Nickel inhibits β-1 adrenoceptor mediated activation of cardiac CFTR chloride channels.

Cardiac ventricular myocytes exhibit a protein kinase A-dependent Cl(-) current (ICl.PKA) mediated by the cystic fibrosis transmembrane conductance regulator (CFTR). There is conflicting evidence regarding the ability of the divalent cation nickel (Ni(2+)), which has been used widely in vitro in the study of other cardiac ionic conductances, to inhibit ICl.PKA. Here the action of Ni(2+) on ICl.PKA activated by β-adrenergic stimulation has been elucidated. Whole-cell patch-clamp recordings were made from rabbit isolated ventricular myocytes. Externally applied Ni(2+) blocked ICl.PKA activated by 1 μM isoprenaline with a log IC50 (M) of -4.107 ± 0.075 (IC50=78.1 μM) at +100 mV and -4.322 ± 0.107 (IC50=47.6 μM) at -100 mV. Thus, the block of ICl.PKA by Ni(2+) was not strongly voltage dependent. Ni(2+) applied internally via the patch-pipette was ineffective at inhibiting isoprenaline-activated ICl,PKA, but in the same experiments the current was suppressed by external Ni(2+) application, indicative of an external site of Ni(2+) action. In the presence of 1 μM atenolol isoprenaline was ineffective at activating ICl.PKA, but in the presence of the β2-adrenoceptor inhibitor ICI 118,551 isoprenaline still activated Ni(2+)-sensitive ICl.PKA. Collectively, these data demonstrate that Ni(2+) ions produce marked inhibition of β1-adrenoceptor activated ventricular ICl.PKA at submillimolar [Ni(2+)]: an action that is likely to involve an interaction between Ni(2+) and β1-adrenoceptors. The concentration-dependence for ICl.PKA inhibition seen here indicates the potential for confounding effects on ICl,PKA to occur even at comparatively low Ni(2+) concentrations, when Ni(2+) is used to study other cardiac ionic currents under conditions of β-adrenergic agonism.

Introduction

A number of distinct chloride conductances have been identified that may contribute to the normal and pathological function of cardiac myocytes [1,2]. These include swelling-activated Cl− current [1,2], Ca2+-activated Cl− current [1,2], anionic background current [3,4] and cAMP/PKA-activated Cl− current (ICl.PKA) activated by β-adrenergic agonists [2,5,6]. The channels that carry ICl.PKA are mediated by a cardiac isoform of the cystic fibrosis transmembrane conductance regulator protein (CFTR: [2,7-9]). Sympathetic activation of ICl.PKA may act to counter the effects of β-adrenergic stimulation of L-type calcium current (ICa,L); consequently ICl.PKA may contribute to the rate-dependent shortening of ventricular action potentials [10,11]. However, the direct measurement of ICl.PKA from cardiac cells and of its modulation of action potentials under physiological recording conditions is confounded by a lack of potent and selective pharmacological CFTR inhibitors [2]. Consequently, cardiac ICl.PKA is usually studied under ‘selective’ recording conditions, with other overlapping conductances inhibited.

The in vitro study of β-adrenergic modulation of some other cardiac ionic conductances is facilitated by the availability of selective pharmacological inhibitors [12,13], which in principle allows these to be separated from β-adrenoceptor activation of ICl.PKA. However, this is not necessarily the case for all the ion currents of cardiac myocytes. The electrogenic Na+–Ca2+ exchanger (NCX) is present throughout the heart and plays an important role in Ca2+ ion handling and in shaping cardiac action potentials [14,15]. Similar to ICl.PKA, cardiac NCX current (INCX) is difficult to study under normal physiological conditions due to a lack of NCX-selective pharmacology. Direct measurements of INCX have therefore tended to involve the inhibition of overlapping voltage and time-dependent conductances and INCX measurement as current sensitive to millimolar concentrations of nickel ions (Ni2+) [16-18]. Selective measurement conditions for cardiac INCX exclude overlapping ICl.PKA in the absence of PKA stimulation, but in the presence of such stimulation there is potential for both currents to be activated [15,19,20]. The results from some studies are suggestive that the use of Ni2+ to study INCX under conditions of β-adrenergic agonism may be complicated by an inhibitory effect of Ni2+ on β-adrenoceptor activated ICl.PKA [19,20], although other data appear inconsistent with this possibility [21]. The present study was therefore undertaken to determine, under CFTR-selective recording conditions, the response of β-adrenoceptor activated cardiac ICl.PKA to Ni2+. The results obtained demonstrate a marked, concentration-dependent inhibitory modulation by Ni2+ of β1-adrenoceptor mediated ICl.PKA.

