β-Adrenoceptor/PKA-stimulation, Na⁺–Ca²⁺ exchange and PKA-activated Cl⁻ currents in rabbit cardiomyocytes: A conundrum

Abstract

Investigations into the functional modulation of the cardiac Na⁺–Ca²⁺ exchanger (NCX) by acute β-adrenoceptor/PKA stimulation have produced conflicting results. Here, we investigated (i) whether or not β-adrenoceptor activation/PKA stimulation activates current in rabbit cardiac myocytes under NCX-‘selective’ conditions and (ii) if so, whether a PKA-activated Cl⁻-current may contribute to the apparent modulation of NCX current (I_NCX). Whole-cell voltage-clamp experiments were conducted at 37 °C on rabbit ventricular and atrial myocytes. The β-adrenoceptor-activated currents both in NCX-‘selective’ and Cl⁻-selective recording conditions were found to be sensitive to 10 mM Ni²⁺. In contrast, the PKA-activated Cl⁻ current was not sensitive to Ni²⁺, when it was activated downstream to the β-adrenoceptors using 10 μM forskolin (an adenylyl cyclase activator). When 10 μM forskolin was applied under NCX-selective recording conditions, the Ni²⁺-sensitive current did not differ between control and forskolin. These findings suggest that in rabbit myocytes: (a) a PKA-activated Cl⁻ current contributes to the Ni²⁺-sensitive current activated via β-adrenoceptor stimulation under recording conditions previously considered selective for I_NCX; (b) downstream activation of PKA does not augment Ni²⁺-sensitive I_NCX, when this is measured under conditions where the Ni²⁺-sensitive PKA-activated Cl⁻ current is not present.

1. Introduction

Na⁺–Ca²⁺ exchanger (NCX) proteins are widely distributed in different tissue types and play important roles in Na⁺ and Ca²⁺ ion homeostasis [1–3]. The stoichiometry of cardiac NCX1 exchange is generally held to be 3 Na⁺ ions for 1 Ca²⁺ ion such that the exchanger can function in both the ‘forward’ (Na⁺-entry, Ca²⁺-efflux) and ‘reverse’ (Na⁺-entry, Ca²⁺-efflux) modes, with the direction of transport determined by the intracellular and extracellular concentrations of the transported ions and transmembrane voltage gradient [1,4,5]. During ventricular action potentials (APs), initial membrane depolarisation may drive reverse-mode NCX activity, providing Ca²⁺ entry which can supplement that occurring via L-type calcium channels to facilitate calcium-induced calcium release from the sarcoplasmic reticulum (SR) [6–8]. Subsequent forward mode exchange acts to extrude Ca²⁺ across the sarcolemmal membrane; consequently, along with SR Ca²⁺ reuptake, NCX1 activity is a major mechanism for restoring low diastolic [Ca²⁺]_i levels [1,7]. Due to its stoichiometry, NCX activity is electrogenic and can influence AP profiles as well as cellular calcium handling [1,9]. Given these important roles, it is perhaps unsurprising that altered NCX activity can contribute to impaired excitation–contraction coupling and arrhythmogenesis in disease states [10]. The properties and regulation of NCX have been studied extensively by the voltage-clamp recording of NCX currents (I_NCX) in isolated cardiac myocytes under selective conditions in which other major cation currents (e.g., Ca²⁺, K⁺, Na⁺, Na⁺/K⁺ ATPase) have been removed or inhibited [11–15]. Extracellular Ni²⁺ is used conventionally to measure I_NCX as it blocks both the inward and the outward I_NCX reliably and reversibly with no preference for either mode of transport [11–13].

