Local calcium gradients during excitation–contraction coupling and alternans in atrial myocytes

Abstract

Subcellular Ca²⁺ signalling during normal excitation-contraction (E-C) coupling and during Ca²⁺ alternans was studied in atrial myocytes using fast confocal microscopy and measurement of Ca²⁺ currents (I_Ca). Ca²⁺ alternans, a beat-to-beat alternation in the amplitude of the [Ca²⁺]_i transient, causes electromechanical alternans, which has been implicated in the generation of cardiac fibrillation and sudden cardiac death. Cat atrial myocytes lack transverse tubules and contain sarcoplasmic reticulum (SR) of the junctional (j-SR) and non-junctional (nj-SR) types, both of which have ryanodine-receptor calcium release channels. During E-C coupling, Ca²⁺ entering through voltage-gated membrane Ca²⁺ channels (I_Ca) triggers Ca²⁺ release at discrete peripheral j-SR release sites. The discrete Ca²⁺ spark-like increases of [Ca²⁺]_i then fuse into a peripheral ‘ring’ of elevated [Ca²⁺]_i, followed by propagation (via calcium-induced Ca²⁺ release, CICR) to the cell centre, resulting in contraction. Interrupting I_Ca instantaneously terminates j-SR Ca²⁺ release, whereas nj-SR Ca²⁺ release continues. Increasing the stimulation frequency or inhibition of glycolysis elicits Ca²⁺ alternans. The spatiotemporal [Ca²⁺]_i pattern during alternans shows marked subcellular heterogeneities including longitudinal and transverse gradients of [Ca²⁺]_i and neighbouring subcellular regions alternating out of phase. Moreover, focal inhibition of glycolysis causes spatially restricted Ca²⁺ alternans, further emphasising the local character of this phenomenon. When two adjacent regions within a myocyte alternate out of phase, delayed propagating Ca²⁺ waves develop at their border. In conclusion, the results demonstrate that (1) during normal E-C coupling the atrial [Ca²⁺]_i transient is the result of the spatiotemporal summation of Ca²⁺ release from individual release sites of the peripheral j-SR and the central nj-SR, activated in a centripetal fashion by CICR via I_Ca and Ca²⁺ release from j-SR, respectively, (2) Ca²⁺ alternans is caused by subcellular alterations of SR Ca²⁺ release mediated, at least in part, by local inhibition of energy metabolism, and (3) the generation of arrhythmogenic Ca²⁺ waves resulting from heterogeneities in subcellular Ca²⁺ alternans may constitute a novel mechanism for the development of cardiac dysrhythmias.

In the heart, the process of excitation-contraction (E-C) coupling links electrical activation to the mechanical activity of cardiac muscle cells (for review see Bers, 2001). The process of E-C coupling refers to the mechanism by which membrane depolarization (i.e. action potentials, APs), and subsequent Ca²⁺ entry (I_Ca) through L-type Ca²⁺ channels (or dihydropyridine receptors, DHPRs) triggers Ca²⁺ release from the sarcoplasmic reticulum (SR), ultimately causing contraction of the myocyte. Ca²⁺ entering through DHPRs triggers intracellular Ca²⁺ release by activating calcium-sensitive Ca²⁺ release channels (ryanodine receptors, RyRs) in the SR membrane. This mechanism is termed calcium-induced Ca²⁺ release (CICR; Fabiato, 1983). Through CICR, a relatively small amount of Ca²⁺ entering through voltage-gated sarcolemmal Ca²⁺ channels elicits Ca²⁺ release from individual clusters of calcium-sensitive RyRs, which in turn dramatically amplifies further Ca²⁺ release by activating additional calcium-sensitive RyRs. Relaxation of cardiac cells is critically dependent upon mechanisms that lower [Ca²⁺]_i through reuptake into the SR and/or extrusion across the sarcolemma. Reuptake of Ca²⁺ provides the necessary filling of the SR to allow sufficient Ca²⁺ for release during the next contraction or, at the level of the intact heart, for the next heart beat.

