Beat to beat Ca²⁺-dependent regulation of sinoatrial nodal pacemaker cell rate and rhythm

Abstract

Introduction

Whether intracellular Ca²⁺ regulates sinoatrial node cell (SANC) action potential (AP) firing rate on a beat-to-beat basis is controversial.

Objective

To directly test the hypothesis of beat-to-beat intracellular Ca²⁺ regulation of the rate and rhythm of SANC.

Methods and results

We loaded single isolated SANC with a caged Ca²⁺ buffer, NP-EGTA, and simultaneously recorded membrane potential and intracellular Ca²⁺. Prior to introduction of the caged Ca²⁺ buffer, spontaneous local Ca²⁺ releases (LCRs) during diastolic depolarization (DD) were tightly coupled to rhythmic APs (r²=0.9). The buffer markedly prolonged the decay time (T₅₀) and moderately reduced the amplitude of the AP-induced Ca²⁺ transient and partially depleted the SR load, suppressed spontaneous diastolic LCRs and uncoupled them from AP generation, and caused AP firing to become markedly slower and dysrhythmic. When Ca²⁺ was acutely released from the caged compound by flash photolysis, intracellular Ca²⁺ dynamics were acutely restored and rhythmic APs resumed immediately at a normal rate. After a few rhythmic cycles, however, these effects of the flash waned as interference with Ca²⁺ dynamics by the caged buffer was reestablished.

Conclusions

Our results directly support the hypothesis that intracellular Ca²⁺ regulates normal SANC automaticity on a beat-to-beat basis.

Introduction

A novel hypothesis for the basis of sinoatrial nodal cell (SANC) normal automaticity, supported by experiments in numerous species generated by many labs (cf [1] for review) postulates that surface membrane electrogenic proteins of SANC, functioning as a voltage oscillator (M clock), and intracellular proteins, generating rhythmic Ca²⁺ oscillations (Ca²⁺ clock), interact as a system of symmetrically entrained, inter-dependent oscillators.

The hypothesis embraces several key interactions resulting in clock entrainment. Following an action potential (AP), SERCA pumps Ca²⁺ from the cytosol into a sarcoplasmic reticulum (SR) that is partially depleted of Ca²⁺, due to periodic Ca²⁺ release triggered by the AP. Shortly after the maximum diastolic potential, when sufficient Ca²⁺ has been pumped to achieve a threshold SR Ca²⁺ load required for spontaneous ryanodine receptor (RyR) activation, RyRs begin to generate spontaneous local Ca²⁺ releases (LCRs) beneath the cell surface membrane. LCRs grow in number and size during the diastolic depolarization (DD). Their contribution to the DD, via activation of Na⁺-Ca²⁺ exchange (NCX) current, also grows exponentially, and this is a major mechanism that drives the DD to threshold for activation of L-type current, which initiates the next AP. The AP, in turn, by (1) inducing Ca²⁺ release from SR, synchronizes the SR in a relatively Ca²⁺ depleted state and (2) by effecting Ca²⁺ influx, restores SR Ca²⁺ lost from the cell via NCX. Thus, during each cycle the AP “resets” and “rewinds” the SR function within the coupled-clock system.

The coupled-clock system pacemaker hypothesis challenges a long-standing pacemaker research community tenet that the membrane clock is the sole or dominant mechanism for normal automaticity of SANC [2, 3]. Acute photolysis of intracellular caged Ca²⁺ compound has provided novel mechanistic insights into beat-to-beat Ca²⁺ cycling in ventricular myocytes [4–7]. Experiments to determine whether impaired SR Ca²⁺ cycling dynamics by an intracellular caged Ca²⁺ buffer, or an acute increase in intracellular Ca²⁺ via photolysis of the caged-Ca²⁺ buffer regulates SANC AP firing, would be a stringent test of the hypothesis that a coupled clock system regulates normal automaticity specifically on a beat-to-beat basis.