Although it has been established that NCX1 can be phosphorylated by protein kinase A (PKA) and protein kinase C (PKC) at sites residing in the large intracellular loop of the NCX protein [16–18], functional regulation of mammalian NCX1 by the β-adrenoceptor/PKA pathway is controversial, with both reports of no alteration [19–24] or upregulation [2,14,15,25–27] of NCX activity in different studies. For example, previous work from this laboratory has provided evidence for an increase in guinea-pig ventricular NCX activity at 37 °C through the β-adrenoceptor/PKA pathway [14,15,25]: the currents recorded during voltage-ramps in NCX recording conditions were stimulated by the β-adrenoceptor agonist, isoprenaline (ISO), by the adenylyl cyclase activator, forskolin, or by raised [cAMP]_i and the rate of decay of caffeine-induced [Ca²⁺]_i transients were accelerated by high [cAMP]_i [15]. There are also reports of stimulation of NCX1 activity measured as ⁴⁵Ca²⁺ uptake for both recombinant NCX1 and native rat ventricular NCX [28,29] and of I_NCX for swine ventricular myocytes [30,31]. On the other hand, it has been suggested that β-adrenoceptor activated changes in ionic current under I_NCX-‘selective’ conditions could result instead from activation of the CFTR (cystic fibrosis transmembrane conductance regulator)-mediated I_Cl,PKA, as this current in guinea-pig myocytes has been reported to be sensitive to extracellular Ni²⁺-application when it is activated by isoprenaline, though not when activated by forskolin [24]. Furthermore, in a study of rabbit ventricular myocytes, Ginsburg and Bers reported no alteration to NCX activity estimated from caffeine-induced Ca²⁺-transient decay or of caffeine induced [Ca²⁺]_i and inward I_NCX under perforated-patch conditions used to assess rate of Ca²⁺ extrusion via NCX [23]. Also, in the same study [23] rabbit ventricular I_NCX elicited by a voltage-ramp protocol was not enhanced by a brief period of isoprenaline application, the authors concluded that under those conditions rabbit ventricular I_NCX was not modulated by isoprenaline [23]. Differences between experimental preparations/species, protocols and conditions may contribute to the differing conclusions regarding PKA-modulation of NCX in the literature [5]. Consequently, the present study was undertaken to determine whether or not I_NCX from rabbit ventricular myocytes can be augmented by the β-adrenoceptor/PKA pathway during sustained application of voltage-ramp protocols using conditions that have been extensively used in prior studies of I_NCX at mammalian body temperature [12–15]. To investigate the contribution of I_Cl,PKA to the activated currents, recordings were also made from rabbit ventricular myocytes under Cl⁻-selective conditions [32] and from rabbit atrial myocytes, which lack PKA-dependent Cl⁻ current [33], under I_NCX recording conditions.

2. Materials and methods

2.1. Cell isolation

Right ventricular and left atrial myocytes were isolated from Langendorff-perfused rabbit hearts by enzymatic and mechanical dispersion, as described previously [34]. All procedures were approved by the Ethics Committee of the University of Bristol and were performed according to the UK Animals (Scientific Procedures) Act, 1986. Male New Zealand White rabbits (2.0–3.0 kg) were killed by cervical dislocation and the heart removed promptly via median thoracotomy. The excised heart was placed in a cold (4 °C) heparinised Tyrode's isolation solution (Table 1) to which 750 μM Ca had been added and the aorta was then cannulated quickly into a Langendorff perfusion system. The heart was perfused retrogradely at 37 °C first with isolation solution containing 750 μM Ca²⁺ for 2 min, followed by 4 min perfusion with the isolation solution containing 100 μM EGTA. Perfusate was then switched to an isolation solution containing a low concentration of Ca²⁺ (240 μM) supplemented with collagenase (Worthington type 1, 1 mg/ml) and protease (Sigma type XIV, 0.1 mg/ml) for 6 min. The heart was then cut off from the cannula and the right ventricular and left atrial free wall were excised into two separate sterile Petri dishes. Using a pair of fine scissors, these excised tissues were chopped into smaller pieces. The chopped tissue was then shaken in a water-bath with re-circulation enzyme and 2% bovine serum albumin (BSA) at 37 °C for 5–10 min. It was then filtered through nylon gauze (200 μm diameter mesh) into a solution containing 3% BSA and 1.5 ml of Kraft-Brühe (KB) medium [35]. The filtrate was centrifuged for 40 s and the supernatant was discarded. The pellet was re-suspended in fresh KB medium and stored at 4 °C. Cells were used within 8 h after isolation and only cells with clear rod shape and striations were used for recordings.

