Modern Concepts Concerning the Origin of the Heartbeat

A panoramic review of the development, maturation, and future of concepts surrounding cardiac pacemaking.

Abstract

Physiological processes governing the heart beat have been under investigation for several hundred years. Major advances have been made in the recent past. A review of the present paradigm is presented here, including a look back at important steps that led us to where we are today, alongside a glimpse into the exciting future of pacemaker research.

The mechanisms initiating and perpetuating the spontaneous heartbeat in living organisms have occupied scientists' thoughts for several hundred years. Many mysteries remain, despite major advances in the last 20 years. The cells comprising the small, localized region of right atrium from which the cardiac impulse originates was discovered in 1907 by English medical student Martin Flack and his mentor Sir Arthur Keith during studies in small mammal hearts (63). They termed this heterogeneous yet distinctive collection of cells the “sino-auricular node” [now the “sino-atrial node” (SAN)] and sparked the deluge of scientific papers investigating why sino-atrial nodal cells (SANC) should behave in such a marvelous way, spontaneously depolarizing 2.8 billion times in the average human lifetime. This review presents an overview of modern concepts of the origin of the heart beat and provides a glimpse into the exciting future of pacemaker research.

Diastolic Depolarization is the Essence of Pacemaking

Spontaneous electrical depolarization during the diastolic phase [“phase 4,” diastolic depolarization (DD); see FIGURE 1] following an action potential is the hallmark of cells comprising pacemaking tissues of the heart (sino-atrial node, atrioventricular node, bundle of His, bundle branches, and His-Purkinje system). Diastolic depolarization causes a slow increase in membrane potential toward an excitation “threshold” (see FIGURE 1), at which point a stereotyped, all-or-nothing action potential fires. The production of repetitive, rhythmical action potentials is the primary responsibility of sino-atrial nodal cells, because their intrinsic rate of diastolic depolarization is (normally) the most rapid. Recent work in the atrioventricular node and Purkinje cells suggests that the mechanisms governing sino-atrial node automaticity are likely to be similar in all pacemaking tissues.

FIGURE 1.

A schematic interplay of different membrane clock and calcium clock mechanisms

An example of a typical action potential of spontaneously beating rabbit SANC (red trace; top). The different phases of the AP are labelled. Special note should be taken of phase 4, representing diastolic depolarization, which is the defining feature of cells capable of pacemaking. Schematic representation of the timing and magnitude of the different components of the “membrane clock” is shown in the middle, whereas the timing and magnitude of the different components of the “Ca²⁺ clock” are shown at the bottom. It can be seen from the bottom (dark blue) that, during phase 4, total cytosolic [Ca²⁺] gradually builds due to accumulation of Ca²⁺ in the form of spontaneous local Ca²⁺ releases emanating from the sarcoplasmic reticulum. Toward the end of diastole, activation of L-type Ca²⁺ channels causes Ca²⁺-induced Ca²⁺ release from the sarcoplasmic reticulum via ryanodine receptors, resulting in the whole cell Ca²⁺ transient. Cytoplasmic Ca²⁺ is then removed by both the SR Ca²⁺ pump, SERCA, and the sarcolemmal sodium-calcium exchanger. MDP, maximum diastolic potential; DD, diastolic depolarization; I_Ca,T, T-type voltage-dependent Ca²⁺ current; I_Ca,L, L-type voltage-dependent Ca²⁺ current; I_NCX, sodium-calcium exchange current; I_K, delayed rectifier potassium current; I_f, funny current; SERCA, sarco-endoplasmic reticulum ATPase; LCRs, local Ca²⁺ releases.

Action potentials generated by pacemaking tissues at appropriate times capture excitable cells of the working atrial and ventricular myocardium, which in turn perform their duty cycle, inducing timely and proportionate expulsion of blood from the heart. Individual sino-atrial nodal cells must continually respond to complex signals generated intrinsically at the subcellular, cellular, and tissue levels, as well as extrinsically from the autonomic nervous system, other hormonal systems, and tissue stretch, and transduce these signals into a beating rate that reflects the integral of the myriad inputs and satisfies the body's demands.

Ion Channels and the Membrane Clock

Since diastolic depolarization is electric in nature (discharging the membrane capacitor), the ensemble of ion channels spanning the sino-atrial nodal cell plasma membrane (sarcolemma) drew attention first in the search for mechanisms underlying spontaneous pacemaking (see FIGURES 1, 2, AND 4). These proteins facilitate selective passage of specific charged species through aqueous pores down concentration gradients created by diverse mechanisms, including exchangers and pumps (50). Ion channels can produce voltage-sensitive membrane currents, which vary in direction and amplitude depending on the membrane potential at a given moment, as well as the ionic constitution of the intra- and extracellular milieu. The currents are activated and then undergo time-dependent decay (see FIGURE 1) due to channel inactivation, with variable times to recovery before they may function once more.

FIGURE 2.

Schematic figure illustrating the cross talk between the membrane clock (right side of schematic SANC) and the Ca²⁺ clock (lying in the center of the schematic SANC)

Online animation reveals schematic behavior of individual ion channels, including the time-dependent nature of the currents that they produce and the coupling of this time- and voltage-dependent behavior to the intracellular Ca²⁺ clock (see animation at Physiology's website). The movie was created by Victor Maltsev.

FIGURE 4.

Schematic diagram of the individual elements of the membrane clock (top right) and the Ca²⁺ clock (bottom left) and how they come together to make up the coupled clock system (middle)

The situation under baseline conditions is shown, whereby constitutively active Ca²⁺-activated ACs produce relatively high levels of cAMP, which go on to activate PKA, in turn leading to relatively high levels of phosphorylation of critical mediators of automaticity belonging to both membrane (I_Ca,L, I_K,r, I_K,s) and Ca²⁺ clocks (RyR, PKA), with the overall result of higher levels of intracellular Ca²⁺. This elevated Ca²⁺ then further activates the ACs, and also calmodulin, leading to activation of CAMKII, and so further phosphorylation of critical mediators of automaticity, with the ultimate result of a further increase in intracellular Ca²⁺; Ca²⁺ release begets Ca²⁺ release. This is only maintained at mid-range by constitutively highly active phosphodiesterases and protein phosphatases (see FIGURE 5B).

Several ionic currents have important roles in spontaneous pacemaking, and because of their time-dependent behavior and sarcolemmal localization, their ensemble behavior has been termed “the membrane clock.”

Decay of I_K conductance (g_K decay)

Spontaneous diastolic depolarization was initially attributed to the decay of I_K2, a theoretical outward current that had been activated by the action potential (see FIGURE 1, purple segments). This gradual decrease in delayed rectifier K⁺ current due to time- and voltage-dependent inactivation of K⁺ channels following an action potential was said to “uncover” theoretical inward (“background”) currents after repolarization (53). The inward current was believed to be mediated by intracellular movement of Na⁺ ions and referred to as I_b,Na, although its molecular basis remains unclear.