In ventricular myocytes the presence of a well-developed three-dimensional network (Soeller & Cannell, 1999) of surface membrane invaginations (transverse tubular network, t-tubules; see Fig. 1A) ensures close proximity of SR release channels with surface membrane DHPRs throughout the entire cell volume. RyRs are organized in clusters in the SR membrane and DHPRs are contained within the t-tubular membranes. The localized apposition of these two membranes and their respective receptor channels form diads or triads, the fundamental functional entities for E-C coupling (Sun et al. 1995; Flucher & Franzini-Armstrong, 1996; Franzini-Armstrong & Protasi, 1997). Release from individual clusters of RyRs has been visualized using laser-scanning confocal microscopy as Ca²⁺ sparks in ventricular (Cheng et al. 1993) and atrial myocytes (Hüser et al. 1996; Blatter et al. 1997). Ca²⁺ sparks are thought to represent elementary events of Ca²⁺ release in cardiac cells. According to the ‘local control’ model of cardiac E-C coupling (Stern, 1992; Rios & Stern, 1997) Ca²⁺ sparks are recruited independently from each other and represent the building blocks of larger, more global Ca²⁺ signals. For example, AP-induced [Ca²⁺]_i transients or propagating Ca²⁺ waves result from the temporal and spatial summation of elementary Ca²⁺ release events (Cannell et al. 1994, 1995; Lopez-Lopez et al. 1994). In ventricular cells the presence of t-tubules results in the close physical association of virtually all SR Ca²⁺ release units (RyRs) with DHPRs in the surface membrane (Scriven et al. 2000). This ensures simultaneous activation of SR Ca²⁺ release throughout the entire ventricular myocyte during an AP. This arrangement contrasts sharply with the spatiotemporal organization of the [Ca²⁺]_i transient in atrial myocytes.

Figure 1. Ca²⁺ signalling during excitation-contraction (E-C) coupling in cat atrial myocytes.

A, confocal images of a ventricular (left) and an atrial (right) myocyte from the same cat heart stained with the membrane-bound fluorescent dye Di-8-ANEPPS. The regular structures spaced in a sarcomeric pattern in the ventricular cell represent t-tubules. In contrast, the atrial myocyte is devoid of any t-tubular staining. B, [Ca²⁺]_i transient recorded in the confocal line-scan mode (scanning frequency was 250 Hz). The scanned line was positioned perpendicular to the longitudinal axis of the cell (c). Electrical stimulation of the cell during acquisition of the line-scan image triggered a ‘U’-shaped [Ca²⁺]_i transient (b), indicating that [Ca²⁺]_i increased first at the periphery of the cell (a) before propagating towards the centre of the myocyte. Panel d shows local [Ca²⁺]_i transients (top) measured in the subsarcolemmal space (ss) and the centre of the cell (ct) as well as averaged over the entire width of the cell (bottom). The arrow indicates the two components of the whole-cell [Ca²⁺]_i transient (see text). C, spatiotemporal pattern of an action potential (AP)-induced atrial [Ca²⁺]_i transient visualized by two-dimensional confocal microscopy using a Nipkow tandem-disk confocal unit in conjunction with a microchannel plate-intensified CCD camera. The images represent cellular fluo-4 fluorescence recorded at a temporal resolution of 60 Hz. [Ca²⁺]_i transients were elicited by extracellular electrical field stimulation. ‘0 ms’ refers to the image immediately preceding the first changes in fluo-4 fluorescence. [Ca²⁺]_i signals in B and C were recorded from different atrial myocytes. (A and B are modified from Hüser et al. 1996.)

Ca²⁺ signalling during E-C coupling in atrial myocytes

In atrial myocytes, the mechanism of E-C coupling and [Ca²⁺]_i regulation appears to be more complex. Ultrastructural studies (McNutt & Fawcett, 1969) have shown that atrial cells lack t-tubules (see Fig. 1A) and exhibit two types of SR membranes: junctional (j-SR) and non-junctional (nj-SR). These terms refer to SR membranes located at the cell periphery in close apposition to the surface membrane, and those located more centrally within the cell interior, respectively. RyRs are found anchored in the membrane of both types of SR (Jorgensen et al. 1993; Lewis Carl et al. 1995) and are therefore likely to participate in atrial E-C coupling (Hatem et al. 1997; Kockskämper et al. 2001). These ultrastructural differences with ventricular myocytes have direct consequences for [Ca²⁺]_i regulation and E-C coupling in atrial myocytes. High temporal and spatial resolution laser-scanning confocal microscopy in conjunction with highly sensitive fluorescent Ca²⁺ indicators has provided novel and exciting insights into the dynamics of Ca²⁺ signalling in atrial myocytes. The key features of atrial Ca²⁺ signalling during E-C coupling are illustrated in Fig. 1B and C. Figure 1B shows the spatial patterns of an AP-induced [Ca²⁺]_i transient in an atrial myocyte, as revealed by recordings obtained in the confocal line-scan mode (x-t confocal imaging). The cell was scanned repetitively in a transverse direction (i.e. perpendicular to the longitudinal axis of the cell; Fig. 1Bc). When electrically stimulated by an AP, [Ca²⁺]_i initially increased immediately under the surface membrane, without any rise of [Ca²⁺]_i in the centre of the cell (Fig. 1Ba). Subsequently, the peripheral region of high [Ca²⁺]_i expanded towards the centre of the myocyte causing the characteristic ‘U’ shape of the [Ca²⁺]_i transient in the line-scan image (Fig. 1Bb). The [Ca²⁺]_i gradients associated with E-C coupling in the atrial cell are also prominent in Fig. 1Bd (top), which shows the time course of [Ca²⁺]_i at a subsarcolemmal and a central location within the cell. The local [Ca²⁺]_i transient measured at the cell periphery (subsarcolemmal, ss) displayed a rapid upstroke and started to decrease before the [Ca²⁺]_i transient recorded in the centre (ct) reached its maximum. These spatial inhomogeneities are also reflected in the [Ca²⁺]_i profile averaged over the entire width of the cell (Fig. 1Bd, bottom), which shows two components (arrow), an initial fast rise of [Ca²⁺]_i followed by a slower further increase of [Ca²⁺]_i. The secondary component is primarily due to release of Ca²⁺ from central regions of the cell (i.e. from nj-SR).