Table 1.

Compositions of solutions (mM).

Constituents	Isolation solution	External solutions				Internal solutions
		NT	INCX	ICl,PKA	Low Cl−	INCX	ICl,PKA
NaCl	130	140	140	150	18	10	–
KCl	4.5	4	–	–	–	–	–
HEPES	5	5	5	10	10	10	10
MgCl2	3.5	1	1	0.5	0.5	0.4	0.5
Glucose	10	10	10	11	11	5	5.5
NaH2PO4	0.4	–	–	–	–	–	–
TEACl	–	–	–	–	–	20	20
EGTA	–	–	–	–	–	5	10
CaCl2	–	2.5	2.5	–	–	1	–
CsCl	–	–	–	–	–	110	–
BaCl2	–	–	1	–	–	–	–
l-Aspartic acid	–	–	–	–	132	–	85
MgATP	–	–	–	–	–	–	10
Na–creatine PO4	–	–	–	–	–	–	5
Na–GTP	–	–	–	–	–	–	0.1
CdCl2	–	–	–	1	1	–	–
Nitrendipine	–	–	0.01	–	–	–	–
Strophanthidin	–	–	0.01	–	–	–	–

pH of the external solutions 7.45 at room temperature with NaOH and for the internal solution 7.2 with CsOH. NT: normal Tyrode's solution.

2.2. Electrophysiological recording and data acquisition

Whole cell patch clamp experiments were performed at 37 °C. The compositions of all experimental solutions used in the patch-clamp recordings are shown in Table 1. An aliquot of cell suspension was placed onto a glass chamber (∼1 ml volume) on the stage of an inverted microscope (Olympus CK40) and the cells were superfused with standard external Tyrode's solution. Experiments were carried out with an Axopatch 200B amplifier (Axon Instruments, USA). Protocols were generated and recorded with PulseFit software (HEKA Elektronik GmbH, Germany) via an analogue-to-digital converter (ITC-16 computer interface, Instrutech Corporations, USA). Patch-pipettes (Corning 7052 glass, AM Systems Inc., USA) were pulled on a horizontal micropipette puller (Model P-97, Sutter Instruments Company) to a resistance of 2.0–3.5 MΩ. Series resistance and membrane capacitance were compensated prior to acquiring any recordings. Data were recorded at 1 kHz. External solutions were applied using a temperature-controlled rapid solution application device [36].

For I_NCX recordings, once the whole-cell configuration was achieved, the external superfusate was changed to an NCX-recording solution as used in previous studies (e.g., [13,37]). This K⁺-free solution (to abolish inward rectifier K⁺-current), contained 10 μM each of nitrendipine (to inhibit the L-type Ca²⁺ currents) and strophanthidin to block the Na⁺–K⁺-pump (Table 1). I_NCX was measured as the current sensitive to rapid application of 10 mM Ni²⁺. This concentration of Ni²⁺ has previously been shown to be maximally effective in blocking I_NCX [13]. A voltage ramp protocol was applied from a holding potential of −80 mV once every 10 s; voltage was stepped to +80 mV for 100 ms before a downward ramp to −120 mV over 2 s [12]. For Ni²⁺-sensitive difference currents, current in the presence of Ni²⁺ was subtracted from current in control, ISO, noradrenaline (NA) or forskolin, as appropriate.

