectively activated by an identical voltage step after a 10 s prepulse to −20 mV has fully inactivated the A-type current (Fig. 1A, right traces). Occasionally, small (< 10%) differences in I_AC amplitudes determined by these two methods were observed in this study and were attributed to incomplete inactivation of I_A during the 300 ms voltage step.

Figure 1. Time- and concentration-dependent inhibition of I_AC by adenosine.

Whole-cell K⁺ currents were recorded from bovine AZF cells at 30 s intervals in response to voltage steps to +20 mV applied from a holding potential of −80 mV with or without 10 s prepulses to −20 mV which inactivate A-type K⁺ current. After I_AC reached maximum amplitude, cells were superfused with adenosine at various concentrations. A, K⁺ current records made with patch pipettes containing the standard pipette solution with (right) or without (left) 10 s prepulses to −20 mV. Numbers correspond to currents immediately after initiating whole-cell recording (1), after I_AC had reached maximum amplitude (2), and after inhibition by 10 μM adenosine (3). B, I_AC amplitudes recorded with (○) or without (•) depolarizing prepulses, plotted against time. Adenosine (10 μM) was superfused as indicated. C, inhibition curve. The fraction of I_AC remaining is plotted against adenosine concentration. Data are means ±s.e.m. of the indicated number of determinations and were fitted with an equation of the form: I/I_max= 1/[1 + (B/K_d)^n_H], where B is adenosine concentration, K_d is the equilibrium dissociation constant and n_H is the Hill coefficient.

Selective inhibition of I_AC by adenosine and NECA

Adenosine, when applied externally to AZF cells by bath perfusion, produced a selective concentration-dependent inhibition of I_AC (Fig. 1). In these experiments, I_AC was activated at 30 s intervals. When this current reached a stable maximum amplitude, the cell was superfused with adenosine at concentrations ranging from 0·01 to 100 μM. Adenosine produced significant inhibition of I_AC in 65·4% (17/26) of cells tested. Inhibition by adenosine began after a delay of one to several minutes, while as many as ten additional minutes were required for maximum inhibition (Fig. 1A). Adenosine inhibited I_AC half-maximally at an estimated concentration (IC₅₀) of 71 nM (Fig. 1A). At maximally effective concentrations, adenosine did not eliminate I_AC, but reduced this current by 78·4 ± 4·6% (n = 8). The rapidly inactivating I_A K⁺ current was not altered by adenosine at concentrations up to 10⁻⁴ M (Fig. 1A). Inhibition of I_AC by adenosine could not be reversed even with prolonged washing (Fig. 1A).

NECA, a non-selective adenosine receptor agonist that does not discriminate among adenosine receptor subtypes (Cusack & Hourani, 1981) potently inhibited I_AC K⁺ current. In the experiment illustrated in Fig. 2A and B, I_AC was allowed to increase to a maximum value (trace 2, left and right) before the cell was superfused with 100 nM NECA. After a delay of 2–2·5 min, I_AC inhibition commenced, reaching a maximum of greater than 95% after a 7 min exposure to the drug (Fig. 2A and B). The rapidly inactivating A-type current was not reduced. Overall, NECA significantly inhibited I_AC in 63·6% (28/44) of cells tested. At concentrations of 10 and 100 nM, NECA inhibited I_AC by 45·5 ± 4·0% (n = 9) and 79·3 ± 2·9% (n = 24), respectively (Fig. 2A). By comparison, adenosine at concentrations of 10 and 100 μM inhibited I_AC by 65·8 ± 5·5% (n = 8) and 78·4 ± 4·6% (n = 8), respectively. The inhibition of I_AC by NECA was also not easily reversed. When I_AC was inhibited by 100 nM NECA, washing reduced inhibition from 79·3 ± 2·9 to 75·8 ± 6·6% (n = 14).

Figure 2. Inhibition of I_AC by NECA.

Whole-cell K⁺ currents were recorded from AZF cells as described in the legend to Fig. 1. After I_AC reached maximum amplitude, cells were superfused with NECA at concentrations of 10 or 100 nM. A, K⁺ current records made with patch pipettes containing standard solution with (right) or without (left) 10 s prepulses to −20 mV. Numbers correspond to currents immediately after initiating whole-cell recording (1), after I_AC had reached maximum amplitude (2), and after inhibition by 100 nM NECA (3). B, I_AC amplitudes recorded with (○) or without (•) depolarizing prepulses, plotted against time. C, summary of results from experiments measuring inhibition of I_AC by NECA (10 and 100 nM) and adenosine (10 and 100 μM). Bars indicate the fraction of I_AC remaining after steady-state block by each nucleotide at the indicated concentration. Values are means and s.e.m. for the indicated number of determinations.