Even though the confocal line-scan imaging technique allows the recording of cardiac Ca²⁺ signals at high temporal resolution, the spatial information is limited since imaging is restricted to a single spatial dimension within the three-dimensional structure of the cell. To gain further insight into the spatial organization of Ca²⁺ signalling during E-C coupling in atrial cells, a fast (60 frames s⁻¹), two-dimensional confocal imaging technique was applied, using a Nipkow tandem-disk confocal unit (Genka et al. 1999; Kockskämper et al. 2001) in conjunction with a microchannel plate-intensified CCD camera. The characteristic two-dimensional features of atrial AP-induced [Ca²⁺]_i transients are illustrated in Fig. 1C, which shows a series of two-dimensional images of the resulting [Ca²⁺]_i transient at a central axial depth of the cell. Electrical stimulation of the myocyte resulted in an increase in [Ca²⁺]_i that started at discrete sites along the periphery. These sites subsequently fused to form a subsarcolemmal ‘ring’ of elevated [Ca²⁺]_i that propagated to the cell centre. Marked inhomogeneities of the [Ca²⁺]_i signals were observed during this process both in the periphery and in the cell centre. The detailed analysis of these [Ca²⁺]_i images revealed the following key features (Kockskämper et al. 2001).

1. APs trigger release of Ca²⁺ from distinct subsarcolemmal calcium release sites.

2. These sites are spaced ≈2 μm apart and are located directly beneath the cell membrane, and therefore represent calcium release sites from j-SR.

3. Local calcium release fluxes are highest from j-SR, identifying them as a localized source of Ca²⁺ release.

4. Diffusion of Ca²⁺ away from the peripheral release sites activates, in a time-dependent fashion, additional release sites located in the membranes of nj-SR. This results in a centripetal propagation of Ca²⁺ release from the cell periphery towards the cell centre by the mechanism of CICR.

5. Mapping of individual release sites within the confocal imaging plane reveals a two-dimensional grid of release sites that is consistent with the three-dimensional distribution of RyR calcium release channels visualized by immunocytochemistry.

6. Combining the whole-cell voltage-clamp technique with confocal [Ca²⁺]_i imaging shows (Sheehan & Blatter, 2003) that Ca²⁺ release from j-SR, in contrast to release from nj-SR, is under tight control of Ca²⁺ entry through L-type Ca²⁺ channels. In these experiments, depolarizing voltage steps from a holding potential of −40 mV to a test potential of +10 mV elicits typical L-type Ca²⁺ currents and the characteristic spatial inhomogeneities of the [Ca²⁺]_i transient identical to those shown in Fig. 1B. Ca²⁺ release from the j-SR is terminated rapidly by deactivation of Ca²⁺ influx (repolarization of the membrane to −40 mV during the rising phase of the subsarcolemmal [Ca²⁺]_i transient) or by reducing the driving force for Ca²⁺ entry (further depolarization of the membrane to +100 mV). In contrast, Ca²⁺ release from nj-SR continues to develop once initiated, and is independent of the duration of membrane depolarization.

7. When the intracellular Ca²⁺ buffer capacity is increased by intracellular dialysis with the exogenous Ca²⁺ buffer EGTA, the propagation of CICR from j-SR to nj-SR is blocked, further supporting the notion that release of Ca²⁺ from nj-SR depends on the preceding elevation of [Ca²⁺] in the subsarcolemmal space (Sheehan & Blatter, 2003).