Recordings of I_Cl,PKA were made as described previously [32]. The K⁺- and Ca²⁺-free external and internal solutions used are shown in Table 1. For experiments using low external [Cl⁻] (21 mM), Cl⁻ was replaced with aspartate. In order to avoid any change in the liquid junction potential with the low external Cl⁻ solutions, an agar bridge with 3 M KCl was used. To record I_Cl,PKA, a sawtooth voltage ramp was applied from a holding potential of 0 mV incorporating a downward ramp from +100 mV to −100 mV over 1 s [32]. Recordings of I_Ca,L were made using the internal solution as used for I_Cl,PKA recording and a K⁺-free Tyrode's external solution in which K⁺ was replaced with equimolar Cs⁺.

2.3. Chemicals

Chemicals used in the cell-isolation were obtained from British Drug House (BDH, Aristar-grade) or Sigma–Aldrich. Strophanthidin and nitrendipine were dissolved in ethanol as 10 mM stock and appropriate dilutions were added to the external solutions used in the experiments to get a final concentration of 10 μM each. Each of 1 μM isoprenaline (ISO), 1 μM noradrenaline (NA) and 10 μM forskolin (Sigma–Aldrich) were used to activate β-adrenoceptor/PKA in separate experiments; previous studies have used similar concentrations of these compounds to activate maximally the exchanger and the Cl⁻ channel currents [20,24,25]. Stock solutions were made fresh in de-ionised water (ISO and NA) or ethanol (forskolin) on the day of experiment and appropriate dilutions were added to the external solutions as needed.

2.4. Data analysis and presentation

Data were analyzed using Igor Pro (WaveMetrics, Inc., USA), Excel 2003 and GraphPad Prism software. Data are presented as mean ± standard error of the mean (SEM), ‘n’ values refer to the number of cells for particular observations (typically from at least two hearts). Membrane current and membrane potential are referred to as I_m and V_m, respectively, in subsequent text and figures. Statistical comparisons were made using a paired Student's t test or one or two-way ANOVA as appropriate. P values of less than 0.05 were considered to be statistically significant.

3. Results

3.1. Isoprenaline activates a current under NCX-recording conditions that is Ni²⁺-sensitive

As shown in Fig. 1A, superfusion of the cells with solution containing 1 μM ISO increased the control current several fold (e.g., at +80 mV, ISO activated the control current 4.0 ± 1.2 fold; n = 7). As reported previously, using similar recording conditions in guinea pig ventricular myocytes [15], the ISO-activated current was completely inhibited by further application of 10 mM Ni²⁺ in the presence of 1 μM ISO. Fig. 1B shows the mean I–V relationships of the net whole-cell currents; the control currents were significantly increased by 1 μM ISO at extremes of voltages tested in the ramp protocol and 10 mM Ni²⁺ in the presence of same concentration of ISO abolished the ISO-activated current.

Fig. 1.

Effect of 1 μM ISO and 10 mM Ni²⁺ on whole-cell currents under NCX-recording conditions. (A) Representative traces of the net whole-cell currents in control solution, following superfusion with ISO-containing solution and in the presence of both ISO and Ni²⁺ (upper panel); voltage ramp protocol is shown in the lower panel. (B) Mean current density–voltage (I–V) relationships of the net whole-cell currents (n = 7). *P < 0.05, **P < 0.01, ***P < 0.001 (Bonferroni's post hoc test).

3.2. I_Cl,PKA activated via β-adrenoceptor stimulation is Ni²⁺-sensitive

To determine the Ni²⁺ sensitivity of the β-adrenoceptor-activated I_Cl,PKA, we used Cl⁻-recording conditions and each of 1 μM ISO or 1 μM NA as the β-agonists. Fig. 2A (upper panel) shows the representative net whole-cell currents; as was found under NCX recording conditions, the 1 μM ISO-activated I_Cl,PKA was completely inhibited by 10 mM Ni²⁺. Fig. 2B (i) shows the I–V relationships of the ISO-activated difference current and the Ni²⁺-sensitive current in the presence of ISO, demonstrating that the two currents were equivalent to each other; the mean data from 4 cells are shown in Fig. 2B (ii). Similarly, the I_Cl,PKA activated by the physiological β-adrenoceptor agonist NA was also completely inhibited by Ni²⁺, as shown in Fig. 2C (i) and (ii).

