The funny current If is essential for the fight-or-flight response in cardiac pacemaker cells

Abstract

The sympathetic nervous system fight-or-flight response is characterized by a rapid increase in heart rate, which is mediated by an increase in the spontaneous action potential (AP) firing rate of pacemaker cells in the sinoatrial node. Sympathetic neurons stimulate sinoatrial myocytes (SAMs) by activating β adrenergic receptors (βARs) and increasing cAMP. The funny current (If) is among the cAMP-sensitive currents in SAMs. If is critical for pacemaker activity, however, its role in the fight-or-flight response remains controversial. In this study, we used AP waveform analysis, machine learning, and dynamic clamp experiments in acutely isolated SAMs from mice to quantitatively define the AP waveform changes and role of If in the fight-or-flight increase in AP firing rate. We found that while βAR stimulation significantly altered nearly all AP waveform parameters, the increase in firing rate was only correlated with changes in a subset of parameters (diastolic duration, late AP duration, and diastolic depolarization rate). Dynamic clamp injection of the βAR-sensitive component of If showed that it accounts for ∼41% of the fight-or-flight increase in AP firing rate and 60% of the decrease in the interval between APs. Thus, If is an essential contributor to the fight-or-flight increase in heart rate.

Figure S1.

Axon Axopatch 200B validation. (A) Schematic of the circuit used to test the response of the Axopatch 200B amplifier in response to the voltage output from a waveform generator. The signal from the waveform generator was split to allow direct comparison of the amplifier output to the original signal. (B and C) Representative sinoatrial APs originally recorded in 1 nM (B) and 1 µM Iso (C) as measured through the model circuit by the Axopatch 200B (green) compared to the original signal (black). (D) Average frequency content of sinoatrial APs recorded in 1 nM (black, n = 13) and 1 µM Iso (red, n = 13) determined by fast Fourier transform. (E and F) Bode plots showing the differences in magnitude (E) and phase (F) for 100 mV sine waves of different frequencies as recorded by the Axopatch 200B compared to the original signals. Gray boxes in D–F indicate the frequency range (1–100 Hz) that contains >99% of the sinoatrial AP signal.

Figure S2.

Dynamic clamp approach. (A) Schematic of the dynamic clamp technique. (B) Membrane voltages recorded from a patch-clamped SAM are used by a Teensy microprocessor to simulate currents using models of I_f in 1 nM and 1 µM Iso (top right). The calculated dynamic clamp current (I_dyn), corresponding to the Iso-sensitive component of I_f (middle right), is then injected back into the SAM in a fast feedback loop, and the resulting changes to AP firing are recorded (bottom right).

Figure S3.

ParamAP automates determination of sinoatrial node AP waveform parameters and fully parameterizes the AP. (A) Schematic illustration of ParamAP waveform parameters, including the new parameter LAPD (purple). Figure adapted from Rickert and Proenza (2017). (B) Sinoatrial AP waveforms modeled by line segments representing average waveform parameters determined by ParamAP in 1 nM or 1 µM Iso indicate that ParamAP adequately parameterizes the sinoatrial AP.

Figure 1.

Sinoatrial action potentials are heterogeneous. (A) Overlays of individual (n = 50) sinoatrial AP waveforms (grey) and the average of all waveforms (black) recorded in 1 nM Iso. (B) Overlays of individual (n = 50) sinoatrial AP waveforms (pink) and the average of all waveforms (red) recorded in 1 µM Iso. (C–E) Correlation plots of individual AP waveform parameters in 1 nM Iso (grey) and 1 µM Iso (pink) as a function of the AP firing rate. Significant correlations are shown as black (1 nM) and red (1 µM) lines, and the Pearson correlation coefficients are noted. All correlation coefficients and details of statistical tests can be found in Table 2.

Figure 2.

Iso affects most sinoatrial AP waveform parameters. (A) Representative sinoatrial APs recorded from a single cell in 1 nM Iso (black) and following wash-on of 1 µM Iso (red). (B) Average (±SEM) percentage change in sinoatrial AP firing rate and waveform parameters induced by perfusion of 1 µM Iso. Red bars denote parameters that are significantly different after βAR stimulation. Since MDP and MRR are negative numbers, a negative percentage change is equivalent to hyperpolarization of the MDP and an increase in the magnitude of MRR. (C1–E4) Box plots of median waveform parameters associated with the diastolic depolarization (C1–C4), action potential voltages and upstroke (D1–D4), and action potential duration and repolarization (E1–E4) in 1 nM Iso (grey) and 1 µM Iso (pink). Individual recordings in C–E are shown as circles. Asterisks indicate significant differences between 1 nM and 1 µM Iso. n = 50 for all cases. Details of statistical tests are found in Table 2.

Figure 3.