A number of detailed studies have helped to elucidate the complex spatiotemporal pattern of [Ca²⁺]_i signalling during E-C coupling in atrial myocytes, and to depict a model of E-C coupling. This model of atrial E-C coupling reveals distinct differences compared to ventricular cells, resulting primarily from the lack of t-tubules in atrial myocytes. For example, in earlier work, Lipp et al. (1990) simultaneously measured the Na⁺-Ca²⁺ exchange current (I_Na-Ca) and bulk [Ca²⁺]_i in atrial myocytes. They postulated that I_Ca triggered Ca²⁺ release from the peripheral SR, which in turn caused further Ca²⁺ release from deeper regions of the SR. Using high-resolution calcium-imaging techniques, Berlin (1995) and Hüser et al. (1996) demonstrated that in guinea-pig and cat atrial myocytes, respectively, APs induced by electrical stimulation elicited highly inhomogeneous [Ca²⁺]_i transients, as shown in Fig. 1B. Subsequent studies obtained similar results in rat atrial myocytes (Tanaka et al. 2001), indicating that these distinct [Ca²⁺]_i gradients are typical for E-C coupling in atrial muscle. Additional studies by Hatem et al. (1997) showed that in human atrial myocytes, [Ca²⁺]_i transients have two components that are not entirely controlled by I_Ca, suggesting that Ca²⁺ release from a subpopulation of RyRs, which are not coupled to L-type Ca²⁺ channels, contributes to Ca²⁺ release during E-C coupling. More recent studies in cat (Kockskämper et al. 2001) and rat (Mackenzie et al. 2001) atrial myocytes using rapid two-dimensional Ca²⁺ imaging provided detailed novel insights into the differential regulation of Ca²⁺ release from j-SR and nj-SR. Both studies demonstrated that electrically evoked [Ca²⁺]_i transients started at distinct locations in the periphery. Release from these subsarcolemmal sites subsequently fused to form a peripheral ring of elevated [Ca²⁺]_i. In cat atrial myocytes, release propagated in a saltatory fashion from the periphery towards the cell centre as a result of CICR from nj-SR release sites. Ca²⁺ release from the central nj-SR was somewhat smaller than from the subsarcolemmal j-SR and reached ≈75 % of the amplitude recorded in the subsarcolemmal space (although in some cells central release slightly exceeded peripheral release). The robust Ca²⁺ release from central SR compartments observed in cat and guinea-pig atrial myocytes (Berlin, 1995; Hüser et al. 1996, 2000; Kockskämper et al. 2001) differs from the results obtained in rat atrial cells where, at least under basal conditions, no significant Ca²⁺ release from the central nj-SR occurred (Mackenzie et al. 2001). Additional differences in E-C coupling between rat and cat atrial myocytes involve the activation order of Ca²⁺ release sites of the j-SR. In cat cells, the magnitude of Ca²⁺ release from single j-SR sites as well as the activation order of individual sites varied significantly during consecutive electrical stimulations, suggesting that Ca²⁺ release from these sites was triggered by stochastic openings of sarcolemmal DHPRs. This is at variance with the observations made in rat cells, where a subpopulation of ‘eager’ release sites responded in a reproducible fashion to repetitive depolarizations. The ‘eager’ sites were the first to show a [Ca²⁺]_i increase upon stimulation and had the highest frequencies of spontaneous Ca²⁺ sparks. Based on these observations, the authors speculated that the strict activation sequence of the subsarcolemmal ‘eager’ release sites may be due to a higher intrinsic Ca²⁺ sensitivity of those sites and their close coupling to adjacent DHPRs (Mackenzie et al. 2001). The spatiotemporal inhomogeneities of the AP-induced [Ca²⁺]_i transient do not appear to be unique to atrial myocytes. Patterns similar to those recorded in atrial muscle have been demonstrated in cardiac Purkinje cells (Cordeiro et al. 2001) and neonatal cardiac myocytes (Haddock et al. 1999), both cell types that, to a large extent, lack a t-tubular system, and even in ventricular myocytes that underwent a reduction of the t-tubular system (Lipp et al. 1996).

With regard to Ca²⁺ signalling and E-C coupling, recent findings have rekindled the interest in the role of inositol-trisphosphate (IP₃) signalling cascade and IP₃-dependent Ca²⁺ release in atrial myocytes. Although it is generally believed that, particularly in ventricular cells, IP₃-induced Ca²⁺ release plays a merely modulatory role in E-C coupling (for reviews see Marks, 2000; Bers, 2001), the situation in the atria might be different. Evidence is accumulating that atrial myocytes express functional IP₃ receptors (IP₃Rs) at higher levels than in ventricular myocytes, and IP₃Rs colocalize with RyRs in the subsarcolemmal space (Lipp et al. 2000). Atrial myocytes have a higher IP₃ turnover rate and higher levels of IP₃ (Woodcock et al. 1995), and IP₃-dependent Ca²⁺ release has been shown to enhance Ca²⁺ spark frequency and twitch [Ca²⁺]_i transient amplitude in atrial myocytes (Lipp et al. 2000). Furthermore, it has been suggested that IP₃-induced Ca²⁺ release may play a crucial role in atrial arrhythmias by facilitating the generation of delayed afterdepolarizations and premature APs (Mackenzie et al. 2002). Thus it seems likely that IP₃-dependent Ca²⁺ signalling may play an important, but yet undetermined role in atrial E-C coupling under physiological as well as pathological conditions (see e.g. Woodcock et al. 1998).