Fig. 2.

Effect of β-adrenoceptor agonists and Ni²⁺ on I_Cl,PKA. (A) Representative traces of net whole-cell current in control and the current activated by 1 μM ISO and the current in the presence of same concentration of ISO + Ni²⁺ (10 mM). Lower panel shows the voltage ramp protocol. (B) I–V relationships of the current activated by ISO and the current blocked by Ni²⁺ in the presence of ISO: (i) representative traces (ii) mean (n = 4) of ISO-activated currents (open squares) and Ni²⁺-sensitive currents (filled squares). (C) I–V relationship of the current activated by NA and the current blocked by Ni²⁺ in the presence of NA: (i) representative traces (ii) mean (n = 5) of NA-activated currents (open triangles) and Ni²⁺-sensitive currents (filled triangles).

3.3. Forskolin-activated I_Cl,PKA is not Ni²⁺-sensitive

In contrast to the currents activated by the β-agonists, when activated downstream to the β-adrenoceptors using 10 μM forskolin, I_Cl,PKA was insensitive to 10 mM Ni²⁺. Fig. 3A shows the time-course of an experiment done under Cl⁻ recording conditions, presenting the net whole-cell currents measured at −100 mV and +100 mV; the forskolin activated current was insensitive to Ni²⁺. On the other hand, lowering the external solution Cl⁻ concentration (to 21 mM) attenuated the current and shifted the reversal potential to a more positive value (trace iv, Fig. 3B), confirming the forskolin-activated current to be Cl⁻-selective. Fig. 3C shows the mean I–V relationships of the Ni²⁺-sensitive difference currents in control and the Ni²⁺-sensitive difference currents in the presence of forskolin. In summary, there was no Ni²⁺-sensitive element to the forskolin-activated I_Cl,PKA.

Fig. 3.

Effect of 10 μM forskolin and 10 mM Ni²⁺ on I_Cl,PKA. (A) Representative time-course of changes in currents with superfusion of forskolin and Ni²⁺ sampled at +100 mV (open circles) and −100 mV (filled circles). (B) I–V relationships of the net whole-cell currents as shown in A. (C) Mean (n = 6) I–V relationships of the currents activated by forskolin (open triangles) and the currents blocked by Ni²⁺ in the presence of forskolin (filled triangles) *P < 0.05, **P < 0.01, ***P < 0.001 (Bonferroni's post hoc test).

3.4. Forskolin does not upregulate Ni²⁺-sensitive I_NCX in either rabbit ventricular or atrial myocytes

Having established that I_Cl,PKA activated by forskolin was insensitive to Ni²⁺, we investigated the effect of 10 mM Ni²⁺ on the current activated by the same concentration of forskolin (10 μM) under NCX recording conditions. Fig. 4A shows the time-course of a typical experiment on a rabbit ventricular myocyte with net whole-cell currents sampled at −120 mV and +80 mV. Following the initial run-up in the current, Ni²⁺ was applied to obtain a measure of I_NCX under control conditions. After washout of the Ni²⁺, application of forskolin resulted in a further increase in the current. There was indeed a Ni²⁺-sensitive component to the current in the presence of forskolin under the NCX recording conditions. However, the I–V relationships of the Ni²⁺-sensitive difference currents in control and the Ni²⁺-sensitive currents in the presence of forskolin as shown in Fig. 4B (traces ii and iii) were effectively identical. The I–V relationships of the mean data are shown in Fig. 4C. Thus, forskolin did not increase I_NCX under these recording conditions.

Fig. 4.