The Funny Current

The most notorious ionic current contributing to the membrane clock is the hyperpolarization-activated cation current, or “funny” current (I_f), carried by HCN channels (mainly HCN4 in the human sino-atrial node) (see FIGURE 1, pink segments) (3). This current is in fact a mixed inward Na⁺ and K⁺ current, activated with slow kinetics on hyperpolarization at membrane potential −50 to −65 mV (4, 8). It was originally labeled as funny because ionic currents in cardiac cells are usually activated by depolarization rather than by hyperpolarization.

The role of I_f in pacemaking was initially speculated to be mainly protective (78), shielding central sino-atrial nodal cells from the hyperpolarizing, bradycardic effect of surrounding atrial myocardium. Later, the focus shifted to I_f being “the pacemaker current” (53), controlling basal normal automaticity and contributing to the response of the sino-atrial node to autonomic inputs. Specifically, I_f was proposed to be primarily responsible for the process of diastolic depolarization, moving membrane potential from its nadir following repolarization [the maximum diastolic potential (MDP)] to potentials at which voltage-gated Ca²⁺ channels activate (approximately −40 mV) to complete diastolic depolarization and achieve the “threshold” potential (see FIGURE 1) (40). This purported central role of I_f as a general pacemaking mechanism was questioned when it was shown that some species do not express I_f at all in cardiac regions leading pacemaking [e.g., the bullfrog sinus venosus (61)]. In those species that do express I_f, in vitro blockade of I_f variably increases cycle length between 5 and 20% (11, 46, 76) but never stops pacemaking completely.

I_f-carrying HCN channels are exclusively expressed in pacemaking regions of the heart. The presence of the Hcn4 gene is often utilized as a marker of pacemaking tissue (9). Geographical colocalization of HCN channels with pacemaking regions is taken at times as evidence of I_f's central importance in automaticity (3), although the concept that the mere presence of a channel guarantees its importance in the physiological function of that region is debatable (34).

HCN channel activity has been shown to be modulated by direct interaction with cyclic adenosine monophosphate (cAMP), produced at the sino-atrial nodal cell sarcolemma from adenosine triphosphate (ATP) by both Ca²⁺-activated and Ca²⁺-inhibited adenylate cyclases (ACs; see FIGURE 4). Activation or inhibition of these ACs is a critical pathway by which the sympathetic and parasympathetic arms of the autonomic nervous system control pacemaking rate (see FIGURE 6) (12). cAMP produced as a consequence of β-adrenergic stimulation causes a depolarizing shift in the voltage dependence of the I_f activation curve (13), meaning that more I_f current is available at diastolic potentials, so causing a steeper slope to diastolic depolarization and shorter time between maximum diastolic potential and threshold, i.e., a faster beating rate (so long as the duration of the action potential is assumed to remain constant) (12). Muscarinic receptor activation acts on I_f in the opposite manner (12, 14). The physiological relevance of this rate modulation in vivo has recently been questioned by data demonstrating that a full range of responses to autonomic agonists (affecting both β-adrenoceptors and muscarinic receptors) is possible even in the complete absence of I_f or when I_f function is deficient (1, 19, 21, 22, 37, 60).

FIGURE 6.

Effects of autonomic nervous system activity on the coupled clock system of pacemaking

A: schematic illustration of the effects of adrenergic stimulation on the coupled clock system of pacemaking. In vivo, epinephrine produced from the adrenal glands and norepinephrine from sympathetic nerve terminals binds to β-adrenoceptors, also known as G-protein-coupled receptors. This leads to the Gs subunit activating G-protien-coupled receptor-activated, Ca²⁺-inhibited adenylate cylases (ACs), which produce cAMP from ATP in additional quantities to that being made by constitutively active, Ca²⁺-stimulated ACs. This additional cAMP further activates PKA (and I_f) to lead to an increase in levels of phosphorylation of critical mediators of automaticity belonging to both the membrane and Ca²⁺ clocks. This in turn leads to the overall levels of intracellular Ca²⁺ being higher, with subsequently higher levels of activation of calmodulin and thence CAMKII, with more phosphorylation. The ultimate effect is more rapid cycling of greater amounts of Ca²⁺, and hence an increase in the ticking speed of the coupled clock system. B: schematic illustration of the effects of muscarinic stimulation on the coupled clock system of pacemaking. Acetylcholine (ACh) produced from parasympathetic nerve terminals binds to muscarinic receptors (MR) on SANC (MR also being a form of G-protein-coupled receptors), leading to two predominant effects. The first of these is the direct interaction between the gβγ subunit and I_K,ACh, which leads to an outward, repolarizing K⁺ current. The second effect is that the giα subunit interacts directly with G-protein-coupled receptor-coupled ACs to inhibit these enzymes, causing a decrease in the amount of cAMP produced, with less activation of PKA (and I_f). Less activation of PKA means that levels of phosphorylation fall, with levels of intracellular Ca²⁺ falling too. Less intracellular Ca²⁺ means less activation of CAMKII, so even less phosphorylation. The combined effects of I_K,ACh activation and AC inhibition means a slower rate of ticking of the coupled clock system.

Thus the views on I_f's exact role in the generation of automaticity have shifted with time. I_f has now come full circle and is once more generally viewed as a safety net (provider of “depolarization reserve,” preventing excessive hyperpolarization) (21) for the prevention of excessive bradycardias (78). The more bradycardic a sino-atrial nodal cell, the more important I_f is likely to be, due to its slow activation kinetics. The effects of I_f may be greatest of all in cells with more negative maximum diastolic potentials than sino-atrial nodal cells (e.g., Purkinje cells), since its activation voltage is less than −65 mV, lower than that typically seen in sino-atrial nodal cells.

L-Type Voltage-Dependent Ca²⁺ Channels

The L-type Ca²⁺ current, I_Ca,L, generates the upstroke of the action potential in central sino-atrial nodal cells, where the fast sodium current (usually responsible for this part of the action potential in atria and ventricles) is absent (see FIGURE 1, orange segments). I_Ca,L activates at voltages more negative than the threshold potential (see FIGURE 1), i.e., −50 to −40 mV (18), suggesting additional involvement in later diastolic depolarization as well as in the action potential upstroke. It has been suggested from mouse experiments that mainly L-type Ca²⁺ channels containing the pore-forming α₁-subunit Ca_v1.3 are involved in diastolic depolarization, whereas Ca_v1.2-containing channels are preferentially involved in the depolarizing upstroke of the sino-atrial nodal cell action potential (45).

T-Type Voltage-Dependent Ca²⁺ Channels

The T-type Ca²⁺ current, I_Ca,T, activates at more negative potentials (approximately −60 mV) than its L-type cousin (see FIGURE 1, yellow segments) (47). It has a postulated role in bringing about early diastolic depolarization, based on evidence that loss of I_Ca,T slows spontaneous pacemaking (18).