Methods

Right ventricular cardiomyocytes were isolated from the hearts of Langendorff-perfused male New Zealand White rabbits as described previously [22]. All procedures were approved by the Ethics Committee of University of Bristol and conformed to the UK Animals (Scientific Procedures) Act, 1986. Prior to use, myocytes were stored at 4 °C in Kraft–Brühe (KB) solution [22,23].

Electrophysiological recording and data acquisition

Whole-cell patch-clamp recordings were made at 37 °C. The data acquisition and recording methods used here have been reported previously [20,24]. Cells were superfused with normal Tyrode’s solution containing (in mM): 140 NaCl, 5 HEPES, 10 d-glucose, 4 KCl, 1 CaCl2, 1 MgCl2, 1 BaCl2, pH 7.45 with NaOH. CFTR-mediated ICl.PKA was recorded as reported previously [20] using a Ca2+, K+-free external solution containing 1 mM CdCl2 and was activated using 1 μM isoprenaline (ISO) [20,25].

Atenolol (Sigma–Aldrich) was made up as a 10 mM stock solution in distilled deionised (Milli-Q) water and was used at a final concentration of 1 μM in external solutions. ICI 118,551 (Tocris, Bristol, UK) was made up as a 4 mM stock solution in deionised water and was used at a final concentration of 100 nM.

Data analysis and presentation

Data were analyzed using Igor Pro (WaveMetrics, Inc., USA), Clampfit from the pClamp 10.0 software suite (Molecular Devices), Excel 2007 and GraphPad Prism 5 software. Data are presented as mean ± standard error of the mean (SEM), ‘n’ values refer to numbers of cells for recordings (typically ⩾ two hearts). Statistical comparisons were made using a Student’s paired t test and one- or two-way repeated measures (RM) ANOVA. P < 0.05 was considered to be statistically significant.

The fractional block of ICl.PKA by Ni2+ ions was calculated from the fraction of β-adrenoceptor-activated current remaining in the presence of Ni2+ compared to that activated prior to Ni2+ addition, as follows Eq. (1):where the I, I, and I represent currents in the presence of control, isoprenaline and Ni2+ with isoprenaline, respectively.

The half-maximal inhibitory concentration (IC50) of Ni2+ was calculated by plotting the mean ± SEM fractional block of ICl.PKA against the Ni2+ concentration and fitting the data with a logistic equation:where Y, Top and Bottom represent the response, maximal and minimum response to the drug respectively; X represents the logarithm of [Ni2+] (μM).

Results and Discussion

Fig. 1(Ai) shows the ‘saw-tooth’ voltage command protocol used to record CFTR-mediated ICl.PKA[20,25]. It was applied continuously (at a frequency of 0.1 Hz) first in control solution, then in the presence of 1 μM isoprenaline and following subsequent Ni2+ ion application in the maintained presence of isoprenaline. Current between +100 mV and −100 mV was measured during the descending ramp phase of the voltage command. Fig. 1(Aii) shows representative currents during this ramp phase, from an individual experiment, plotted against voltage under the three conditions. Both inward and outward current components were increased markedly by isoprenaline, with an accompanying negative shift in zero-current potential, as anticipated for ICl.PKA activation. In the presence of Ni2+ (3 mM), current was restored toward control values, indicative of substantial inhibition of the isoprenaline-activated current component. Fig. 1(B) shows a time-course plot (same experiment as Fig. 1A) of currents at +100 and −100 mV, illustrating (i) the rapid onset and reversal of Ni2+ effects and (ii) the fact that Ni2+ inhibition of isoprenaline-activated ICl.PKA exhibited concentration-dependence. Fig. 1C shows concentration–response relations for inhibition by Ni2+ of isoprenaline-activated ICl.PKA. The IC50 values derived from the plots in Fig. 1(C) were 78.1 μM at +100 mV (95% confidence intervals of 55.0–111.0 μM) and 47.6 μM at −100 mV (95% confidence intervals of 29.7–76.3 μM); thus ICl,PKA inhibition by Ni2+ was not strongly voltage dependent.

Fig. 1

The effect of extracellular Ni2+ on isoprenaline-activated ICl.PKA. (A) Panel (Ai) shows the voltage-ramp protocol (holding potential = 0 mV, frequency of application 1/10 s) used for recording Cl− currents. Panel (Aii) shows representative currents, plotted against voltage, obtained during the descending phase of the ramp saw-tooth. Letters indicate traces obtained from the time-points indicated in panel (B). (B) Representative time course of an experiment with currents sampled at +100 mV (open circles) and −100 mV (filled circles) during saw-tooth voltage-ramps; the solid bars at the top indicate application of 1 μM isoprenaline (ISO) and Ni2+ at the concentrations indicated. (C) Concentration–response relationship of the effect of Ni2+ on ICl.PKA. Concentration–responses are shown at +100 mV (open circles) and −100 mV (filled circles). The ‘n’ numbers at each respective concentration are shown in parentheses. Solid and dashed lines represent fits to the data with Eq. (2) at −100 mV and +100 mV respectively. The fitted logIC50 (M) at +100 and −100 mV were respectively −4.107 ± 0.075 and −4.322 ± 0.101; the nH values for the fits were 1.145 ± 0.187 at +100 mV and 1.019 ± 0.214 at −100 mV.