Given the potential role of IP₃-dependent Ca²⁺ signalling in the atria, in summarizing the available data, the following model of E-C coupling is proposed for atrial myocytes. Two distinct mechanisms of SR Ca²⁺ release coexist during normal E-C coupling, both of which depend on a form of CICR. Ca²⁺ release is activated initially from the j-SR located in the subsarcolemmal region of the cell and is triggered by Ca²⁺ influx (I_Ca) through L-type Ca²⁺ channels. Since gating of L-type Ca²⁺ channels is controlled by voltage, Ca²⁺ release from junctional sites is tightly controlled by membrane voltage (similar to ventricular cells). The subsarcolemmal elevation of [Ca²⁺]_i provides enough Ca²⁺ for subsequent diffusion towards the closest nj-SR release sites where it activates additional Ca²⁺ release. Once initiated, Ca²⁺ release from nj-SR then proceeds with little further dependence on membrane Ca²⁺ influx (and thus membrane voltage) via propagating CICR in a saltatory, centripetal fashion. Propagation of CICR from nj-SR behaves in a nearly all-or-none fashion and is reminiscent of the Ca²⁺ waves observed in ventricular cells under Ca²⁺ overload conditions (Takamatsu & Wier, 1990; Wier & Blatter, 1991). In this way, the signal for Ca²⁺ release and subsequent contraction is transmitted reliably from the sarcolemma throughout the atrial myocyte in the absence of t-tubules.

Electromechanical and Ca²⁺ alternans in the heart: failure of E-C coupling?

During electrical excitation of cardiac myocytes at constant frequency [Ca²⁺]_i transients are fairly uniform in magnitude and kinetics. Abrupt variations of the excitation patterns (change in frequency, pause in stimulation, extra beats) typically result in beat-to-beat changes of the [Ca²⁺]_i transient characteristics and cause phenomena such as positive and negative staircase, post-rest potentiation or post-extrasystolic potentiation (for review see Bers, 2001). These variations are commonly explained by changes in the available amount of releasable Ca²⁺ within the SR. In contrast, cardiac alternans describes the cyclic, beat-to-beat variations in contraction amplitude (mechanical alternans) and AP duration (electrical alternans) at constant stimulation frequency (e.g. Wolfahrt, 1982). Alternans occurs in cardiac failure (e.g. Dumitrescu et al. 2002b) during myocardial ischaemia, and is believed to be an important factor in the pathogenesis of arrhythmias (Dilly & Lab, 1988; Konta et al. 1990) and the development of re-entry phenomena (Rubenstein & Lipsius, 1995; Berger, 2000; Pastore & Rosenbaum, 2000). Clinically, alternans is manifest as pulsus alternans and specific changes in the ST segment of the electrocardiogram (e.g. Uno, 1991; Surawicz & Fish, 1992; Lab & Seed, 1993; Pastore et al. 1999). T-wave alternans has been linked to ventricular fibrillation and sudden cardiac death under various conditions including acute myocardial ischaemia (Smith et al. 1988; Verrier & Nearing, 1994) and long QT syndrome (Shimizu & Antzelevich, 1999). In cardiac muscle preparations or isolated myocytes, alternans can be induced experimentally by pacing (for references see review by Euler, 1999), during acidosis (Orchard et al. 1991), low extracellular Ca²⁺ concentration (Badeer et al. 1967), hypothermia (e.g. Spencer et al. 1992) and, as demonstrated most recently (Hüser et al. 2000), by direct interference with cellular energy metabolism through inhibition of glycolysis (see also below). Furthermore, β-adrenergic stimulation, hypercalcaemia, cardiac glycosides and calcium channel blockers have been shown to counteract the likelihood for the occurrence of alternans by increasing the critical pacing threshold leading to alternans (for references see Euler, 1999).

