All experiments were carried out in accordance with guidelines of the local institutional animal care and use committee. In addition, the investigation complied with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996). Both SA nodal and atrial cells were isolated from hearts of male New Zealand White rabbits of 4 months old.

Single SA nodal cells were enzymatically isolated from the entire SA nodal region as described previously (Verkerk et al., 2009). Single atrial cells were isolated by enzymatic dissociation from the left atrium using the same protocol as described previously for the isolation of ventricular myocytes (Den Ruijter et al., 2010), with the adaptation that the enzymatic solution was complemented with 6.6 μg/mL protease. Small aliquots of either SA nodal or atrial cell suspension were put in a recording chamber on the stage of an inverted microscope. Cells were allowed to adhere for 5 min after which superfusion with Tyrode’s solution was started. Tyrode’s solution had a temperature of 36 ± 0.2°C and contained (in mM): NaCl 140, KCl 5.4, CaCl₂ 1.8, MgCl₂ 1.0, glucose 5.5, and HEPES 5.0; pH was set to 7.4 with NaOH. Spindle and elongated spindle-like cells displaying regular contractions were selected for recordings from SA nodal myocytes. Intrinsically quiescent rod-shaped cross-striated myocytes with a smooth surface were selected for recording from atrial myocytes.

Action potentials were recorded by the amphotericin-perforated patch-clamp technique using an Axopatch 200B amplifier (Molecular Devices, Sunnyvale, CA, USA). SA nodal action potentials were low-pass filtered (cut-off frequency 1 kHz) and digitized at 2 kHz; atrial action potentials at 5 and 40 kHz, respectively. Potentials were corrected for the estimated liquid junction potential (Barry and Lynch, 1991). Data acquisition and analysis were accomplished using custom software. For recording from SA nodal myocytes, pipettes (borosilicate glass; resistance 2–3 MΩ) were heat polished and filled with solution containing (in mM): K-gluc 125, KCl 20, NaCl 10, amphotericin-B 0.22, and HEPES 10; pH was set to 7.2 with KOH. For recording from atrial myocytes, the pipette solution contained (in mM): K-gluc 110, KCl 30, NaCl 5, MgCl₂ 1, amphotericin-B 0.22, and HEPES 10; pH was set to 7.28 with KOH. Action potentials in atrial cells were elicited at 1–4 Hz by 2-ms, ≈1.5× threshold current pulses through the patch pipette.

Action potentials were characterized by duration at 20, 50, and 90% repolarization (APD₂₀, APD₅₀, and APD₉₀, respectively), maximum diastolic potential (MDP), action potential amplitude, maximum upstroke velocity, and, in case of SA nodal myocytes, cycle length and diastolic depolarization rate measured over the 50-ms time interval starting at MDP +1 mV. Parameter values obtained from 10 consecutive action potentials were averaged. Action potentials were measured in the absence and presence of noradrenalin, acetylcholine, or a combination of both in the same myocyte. In case of atrial myocytes, the myocyte was stimulated at a frequency of 2 Hz, unless otherwise stated. In order to obtain steady-state conditions, action potential recordings were started 5 min after the addition of noradrenalin or acetylcholine to the Tyrode’s solution.

Data are presented as mean ± SEM. Paired comparisons were made using a paired t-test. Group comparisons were made using either an unpaired t-test or one-way ANOVA with the Holm–Sidak post hoc test. Frequency dependence and dose dependence were assessed with the Friedman repeated measures ANOVA on ranks, followed by pairwise comparisons with the Student–Newman–Keuls post hoc test. P < 0.05 defined statistical significance.