Inhibition of I_AC by adenosine receptor subtype-selective agonists

Experiments with adenosine and the non-selective adenosine agonist NECA clearly demonstrated that bovine AZF cells express adenosine receptors that are coupled to the inhibition of I_AC K⁺ channels. This inhibition could occur through activators of any of several adenosine receptor subtypes. Adenosine receptor agonists with subtype selectivity for A₁, A_2A and A₃ receptors have been developed (Ralevic & Burnstock, 1998). Three of these agents were used to identify the adenosine receptors mediating I_AC inhibition.

CCPA is an adenosine derivative that potently (K_d < 1 nM) and selectively activates A₁ receptors (Lohse et al. 1988). CCPA inhibited I_AC at concentrations as low as 1 nM. Figure 3A (left panel) illustrates an experiment in which I_AC grew to a maximum amplitude of nearly 1400 pA (trace 2, top and bottom) before superfusion of the cells with 10 nM CCPA, which inhibited I_AC by a maximum of approximately 82% (trace 3). In contrast to the inhibition produced by adenosine and NECA, inhibition of I_AC by CCPA was significantly reversed by washing with control saline (Fig. 3A, trace 4 and graph).

Figure 3. Inhibition of I_AC by adenosine receptor agonists.

Whole-cell K⁺ currents were recorded from AZF cells as described in the legend to Fig. 1. After I_AC reached a stable amplitude, cells were superfused with the A₁ agonist CCPA, the A_2A agonist CGS 21680, or the A₃ agonist IB-MECA. A, current traces and corresponding plots of I_AC amplitude against time for recordings with (bottom traces and ○) and without (top traces and •) 10 s prepulses to −20 mV for 3 representative cells. Numbers on traces correspond to currents immediately after initiating whole-cell recording (1), after I_AC reached maximum amplitude (2), after inhibition by the indicated adenosine agonist (3), and after wash with control saline (4). B, summary of results from experiments as in A. Bars indicate the fraction of I_AC remaining after steady-state block by each agent at the indicated concentration. Values are means and s.e.m. for the indicated number of determinations.

Overall, CCPA significantly inhibited I_AC in 57·9% (11/19) of cells tested. At concentrations of 1 and 10 nM, CCPA inhibited I_AC by 38·1 ± 4·9% (n = 3) and 64·8 ± 8·4% (n = 8), respectively (Fig. 3A). Increasing the CCPA concentration to 100 nM produced no additional inhibition of I_AC. In four cells, washing with control saline reversed block by CCPA from 79·6 ± 8·4 to 51·6 ± 6·5%.

ACTH and its primary intracellular messenger cAMP both inhibit I_AC in AZF cells (Mlinar et al. 1993; Enyeart et al. 1996b). A_2A adenosine receptor activation is linked through G_S to adenylate cyclase activation and the synthesis of cAMP (Libert et al. 1989; Palmer & Stiles, 1995). We used the A_2A-selective agonist CGS 21680 to determine whether adenosine-mediated inhibition of I_AC was produced in part through activation of A_2A receptors present on AZF cells. CGS 21680 activates A_2A receptors in various cells with an EC₅₀ of < 20 nM (Hutchinson et al. 1990). CGS 21680 significantly inhibited I_AC at concentrations ranging from 10 to 200 nM in 72% (13/18) of cells exposed to this drug (Fig. 3). At a concentration of 100 nM, CGS 21680 inhibited I_AC by 75·9 ± 3·1% (n = 8). By comparison, adenosine at a maximally effective concentration of 100 μM inhibited I_AC by 78·4 ± 4·6% (n = 8). At a still higher concentration of 200 nM, CGS 21680 inhibited I_AC by 85·7 ± 4·2% (n = 3). However, because CGS 21680 activates A₃ receptors with an EC₅₀ of approximately 500 nM (Feoktistov & Biaggioni, 1997), inhibition of I_AC at concentrations of 100 nM and above may reflect interaction with A₃, as well as A_2A receptors.

As with adenosine and other agonists, inhibition of I_AC by CGS 21680 was selective for I_AC; the rapidly inactivating I_A current was not reduced (Fig. 3A, top traces). Compared to CCPA, inhibition by CGS 21680 was poorly reversible. In a total of five cells, washing with control saline reduced block from 76·4 ± 4·6% to only 69·6 ± 5·7%.

IB-MECA is a potent and relatively selective agonist of A₃ adenosine receptors. The drug activates A₃ adenosine receptors with an EC₅₀ of approximately 1 nM, while at least 50-fold higher concentrations are required for activation of A_2A or A₁ receptors (Gallo-Rodriguez et al. 1994). IB-MECA inhibited I_AC in 66·7% (10/15) of cells tested. At a concentration of 10 nM, where it interacts almost exclusively with A₃ receptors, IB-MECA inhibited I_AC by 69·3 ± 6·9% (n = 8) (Fig. 3A and B). Inhibition of I_AC by this drug was partially reversible, and was reduced from 80·2 ± 5·8 to 57·7 ± 11·6% (n = 4) upon washing with control saline.