The rapid onset and reversal of Ni2+ effects evident in Fig. 1(B) indicates that Ni2+ was able to reach its site of action rapidly on application. One possible explanation for this rapidity of action is that Ni2+ ions may interact directly with an extracellular target to inhibit ICl.PKA. If this is the case, then internally applied Ni2+ ions should be ineffective at inhibiting ICl.PKA. This possibility was investigated by including Ni2+ in intracellular pipette dialysate. The Ni2+ concentration used (300 μM) was selected as it was sufficient to inhibit ICl.PKA by >80% when applied externally (Fig. 1C). The mean current–voltage (I–V) plots in Fig. 2(Ai) and (Bi) show that there was little difference in control currents, those in isoprenaline or those in isoprenaline + externally applied Ni2+ between cells dialysed with Ni2+-free and Ni2+-containing pipette solutions. Fig. 2(Aii) and (Bii) show close similarities between isoprenaline-sensitive currents in cells dialysed with Ni2+-free and Ni2+-containing solutions and also in the remaining isoprenaline-sensitive current in the presence of externally applied Ni2+. Thus, without Ni2+ in the pipette solution, the isoprenaline-sensitive current at +100 mV was 12.46 ± 1.90 pA/pF (n = 6), and decreased to 1.82 ± 0.91 pA/pF with 10 mM external Ni2+. When 300 μM Ni2+ was included in the pipette solution, the isoprenaline-sensitive current at +100 mV was 13.53 ± 2.04 pA/pF (n = 6; P > 0.05, compared with pipette solution without Ni2+), and decreased to 2.63 ± 0.99 pA/pF with 10 mM external Ni2+ (P > 0.05, compared with pipette solution without Ni2+). These data demonstrate an external site of Ni2+ action to inhibit ICl.PKA. A direct effect of external Ni2+ on the CFTR channel mediating ICl.PKA is precluded from previous observations that when ICl.PKA is activated by forskolin (to activate adenylate cyclase downstream of the β-adrenoceptor), externally applied Ni2+ does not inhibit the current [19,20]. Thus, Ni2+ is most likely to exert its inhibitory action by direct interaction with β-adrenoceptors.

Fig. 2

Intracellular Ni2+ does not inhibit ISO-activated ICl.PKA. (Ai) Mean current–voltage relations (mean ± SEM; solid lines shows the mean values and the surrounding dotted lines show SEMs) recorded using pipette solution without intracellular Ni2+. Currents were recorded in control extracellular solution (a), in the presence of 1 μM isoprenaline (ISO) (b) and in the presence of 1 μM ISO plus 10 mM Ni2+ (c) (n = 6). (Aii) Mean ISO-activated ICl.PKA calculated from the data shown in panel (Ai). (Bi) Mean current–voltage relations (mean ± SEM; solid lines show the mean values and the surrounding dotted lines shows SEMs) recorded using pipette solution containing intracellular Ni2+ (300 μM). Currents were recorded in control extracellular solution (a), in the presence of 1 μM ISO (b) and in the presence of 1 μM ISO plus 10 mM Ni2+ (c) (n = 6). (Bii) Mean ISO-activated ICl.PKA calculated from the data shown in panel (Bi).

In order to determine whether β1 or β2 adrenoceptors are the likely targets of Ni2+, isoprenaline was applied in the presence of β1 or β2 adrenoceptor inhibitors. Fig. 3 shows the results of experiments with the β1-adrenoceptor inhibitor atenolol. Fig. 3(A) and (B) show that 1 μM atenolol application produced a rapid inhibition of ICl,PKA that was similar to that produced by 10 mM Ni2+, with current in the presence of each agent close to that in control solution. Fig. 3(C) shows that mean I–V relations for Ni2+-sensitive and atenolol-sensitive currents (i.e. the isoprenaline-activated ICl.PKA inhibited by atenolol and Ni2+) were closely superimposed. These observations implicate β1 adrenoceptors in the activation of Ni2+ sensitive ICl.PKA. In further experiments, isoprenaline was applied subsequent to application of the β2 adrenoceptor antagonist, ICI 118,551. As shown in Fig. 4(A) and (B), ICI 118,551 (at a substantial concentration of 100 nM) did not inhibit the ability of isoprenaline to activate ICl.PKA, nor was the ability of Ni2+ to inhibit isoprenaline-activated current impaired. Fig. 4(C) compares fractional inhibition of isoprenaline-activated ICl.PKA by 10 mM Ni2+ in samples of cells treated with ICI 118,551 to cells not exposed to this agent (‘control’ in Fig. 4C) at both a positive and negative voltage during the descending ramp of the saw-tooth command: there was no significant difference between ICI 118,551 and control. Considered together with the data in Fig. 3, these observations demonstrate that Ni2+-sensitive isoprenaline-activated ICl.PKA was mediated by β1 adrenoceptor activation.