First, we characterized the action potentials of the SA nodal and left atrial myocytes that we used for our study. SA nodal cells were spontaneously active, while atrial cells were quiescent. In atrial cells, action potentials could be elicited by current pulses through the patch pipette. Figure 1A shows typical spontaneous action potentials of an SA nodal myocyte as well as typical action potentials recorded from an atrial myocyte that was stimulated at 3 Hz, i.e., with a cycle length similar to that of the SA nodal myocyte. Average action potential parameters are summarized in Figure 1B. All parameters, except APD₉₀, differed significantly between both cell types. Atrial myocytes had a stable resting membrane potential of −79.2 ± 1.1 mV (n = 12), while SA nodal cells showed a spontaneous diastolic depolarization with a maximum diastolic potential of −63.5 ± 1.7 mV (n = 12). In SA nodal cells, this diastolic depolarization resulted in pacemaker activity with an intrinsic cycle length of 302 ± 10 ms. In SA nodal cells, the maximum upstroke velocity was typically low (6.6 ± 1.1 V/s) as opposed to atrial myocytes (297 ± 39 V/s). In both cell types, action potentials overshot the zero potential value, but the APA was higher in atrial myocytes. Action potentials of atrial myocytes repolarized earlier and faster, resulting in shorter APD₂₀ and APD₅₀; however, APD₉₀ did not differ significantly. Atrial action potentials showed a frequency dependence in maximum upstroke velocity, APD₂₀ and APD₅₀. Both APD₂₀ and APD₅₀ increased with an increase in stimulus frequency, whereas maximum upstroke velocity tended to decrease with an increase in frequency (Figure 1C).

Next, we assessed the effects of noradrenalin, acetylcholine, or a combination thereof on the action potentials of our SA nodal myocytes. As illustrated in Figure 2A, application of noradrenalin resulted in an increase in the spontaneous beating frequency, associated with an increase in the diastolic depolarization rate without any striking differences in other action potential parameters. Application of acetylcholine, on the other hand, resulted in a clear hyperpolarization of the maximum diastolic potential, a decrease in the diastolic depolarization rate and a dramatic increase in cycle length (Figure 2B). If acetylcholine was applied in the presence of noradrenalin, there was no substantial hyperpolarization of the maximum diastolic potential and the increase in cycle length was less prominent (Figure 2C).

Figure 3 summarizes the effects on action potential parameters for a total of 13 cells. These data confirm the aforementioned effects on cycle length (Figure 3A), diastolic depolarization rate (Figure 3B), and maximum diastolic potential (Figure 3C). In addition, Figure 3 shows that both noradrenalin and acetylcholine caused an increase in maximum upstroke velocity, which was larger in case of acetylcholine (Figure 3E). Figure 3 may suggest that acetylcholine increases action potential amplitude (Figure 3D) and action potential duration (Figures 3F–H), but it should be noted that these data are not statistically significant. According to Figure 3, and also Figure 2C, intermediate results were obtained in the combined presence of noradrenalin and acetylcholine, but it should be noted that these observations are based on data from only one cell, as emphasized by the use of “bleached” bars for “NA + ACh” in Figure 3.

We also assessed the effects of noradrenalin, acetylcholine, or a combination thereof on the action potentials of our atrial myocytes. The results are shown in Figures 4–7.

As illustrated in Figure 4A, application of noradrenalin resulted in a slight hyperpolarization of the maximum diastolic potential. The same was found for acetylcholine (Figure 4B), but not so much for the combination of noradrenalin and acetylcholine (Figure 4C). In all cases, i.e., application of noradrenalin, acetylcholine, or a combination thereof, the action potential amplitude was increased. Effects on action potential duration differed between noradrenalin and acetylcholine, with an increase in case of noradrenalin and a decrease for acetylcholine (Figures 4A,B) and a less pronounced effect for the combination of noradrenalin and acetylcholine (Figure 4C).

Figure 5 summarizes the effects on action potential parameters for a total of 23 cells. As already observed in the typical examples of Figure 4, there is a slight hyperpolarization of the maximum diastolic potential (Figure 5A) and an increase in action potential amplitude (Figure 5B). Also, under all conditions there is an increase in maximum upstroke velocity (Figure 5C). The effects on APD₅₀ and APD₉₀ are different between the three conditions, with a decrease in case of acetylcholine, an increase in case of noradrenalin and a smaller, but still significant increase in case of acetylcholine combined with noradrenalin (Figures 5E,F).