I_Na,K

This sarcolemmal ATPase pump (see FIGURE 4), extruding three Na⁺ ions and transporting two K⁺ ions intracellularly, generates a net outward, hyperpolarizing current (I_Na,K or I_p) that influences pacemaking by slowing the rate of automaticity (46). In sino-atrial nodal cells, I_Na,K contributes to setting the maximum diastolic potential (55), and I_f-derived Na⁺ ions are considered important in determining the magnitude of I_Na,K due to the exquisite sensitivity of the sodium-potassium pump to fluctuating concentrations of Na⁺ (58).

The Integrated Ion Current Membrane Clock

Using reductionist experimental techniques, it has been possible to define time- and voltage-dependent behaviors of individual ionic currents. Harnessing the resultant ion channel biophysical data, diverse in silico theoretical models of sino-atrial nodal cell behavior have been produced, capable of recapitulating the shapes of spontaneous action potentials. Since the ensemble of surface membrane ion channels can generate spontaneous and self-perpetuating action potentials in silico (76), it was dubbed a “membrane clock” (44). Thus the membrane clock theory of pacemaking proposes that dynamic time- and voltage-dependent interaction of the above membrane-delimited electrogenic ion channels leads diastolic depolarization from maximum diastolic potential to threshold potential, so activating L-type voltage-dependent Ca²⁺ channels to effect the rapid upstroke of the action potential. Repolarization to the maximum diastolic potential is then effected by the array of voltage-dependent K⁺ channels that yield transient outward (I_to), fast delayed rectifier (I_K,r), and slow delayed rectifier (I_K,s) K⁺ currents, after which point the whole process starts again (see FIGURES 1, 2, AND 4).

Intracellular Ca²⁺ Cycling and the Ca²⁺ Clock

The existence of a second independent (yet entrainable) clock generating diastolic intracellular Ca²⁺ releases from a major intracellular store [the sarcoplasmic reticulum (SR); see FIGURE 4, BOTTOM LEFT] proved to be an intriguing hypothesis. Intracellular Ca²⁺ releases in cardiac cells were, for many years, considered to represent pathological phenomena, since they could only be recreated experimentally under conditions of abnormal electrolyte concentrations or toxic doses of digitalis (52).

Initial suspicions regarding the importance of Ca²⁺ signals in sino-atrial nodal cells developed from observations that blocking intracellular SR Ca²⁺ release with the plant alkaloid ryanodine in cat latent atrial pacemaker cells led to a progressive decrease in the rate of the latter part of diastolic depolarization (57). This suggested occurrence of a late diastolic phase of Ca²⁺ release from the SR. Confocal microscopy and Ca²⁺-sensitive fluorescent probes allowed the identification of spontaneous, roughly periodic, diastolic Ca²⁺ releases localized to the subsarcolemmal region of pacemaking cells. These were dubbed “local Ca²⁺ releases” (LCRs), and the timing of their occurrence relative to the prior peak of the action potential-induced whole cell Ca²⁺ transient was referred to as “local Ca²⁺ release period” (see FIGURE 3).

FIGURE 3.

Schematic of computer modeling simulations of the intricate relationship between critical components of the membrane and Ca²⁺ clocks, with respect to phases of the action potential

Two consecutive whole action potentials are illustrated in black at the top. The critical components of the membrane clock are illustrated in blue at the top. I_Ca,L is depicted as an inward current because its waveform is shown as occurring below the isoelectric line. It can be seen that, during the latter parts of diastolic depolarization, I_Ca,L activates and rapidly increases to cause the rapid upstroke of the SANC action potential. I_Ca,L then gradually fades during repolarization until it becomes negligible at the maximum diastolic potential. I_NCX is also depicted as a blue waveform below the isoelectric line, clarifying that it is an inward current. I_NCX functions throughout the cardiac cycle, gradually increasing as diastolic depolarization proceeds and peaking in response to the large increase in I_Ca,L that occurs concomitant with Ca²⁺-induced Ca²⁺ release from the SR. The critical components of the Ca²⁺ clock are illustrated in red at the bottom. SR Ca²⁺ release flux increases during diastolic depolarization as the signal mass of diastolic LCRs grows. This then peaks as there is en masse activation of RyRs due to their interaction with cytosolic Ca²⁺ derived from L-type Ca²⁺ channels, leading to Ca²⁺-induced Ca²⁺ release. SR Ca²⁺ release then falls during repolarization, reaching a nadir at a time similar to (but not the same as) the maximum diastolic potential (Ca²⁺ nadir usually lagging behind the maximum diastolic potential). Ca²⁺ release flux then grows once more as the SR is refilled with Ca²⁺ by SERCA and the threshold for RyR Ca²⁺ release in the form of LCRs is reached. The concept of the “LCR period” is illustrated as the time taken from the peak of the prior action potential-induced Ca²⁺ transient to the timing of first release of LCRs (corresponding to the timing of maximal junctional SR Ca²⁺ content). “Ignition” is the term used to describe the effect of diastolic LCRs that interact with NCX to cause the membrane potential to increase toward the threshold for activation of the L-type Ca²⁺ current, which “ignites” an action potential. Finally, Ca²⁺ concentration within the “junctional” SR is depicted in the lowermost red waveform. Junctional refers to the part of the SR that is closest to the sarcolemma (so-called jSR). jSR Ca²⁺ concentration is at its greatest just before RyRs begin to release diastolic LCRs in early diastole. As the signal mass of LCRs produced increases, so the jSR Ca²⁺ concentration decreases, reaching a nadir when L-type Ca²⁺ channel-derived Ca²⁺ prompts Ca²⁺-induced Ca²⁺ release. Following this, the SR gradually refills, courtesy of the action of SERCA, until it once more reaches a peak in early diastole just before the resumption of production of LCRs. Of greater importance than any of the individual component waveforms in this model is the sum of the complex interaction between them. There is a yin-yang relationship between behaviors of the two clocks. For example, as jSR Ca²⁺ concentration peaks, RyR Ca²⁺ release flux in the form of LCRs also begins to increase, leading to an increase in I_NCX and a quickening of the rate of diastolic depolarization (red arrow). As I_Ca,L reaches its peak just before the peak of the action potential, I_NCX is also reaching its maximum, as is RyR Ca²⁺ release flux (CICR), whereas jSR Ca²⁺ concentration is about to hit rock bottom (blue arrow). These relationships rock back and forth on a beat-to-beat basis to ensure continued automaticity in the coupled clock model.

Local Ca²⁺ releases emerge spontaneously in sino-atrial nodal cells under basal conditions in the form of sparks from SR-based Ca²⁺ release channels, ryanodine receptors (RyRs) (23, 27) (see FIGURES 1 AND 3). The presence of Ca²⁺ overload or sympathetic stimulation is not required for their production. Confocal and two-dimensional fluorescence imaging techniques demonstrate that local Ca²⁺ releases begin to tentatively emerge from the SR following the dissipation of the global, action potential-induced Ca²⁺ transient, as a hypothetical intra-SR Ca²⁺ threshold for spontaneous RyR activation is achieved. During the ensuing diastole, the integrated local Ca²⁺ release Ca²⁺ signal increases at a variable rate, reaching a crescendo immediately before the onrushing action potential. This, in turn, triggers en masse emptying of the SR of its Ca²⁺ (see FIGURE 1), completely engulfing the Ca²⁺ being produced in the form of local Ca²⁺ releases. The SR is then refilled by its ATPase pump SERCA, and the cycle begins once more.