Adenosine receptor antagonists

Experiments with adenosine and adenosine receptor subtype-selective agonists indicated that I_AC inhibition was mediated through multiple P₁ receptors. Specific P₁ receptor subtype antagonists were used to further establish the identity of the adenosine receptors that mediate inhibition of I_AC.

DPCPX is a selective A₁ adenosine receptor antagonist with an IC₅₀ near 1 nM and 700-fold selectivity for A₁ over A₂ receptors (Bruns et al. 1987; Lohse et al. 1988). At a concentration of 100 nM, DPCPX had no effect on I_AC, but completely prevented inhibition of this K⁺ current by CCPA (10 nM). In the experiment illustrated in Fig. 4A and B, I_AC grew from its control amplitude (trace 1) to a maximum of over 800 pA (trace 2) within 15 min. Superfusion of 100 nM DPCPX alone had no effect on the amplitude of I_AC (trace 3), but completely prevented inhibition of I_AC by the A₁ agonist CCPA (trace 4).

Figure 4. The A₁ antagonist DPCPX suppresses I_AC inhibition by CCPA.

Whole-cell K⁺ currents were recorded from AZF cells as described in the legend to Fig. 1. After I_AC reached a stable amplitude, cells were superfused with DPCPX (100 nM) followed by CCPA (10 nM). A, K⁺ current records with (right) and without (left) 10 s prepulses to −20 mV. Numbers correspond to currents immediately after initiating whole-cell recording (1), after I_AC reached maximum amplitude (2), after superfusion of DPCPX (3), and after superfusion of CCPA in the presence of DPCPX (4). B, I_AC amplitudes recorded with (○) or without (•) depolarizing prepulses plotted against time with numbers corresponding to traces in A. C, summary of results of experiments illustrated in A and B. Bars indicate the fraction of I_AC remaining at steady state. Values are means and s.e.m. for the indicated number of separate determinations.

Overall, DPCPX by itself failed to alter I_AC amplitude but completely prevented the inhibition of I_AC by CCPA (10 nM) in each of four cells. By comparison, CCPA (10 nM) alone inhibited I_AC by 64·8 ± 8·4% (n = 9) (Fig. 4A). These experiments using DPCPX in combination with CCPA confirm that bovine AZF cells express A₁ receptors that are coupled to inhibition of I_AC.

MRS 1191 is an adenosine receptor antagonist that blocks A₃ receptors at concentrations below 100 nM but inhibits A₁ and A_2A receptors only at concentrations above 1 μM (Jacobson et al. 1997). At a concentration of 500 nM, MRS 1191 had no effect on I_AC, but completely eliminated inhibition of I_AC K⁺ current by 10 nM IB-MECA (Fig. 5). In the experiment illustrated in Fig. 5A and B, I_AC grew from its control amplitude of less than 100 pA (trace 1) to approximately 750 pA (trace 2) during 12 min of whole-cell recording. Superfusion of 500 nM MRS 1191 produced no inhibition of I_AC (trace 3), but completely prevented inhibition of I_AC by the A₃ agonist IB-MECA (trace 4). In contrast, the non-selective adenosine receptor antagonist NECA (100 nM) inhibited I_AC almost completely in the continued presence of MRS 1191 (trace 5).

Figure 5. The adenosine A₃ receptor antagonist MRS 1191 suppresses I_AC inhibition by IB-MECA.

Whole-cell K⁺ currents were recorded from AZF cells as described in the legend to Fig. 1. After I_AC reached maximum amplitude, cells were superfused with MRS 1191 (500 nM), followed sequentially by IB-MECA (10 nM) and NECA (100 nM). A, K⁺ current records with (right) and without (left) 10 s prepulses to −20 mV. Numbers correspond to currents immediately after initiating recording (1), after I_AC had reached maximum amplitude (2), after superfusion of MRS 1191 (3), after superfusion of IB-MECA and MRS 1191 (4), and after superfusion of NECA and MRS 1191 (5). B, I_AC amplitudes recorded with (○) or without (•) depolarizing prepulses plotted against time with numbers corresponding to traces in A. C, summary of results of experiments as illustrated in A and B. Values are means and s.e.m. of the indicated number of separate determinations.

Overall, MRS 1191 (500 nM) by itself produced no inhibition of I_AC in any of five cells. However, this drug reduced inhibition of I_AC by IB-MECA (10 nM) from 69·3 ± 6·9% (n = 8) to 1·1 ± 1·0% (n = 4). In contrast, MRS 1191 did not alter inhibition of I_AC by the non-selective adenosine agonist NECA. In the presence of MRS 1191, NECA (100 nM) inhibited I_AC by 82·6 ± 1·4% (n = 5), compared to the control value of 79·3 ± 2·9% (n = 24) (Fig. 5A).