In addition to the experiments of Figures 4 and 5, in which the cells were stimulated at 2 Hz, we carried out experiments in which we varied the stimulation frequency over the range of 1–4 Hz. The data, obtained from a total of 12 cells, did not reveal a clear-cut frequency dependence in the effects of acetylcholine or noradrenalin on action potential duration (Figure 6). Data obtained in the combined presence of noradrenalin and acetylcholine are from only two cells and therefore appear “grayed out” in Figure 6.

The above data were obtained with noradrenalin and acetylcholine concentrations of 1 μM (1000 nM). In a total of five cells, we varied these concentrations over the range of 1–1000 nM. The effects on action potential duration showed a dose dependence, in particular for APD₅₀, with considerably smaller effects at 1 nM than at 1 μM (Figure 7). However, it should be noted that the data with acetylcholine are from one cell, as are the data in the combined presence of noradrenalin and acetylcholine. These data therefore appear “grayed out” in Figure 7.

The beating rate of our SA nodal cells increases upon application of noradrenalin and decreases upon application of acetylcholine (Figures 2A,B and 3A), which is associated with an increase and a decrease in the rate of diastolic depolarization, respectively (Figure 3B). It is tempting to explain this change in depolarization rate by a cAMP-mediated upregulation or downregulation of the hyperpolarization-activated “pacemaker current” or “funny current” I_f (DiFrancesco, 1995). However, one should be cautious when drawing conclusions regarding underlying mechanisms from changes in action potentials. This may be illustrated by the increase in maximum upstroke velocity that is observed in response to noradrenalin as well as acetylcholine (Figure 3E). In case of noradrenalin, this increase may well result from an upregulation of the L-type calcium current I_Ca,L (Hagiwara et al., 1988), but in case of acetylcholine the even larger increase may be the effect of less voltage-dependent inactivation of I_Ca,L due to the larger recovery time between consecutive upstrokes and less calcium-dependent inactivation of I_Ca,L due to the smaller calcium transients (van Borren et al., 2010). Additionally, the activation of a small number of fast sodium channels may have contributed to the increase in maximum upstroke velocity as a result of the acetylcholine-induced hyperpolarization of the maximum diastolic potential which results in incomplete voltage-dependent inactivation of these channels (Veldkamp et al., 2003). A full exploration of the autonomic modulation of the complex dynamic interactions of voltage- and calcium-dependent processes within the SA nodal pacemaker cell may require in silico experiments (Maltsev and Lakatta, 2010). Such experiments have already proven useful in a computer simulation study of the autonomic modulation of the electrical activity of bullfrog atrial myocytes (Shumaker et al., 1991). Figures 3F–H suggest an increase in action potential duration in response to acetylcholine, but this may reflect, at least in part, the hyperpolarization of the maximum diastolic potential (Figure 3C) rather than a “true” increase in action potential duration. This hyperpolarization can be readily explained by the activation of I_K,ACh (van Borren et al., 2010).

A reduction in beating frequency as well as a hyperpolarization of the maximum diastolic potential was also reported for SA nodal myocytes isolated from wild-type mice in response to carbachol (Cifelli et al., 2008). These effects were both larger for myocytes from RGS4^−/− mice. In the top panel of their Online Figure 5, Yang et al. (2010) show “representative recordings of spontaneous action potential firing in SA nodal myocytes isolated from wild-type mice.” With an increase in cycle length, a decrease in diastolic depolarization rate, a hyperpolarization of the maximum diastolic potential and an increase in action potential amplitude, the changes in action potential configuration in response to carbachol are consistent with our data for rabbit SA nodal myocytes in response to acetylcholine (Figure 3).