Effect of 10 μM forskolin and 10 mM Ni²⁺ on whole-cell currents under NCX-recording conditions in rabbit ventricular cells. (A) Representative time-course of changes in currents with superfusion of forskolin and Ni²⁺ sampled at +80 mV (open circles) and −120 mV (filled circles). (B) I–V relationships of the currents blocked by Ni²⁺ in control (i and ii) and in the presence of forskolin (iii and iv). (C) Mean (n = 5) I–V relationships of the Ni²⁺-sensitive current in control (filled diamonds) and Ni²⁺-sensitive current with forskolin (filled squares).

Since rabbit atrial myocytes have been shown to lack I_Cl,PKA [33], similar experiments were carried out using atrial cells as shown in Fig. 5. 10 μM forskolin did not activate the basal net whole-cell current and the Ni²⁺-sensitive difference currents in control solution were identical to the Ni²⁺-sensitive currents in the presence of forskolin (I–V relationships of the representative difference currents and mean data are shown in Fig. 5B and C). In order to verify whether the concentration of forskolin being used was effective in activating PKA-dependent pathways, we tested the effect of 10 μM forskolin on the L-type Ca²⁺ current (I_Ca,L) in rabbit atrial cells isolated from the same hearts; forskolin increased I_Ca,L by 2.57 ± 0.71 fold (n = 8, data not shown).

Fig. 5.

Effect of forskolin and Ni²⁺ in whole-cell currents under NCX-recording conditions in rabbit atrial cells. (A) Representative time-course of changes in currents with superfusion of forskolin and Ni²⁺ sampled at +80 mV (open circles) and −120 mV (filled circles). (B) I–V relationships of the currents blocked by Ni²⁺ in control (i and ii) and in the presence of forskolin (ii and iii). (C) Mean (n = 4) I–V relationships of the Ni²⁺-sensitive difference current in control (filled upward triangles) and Ni²⁺-sensitive current with forskolin (filled downward triangles).

4. Discussion

We have investigated the effects of acute β-adrenoceptor/PKA-stimulation on currents in rabbit ventricular and atrial myocytes under conditions previously considered ‘selective’ for NCX; we also tested the sensitivity of the PKA-activated Cl⁻ current, I_Cl,PKA, to Ni²⁺. Our results provide novel evidence that the apparent increase in I_NCX with β-adrenoceptor stimulation in rabbit cardiac myocytes is likely due to activation of a PKA-activated Cl⁻ current, which is also sensitive to Ni²⁺ when activated proximal to the β-adrenoceptors. Importantly, data from rabbit atrial myocytes, which do not express PKA-dependent CFTR Cl⁻ channels, demonstrate that I_NCX was not increased in the absence of I_Cl,PKA; to the best of our knowledge, this represents the first investigation of the effects of PKA-stimulation on I_NCX in rabbit atrial myocytes. In addition, Ni²⁺ acts to inhibit the response to β-adrenoceptor stimulation at the level of the receptor, but not downstream.

Consistent with previous studies [24], β-adrenoceptor stimulation, using either isoprenaline or noradrenaline, elicited an increase in current in the NCX-recording conditions. However this apparent increase in current is suggested to reflect activation of I_Cl,PKA as both the I_NCX and I_Cl,PKA were found to be sensitive to Ni²⁺ when activated by β-adrenoceptor stimulation: similar to the report by Lin et al. in guinea pig myocytes [24], I_Cl,PKA in rabbit ventricular myocytes was sensitive to extracellular Ni²⁺ when activated by β-adrenoceptor agonists, but not when activated downstream to the receptor by forskolin. This suggests that Ni²⁺ inhibits I_Cl,PKA upstream to adenylyl cyclase, most likely at the level of β-adrenoceptors. Thus, since the NCX recording conditions in this study did not preclude the contribution of Cl⁻ currents [5,6], I_Cl,PKA is very likely to have contributed to the Ni²⁺-sensitive currents activated by β-adrenoceptor stimulation under NCX recording conditions. On the other hand, the effects on I_NCX of PKA stimulation via forskolin can be measured reliably as the change in Ni²⁺-sensitive current. Since I_Cl,PKA is absent from rabbit atrial myocytes [33], it was possible to examine the modulation of the Ni²⁺-sensitive I_NCX in these cells with minimal contamination by I_Cl,PKA. Consistent with this, the basal net whole-cell current was not increased by 10 μM forskolin in rabbit atrial cells (Fig. 5A). Importantly the Ni²⁺-sensitive I_NCX was not stimulated by forskolin (Fig. 5C). Thus, under NCX recording conditions, the Ni²⁺-sensitive difference current in control and the Ni²⁺-sensitive current in the presence of forskolin were superimposable, both in rabbit ventricular (Fig. 4B) and atrial myocytes (Fig. 5B). This demonstrates that activation of the PKA-pathway via forskolin does not increase I_NCX under these conditions.