To insure that I_AC inhibition by IB-MECA was mediated exclusively through activation of A₃ receptors, we studied I_AC inhibition by IB-MECA in the presence of the A₁ antagonist DPCPX and the A_2A antagonist ZM 241385 in combination. ZM 241385 is a potent A_2A antagonist that inhibits these receptors with an IC₅₀ of 1 nM and does not interact with A₃ receptors (Poucher et al. 1995). DPCPX (100 nM) and ZM 241385 (100 nM) failed to alter inhibition of I_AC by IB-MECA (10 nM).

In the experiment illustrated in Fig. 6, a cell was superfused with saline containing DPCPX and ZM 241385 before being exposed to saline containing these two antagonists plus IB-MECA (10 nM). Under these conditions, IB-MECA reduced the maximum current (trace 2) by 72·8% (trace 4). Overall, in three similar experiments, IB-MECA inhibited I_AC by 71·2 ± 2·7% (n = 3) in the presence of the A₁ and A_2A antagonists compared to 69·3 ± 6·9% (n = 8) under control conditions.

Figure 6. Effect of DPCPX and ZM 241385 on I_AC inhibition by IB-MECA.

Whole-cell K⁺ currents were recorded from AZF cells as described in the legend to Fig. 1. After I_AC had reached maximum amplitude, cells were superfused with saline containing DPCPX (100 nM) and ZM 241385 (100 nM). After 3 min, cells were superfused with saline containing the two antagonists and IB-MECA (10 nM). A, K⁺ current records with (right) and without (left) 10 s prepulses to −20 mV. Numbers correspond to current recorded immediately after initiating whole-cell recording (1), after I_AC had reached maximum amplitude (2), after superfusion of DPCPX and ZM 241385 (3), and after superfusion of the two antagonists and IB-MECA (4). B, I_AC amplitudes recorded with (○) and without (•) depolarizing prepulses plotted against time with numbers corresponding to traces in A.

Results of experiments with the A_2A-selective agonist CGS 21680 indicated that AZF cells express A_2A receptors that are coupled to I_AC inhibition. Although CGS 21680 displays 40-fold selectivity for A_2A over A₃ receptors with reported IC₅₀ values of 15 and 580 nM, respectively, the inhibition of I_AC that was observed at maximally effective concentrations of CGS 21680 (> 100 nM) could have been mediated in part through A₃ receptors. We used the A_2A-selective antagonist ZM 241385 to investigate this possibility.

At a concentration of 500 nM, ZM 241385 significantly blunted the inhibition of I_AC by CGS 21680 (100 nM). In the experiment illustrated in Fig. 7A (left panel), CGS 21680 inhibited I_AC by only 45% (trace 3) in the presence of ZM 241385. Overall, ZM 241385 reduced I_AC inhibition by CGS 21680 from 75·9 ± 3·1% (n = 8) to 40·7 ± 7·7% (n = 3) (Fig. 7A).

Figure 7. CGS 21680 inhibits I_AC through activation of A_2A and A₃ receptors.

Whole-cell K⁺ currents were recorded from AZF cells as described in the legend to Fig. 1. After I_AC reached maximum amplitude, cells were superfused with saline containing ZM 241385 (500 nM) or ZM 241385 and MRS 1191 (500 nM), followed by CGS 21680 (100 nM) and NECA (100 nM). A, current traces and corresponding plots of I_AC amplitude against time for recordings with (bottom traces and ○) and without (top traces and •) 10 s prepulses to −20 mV for 2 representative cells pretreated with ZM 241385 (left) or ZM 241385 + MRS 1191 (right). Numbers on traces correspond to currents immediately after initiating whole-cell recording (1), after I_AC reached maximum amplitude (2), and after inhibition by CGS 21680 (3) or NECA (4). B, summary of results from experiments as in A. Bars indicate the fraction of I_AC remaining after steady-state block by CGS 21680 or NECA. Values are means and s.e.m. for the indicated number of determinations.

The blunting of CGS 21680-mediated inhibition of I_AC by ZM 241385 indicates that A_2A receptors are present on AZF cells and are coupled to I_AC inhibition. In addition, these results suggested that the response to CGS 21680 also involved the activation of A₃ receptors. Accordingly, MRS 1191 in combination with ZM 241385 was much more effective in preventing inhibition of I_AC by CGS 21680 (Fig. 7A, right panel, and B). In a total of six cells exposed to both MRS 1191 (500 nM) and ZM 241385 (500 nM), CGS 21680 reduced I_AC by only 15·8 ± 4·1% (n = 6). In contrast, NECA (100 nM), which activates A₁ as well as A₂ and A₃ receptors, inhibited I_AC by 63·6 ± 5·8% (n = 6) in the combined presence of the A_2A- and A₃-selective antagonists.

Experiments using adenosine receptor agonists and antagonists indicate that adenosine inhibits I_AC K⁺ channels through the activation of multiple adenosine receptor subtypes. Accordingly, we found that the combination of A₁, A_2A and A₃ antagonists DPCPX, ZM 241385 and MRS 1191 completely eliminated inhibition of I_AC produced by adenosine.