RyRs are critical to the generation of cardiac automaticity; knockout of RyR2 is embryonically lethal in mice (66). Inducible, cardiac-specific knockout of RyR in the more developmentally advanced mouse decreases RyR protein expression by ∼50% and leads not only to features of cardiomyopathy but also to bradycardia and severe arrhythmias (7).

To appreciate the intrinsic behavior of the Ca²⁺ clock, it was necessary to once more adopt a reductionist approach, disengaging it from the membrane clock, preventing ionic/chemical or electrical cross-contamination. Studying the Ca²⁺ clock in isolation, it has become clear that the phenomenon of spontaneous diastolic Ca²⁺ release is an almost rhythmic (1–5 Hz) process, not dependent on the functioning of membrane-bound ion channels or even the presence of the sarcolemma for its perpetuation. Specifically, local Ca²⁺ releases continue in chemically “skinned” sino-atrial nodal cells in which the membrane has been permeabilized, and effectively “removed.” Local Ca²⁺ releases are also capable of continuing to occur in sino-atrial nodal cells with intact sarcolemmal functioning in the presence of voltage clamp at voltages preventing loss of Ca²⁺ from the cell (−10 mV), i.e., voltages where the sodium-calcium exchange current (I_NCX) is negligible (41, 44, 74).

I_NCX

Clarification of the relevance of local Ca²⁺ releases to automaticity came with a critical discovery: inhibiting SR Ca²⁺ release with ryanodine led to a decrease in I_NCX (36), suggesting that local Ca²⁺ release interaction with this membrane-bounded exchanger could affect membrane potential–the SR could “talk to” the sarcolemma (see FIGURE 3)!

I_NCX is generated by the sarcolemmal sodium-calcium exchanger (NCX), a high-capacity, low-Ca²⁺ affinity, voltage-dependent, time-independent electrogenic protein that functions throughout the cardiac cycle (62). NCX's importance to pacemaking was largely overlooked for many years because of its very nature, being an exchanger and not an ion channel. “Forward mode” NCX exchanges one intracellular Ca²⁺ for three extracellular Na⁺ ions (see FIGURE 4), generating net inward current, estimated during diastole at between 9 and 48 pA for a “standard” 30-pF sino-atrial nodal cell (35). This is similar or larger than the magnitude of I_f, suggesting that both mechanisms are important for sino-atrial nodal cell automaticity.

NCX has important features distinguishing it from ion channels. NCX has no specific time-dependent gating or clocking mechanism, its time-varying behavior being derived from the rhythmicity of local Ca²⁺ releases from the SR. This means that, in contrast to I_f, for example, which takes a significant amount of time to activate upon sino-atrial nodal cell repolarization, NCX is instantaneously activated by exposure to intracellular Ca²⁺. As diastolic depolarization proceeds, I_NCX grows as the signal mass of local Ca²⁺ released from the SR increases, whereas nonselective I_f becomes smaller in magnitude as it approaches its reversal potential of approximately −30 mV.

The central tenet of the Ca²⁺ clock is that diastolic local Ca²⁺ releases interact with sarcolemmal NCX to yield an inward current, progressively increasing the rate of diastolic depolarization toward threshold for the firing of an action potential via activation of L-type Ca²⁺ channels (5). The process is aided by tight geographical colocalization of RyR and NCX across the subsarcolemmal gap within sino-atrial nodal cells (38). As previously noted, if I_NCX is rendered negligible by clamping it at its reversal potential (−10 mV), rhythmical local Ca²⁺ releases persist indefinitely because Ca²⁺ loss from the cell is prevented (74). Contrastingly, if sino-atrial nodal cell action potentials are prevented from occurring by voltage clamping the cell at its maximum diastolic potential, rhythmical local Ca²⁺ releases occur only transiently before dying out after a few cycles due to cell Ca²⁺ depletion due to ongoing functioning of NCX (74).

Loss of NCX is devastating; targeted inactivation is embryonically lethal in mice (31). Atrial (including sino-atrial node)-specific NCX knockout in mice causes failure of normal sino-atrial node pacemaking, although these animals survive to adulthood due to the presence of a bradycardic junctional escape rhythm (16, 17).

Coupling of the Clocks → Mutual Entrainment

Recent studies suggest that there are myriad points where the function of the two clocks overlap; these so-called nodes are emphasized in the coupled clock system illustrated in FIGURE 4 (35). These points have assumed increasing importance in pacemaking fidelity as new in silico models of pacemaker function have emerged (42) (see FIGURE 3). It is now accepted by most in the pacemaker field that the two clocks function synergistically in a coupled system involving their mutual entrainment, ensuring robustness alongside reliability of function through multiple functional redundancies. The important coupling nodes are described below.

Interaction of Local Ca²⁺ Release-Derived Ca²⁺ and NCX

NCX has a dual affiliation as a member of both membrane and Ca²⁺ clocks. The link between SR-derived, RyR-generated local Ca²⁺ releases and membrane-bounded NCX is a critical “node” between the SR and sarcolemma (see FIGURES 3 AND 4). If forward mode NCX is blocked (e.g., by acute removal of extracellular Na⁺), sino-atrial nodal cell automaticity abruptly ceases, because the rate of diastolic depolarization is insufficient to activate sarcolemmal I_Ca,L, even though local Ca²⁺ releases persist (5, 59). Similarly, buffering intracellular Ca²⁺ impairs the ability of local Ca²⁺ releases to occur and interact with NCX and can lead to abolition of spontaneous action potential firing rate (38). Crucially, blockade of I_NCX causes failure of pacemaking even when other critical mechanisms of pacemaking, including I_Ca,L, are intact (5) or even enhanced [as illustrated in cases of simulated ischaemia (39)].

L-Type Channel-Derived Ca²⁺ and Ca²⁺-Induced Ca²⁺ Release from the SR

L-type Ca²⁺ channels are also dually affiliated members of both clocks. The Ca²⁺ influx from these ion channels induces en masse Ca²⁺ release from the SR via RyRs (termed Ca²⁺-induced Ca²⁺ release; see FIGURE 4). This resets the SR every cycle in a Ca²⁺-depleted state, preventing SR Ca²⁺ overload and synchronizing (in a depleted state) SR Ca²⁺ content across sino-atrial nodal cells. In time, with SERCA-mediated refilling, SR Ca²⁺ content reaches a threshold level when RyRs will begin to produce local Ca²⁺ releases once more, with subsequent activation of NCX and perpetuation of the next cycle.

The en masse Ca²⁺ depletion of SR caused by Ca²⁺ induced Ca²⁺ release ensures that subsequent timing of local spontaneous RyR activation to generate local Ca²⁺ releases is precise and occurs at a time when interaction with NCX will yield an optimal depolarizing effect to drive membrane potential toward threshold. Through ensuring robust Ca²⁺ influx every cycle, the membrane clock (via both L- and T-type Ca²⁺ channels) guarantees that there will be sufficient Ca²⁺ for SERCA to replenish the SR store, even in the presence of ongoing NCX functioning, which would otherwise deplete the cell's Ca²⁺ stores.

