cAMP-Dependent Signaling Restores AP Firing in Dormant SA Node Cells via Enhancement of Surface Membrane Currents and Calcium Coupling

Abstract

Action potential (AP) firing rate and rhythm of sinoatrial nodal cells (SANC) are controlled by synergy between intracellular rhythmic local Ca²⁺ releases (LCRs) ("Ca²⁺ clock") and sarcolemmal electrogenic mechanisms ("membrane clock"). However, some SANC do not fire APs (dormant SANC). Prior studies have shown that β-adrenoceptor stimulation can restore AP firing in these cells. Here we tested whether this relates to improvement of synchronization of clock coupling. We characterized membrane potential, ion currents, Ca²⁺ dynamics, and phospholamban (PLB) phosphorylation, regulating Ca²⁺ pump in enzymatically isolated single guinea pig SANC prior to, during, and following β-adrenoceptor stimulation (isoproterenol) or application of cell-permeant cAMP (CPT-cAMP). Phosphorylation of PLB (Serine 16) was quantified in the same cells following Ca²⁺ measurement. In dormant SANC LCRs were small and disorganized at baseline, membrane potential was depolarized (-38 ± 1 mV, n = 46), and I_CaL, I_f, and I_K densities were smaller vs SANC firing APs. β-adrenoceptor stimulation or application of CPT-cAMP led to de novo spontaneous AP generation in 44 and 46% of dormant SANC, respectively. The initial response was an increase in size, rhythmicity and synchronization of LCRs, paralleled with membrane hyperpolarization and small amplitude APs (rate ∼1 Hz). During the transition to steady-state AP firing, LCR size further increased, while LCR period shortened. LCRs became more synchronized resulting in the growth of an ensemble LCR signal peaked in late diastole, culminating in AP ignition; the rate of diastolic depolarization, AP amplitude, and AP firing rate increased. I_CaL, I_K, and I_f amplitudes in dormant SANC increased in response to β-adrenoceptor stimulation. During washout, all changes reversed in order. Total PLB was higher, but the ratio of phosphorylated PLB (Serine 16) to total PLB was lower in dormant SANC. β-adrenoceptor stimulation increased this ratio in AP-firing cells. Thus, transition of dormant SANC to AP firing is linked to the increased functional coupling of membrane and Ca²⁺ clock proteins. The transition occurs via (i) an increase in cAMP-mediated phosphorylation of PLB accelerating Ca²⁺ pumping, (ii) increased spatiotemporal LCR synchronization, yielding a larger diastolic LCR ensemble signal resulting in an earlier increase in diastolic I_NCX; and (iii) increased current densities of I_f, I_CaL, and I_K.

Keywords: cardiac automaticity; coupled-oscillator system; dormant cells; pacemaker mechanism; sinoatrial nodal cells.

FIGURE 1

(A) The proportion of baseline behavior of freshly isolated guinea pig SANC we studied total 34 cells. About 1/3 and 2/3 were AP present (firing) and AP absent (dormant), respectively. (B) Illustrative examples of simultaneous voltage (magenta), global cytosolic Ca²⁺ (green) and LCR ensemble area (blue) in dormant SANC (left) and firing SANC (right) at baseline. The corresponding frequency analysis by fast Fourier transform (FFT) and autocorrelation analysis (AC) are shown below. (C) Major ion current density of voltage-dependent L-type Ca²⁺ channel, voltage-dependent K channel and the funny current in dormant (black) and firing (yellow) SANC at baseline; those of dormant SANC were reduced compared to those of firing SANC. The protocol is shown in the right panel.