In our atrial myocytes, we found a hyperpolarization of the maximum diastolic potential (or “resting” membrane potential) in response to acetylcholine (Figure 5A), as we did in our SA nodal myocytes (Figure 3C). However, in contrast to our SA nodal myocytes, such hyperpolarization was also observed in response to noradrenalin. This may be explained by the β-adrenergic activation of I_K,ACh in atrial myocytes (Sorota et al., 1999) and by the β-adrenergic stimulation of the atrial-specific potassium current I_KH (Ehrlich et al., 2004; Yeh et al., 2007). Furthermore, we found an increase in action potential amplitude and, as in our SA nodal myocytes, an increase in maximum upstroke velocity both in response to noradrenalin and in response to acetylcholine (Figures 5B,C). This may be explained by an increase in the fast sodium current, which is responsible for the fast upstroke in atrial cells, due to the more negative maximum diastolic potential and thus larger availability of fast sodium channels (less voltage-dependent inactivation). On the other hand, increased intracellular calcium levels in response to noradrenalin may have inhibitory effects on these sodium channels (Casini et al., 2009), but such effects are not apparent in our current data.

Action potential duration is increased in response to noradrenalin and decreased in response to acetylcholine, with more or less additive effects if both noradrenalin and acetylcholine are present (Figures 5D–F). The shortening of the action potential in response to acetylcholine is in line with data from literature (Koumi et al., 1995; Patterson et al., 2008) and may be attributed to the activation of I_K,ACh. Of note, acetylcholine does not inhibit the L-type calcium current of atrial myocytes under basal conditions, in contrast to that of SA nodal myocytes (Petit-Jacques et al., 1993).

The increase in action potential duration in response to noradrenalin (Figures 5D–F) may be attributed to the increase in I_Ca,L in response to β-adrenergic stimulation (Bean, 1985; Dun et al., 2003), which would then be sufficiently large to counteract the action potential shortening effects of the aforementioned β-adrenergic activation of I_K,ACh and I_KH. However, Yeh et al. (2007) observed a prominent decrease in action potential duration in a canine left atrial tissue preparation in response to specific β-adrenergic stimulation by isoproterenol (isoprenaline, 5 μM). This suggests that the α-adrenergic action of noradrenalin is essential for the increase in action potential duration that we observed in our experiments. A clear increase in action potential duration in response to the selective α₁-adrenergic receptor agonist phenylephrine (100 μM) has been observed in left, but not right, atrial preparations from rabbit heart (Jahnel et al., 1992). In isolated rat atria, Northover (1994) also observed a significant increase in action potential duration in response to phenylephrine (50 μM), whereas the effects of isoproterenol (0.5 μM) were small and rather variable. In contrast, phenylephrine (100 μM) only had a minor effect on action potential duration in the experiments by Yeh et al. (2007). The apparent discrepancies in experimental results may be explained by species- and tissue-dependent differences and/or by differences in cAMP levels due to differences in basal adenylyl cyclase activity (Méry et al., 1997; Harvey and Belevych, 2003). Furthermore, a dose dependence in the effects of isoproterenol and phenylephrine on the action potential and the underlying transmembrane currents may play a role (Shumaker et al., 1991).

The data on action potential duration of Figure 5 were obtained at a stimulus frequency of 2 Hz. Figure 6 demonstrates that similar effects were obtained at other frequencies in the range of 1–4 Hz, without a clear-cut frequency dependence, although the increase in APD₅₀ at 1 Hz in the presence of noradrenalin is significantly different from that at 4 Hz. The data at 2 Hz of Figure 6A are in line with the data of Figure 5E. Also, there is consistency between the data at 2 Hz of Figure 6B and the data of Figure 5F. However, the difference between “NA” and “NA + ACh” in Figure 6B is smaller than expected from Figure 5F, where this difference is almost significant (P = 0.06).

In the present study, we have used the amphotericin-perforated patch-clamp technique to test the effects of noradrenalin, acetylcholine, or a combination thereof on the action potential configuration of isolated sinoatrial and atrial myocytes under close-to-physiological conditions. Whereas this approach allows a detailed study of the effects of a well-controlled dose of these neurotransmitters on the intrinsic action potential configuration of these isolated myocytes, it does not take into account the complexity and heterogeneity of the intact myocardium. Therefore, caution should be applied when translating our in vitro results to the in vivo situation. Furthermore, although human action potential morphology is better resembled by rabbit than by mouse, there may be species differences in the effects of noradrenalin and acetylcholine.