Voltage-Dependent K⁺ Channels and Intracellular Ca²⁺

Sarcolemmal voltage-dependent K⁺ channels that are activated upon depolarization and subsequently effect sino-atrial nodal cell repolarization also affect cell Ca²⁺ balance indirectly. They do this via their hyperpolarizing effect, which activates NCX and inactivates I_Ca,L (35), thereby affecting the amount of Ca²⁺ available for pumping back into the SR, thus forming another node point between membrane and Ca²⁺ clocks. In turn, these K⁺ channels can themselves be affected by levels of cytosolic Ca²⁺ (20, 67), thus illustrating the complexity of cross talk between the two coupled clocks.

The Function of the Two Clocks is Modulated by the Same Biochemical Mechanisms: cAMP, PKA, and CaMKII

A major step forward supporting the coupling of the clocks came with the demonstration that, under baseline conditions, there exists high levels of cAMP in sino-atrial nodal cells, and, further, high levels of phosphorylation of critical proteins belonging to both membrane and Ca²⁺ clocks (high relative to working myocardial cells) (70). ACs, responsible for cAMP generation in sino-atrial nodal cells, are constitutively highly active (FIGURE 4). Some are Ca²⁺ inhibited (77) (AC5 and AC6, also found in ventricular myocytes), whereas others are Ca²⁺ activated (48) (AC1 and AC8). Others (AC2) are Ca²⁺ independent. Ca²⁺ itself also binds to calmodulin, which activates ACs even further (FIGURE 4).

One of the major roles of cAMP in sino-atrial nodal cells is protein kinase A (PKA) activation. PKA, alongside calmodulin-activated Ca²⁺-calmodulin kinase II (CaMKII), phosphorylates L-type Ca²⁺ channels, RyRs, phospholamban, and voltage-dependent K⁺ channels (together referred to as “key protein mediators of automaticity”; see FIGURE 4) (70, 75). Phospholamban is the molecular “brake” on SERCA, and its phosphorylation disengages its inhibitory action. Such high basal levels of phosphorylation of the aforementioned proteins appear critical for pacemaker function: inhibition of either PKA or CAMKII leads to cessation of automaticity (70, 73). The phosphorylation state of phospholamban has been used as an indicator of global sino-atrial nodal cell phosphorylation status (35, 68), since the mechanisms by which it is phosphorylated share a commonality with the mechanisms responsible for the phosphorylation of other key protein mediators of automaticity, i.e., via PKA, CAMKII.

PKA- and CAMKII-mediated phosphorylation promote Ca²⁺ influx via I_Ca,L and lead to more rapid kinetics of SERCA Ca²⁺ pumping, and possibly a lower threshold for release of Ca²⁺ by RyRs, with significant amplification of local Ca²⁺ release signal mass (70). Phosphorylation synchronizes RyR activation (65) and also increases I_K kinetics to speed up repolarization. Higher cytoplasmic levels of Ca²⁺ during Ca²⁺ transients and more substantial diastolic releases further activate Ca²⁺-sensitive adenylate cyclases, which produce more cAMP to further activate PKA, calmodulin, and CAMKII.

This positive feedback system, whereby Ca²⁺ release begets more Ca²⁺ release (FIGURE 5A), is critically kept in check by constitutively high levels of phosphodiesterase (PDE) (71) and protein phosphatase (PP) (84) activity. Together, these constantly antagonize the phosphorylating effects of PKA/CAMKII (FIGURE 5B) (72). The net result is that the level of phosphorylation of proteins of the coupled-clock system in the basal state is naturally set to a level that keeps the action potential firing rate at about the midpoint of its functional range. This allows regulation of the action potential firing rate by gradations in the phosphorylation status (see below). The fact that processes involved in both clocks are regulated by the same systems of phosphorylation and dephosphorylation strongly suggests that the ticking speed of the two clocks will be inextricably coupled in vivo and represents another node point.

FIGURE 5.

The coupled clock system

A: Ca²⁺ release begets Ca²⁺ release. A stripped down schematic figure illustrating how, without highly constitutively active phosphodiesterases and protein phosphatases, the coupled clock system would inexorably accelerate itself toward its fastest rate. B: schematic demonstration of the sites of action of the highly constitutively active phosphodiesterases (PDEs) and protein phosphatases (PPs) that keep the coupled clock system in check, allowing it to operate approximately in the mid-range of its capabilities. Note that protein kinase A phosphorylates and activates PDEs, increasing their activity; this is antagonized by the activity of PPs. Protein kinase A also activates PPs directly. Ca²⁺-calmodulin kinase II phosphorylates and activates protein phosphatase inhibitors, which go on to inhibit PPs, which in turn antagonize the phosphorylating effect of CAMKII on PP inhibitors.

Accelerator and Brake: The Response to the Autonomic Nervous System and the Importance of Timing of Local Ca²⁺ Releases

The phosphorylation status of phospholamban (and by deduction the phosphorylation status of all targets of PKA and CAMKII) can be pharmacologically increased or decreased by substances that mimic the effects of G-protein-coupled receptor signaling (via β-adrenergic receptors of the sympathetic nervous system and muscarinic receptors of the parasympathetic nervous system) to decrease or increase cycle length, respectively, via an increase or decrease in phosphorylation of the critical nodes of the coupled clock system described above in the basal state (35) (FIGURE 6). Such signaling has the remarkable ability to facilitate the wide range of pacing frequencies necessary in vivo in humans, yielding heart rates ranging from 30 to 240 beats/min. The ability of small modifications in phosphorylation status of key mediators of automaticity (as indexed by SR Ca²⁺ pumping rate, reflecting the degree of phosphorylation of phospholamban controlling SERCA behavior) has been modeled in biophysically detailed in silico models of sino-atrial nodal cells, demonstrating that such variability in phosphorylation levels are capable of producing the whole gamut of beating rates seen in vivo (see FIGURE 7).

FIGURE 7.