FIGURE 2

(A) Double immunolabeling of phosphorylated phospholamban (p-PLB) at Serine 16 (green) and total PLB (red) and the merged images of SANC that were dormant at baseline (top), firing at baseline (middle) and dormant at baseline but began firing in response to isoproterenol (bottom). (B) The p-PLB/total PLB ratio of SANC firing at baseline (N = 3/n = 10) vs dormant at baseline (N = 3/n = 10) (At baseline, firing cells have much higher pPLB S16/PLB ratio than non-firing cells)] and (C) The p-PLB/total PLB ratio of SANC firing in isoproterenol (N = 5/n = 15) vs firing at baseline (N = 3/n = 10) (Firing cells in ISO have higher pPLB S16/PLB ratios than firing cells at BL). t-test, ***p < 0.001, **p < 0.01.

FIGURE 3

(A) Membrane potential tracings of SANC that fire spontaneous APs at baseline (black) that accelerated in response to isoproterenol (magenta). (B) Among 12 initially dormant SANC, 44% fired APs in response to β-AR stimulation (“responder”) while 56% failed to do so (“non-responder”). This was significantly different from the corresponding time control group (n = 8) in which spontaneous AP firing did not occur (p < 0.05).) When isoproterenol was washed out, the β-AR stimulation-induced APs stopped. Detail of the beginning (C), steady state (D) and wash-out (E) phases of the β-AR stimulation-induced AP in a SANC that was dormant under initial baseline conditions. (F) The cell permeable CPT-cAMP recapitulated the β-AR stimulation-induced APs in other set of SANC that were dormant at baseline. Note the similarity to panel (B–E).

FIGURE 4

Time domain (left column) and the Ca²⁺-voltage phase-plot loops (right column) of simultaneous membrane potential (red), whole-cell Ca²⁺ transient (green) and LCR ensemble area (blue) measurement at the early stage of AP induction in response to β-AR stimulation in a SANC that was electrically dormant at baseline. Note that, to minimize both phototoxicity and bleaching of the Ca²⁺ indicator, Ca²⁺ signals could be recorded only for short time window, although membrane potential could be recorded for prolonged periods: (A) Baseline, (B) The first AP, and (C) Following early phase that is characterized by APs with a small amplitude and failed ignition (arrows). Blue line shows the integrated area of detected individual LCRs by our image analysis program (Maltsev et al., 2017).

FIGURE 5

Later stage of the β-AR stimulation-induced AP induction sequence observed in other SANC that was initially dormant at baseline. Time domain (left), Ca^{2 +} -Vm loop (middle) and FFT of Ca²⁺ signals (right) during the baseline dormancy (A), early stages (B,C), and steady state (D) of AP induction in response to β-AR stimulation. The Ca-Vm loop evolves to expand as the AP cycle length shortens in response to β-AR stimulation.

FIGURE 6

When the intracellular cAMP level decreased by washout of BARs/CPT-cAMP, the changes described in Figure 5 reversed in order. Compared to the steady state AP firing (panel A), the AP cycle length is prolonged and LCR ensemble is decreased (panel B), suggesting the disengagement of M- and Ca²⁺-clocks. Shortly after cAMP washout (the panel B), spontaneous AP firing ceased, and the cell returned to dormant state again (panel C) in which the membrane potential drifted around –40 mV, ensemble LCR signal area decreased, and the voltage-Ca²⁺ loop demonstrated only very small fluctuation around one point, rather than crating a loop.

FIGURE 7

Consecutive in-series measurements of ion-current density of L-type Ca²⁺ current (left panels), K⁺ current (center panels) and funny current (right panels) from SANC revealed that M-clock function of both responder (magenta) and non-responder (purple) dormant SANC increases in response to β-AR stimulation by 100 nM isoproterenol. The voltage protocol (same to those for Figure 1C) is shown above the I-V curves. Example I–V curves for each current is shown in the bottom raw. The data shown are from different cells, one without isoproterenol (black) and one with isoproterenol (magenta).

FIGURE 8

The phosphorylated-PLB/total PLB ratios (A) and total PLB number (B) and were positively (n = 32; Y = 0.004982X + 0.3209, p = 0.0132; R² = 0.188) and negatively (n = 31; Y = –123.5X + 30723, p = 0.0161; R² = 0.184) associated with the firing rate, respectively.