There is strong evidence that alternans is ultimately linked to alterations in myocardial Ca²⁺ homeostasis. Indeed, during alternans, changes in contraction amplitude are paralleled by changes in the amplitude of the underlying [Ca²⁺]_i transient (e.g. Lab & Lee, 1990; Kihara & Morgan, 1991; Orchard et al. 1991; Uno, 1991; Hüser et al. 2000), which can show marked spatial heterogeneities in amplitude and phase of alternans in the intact heart under ischaemic conditions (Qian et al. 2001). Furthermore, during alternans, mechanical restitution is significantly slowed after the large-amplitude twitch (e.g. Spencer et al. 1992; Rubenstein & Lipsius, 1995), which is indicative of impaired [Ca²⁺]_i regulation (Hirayama et al. 1993). Mechanical restitution, however, is the result of a sequence of intracellular processes involving Ca²⁺ uptake into the SR, intra-SR redistribution of Ca²⁺, and recovery of SR release channels from inactivation (or adaptation). Thus, the mechanism underlying alternans may reside anywhere in this multi-step process, or may possibly even involve the trigger signal for SR Ca²⁺ release (i.e. I_Ca; Dumitrescu et al. 2002a, b).

Figure 2 shows the characteristic features of electromechanical and Ca²⁺ alternans in cat atrial myocytes. Alternans was initiated by electrical stimulation at frequencies of 0.5-1.5 Hz at room temperature. The simultaneous recordings of APs and cell shortening reveal the typical characteristics of discordant electromechanical alternans (i.e. the short AP correlates with the strong contraction, and vice versa). The spatial and temporal patterns of [Ca²⁺]_i transients during alternans was visualized using the line-scan mode of the confocal microscope. A single atrial myocyte (Fig. 2B) was scanned repetitively along a line positioned perpendicular to the longitudinal cell axis (transverse scanning). The whole-cell [Ca²⁺]_i transient, averaged over the entire width of the cell (Fig. 2Ba) revealed regularly alternating amplitudes of [Ca²⁺]_i. The line-scan images (Fig. 2Bb) and local [Ca²⁺]_i transients (Fig. 2Bc) showed marked [Ca²⁺]_i gradients underlying the electrically evoked [Ca²⁺]_i transients. During the early phase of both the large- and small-amplitude [Ca²⁺]_i transients, the rise of [Ca²⁺]_i occurred initially at the cell periphery (in the subsarcolemmal space, ss) and subsequently expanded towards the centre (ct) of the myocyte. During alternans, marked differences in local [Ca²⁺]_i gradients were observed between the large- and small-amplitude [Ca²⁺]_i transients. During the large-amplitude transient, Ca²⁺ release from the SR occurred throughout the entire cell. In contrast, during the small-amplitude transient, Ca²⁺ release in central regions of the cell consistently failed, leading to marked [Ca²⁺]_i gradients in centripetal directions. These spatially heterogeneous changes in [Ca²⁺]_i were in sharp contrast with [Ca²⁺]_i transients observed during alternans in ventricular myocytes, which were homogeneous throughout the entire cell (Hüser et al. 2000).

Figure 2. Electromechanical and Ca²⁺ alternans in atrial myocytes.

A, simultaneous recordings of APs (current-clamp method) and cell shortening (video edge detection) from a single cat atrial myocyte. Electrical stimulation at frequencies of 0.5-1.5 Hz evoked discordant electrical (top trace) and mechanical (bottom trace) alternans. To the right, two APs recorded during successive small- (open circle) and large-amplitude (filled circle) shortenings are superimposed to illustrate the differences in duration and kinetics. B, spatiotemporal characteristics of [Ca²⁺]_i transients during alternans in an atrial myocyte. Alternans (a) in atrial cells is characterized by marked variations in the spatial profile of the [Ca²⁺]_i transients (b). Local [Ca²⁺]_i profiles (panel c), recorded during large- and small-amplitude [Ca²⁺]_i transients, revealed [Ca²⁺]_i gradients directed from subsarcolemmal (ss) regions towards central regions (ct) of the cell. The small-amplitude [Ca²⁺]_i transients were spatially restricted to the subsarcolemmal regions. (Modified from Hüser et al. 2000.)