Computer model simulation of a biophysically detailed SANC

Computer model simulation of a biophysically detailed SANC incorporating coupled membrane and Ca²⁺ clock mechanisms of automaticity, illustrating the wide variety of pacing rates that can be achieved simply by varying the pumping rate (Pup) of Ca²⁺ back into the SR (in vivo, this occurs via the action of the SR ATPase SERCA). Variable SR Ca²⁺ pumping rates are an index of the degree of phosphorylation of the regulatory molecule phospholamban, which acts as a molecular brake on the pumping rate of SERCA. With increasing degrees of phospholamban phosphorylation, the brake on SERCA functioning is released, and Ca²⁺ pumping proceeds at a greater rate. The inverse is true when levels of phosphorylation decrease: rate of Ca²⁺ pumping by SERCA back into the SR decreases. It is likely that the level of phospholamban phosphorylation also reflects the degree of phosphorylation (and therefore activation) of other key mediators of automaticity, including L-type Ca²⁺ channels, RyRs, and delayed rectifier K⁺ channels. The maximum rates of SERCA Ca²⁺ pumping are shown in A. Baseline rates of pumping are shown by the black waveform and are set at rates of 12 mM/s, although they can be varied in the model between 1 and 30 mM/s. A modeled increase in the rate of SERCA Ca²⁺ pumping causes earlier and larger magnitude Ca²⁺ release flux from RyRs (B), meaning that, at faster pumping rates, I_NCX is activated earlier and with a greater magnitude (C). This means that the time to achieve the threshold potential from maximum diastolic potential for the firing of an action potential is progressively shorter. Hence, greater SR Ca²⁺ pumping rates cause progressively shorter cycle lengths (D). The opposite is true for slower rates of SR Ca²⁺ pumping: later occurring, smaller RyR Ca²⁺ release flux, with later occurring, smaller I_NCX, leading to longer periods of time to achieve threshold potential for firing of an action potential, leading to slower beating rates. Such a mechanism would allow a wide variability in heart rate sufficient to satisfy the majority of demands placed on the system by the organism in vivo.

The change in cycle length that occurs as a result of phosphorylation or dephosphorylation of key mediators of automaticity is linearly related to the shortening or lengthening of local Ca²⁺ release period (see FIGURE 8) (35), generating the flexibility required by the sino-atrial node to operate at heart rates across the physiological spectrum and facilitating rapid response to the demands of the body translated through the opposing arms of the autonomic nervous system. The basal behavior of a sino-atrial nodal cell can be conceptualized to be like a car's engine idling at cruising speed, right in the middle of the range of possible velocities it is capable of producing. A small push on the accelerator increases levels of phosphorylation from baseline, ultimately shortening local Ca²⁺ release period and cycle length. A small push on the brake has the opposite effect. The system is tuned to idle in the middle of its operational capabilities, facilitating maximum efficiency and rapidity of responsiveness.

FIGURE 8.

LCR period as the master integrated function of the coupled clock system

Left: the relationship between the amount of phospholamban phosphorylation compared to control and LCR period. It can clearly be seen that decreasing the amount of phospholamban phophorylation relative to that present at control (white square) increases LCR period, and, vice-versa, increasing phospholamban phosphorylation relative to control decreases LCR period. Right: a clear and linear relationship passing through the origin between LCR period and cycle length is shown. Image is adapted from Ref. 35 and used per Circulation Research's permissions policy.

The local Ca²⁺ release period has a unique position at the center of the axes of action of both membrane and Ca²⁺ clocks, and rather than simply describing the behavior of the Ca²⁺ clock in isolation, because of the intricate interrelation of the two clocks through the above-described node points, local Ca²⁺ release period actually summarizes the combined behavior of the membrane and Ca²⁺ clocks working synergistically. Because of this, local Ca²⁺ release period has been described as “the master integrated function” of the coupled clock system (see FIGURE 8) (35).

The local Ca²⁺ release period is not only related to cycle length changes between baseline and following transition to different steady states but also is closely related to cycle length changes on a beat-to-beat basis. Data from rabbit sino-atrial nodal cells (see FIGURE 9, A AND B) have demonstrated that there is beat-to-beat regulation of cycle length that is affected by intracellular Ca²⁺ cycling and reported by local Ca²⁺ release period (82). In these experiments, baseline local Ca²⁺ release period was proportionately related to cycle length on a beat-to-beat basis. When sino-atrial nodal cells are loaded with a Ca²⁺ buffer (FIGURE 9A), the kinetics of cytosolic Ca²⁺ removal markedly slowed, leading to prolongation and irregularity of local Ca²⁺ release period associated with prolongation and irregularity of cycle length (FIGURE 9A, first two cycles exhibiting markedly different duration), and the relationship between local Ca²⁺ release period and cycle length markedly diminished. However, following flash photolysis of the buffer, which causes Ca²⁺ to be released, beating rate increased and became more regular, and the beat-to-beat relationship between local Ca²⁺ release period and cycle length was restored (FIGURE 9A, green arrows).

FIGURE 9.

The critical relationship between local Ca²⁺ release period and cycle length

A: exposure of rabbit SANC to a caged Ca²⁺ buffer causes progressive uncoupling of the tight relationship between LCR period and CL. Note the two long cycles at the start of the trace where LCRs (arrow heads) appear to be randomly scattered between action potentials, which occur at an abnormally slow rate. With UV flashes, Ca²⁺ is released from the caged buffer, and there is a temporary return to rhythmical, more rapidly occurring action potentials, with large and well organized LCRs occurring in the period befopre the upstroke of the action potential. Gradually, the effect of the flash fades, and the Ca²⁺ is once more taken up by the buffer, with a return to slower, more irregular action potentials. Image is from Ref. 77 and used with permission from Physiological Reviews. B: Bi shows simultaneous recordings of membrane potential (top) and confocal linescan measured Ca²⁺ fluorescence (bottom) in a SANC that is acutely released from voltage clamp at −65 mV. The arrow shows the exact moment of release of voltage clamp. It can crudely be seen that there is a progressive increase in beating rate, with a beat-to-beat recovery in LCR signal mass. Bii demonstrates the mean rate of recovery of local Ca²⁺ releases (red squares), cycle length (green diamonds), and diastolic depolarization rate (blue circles) in 5 SANC following acute release of voltage clamp. There is a high degree of correlation between recovery of LCRs and both increasing rate of diastolic depolarization rate (r² = 0.78) and decrease in CL (r² = 0.92). Biii demonstrates the close relationship between LCR period and CL following release of voltage clamp (green triangles). This data is compared with data concerning the relationship between LCR period and CL during spontaneous beating (red squares). Image is from Ref. 69 and used with permission from Biophysical Journal.

Additional evidence of the importance of local Ca²⁺ release period with respect to dictating cycle length during transition from one steady state to another is given in data relating to the effects of the release of sino-atrial nodal cells voltage clamped at −65 mV (FIGURE 9B) (74). The initially quiescent, voltage-clamped cell responds to release of voltage clamp by gradual beat-by-beat amplification of local Ca²⁺ release signal mass (the product of local Ca²⁺ release size, amplitude, and duration; FIGURE 9 Bi) (74). This effect is associated with a gradual increase in diastolic depolarization rate and a concomitant decrease in cycle length until pre-voltage clamp parameters are recapitulated (FIGURE 9 Bii AND Biii) (74).

Ivabradine: A Pure Membrane Clock Antagonist That Also Affects the Ca²⁺ Clock?

The concept of robust coupling between clocks predicts that the bradycardic effects of ivabradine, a pure I_f blocker, will be mediated ultimately by an effect on the Ca²⁺ clock, because blocking I_f will be expected to decrease intracellular Ca²⁺ (a so-called reverse “Bowditch treppe” effect) as action potential firing rate slows, i.e., a direct membrane clock effect and an indirect Ca²⁺ clock effect. Work is presently underway to substantiate this hypothesis and to provide evidence of the inextricable coupling of the clocks (Lakatta et al., unpublished observations).