In the experiment illustrated in Fig. 8A and B, I_AC grew to a stable value (trace 2) before superfusion with saline containing DPCPX (100 nM), ZM 241385 (500 nM) and MRS 1191 (500 nM). These three antagonists had no effect on I_AC (trace 3), but blocked the inhibition of I_AC by adenosine (10 μM) (trace 4). By comparison, inhibition of I_AC by Na₂ATP (10 μM) was unaffected (trace 5). ATP-mediated inhibition of I_AC occurs through activation of a separate nucleotide receptor with a P2Y₃ agonist profile (Xu & Enyeart, 1999).

Figure 8. Effect of combined P₁ antagonists on I_AC inhibition by adenosine and ATP.

Whole-cell K⁺ currents were recorded from AZF cells as described in the legend to Fig. 1. After I_AC reached maximum amplitude, cells were superfused with saline containing a mixture of the 3 adenosine antagonists - DPCPX (100 nM), ZM 241385 (500 nM) and MRS 1191 (500 nM). After 3 min, cells were superfused with saline containing the 3 antagonists and adenosine (10 μM), followed by ATP (10 μM). A, K⁺ current records with (right) and without (left) 10 s prepulses to −20 mV. Numbers correspond to current recorded immediately after initiating whole-cell recording (1), after I_AC had reached a maximum amplitude (2), after superfusion of adenosine antagonists (3), after superfusion of adenosine (4), and after superfusion of Na₂ATP (5). B, I_AC amplitudes recorded with (○) and without (•) depolarizing prepulses plotted against time with numbers corresponding to traces in A. C, bars indicate the fraction of I_AC remaining after steady state was reached for each combination of drugs. Values are means and s.e.m. for the indicated number of determinations.

Overall, in control saline, 10 μM adenosine inhibited I_AC by 65·8 ± 5·5% (n = 8) compared to 2·8 ± 2·8% (n = 3) in the presence of the three antagonists. By comparison, in saline containing the three antagonists, 10 μM ATP inhibited I_AC by 70·5 ± 5·7% (n = 3) compared to 71·3 ± 3·2% (n = 26) in control saline (Fig. 8A) (Xu & Enyeart, 1999). These results show that adenosine and ATP inhibit I_AC by activation of separate P₁ adenosine and P₂ nucleotide receptors.

Adenosine inhibition of I_AC is G-protein dependent and requires ATP hydrolysis

The studies with P₁ subtype-specific agonists and antagonists described above demonstrated that adenosine inhibits I_AC through activation of A₁, A_2A and A₃ receptors. Each of these receptors is coupled to effector molecules through GTP-binding proteins, including G_S and G_i (Ralevic & Burnstock, 1998). To determine whether adenosine-mediated inhibition of I_AC required activation of a G-protein, GTP in the pipette solution was replaced with the inactive guanine nucleotide GDP-β-S. With 1 mM GDP-β-S in the patch pipette, adenosine (10 μM) and the non-selective adenosine agonist NECA (100 nM) each failed to significantly inhibit I_AC (Fig. 9A, left panel, and B). By comparison, under control conditions (200 μM GTP in the pipette) adenosine and NECA inhibited I_AC by 65·8 ± 5·5% (n = 8) and 79·3 ± 2·9% (n = 24), respectively (Fig. 9A).

Figure 9. Inhibition of I_AC by adenosine and NECA is suppressed by GDP-β-S, AMP-PNP and UTP.

K⁺ currents were recorded in AZF cells using the two voltage protocols described in the legend to Fig. 1. Patch pipettes contained standard solution supplemented with 1 mM GDP-β-S instead of GTP, or solution in which 1 mM AMP-PNP or 2 mM UTP was substituted for ATP as indicated. After I_AC reached a stable maximum amplitude, cells were superfused with adenosine or adenosine followed by NECA. A, current traces and associated time-dependent plots of I_AC amplitude recorded with pipettes containing GDP-β-S (left), AMP-PNP (middle), or UTP (right). Current traces were recorded with (bottom) and without (top) 10 s prepulses to −20 mV immediately after initiating whole-cell recording (1), after I_AC reached a stable maximum amplitude (2), after superfusion of adenosine (10 μM) (3), and after superfusion of NECA (100 nM) (4). I_AC amplitudes recorded with (○) or without (•) depolarizing prepulses are plotted for corresponding cells. B, summary of results for experiments as in A. Bars indicate the fraction of I_AC remaining after steady-state block by adenosine (10 μM; ▪) or NECA (100 nM; ) in experiments with pipettes containing 5 mM ATP + 200 μM GTP (control), 5 mM ATP + 1 mM GDP-β-S, 1 mM AMP-PNP + 200 μM GTP, or 2 mM UTP + 200 μM GTP, as indicated. Values are means and s.e.m. for the indicated number of determinations.