Since several studies have provided evidence that the cellular mechanisms underlying alternans are directly linked to cellular Ca²⁺ homeostasis we investigated two key steps in the process of E-C coupling that might be the cause of electromechanical alternans. The reduced release of Ca²⁺ from the SR suggested that alternans is caused by either a reduction of trigger Ca²⁺ (i.e. alternations of the voltage-dependent I_Ca) or possible beat-to-beat alternations of SR Ca²⁺ load. Experimental evidence summarized in Fig. 3 argues against these two possibilities. In a cat atrial myocyte, simultaneous measurements of cell shortening and I_Ca (whole-cell voltage-clamp condition) showed that in contrast to the alternations in cell shortening, peak I_Ca did not alternate, suggesting that the signal that triggers SR Ca²⁺ release is not the key mechanism responsible for the development of alternans. Moreover, during the large-amplitude [Ca²⁺]_i transient, the inactivation of I_Ca was slightly faster, and the current component elicited upon repolarization, attributable to Na⁺-Ca²⁺ exchange, was larger. Both of these observations can be explained by a greater amount of released Ca²⁺ during the large-amplitude contraction, leading to stronger calcium-dependent inactivation of the current and enhanced Ca²⁺ extrusion via Na⁺-Ca²⁺ exchange. Nevertheless, the constant amplitude of peak I_Ca during alternans argues against significant alternations in trigger Ca²⁺ for SR Ca²⁺ release as a primary mechanism. In addition, the occurrence of alternans in voltage-clamped cells demonstrates that the variations in AP duration are not causing the beat-to-beat changes in [Ca²⁺]_i transient and contraction amplitude. In fact, the changes in AP duration are likely to be the result of alternations of the [Ca²⁺]_i-dependent processes responsible for the shape of the AP. This is consistent with the findings that cardiac alternans has been observed during AP clamp (i.e. in the absence of any changes in AP configuration; Chudin et al. 1999), and that the inhibition of SR Ca²⁺ release by ryanodine eliminates alternation of AP duration (Rubenstein & Lipsius, 1995). Thus, the alternations of AP duration appear to be a consequence rather than the cause of electromechanical and Ca²⁺ alternans.

Figure 3. Ca²⁺ currents (I_Ca) and sarcoplasmic reticulum (SR) content during alternans.

A, simultaneous measurements of cell shortening (Δl, top) and I_Ca (bottom) in an atrial myocyte. Holding potential = −40 mV, test potential = 0 mV. Despite large differences in cell shortening during alternans, peak I_Ca remained constant. B, SR Ca²⁺ content during alternans. SR Ca²⁺ content was evaluated with 10 mm caffeine pulses in atrial myocytes. Caffeine was applied twice to the same cell. The amplitude of the caffeine-induced [Ca²⁺]_i transient did not reveal any significant differences after small- (left panel) and large-amplitude (right panel) [Ca²⁺]_i transients, indicating that SR Ca²⁺ content remained constant and did not alternate. Whole-cell [Ca²⁺]_i transients were measured with the calcium-sensitive dye indo-1. Cellular indo-1 fluorescence was measured at 405 nm (F₄₀₅) and 485 nm (F₄₈₅). Changes in [Ca²⁺]_i are expressed as changes in the ratio F₄₀₅/F₄₈₅. (Modified from Hüser et al. 2000.)

An alternative mechanism, proposed in earlier studies (see review by Euler, 1999), for the changes in amplitude of the [Ca²⁺]_i transient during alternans could be alternating beat-to-beat change in the amount of releasable SR Ca²⁺. To test this possibility, individual atrial cells were stimulated with caffeine during alternans to estimate the SR Ca²⁺ content in relation to the small-amplitude and large-amplitude [Ca²⁺]_i transients. As shown in Fig. 3B, there was no significant difference in the amplitude of the caffeine-induced [Ca²⁺]_i transients. These findings indicate that the difference in amplitudes of electrically evoked [Ca²⁺]_i transients, at least in atrial myocytes, is not caused primarily by beat-to-beat alterations of SR Ca²⁺ load or the amount of releasable Ca²⁺. Nonetheless, conditions may occur where the interactions between various Ca²⁺ fluxes lead to beat-to-beat alternations in SR content and, as a consequence, to cytoplasmic Ca²⁺ alternans (Eisner et al. 2000).

Functional coupling between glycolysis and alternans

So far, the results indicate that two pivotal steps of E-C coupling could be eliminated as a primary cause for alternans. This raises the possibility that alternans may result from beat-to-beat alternations in the gain of SR Ca²⁺ release or the efficiency of E-C coupling. It is well established that conditions that favour the occurrence of alternans, such as ischaemia (e.g. Lee et al. 1988; Wu & Clusin, 1997), are also directly related to profound changes in the metabolic state of cardiac myocytes, such as low levels of intracellular ATP and decrease of pH. In the heart, more than 90 % of the cellular ATP is produced by mitochondrial oxidative phosphorylation. Thus, the amount of ATP generated by glycolysis in the normally oxygenated heart is almost negligible in relation to the total energy requirements of the heart. Experimental evidence is accumulating, however, to suggest that glycolytically produced ATP may be the preferential energy source for a variety of cardiac cellular functions (recently reviewed by Goldhaber, 1997; see also Weiss & Hiltbrand, 1985) including ion channels and transport mechanisms. Association of glycolytic enzymes in physical proximity to ATP-dependent proteins might form microenvironments (Masters, 1977) in which the locally generated ATP is kinetically preferred over bulk cytosolic ATP.