Novel Insights into Mechanisms of Automaticity

The Role of Regulators of G-Protein Signaling (RGS) Proteins: G_iα, G_iβγ

Regulator of G-protein signaling proteins (or RGS proteins) have recently been discovered in the heart and inactivate G-proteins, resulting in termination of signaling by both G_α and G_βγ proteins (80). They are believed to be critical for regulation of parasympathetic signaling within the heart. More than 20 such proteins exist. Work with RGS4 (10) and RGS6 (80) has confirmed that these proteins are essential for modification of the response to parasympathetic activity via the acetylcholine-activated potassium current I_K,ACh (a critical hyperpolarizing effector in the response to muscarinic receptor activation; see FIGURE 6), and without them (for example, in knockout mouse models) the response to muscarinic G-protein-coupled receptor stimulation is exaggerated.

Adult rabbit sino-atrial nodal cells in culture beat ∼50% slower than freshly isolated sino-atrial nodal cells and show a longer action potential duration alongside lower levels of phospholamban and RyR phosphorylation (79). Cultured cells also have lower levels of RGS2. Adenovirus infection of sino-atrial nodal cells to increase RGS2 expression is capable of partially rescuing the beating rate of cultured sino-atrial nodal cells. Inactivation of G_i signaling by pertussis toxin leads to action potential firing rates close to levels seen in freshly isolated sino-atrial nodal cells. Cultured sino-atrial nodal cell beating rates can also be completely rescued by exposure to isoprenaline (i.e., stimulation of the β-adrenergic G-protein-coupled receptor). These findings suggest that the lower beating rate of cultured sino-atrial nodal cells compared with freshly isolated sino-atrial nodal cells is due to unregulated G_i signaling, with concomitant negative effects on downstream targets for phosphorylation.

The Influence of Sino-Atrial Nodal Cell Mitochondrial Ca²⁺ Buffering on Automaticity

Given the critical position that intracellular Ca²⁺ occupies at the center of the current paradigm of the coupled clock system of pacemaking, cellular functions that interfere with cell Ca²⁺ balance could be expected to affect pacemaking. Sino-atrial nodal cells have a high density of mitochondria (81). Yaniv et al. (83) recently showed that these have a significant role to play in cytosolic Ca²⁺ buffering in rabbit sino-atrial nodal cells. Blocking Ca²⁺ influx into mitochondria via its uniporter decreases levels of intra-mitochondrial Ca²⁺, while leading to an increase in SR Ca²⁺ content and local Ca²⁺ release size/duration/amplitude, and a shortening of local Ca²⁺ release period, with subsequent shortening of cycle length. Blocking mitochondrial Ca²⁺ efflux via mitochondrial NCX has the opposite effect. Notably, blocking I_f has no effect in preventing the effects of interfering with mitochondrial Ca²⁺ cycling on cycle length.

It is also noteworthy that cAMP, PKA, and CAMKII regulate the production of energy in the form of ATP in sino-atrial nodal cell mitochondria (see FIGURE 4), whereas it is Ca²⁺ that regulates ATP production in ventricular myocyte mitochondria (81).

Other cellular organelles that affect cell Ca²⁺ balance will likely affect automaticity in health and disease; further work is required in this area.

Biological Pacemakers

The quest to replace electronic pacemakers with a biological substitute continues to proceed down many diverse avenues. Based on the ideas discussed above, a successful biological pacemaker design would likely target the integral function of the entire system rather than one particular component (the natural robust pacemaker system has multiple redundant mechanisms; see FIGURES 1–3). Approaches that simplistically attempt to address only a single part of the system [such as targeted upregulation of I_f (49, 56) or downregulation of I_K1 (49)] have turned out to be insufficient. HCN2 gene transfer to upregulate I_f, for example, leads only to unacceptably slow idioventricular rates and moderate autonomic responsiveness (56).

An approach more likely to succeed involves upregulating the whole Ca²⁺ clock in working cardiomyoytes, converting them into pacemaking cells by concomitantly activating the coupled membrane clock. Increasing expression of Ca²⁺-activated ACs is one strategy to achieve this. Indeed, such a strategy has already been used to generate automaticity in neonatal rat ventricular myocytes (32), embryonic stem cell-derived cardiomyocytes (85), and P19-derived pacemaker-like cells (43). Direct addition of cAMP to embryonic stem cell-derived stem cells has also demonstrated flexibility and rhythmicity in pacemaking (85). In these cells, blocking the effects of cAMP and therefore clock coupling with PKA inhibitors causes rhythmicity to disappear (85). Adenoviral introduction of Ca²⁺-activated AC1 into the left bundle branches of atrio-ventricular blocked canines has been shown to be capable of generating sino-atrial nodal cell-like automaticity (6). An alternative approach that also awakens the coupled clock system in non-pacemaking cells is the reactivation of the sino-atrial nodal cell-type gene programme within ventricular myocytes by transcription factors that have been shown to have a key role to play in the embryogenesis of the sino-atrial node: both tbx18 in neonatal rat ventricular myocyte (29, 30) and TBX3 in mature murine cardiomyocytes (2) have shown promise in this regard.

Future Focuses

Each step forward in pacemaker research over the years has subsequently generated more questions than it succeeded in answering. Therefore, despite a hitherto unprecedented degree of biophysical appreciation of pacemaking mechanisms, we continue to face challenging questions in need of intensive further investigation, the most pressing of which are summarized below. One important thing, however, becomes crystal clear: the cardiac pacemaker is a complex system of numerous interacting components localized both to the cell membrane and within the cell. How robust and flexible pacemaking within such a complex system emerges needs further detailed experimental studies, as well as development and application of new approaches of systems biology and complex systems.

Human aging is well known both in vivo and in vitro to be associated with dysfunction of the sinoatrial node (51), along with the rest of the cardiac conduction system, leading to the excessive requirement for implantation of permanent pacing systems in the elderly. The exact mechanisms by which age-related sino-atrial nodal dysfunction occurs, however, remain unclear, although many have been implicated. These include an age-related cell-to-cell disconnection due to decreased connexin 43 expression (26) and a downregulation of I_Ca,L (25). Aging in murine sino-atrial nodal cells has more recently been associated with a decrease in SR Ca²⁺ cycling and responsiveness to phosphodiesterase inhibition (64). Similar changes in Ca²⁺ handling mechanisms have been shown in rodent working cardiomyocytes (24), where there is an established age-related increase in myocyte size, decrease in myocyte number, and increase in matrix connective tissue. Future work to completely elucidate the electrophysiological basis of the age-related remodeling of the sino-atrial node that contributes ultimately to its dysfunction will be important in defining potential new therapeutic measures to retard or prevent such undesirable change, with the ultimate aim of subverting the need for any artificial pacemaker systems, biological or not.