Inhibition of I_AC by the peptide hormones ACTH and AII, and by external ATP in each case requires the presence of hydrolysable ATP in the pipette solution (Mlinar et al. 1995; Enyeart et al. 1997; Xu & Enyeart, 1999). Replacing ATP in the pipette with AMP-PNP or UTP nearly eliminated the inhibition of I_AC by adenosine and NECA (Fig. 9A, middle and right panels, and B). With 1 mM AMP-PNP in the recording pipette, adenosine (10 μM) and NECA (100 μM) inhibited I_AC by only 0·9 ± 0·9% (n = 6) and 11·3 ± 3·7% (n = 5), respectively. Similarly, with 2 mM UTP in the pipette, adenosine (10 μM) and NECA (100 nM) inhibited I_AC by only 1·1 ± 0·9% (n = 4) and 9·2 ± 4·0% (n = 4), respectively. Thus, inhibition of I_AC by activation of A₁, A_2A and A₃ receptors requires both G-proteins and hydrolysable ATP.

Depolarization of AZF cells by adenosine

I_AC K⁺ channels display little voltage dependence and remain open at negative membrane potentials, characteristics of a channel that would set the resting membrane potential (Mlinar et al. 1993; Enyeart et al. 1996b; Enyeart et al. 1997). Agents that selectively inhibit I_AC uniformly depolarize AZF cells (Mlinar et al. 1993, 1995; Enyeart et al. 1996b). Adenosine (10 μM) also significantly depolarized AZF cells.

In the experiment illustrated in Fig. 10, I_AC was recorded at 30 s intervals until it reached a stable maximum amplitude (Fig. 10A, trace 2, left and right panels). The membrane potential was then recorded upon switching to current clamp, and the cell was superfused with 10 μM adenosine as indicated (Fig. 10A). After a delay of approximately 3 min, adenosine produced a continuous depolarization of the membrane from its resting potential of −60 mV, until it reached a stable value at a membrane potential of −27 mV. Upon switching back to voltage clamp, it was observed that I_AC had been inhibited by > 90% (Fig. 10A). Overall, adenosine (10 μM) depolarized AZF cells by 46·2 ± 5·8 mV (n = 6).

Figure 10. Effect of adenosine on I_AC K⁺ current and membrane potential.

Whole-cell K⁺ currents were recorded as described in the legend to Fig. 1. When I_AC reached a maximum value, membrane potential was recorded after switching to current clamp, and cells were superfused with adenosine (10 μM). A, K⁺ currents recorded with (right traces) or without (left traces) depolarizing prepulses immediately after initiating recording (1), after I_AC reached maximum amplitude (2), and after inhibition by adenosine (3). B, recording of membrane potential initiated after I_AC reached maximum amplitude. Adenosine was perfused at the indicated time.

DISCUSSION

In this study, we discovered that bovine AZF cells express A₁, A_2A and A₃ adenosine receptors, each of which is coupled to the inhibition of I_AC K⁺ channels through a G-protein-dependent mechanism requiring ATP hydrolysis. Inhibition of I_AC by adenosine was associated with membrane depolarization.

Multiple adenosine receptor subtypes coupled to I_AC inhibition

In previous studies, adenosine was shown to modulate the synthesis of cAMP and cortisol by adrenal cortical cells, indicating the presence of specific receptors for this purine on the cell membrane (Wolff & Cook, 1977; Cooper & Gleed, 1978; Shima, 1986). The adenosine receptors mediating these actions were not identified. In the present report, we provide convincing evidence that bovine adrenocortical cells express at least three different adenosine receptor subtypes.

Based on the relative potency of P₁ receptor agonists and antagonists, the adenosine receptors of AZF cells are similar to A₁, A_2A and A₃ receptors expressed by many other cells. Specifically, the non-selective synthetic agonist NECA and the specific A₁- and A₃-selective agonists CCPA and IB-MECA effectively inhibited I_AC at concentrations of 10 nM, a result consistent with their reported potency in other cells (Ralevic & Burnstock, 1998). At this low concentration, CCPA and IB-MECA are quite specific and do not interact with other P₁ receptor subtypes (Lohse et al. 1988; Gallo-Rodriguez et al. 1994). Accordingly, the A₁-selective antagonist DPCPX effectively eliminated I_AC inhibition by CCPA while the A₃-selective antagonist MRS 1191 blocked inhibition of I_AC by IB-MECA.

The effective inhibition of I_AC by the A_2A agonist CGS 21680 at concentrations from 10 to 200 nM is consistent with the reported EC₅₀ of 15 nM and indicated that AZF cells also express A_2A receptors (Hutchinson et al. 1990). The partial inhibition of CGS 21680-mediated inhibition of I_AC by the potent and selective A_2A antagonist ZM 241385 confirmed this conclusion.