Whilst our finding of an effect of extracellular Ni²⁺ at the level of β-adrenoceptors is consistent with that reported by Lin et al. in guinea pig ventricular myocytes [24], our results in part contrast with those of Ginsberg and Bers [23]: whilst both studies report a lack of activation of rabbit I_NCX via the PKA-pathway, Ginsburg and Bers reported the isoprenaline-activated I_Cl,PKA in rabbit ventricular myocytes to be insensitive to Ni²⁺ [23], whereas we found it to be Ni²⁺-sensitive. The reasons for this apparent discrepancy are unclear. Whilst the precise mechanism for the action of extracellular Ni²⁺ in our study and that of Lin et al. [24] remains unknown, taken together, the data are compatible with a previously established role for extracellular divalent cations in ligand binding to G-protein-coupled receptors [38].

Although it is clear from our data that, under the conditions of the present study, stimulation of the PKA-pathway downstream to the receptors does not activate rabbit I_NCX, we cannot exclude entirely the possibility that activation of β-adrenoceptors may modulate NCX function in this species. As there is evidence of compartmentalization of cAMP signalling in sub-cellular microdomains, activation of PKA via β-adrenoceptor stimulation may elicit distinct effects to activation of PKA via forskolin [39]. Evidence for a role of the subcellular environment in the NCX response to PKA-phosphorylation is provided by the observation that while NCX1 activity was shown to be upregulated by PKA in rat myocytes and Xenopus oocytes [27], there was no alteration in the NCX1 activity in response to PKA stimulation when NCX1 was heterologously expressed in HEK293 cells [28]. It has been suggested that NCX1 may be part of a ‘macromolecular complex’ along side PKA-subunits, protein kinase C, protein kinase-A anchoring proteins (mAKAP) and protein phosphatases PP1 and PP2A [5,29]. It is also possible that the basal levels of phosphorylation of various PKA-targets including the NCX vary, which will influence the sensitivity to β-adrenoceptor-mediated changes in the I_NCX; this was indeed shown in failing pig myocytes where there was a higher basal level of phosphorylation of NCX by PKA, which lead to a greater basal I_NCX [30]. That study also demonstrated almost 500% increase in I_NCX in non-failing pig myocytes with β-adrenoceptor stimulation [30].

5. Conclusions

We conclude that, under experimental conditions in which major voltage and time-dependent conductances are inhibited: (i) there is contamination by I_Cl,PKA in the apparent activation of rabbit ventricular I_NCX seen with β-adrenoceptor stimulation and (ii) that the Ni²⁺-sensitive I_NCX is not upregulated by PKA-stimulation in conditions where I_Cl,PKA is absent. Our results in respect of effects of Ni²⁺ are consistent with those of Lin et al. [24]. The conclusions of our study are concordant with those of Ginsberg and Bers [23] in that we did not see any upregulation of Ni²⁺-sensitive I_NCX with PKA-stimulation in rabbit myocytes under the conditions used at 37 °C. This does not exclude the possibility of modulation of I_NCX via this pathway under other conditions or in other species. The findings of this study further warn us to be aware of contaminating membrane currents under conditions otherwise thought to be ‘selective’, especially in ventricular cells when I_NCX is measured as the Ni²⁺-sensitive current.