Fig. 3

Involvement of β-1 adrenoceptors in Ni2+-sensitive ICl.PKA. (A) Representative time course of an experiment with currents sampled at +100 mV (open circles) and −100 mV (filled circles) during saw-tooth voltage-ramps; the solid bars at the top indicate application of 1 μM isoprenaline (ISO), 10 mM Ni2+ and 1 μM atenolol. (B) Representative current–voltage relations obtained during the descending phase of the ramp saw-tooth. Letters indicates traces obtained as indicated in panel (A). (C) Mean current–voltage relations for the Ni2+-sensitive (n = 5) and atenolol-sensitive (n = 5) difference currents in the presence of 1 μM ISO. Bars indicate ±SEM. There was no statistically significant difference between the mean Ni2+-sensitive and atenolol-sensitive current–voltage relations.

Fig. 4

β-2 adrenoceptors do not activate Ni2+-sensitive ICl.PKA. (A) Representative time course of an experiment with currents sampled at +100 mV (open circles) and −100 mV (filled circles) during saw-tooth voltage-ramps; the solid bars at the top indicates application of 100 nM ICI 118,551, 1 μM isoprenaline (ISO) and 10 mM Ni2+. (B) Representative current–voltage relations obtained during the descending phase of the ramp saw-tooth. Letters indicate traces obtained as indicated in panel (A). (C) Fractional block by 10 mM Ni2+ of ISO-activated ICl.PKA at +100 mV and −100 mV in the absence (control) and presence of 100 nM ICI 118,551. ICI 118,551 had no statistically significant effect on fractional block at either voltage. Replicate numbers are given in parentheses.

The findings of the present study are inconsistent with a lack of Ni2+ modulation of ICl.PKA[21] and instead support and extend evidence for an inhibitory effect of Ni2+ on CFTR-encoded ICl.PKA, when activated via β-adrenergic stimulation [19,20]. This study reports for the first time the concentration-dependence of this effect and the β-adrenoceptor subtype responsible. Ni2+ has been reported to inhibit ventricular INCX with a KD of ∼290 μM in electrophysiological experiments with a cAMP-free pipette dialysate and of ∼160 μM with raised (100 μM) cAMP [18], values higher than the IC50 values for inhibition of ICl.PKA found in this study. The concentration-dependence of Ni2+ inhibition of ICl.PKA observed here therefore indicates strong overlap between concentration-dependent inhibition of INCX and of β1 adrenoceptor activated ICl.PKA. Thus, an important consideration for the future study of β-adrenoceptor-mediated modulation of INCX is that this would best be undertaken under conditions in which recording solutions are chosen that either preclude CFTR activation, or in which alternative approaches (different NCX inhibitors to Ni2+ or ion replacement) are used to isolate INCX.

Ni2+ inhibits cardiac L-type Ca current (ICa,L) in electrophysiological experiments with a KD of ∼330–530 μM (the higher value with raised cAMP in the pipette dialysate) [26], and produces marked effects at submillimolar concentrations on the magnitude and kinetics of Kv1.5 channel current [27,28] and upon hERG channel activation kinetics [29]. These actions occur in the absence of β-adrenoceptor activation and, in contrast to the effect on β-adrenoceptor activated ICl.PKA shown here, involve direct interactions between Ni2+ ions and the affected channels. However, extracellular divalent ions have also been reported to modify the neuromuscular junction acetylcholine receptor-channel complex [30] and can influence ligand binding to G-protein coupled receptors [31,32]. Zn2+ and Cu2+ ions have been reported to interact with α(1A) adrenoceptors with micromolar affinity and to act as allosteric modulators [33], whilst Mg2+ ions have been reported to decrease agonist affinity for platelet β-adrenoceptors [34]. Thus, the inhibitory effect of Ni2+ on ICl.PKA reported here appears likely to result either from effects of the cation on isoprenaline binding to β1 adrenoceptors, or upon subsequent coupling between the receptor and adenylate cyclase. Future work to determine the underlying mechanism of this effect of extracellular Ni2+ is warranted. Perhaps most significantly, the findings of this study indicate that care is required in the use of Ni2+ in the study of cardiac conductances, particularly under conditions of sympathetic agonism, during which effects on receptor-activation of ionic current may confound interpretation of direct effects of the cation on ion channels or transporters.