The results of these experiments are consistent with the hypothesis that individual bovine AZF cells each express multiple adenosine receptor subtypes. Several lines of evidence support this conclusion. First, adenosine and the non-selective agonist NECA each inhibited I_AC in the presence of one or even two selective adenosine receptor subtype antagonists. However, inhibition by adenosine was eliminated in cells that had been pre-treated with a mixture containing A₁, A_2A and A₃ receptor antagonists.

Second, I_AC was significantly inhibited by the non-selective agonists adenosine and NECA in 65·4 and 63·6% of cells tested. By comparison, adenosine receptor subtype-selective agonists CCPA, CGS 21680 and IB-MECA inhibited I_AC in 57·9, 72·2 and 66·7% of cells tested, respectively. Since non-selective agonists that activate A₁, A_2A and A₃ receptors, and agonists that selectively activate only one of these three P₁ receptors each inhibit I_AC in approximately 2/3 of AZF cells, it appears likely that a cell responsive to any of the three agonists expresses all three of the adenosine receptor subtypes. If this were not the case and each of the P₁ subtypes was randomly expressed by 2/3 of AZF cells, then only (1/3)³ or 3·7% of cells would be expected to be insensitive to non-selective agonists such as NECA or adenosine.

At present, we cannot explain why approximately 2/3 of cells tested express adenosine receptors that are functionally coupled to I_AC inhibition. This might be due to the existence of two different populations of AZF cells, or to downregulation or desensitization of adenosine receptors during the preparation of AZF cells. Downregulation and desensitization have been reported for each of the adenosine receptor subtypes (Ralevic & Burnstock, 1998).

Although our results establish that I_AC inhibition by adenosine is mediated by at least three P₁ receptor subtypes, they do not exclude the possibility that AZF cells might also express A_2B receptors coupled to I_AC inhibition. The lack of specific A_2B receptor agonists or antagonists prevented us from exploring this possibility.

Regardless, it is extremely unlikely that A_2B receptor activation contributed to any of the inhibitory responses that we observed in this study. NECA is among the most potent agonists for A_2B receptors with a K_d of approximately 2 μM (Brackett & Daly, 1994). NECA produced maximum effects in our study at a concentration of 100 nM, where only a small fraction of A_2B receptors would be occupied. The other adenosine agonists used in this study (CCPA, CGS 21680, IB-MECA) bind to A_2B receptors with K_d values 100- to 1000-fold higher than that reported for NECA (Feoktistov & Biaggioni, 1997). Each of these three agents inhibited I_AC at concentrations where an insignificant fraction of A_2B receptors would be occupied. Further, of the adenosine receptor subtype-selective antagonists used in this study, only DPCPX shows any affinity for A_2B receptors, and even this agent is 20-fold selective for A₁ over A_2B receptors (Feoktistov & Biaggioni, 1997).

P₁ and P₂ purinergic receptors

Although the combination of A₁, A_2A and A₃ antagonists completely eliminated adenosine-mediated inhibition of I_AC, these agents did not alter inhibition of I_AC by external ATP. These results conclusively demonstrate that adenosine and ATP inhibit I_AC by separate P₁ and P₂ nucleotide receptors present on AZF cells. Apparently, none of the inhibition of I_AC by ATP occurs through the interaction of this nucleotide with adenosine receptors or by metabolism of this nucleotide to adenosine, since ATP-mediated inhibition of I_AC was not diminished by adenosine receptor antagonists. Further, the P2Y₃ nucleotide receptors that mediate ATP inhibition of I_AC (Xu & Enyeart, 1999) are not activated by adenosine since the effects of adenosine were completely blocked by the combined presence of the three selective P₁ antagonists.

We conclude that bovine AZF cells express a minimum of four nucleotide receptors, including three P₁ adenosine receptors and a previously described P₂ receptor. These receptors are coupled to inhibition of I_AC K⁺ channels and membrane depolarization (Xu & Enyeart, 1999).

Signalling mechanisms coupling P₁ receptors to I_AC channels

The activation of A₁, A_2A or A₃ receptors on AZF cells is sufficient to inhibit I_AC. Since adenosine and NECA each activate all three receptors, the elimination of I_AC inhibition by these two agonists upon substituting GDP-β-S for GTP in the pipette indicates that each of the receptors is coupled to I_AC channels through G-proteins. Similarly, the elimination of I_AC inhibition by adenosine and NECA upon substituting AMP-PNP or UTP for ATP in the pipette solution indicates that inhibition of I_AC through each receptor occurs through a mechanism requiring ATP hydrolysis.