Methods

We loaded NP-EGTA, a caged Ca²⁺ buffer, into single, isolated rabbit SANC and simultaneously recorded membrane potential via a perforated patch, and imaged Ca²⁺ by confocal microscopy to determine the effects on AP firing in SANC, of: (1) Ca²⁺ buffering by the caged EGTA; and (2) the acute increase in intracellular Ca²⁺ induced by flash photolysis of the Ca²⁺ buffer. A detailed expanded methods description is available in the Online Data Supplement.

Results and Discussion

Prior to Ca²⁺ buffer loading, SANC generate spontaneous rhythmic diastolic LCRs that are highly coupled to the subsequent AP occurrence (Fig. 1A). In response to loading of the Ca²⁺ buffer, the kinetics of Ca²⁺ removal from the cytosol are markedly slowed (Fig. 1B and Table S1). The integral of Ca²⁺ release in the form of LCRs (detected by the Ca²⁺ indicator) becomes substantially reduced (Fig. 1C). The LCR period, defined as the time from the post AP-induced Ca²⁺ transient peak to LCR occurrence, becomes prolonged and dysrhythmic. As a result, LCRs become uncoupled from AP generation, which becomes markedly slowed and then dysrhythmic (Fig. 1D). Coefficient of variation (SD/mean) of AP cycle length (CL) prior to NP-EGTA loading was 7±1.5% and increased at 2 min and 5 min following NP-EGTA loading to 9±2% and 20±5%, respectively (n=7 cells). Note that the LCR period closely predicted each AP CL prior to introduction of Ca²⁺ buffer (r²=0.9, p<0.01), but not after (r²=0.4) (Fig. 1D). Average data on AP induced transients, LCR characteristics and AP CL (n=7 cells) before and after NP-EGTA application are provided in the online supplement Table S1.

Figure 1.

(A) Confocal line-scan images of Ca²⁺ (expressed as a peak value (F) normalized to minimal fluorescence (F₀)) of a representative SANC before and following exposure to 15μM NP-EGTA (small white arrow heads marks LCRs). Histograms of (B) T₅₀ of the AP-induced Ca²⁺ transient, and (C) the integral of Ca²⁺ release in the form of LCRs (detected by the Ca²⁺ indicator). (D) Prior to introduction of the Ca²⁺ buffer, rhythmic LCR periods were tightly linked to each AP occurrence; but in the presence of NP-EGTA, the LCR periods become dysrhythmic and uncoupled from the AP cycle length. Regression lines are shown with their respective equations.

Figure 2 illustrates the effect of photolysis to release Ca²⁺ in a cell that had been preloaded with the caged buffer. Prior to flash photolysis, AP firing rate is slow and dysrhythmic and LCRs are small and partially uncoupled from AP generation (Fig.2A). The SR Ca²⁺ load was reduced by NP-EGTA (the amplitude of the Ca²⁺ transient induced by a rapid application of caffeine onto the cells decreased from 1.6±0.1 to 1.3±0.1 F/F₀, n=9 in each group). Four 50 ms flashes were applied within a time interval of about 150 ms (black arrows in Fig. 2B). Although small in magnitude (Fig. S1), the photo-released Ca²⁺ dramatically affects SANC function. The AP-induced Ca²⁺ transient, LCR characteristics/pattern, the DD slope, and the pattern of AP firing are affected (Fig. 2C-D). The initial effect of the flashes to markedly accelerate DD and acutely reduce CL by ~50% occurs after the termination of the last flash (between beats 3 and 4 in Figure 2B and shown in a slower time base in Fig.2F) and corresponds in time to the emergence of augmented and synchronous LCRs, the ensemble of which produces substantial Late Diastolic Ca²⁺ Elevation (“LDCaE” in Fig. 2E-F) (reaching the red color of F/F₀ in the second cycle). This strongly suggests that the marked increase in DD slope, and reduction in CL induced by the flash, do not result directly from Ca²⁺ release during the flash, but rather require SR pumping of Ca²⁺ released by the flash and subsequent SR Ca²⁺ release, resulting in the emergence of large, abundant LCRs. In other words, the acute, transient, marked reduction in CL caused by the flash was likely mediated by its effects to markedly increase LCRs, which accelerate the DD via Ca²⁺-dependent effects on sarcolemmal molecules. It cannot be directly predicted, however, to what extent the reduction in CL is attributable to augmentation in LCRs that resulted from SR Ca²⁺ cycling resulting from the flash, or to a direct effect of flash-released Ca²⁺ on sarcolemmal proteins.