βAR stimulation accelerates AP firing by shortening the diastolic interval. (A1–E4) Correlation plots of the change in waveform parameters associated with diastolic depolarization (A1–A4), action potential upstroke (B1–B4), and action potential duration (C1–C4) versus the change in firing rate in SAMs between 1 nM and 1 µM Iso. Significant correlations are shown as a red line, and the Pearson correlation coefficient is noted. n = 50 for all cases. All correlation coefficients and details of statistical tests can be found in Table S1. (D) Average importance of changes to waveform parameters during βAR stimulation in predicting concomitant changes in AP firing rate as measured by random forest machine learning analysis. Variables that were significantly correlated with changes in firing rate in A–C are shown in red. Error bars show the minimum and maximum values obtained from the 100 forests modeled.

Figure 4.

Dynamic injection of βAR-stimulated I_f changes AP firing rate. (A) Representative sinoatrial APs (top) recorded in 1 nM Iso with no current injection (black) and with dynamic addition of βAR stimulated I_f (bottom) with a conductance of 5 nS (teal), 10 nS (blue), or 40 nS (green). (B) Representative sinoatrial APs recorded in 1 µM Iso (top) without current injection (red) and with dynamic subtraction of βAR-stimulated I_f (bottom) with a conductance of 10 nS (purple). (C) Average (±SEM) fractional increase (%) in sinoatrial AP firing rate with dynamic injection of βAR-stimulated I_f at conductances between 5 and 40 nS. Injections of 10 nS or greater caused significant acceleration of the AP firing rate compared to control (5 nS—P = 0.9376, n = 19; 10 nS—P = 0.0401, n = 19; 20 nS—P = 0.0013, n = 17; 30 nS—P = 0.0003, n = 7; 40 nS—P < 0.0001, n = 14). (D) Average (±SEM) firing rate of cells perfused with 1 nM (black) or 1 µM Iso (red) before (Control) and after dynamic addition (blue) or subtraction (purple) of βAR-stimulated I_f (10 nS). Individual recordings are shown as smaller circles (n = 19 in 1 nM Iso and n = 10 in 1 µM Iso). Dynamic subtraction of βAR-stimulated I_f significantly reduced the AP firing rate compared to 1 µM Iso (P = 0.0402).

Figure 5.

Cell-scaled dynamic clamp approach. (A–C) Protocol for dynamic clamp recordings with simulated conductance scaled to endogenous I_f. (A) Spontaneous APs are recorded from a cell perfused with 1 nM Iso (top) and dynamic clamp current is monitored (bottom). After AP amplitude stabilizes following perforation (green), the set of control APs are recorded (gray box), after which the dynamic clamp circuit is switched on and the maximal I_f conductance used to calculate the injected current (I_dyn) is increased in 2 nS steps (blue). Following the injected conductance series, the dynamic clamp circuit is switched off, a second set of control APs are recorded, and the cell is perfused with 1 µM Iso to measure the full βAR-stimulated response (red). (B) Following the control, dynamic current clamp, and Iso-stimulated recordings of APs in current-clamp mode, the amplifier is changed to voltage-clamp mode. 1 mM Ba²⁺ is perfused to block inward rectifier currents, and the endogenous I_f conductance is calculated from the current elicited by a voltage-step to −130 mV, where the current is >90% activated. In this example, conductance was calculated as 9.58 nS based on the 912-pA current and I_f reversal potential of −30 mV. (C) The endogenous I_f is then used to post hoc determine the dynamic clamp conductance for which the AP firing rate and waveform will be analyzed (blue).

Figure 6.

Cell-scaled dynamic clamp shows βAR stimulation of I_f accounts for 36% of the increase in AP firing rate. (A) Top: Representative sinoatrial APs from one cell in 1 nM Iso with no current injection (black), in 1 nM Iso with dynamic injection of βAR-stimulated I_f (blue), and in 1 µM Iso with no current injection (red). Bottom: Dynamic current injection (blue) during the APs recorded during dynamic clamp. (B) Average (±SEM) firing rate of cells in 1 nM Iso before (black) and after cell-scaled dynamic injection of βAR stimulated I_f (blue) or perfusion of 1 µM Iso (red). Individual recordings are shown as small circles (n = 13 for all cases). Control AP firing rates immediately preceding the dynamic I_f-Iso addition and perfusion of 1 µM Iso did not differ significantly (P = 0.7432).

Figure 7.

βAR stimulation of I_f significantly shortens LAPD and DD. (A) Average (±SE) fractional change (%) in sinoatrial AP firing rate and waveform parameters with perfusion of 1 µM Iso (red) or dynamic injection of βAR stimulated I_f (blue) compared to recordings in 1 nM Iso. (B1–B4) Average (+SEM) waveform parameters (EDD, LDD, DDR, LAPD) of cells in 1 nM Iso before (black) and after cell-scaled dynamic injection of βAR stimulated I_f (blue). Individual recordings are shown as small circles (n = 13 for all cases). Details of statistical tests can be found in Table S2.