With respect to G-protein dependence and the requirement for hydrolysable forms of ATP, inhibition of I_AC by adenosine resembles that by ACTH, AII and external ATP (Mlinar et al. 1993, 1995; Enyeart et al. 1996b; Xu & Enyeart, 1999). These results suggest that inhibition through all six of these receptors may be mediated through common signalling pathways. ACTH and A_2A adenosine receptors both activate adenylate cyclase through G_Sα. cAMP inhibits I_AC through a novel protein kinase A-independent pathway requiring ATP hydrolysis (Enyeart et al. 1996b). Thus, cAMP probably mediates inhibition of I_AC in response to activation of ACTH or adenosine A_2A receptors. However, ACTH inhibits I_AC almost completely in nearly 100% of cells tested (Mlinar et al. 1993; Enyeart et al. 1996b), and is therefore a much more effective agonist than adenosine. A possible explanation for this disparity in effectiveness is that, in addition to activating A_2A receptors, adenosine also activates A₁ and A₃ receptors on AZF cells, which are coupled to the inhibition of adenylate cyclase. The subsequent reduction of cAMP synthesis may lead to the diminished inhibition of I_AC.

In this regard, the maximum inhibition of I_AC by NECA and adenosine was only slightly greater than that provided by CCPA, CGS 21680 or IB-MECA. Thus, activation of multiple adenosine receptors by non-selective agonists does not yield additive effects at the level of I_AC inhibition. Inhibition of A_2A-coupled adenylate cyclase through A₁ and A₃ receptor activation may explain this lack of additivity.

In addition to being coupled to the inhibition of adenylate cyclase, A₁ and A₃ receptors activate phospholipase C (PLC), leading to the synthesis of diacylglycerol and IP₃ and the release of intracellular Ca²⁺ (Ralevic & Burnstock, 1998). AII acting through a losartan-sensitive A₁ receptor, and external ATP acting through a P2Y₃-like receptor also activate PLC in AZF cells (Xu & Enyeart, 1999). However, the inhibition of I_AC by AII and ATP occurs by a mechanism independent of PLC-generated second messengers (Mlinar et al. 1995; Xu & Enyeart, 1999). Inhibition of I_AC by adenosine through A₁ and A₃ receptors may also occur through an alternative pathway.

Adenosine modulates the activity of a number of K⁺ channels in heart, smooth muscle and neurons through activation of A₁ or A_2A receptors. In nearly every case, stimulation of the adenosine receptor leads to channel activation (Dart & Standen, 1993; Belardinelli et al. 1995; Pan et al. 1995). The signalling pathways that link P₁ receptors to these channels are varied and have not been fully described. Irrespective of the signalling pathways involved, I_AC K⁺ channels are distinctive in their inhibition through three separate adenosine receptors.

Adenosine and AZF cell physiology

The role of adenosine in the regulation of AZF cell function has not been determined. In many cells, including neurons, cardiac myocytes and smooth muscle cells, adenosine acts as a protective agent during periods of metabolic stress, reducing the metabolic rate through activation of K⁺ channels and stabilization of the membrane potential near the K⁺ equilibrium potential.

In the endocrine system, adenosine may regulate the secretion of glucoregulatory hormones, thereby conserving glucose for hypoxia-sensitive tissues. Specifically, adenosine acts through an A₁ receptor to inhibit insulin secretion, thereby reducing glucose utilization (Hillaire-Buys et al. 1989). Adenosine stimulates the secretion of the glucose-conserving hormone glucagon through an A_2A receptor (Chapal et al. 1985).

Adenosine has been reported to stimulate adenylate cyclase and steroidogenesis in an adrenal cell line and to potentiate ACTH-stimulated cortisol production by normal rat adrenocortical cells (Wolff & Cook, 1977; Cooper & Gleed, 1978). ATP and other purines are released along with catecholamines from secretory granules of medullary chromaffin cells (Cena & Rojas, 1990). Rays of adrenal medullary tissue traverse the adrenal cortex and clusters of chromaffin cells are scattered through all three regions of the cortex (Nussdorfer, 1996). AZF cells express receptors for ATP and catecholamines which when activated stimulate cortisol secretion (Walker et al. 1988; Hoey et al. 1994). ATP-stimulated cortisol secretion is associated with I_AC inhibition and membrane depolarization (Xu & Enyeart, 1999). Overall, these results suggest that ATP and other nucleotides, along with catecholamines, could act in a paracrine fashion to modulate I_AC current and cortisol secretion, synchronizing the stress-induced release of this glucose-conserving hormone.

In addition to a possible role in the regulation of cortisol secretion, adenosine may act to regulate the growth or differentiation of AZF cells. AII and ACTH, which together activate many of the same signalling pathways as adenosine in these cells, have long-term effects on the growth and differentiated state of AZF cells (Gill et al. 1977, 1982; Arola et al. 1993; Barbara & Takeda, 1995). Accordingly, both of these peptides induce the expression of steroid orphan receptor transcription factors (Enyeart et al. 1996a). It will be interesting to determine the effects of adenosine on AZF cell physiology, including growth, differentiation, gene expression and cortisol secretion.

In conclusion, we have determined that bovine AZF cells express three separate adenosine receptor subtypes, each of which is coupled to the inhibition of I_AC K⁺ channels through a G-protein-dependent mechanism. Through activation of these receptors, adenosine may control membrane potential and related cell functions in AZF cells.