Figure 2.

Photo-released caged Ca²⁺ acutely re-establishes coupling of Ca²⁺ clock and M clocks. (A) Prior to the flash, the Ca²⁺ buffer uncoupled LCRs (line scan image) from APs (brown lines) resulting in a slow, dysrhythmic AP firing rate. (B) Following the last of four 50 ms UV flashes (black arrows), large, synchronous LCRs emerge (small white arrows) and are coupled to the AP occurrence via an acute increase in DD rate, resulting in acute increase in AP firing rate (C). The concomitant acute increase of integral of Ca²⁺ release in the form of LCRs (detected by the Ca²⁺ indicator) (LDCaE) is shown in panel D. (Note that APs continued to be recorded for a number of beats beyond the Ca²⁺ recording). (E, F) The absence of LDCaE pre-flash and its appearance following flash. (G) The acute coupling of the clocks post flash induces a shift in the Ca²⁺-membrane potential phase-plane diagram relates to AP cycles in E and F.

The phase-plane diagram of the respective duty cycles of the Ca²⁺ signal and the membrane potential (Fig. 2G) illustrates that photo-released Ca²⁺ produces a profound change in the coupling of Ca²⁺ and M-clocks. Prior to the flash, the Ca²⁺-membrane-voltage loop cycles counterclockwise throughout the cycle, but following the flash the loop cycles half clockwise/half counterclockwise (Fig. 2G). The flash-induced clockwise cycling at the beginning of the loop at low diastolic potentials reflects the emergence of the strong diastolic Ca²⁺ signal that accelerates the DD. Similar trends were found in nine of the SANC. After three rhythmic APs shown in Fig.2B (AP recording), the effects of the flash waned, likely due to reestablishment of altered/impaired SR Ca²⁺ cycling dynamics by Ca²⁺ by the caged buffer, and SANC AP firing rate again becomes slower and dysrythmic on a beat-to-beat basis.

Numerous prior studies, in a variety of species have demonstrated that intracellular Ca²⁺ buffering by BAPTA-AM or EGTA-AM markedly slows the spontaneous AP firing rate, but intracellular Ca²⁺ was not measured in these cells (review [1]). However, by using caged Ca²⁺ buffer and measuring intracellular Ca²⁺ we examined not only the link between impaired Ca²⁺ cycling dynamics and disturbed AP firing affected by intracellular Ca²⁺ buffering, but also the beat-to-beat relationship between intracellular Ca²⁺ cycling and AP generation by an acute increase in intracellular Ca²⁺ accomplished by flash photolysis.

Of note, recent reports challenge the existence or symmetry of the coupled-clock system [2, 3]. Introduction of BAPTA into guinea pig SANC via acute rupture of a surface membrane patch failed (in the absence of intracellular Ca²⁺ measurements) to detect any acute reduction in AP firing rate [3]. The null result, which conflicts with the present and numerous other studies (cf [1] for review) can be explained on the basis of the method to apply BAPTA via a pipette; i.e. a small, artifactual rupture-induced patch seal leak current that shifts the current balance of tiny DD currents offsetting the true BAPTA effect that would, in the absence of the artifact, suppress the DD and prolong the cycle length [8] as demonstrated by numerical modeling of the null results [8].

In summary, our results provide direct evidence to support the hypothesis that a system of an intracellular Ca²⁺ clock, coupled to a surface membrane clock [1], rather than a dominant surface membrane ion channel clock [2, 3], regulates SANC normal automaticity on a beat-to-beat basis.

Glossary

AP

action potential

DD

diastolic depolarization

LCR

local Ca²⁺ releases

LDCAE

late diastolic Ca²⁺ elevation

NCX

Na⁺-Ca²⁺ exchanger

RyR

ryanodine receptor

SANC

sinoatrial nodal cells

SERCA

Ca²⁺ ATPase sarcoplasmic reticulum pump

SR

sarcoplasmic reticulum