The exact molecular mechanism of Ca²⁺ release by RyRs remains unclear: How does intra- and extra-SR Ca²⁺ influence RyR open probability? Do all RyRs have the same threshold for release of local Ca²⁺ releases, or is this heterogeneous? Is there local variability in Ca²⁺ concentration within the SR that makes certain RyRs more likely to open and release Ca²⁺? Is there local variation in extra-SR cAMP that makes certain RyRs more likely to open? Are certain areas of the cell more likely to exhibit local Ca²⁺ releases than others, and, if so, what is it that controls the generation of these “hotspots”? By what process does stochastic opening and closing of RyRs become synchronized to generate roughly periodic local Ca²⁺ releases? Is there local RyR cross talk, and, if so, by what mechanism?

Protein kinase C (PKC) is becoming recognized as an important factor that influences the phosphorylation status of critical mediators of automaticity in sino-atrial nodal cells, alongside the well characterized behaviors of PKA and CAMKII. Pilot studies have shown that blocking PKC activity leads to a 2.5-fold decrease in the beating rate of rabbit sino-atrial nodal cells (69), with effects on local Ca²⁺ release characteristics and timing that are likely to be mediated by the loss of the effect of PKC on the combination of L-type Ca²⁺ channels, SERCA, and RyR. Exactly how the behavior of PKC fits in with the much more established actions of PKA is the focus of ongoing work.

Newly discovered mechanisms of regulating sino-atrial nodal cell Ca²⁺ balance have generated much recent interest. Sarcolemmal store-operated Ca²⁺ channels (SOCCs) have the ability to affect cell Ca²⁺ balance through a poorly characterized ability to detect the level of Ca²⁺ present within intracellular stores (28), with low detected levels leading to an increase in Ca²⁺ flux intracellularly. Blocking store-operated Ca²⁺ channel activity slows pacemaking in isolated murine sino-atrial node by 65% without affecting L-type Ca²⁺ channels (28). Their contribution to pacemaking automaticity requires further elucidation.

A more complete definition of the role of I_Na,K in sino-atrial nodal cells and a complete characterization of the importance of the regulatory protein phospholemman is required. It is recognized that, in working cardiomyocytes, phospholemman is the major target for PKA- and PKC-mediated phosphorylation and that phosphorylation leads to removal of its inactivating effects on the sodium-potassium ATPase (15). Phospholemman inhibits NCX when phosphorylated (86) and has a similar effect on L-type Ca²⁺ channels (15). The implications of the behavior of I_Na,K and phospholemman in pacemaking cells requires clarification, as does the regulation of cellular Na⁺.

Summary

The present paradigm of sino-atrial nodal cell automaticity involves a coupled clock system, whereby sarcolemmal proteins of the electrical membrane clock interact with the intracellular Ca²⁺ cycling apparatus making up the chemical Ca²⁺ clock and vice versa, with the behavior of each individual clock's components constantly refining and responding to the behavior of the components of the other clock (mutual entrainment). By behaving in this way, one clock guarantees ongoing, accurate, and sufficient functioning of its counterpart and vice versa, and in doing so ensures that this most vital of life's foundation stones is optimally robust and reliable. The diverse nature of the clocks offers heterogeneous redundancy at the single sino-atrial nodal cell level, whereas the diverse characteristics of the sino-atrial nodal cell population comprising the sino-atrial node offers a further layer of heterogeneous redundancy, maximizing robustness. Previous reductionist approaches to pacemaker investigation unwittingly has divorced the clocks, treating them as “asymmetrically” important (54), i.e., feuding partners with one party more important than the other. Such approaches suggested that one clock might exist perfectly well without the other, when all along the happy marriage and symmetrical importance of individual clock functions is what makes the system behave in such an elegantly harmonious manner. Certain crucial mechanisms responsible for automaticity at baseline are simply scaled up or down by phosphorylation/dephosphorylation in response to the G-protein-coupled receptor signaling employed by the autonomic nervous system.

A General Theory of Cardiac Function has Emerged

A general theory of the heart's function has emerged from the concurrent study of sino-atrial nodal cells alongside cells of the working myocardium (see FIGURE 10). It has long been recognized that, as the heart beats faster, it beats stronger [the so-called Bowditch effect (33)]. But, surprisingly, it was only recently realized that these diverse cell types employ similar apparatuses to achieve, first, increased chronotropy, and then increased inotropy. Specifically increased chronotropy is achieved via increased sino-atrial nodal cell Ca²⁺ cycling kinetics via a cross talk of SERCA (SR Ca²⁺ uptake) → RyR (Ca²⁺ release) → NCX (inward current during diastole and Ca²⁺ efflux) → L-type Ca²⁺ channels (action potential and Ca²⁺ influx). Increased inotropy is achieved also via increased Ca²⁺ cycling kinetics utilizing a cross talk of the very same molecules: L-type Ca²⁺ channels (Ca²⁺ influx) → RyR (Ca²⁺ release) → SERCA (SR Ca²⁺ uptake) → NCX (Ca²⁺ efflux). Through exploration of this general theory, our appreciation of the physiology and pathophysiology of cardiac automaticity might allow ongoing fascinating insights into this most critical cardiac function.

FIGURE 10.

A general theory of cardiac function, involving similar mechanisms in SANC and ventricular myocytes

Roughly periodic diastolic Ca²⁺ releases in the form of LCRs in SANC interact with sarcolemmal NCX to produce an inward current that increases the rate of diastolic depolarization toward the threshold potential for activation of L-type Ca²⁺ channels. Ca²⁺ influx through these channels depolarizes SANC and activates RyRs to produce en masse emptying of the SR in a process known as CICR and leads to the action potential-induced Ca²⁺ transient. Depolarization of SANC spreads via cell-to-cell contacts, leading to depolarization of atrial myocytes and, following conduction via the atrioventricular node, to depolarization of ventricular myocytes. This electrical process involves influx of Na⁺ ions into ventricular myocytes via fast sodium channels. Depolarization subsequently causes influx into ventricular myocytes of Ca²⁺ via L-type Ca²⁺ channels, which again activate SR-based RyRs, leading to en masse emptying of the SR of its Ca²⁺, which binds to the sarcomere to cause shortening and the ultimate goal of cardiac inotropism. In this schema, the chemical signal of Ca²⁺ release from the SR in SANC initiates an electrical signal, which is passed to ventricular myocytes. In ventricular myocytes, electrical depolarization then initiates a chemical signal (intracellular Ca²⁺ release) to produce the ultimate goal: a timely and efficient mechanical oscillation, i.e., a heart beat. Thus heart rate (chronotropy) controlled by SANC leads to execution of a heart beat by ventricular myocytes by utilizing an almost identical set of molecules behaving in the same, albeit differently ordered, manner, a general theory of cardiac function. Increasing or decreasing the rate of activity of SANC in turn increases or decreases the force of contraction of ventricular myocytes via phosphorylation/dephosphorylation of the same molecules (phospholamban, RyR, L-type Ca²⁺ channels, and delayed rectifier K⁺ channels) via the same mechanisms involving PKA and CAMKII in these two neighboring yet physiologically diverse cell types (see supplemental movie at Physiology's website; the movie was created by Victor Maltsev).