There are indeed several lines of evidence that glycolytic metabolism is involved in the generation of alternans. Experimental inhibition at different levels of the glycolytic pathway is capable of inducing alternans in cat atrial myocytes. As shown in Fig. 4A, exposure to pyruvate causes reversible [Ca²⁺]_i transient alternans in cat atrial myocytes. With pyruvate as a metabolic fuel, the flux through glycolysis is markedly inhibited. This inhibitory effect is thought to be caused by inhibition of the phosphofructokinase reaction via increased cytosolic concentrations of citrate and ATP (Newsholme et al. 1962). Similar effects were seen with other inhibitors of glycolysis such as iodoacetic acid (which inhibits glycolysis at the level of glyceraldehyde 3-phosphate dehydrogenase; Fig. 4B), the fatty acid β-hydroxy-butyrate (Hüser et al. 2000) and 2-deoxyglucose (Kockskämper et al. 2002). These experiments indicate that selective inhibition of glycolysis is capable of triggering alternans, suggesting that the availability of glycolytically derived ATP plays a crucial role in the development of electromechanical alternans. Furthermore, the induction of alternans by inhibition of glycolysis appears to be a subcellular phenomenon. Focal application of inhibitors of glycolysis to single myocytes caused subcellular Ca²⁺ alternans, which was confined to the subcellular regions exposed to the blocking agent.

Figure 4. [Ca²⁺]_i transient alternans induced by inhibition of glycolysis.

Inhibition of glycolysis by exposure to 10 mm pyruvate (A) and 1 mm iodoacetic acid (B) caused [Ca²⁺]_i transient alternans in an atrial myocyte. Alternans were reversible upon washout of the inhibitors of glycolysis. [Ca²⁺]_i was measured with the calcium-sensitive dye indo-1. (Modified from Hüser et al. 2000.)

Studying Ca²⁺ alternans with fast two-dimensional fluorescence confocal microscopy revealed surprising spatiotemporal [Ca²⁺]_i patterns that may have important ramifications for the understanding of atrial dysrhythmias. When Ca²⁺ alternans was elicited either by increasing the stimulation frequency or by metabolic interventions targeting glycolysis, marked subcellular variations of alternans were observed (Kockskämper & Blatter, 2002), including longitudinal and transverse gradients of Ca²⁺ alternans as well as neighbouring subcellular regions alternating out of phase, as shown in Fig. 5. Interestingly, when two adjacent regions within a myocyte alternated out of phase, steep [Ca²⁺]_i gradients developed at their border causing delayed propagating Ca²⁺ waves. Ca²⁺ waves are typically the result of SR Ca²⁺ overload and/or alterations in the refractoriness of Ca²⁺ release, and are thought to be arrhythmogenic (Stern et al. 1988). The generation of arrhythmogenic Ca²⁺ waves by subcellular variations in the phase of Ca²⁺ alternans, thus may represent a novel mechanism for the development of atrial dysrhythmias.

Figure 5. Subcellular [Ca²⁺]_i alternans.

[Ca²⁺]_i signals recorded with the calcium-sensitive dye fluo-4 from a cat atrial myocyte with the aid of two-dimensional confocal microscopy. The images shown were recorded at intervals of 33 ms and represent surface plots of changes in [Ca²⁺]_i. The traces represent normalized subcellular [Ca²⁺]_i transients recorded from the regions of interest marked by the rectangles. The top and bottom region of the cell reveal Ca²⁺ alternans that is out of phase.

The exact mechanism through which inhibition of glycolysis causes alternans remains to be elucidated. Nonetheless, the hypothesis of compartmentalized energy production based on glycolytically derived ATP remains an intriguing possibility to explain the generation of alternans. It is known that SR Ca²⁺ release and the activity of the RyR release channel (or ancillary proteins in the microenvironment of the RyR) are regulated by phosphorylation processes (Hain et al. 1995). Biochemical studies have suggested that the RyR channel is a substrate for exogenous and endogenous protein kinases (PKA, PKG and the multifunctional calcium- calmodulin-dependent kinase; see e.g. Witcher et al. 1991; Hohenegger & Suko, 1993; Strand et al. 1993). Phosphorylation of the RyR may even be involved in the recruitment of active channels during each contraction cycle of the heart (Hain et al. 1995).

How may RyR phosphorylation, glycolysis and alternans be linked?

Glycogenolytic and glycolytic enzymes have been shown to bind to the sarcolemmal and SR membrane (e.g. Entman et al. 1976; Pierce & Philipson, 1985), where they support the activity of the SR-Ca²⁺-ATPase (Xu et al. 1995), the Na⁺-K⁺-ATPase (Glitsch & Tappe, 1993) or the ATP-sensitive K⁺ channel (Weiss & Lamp, 1989). In skeletal muscle, for example, compartmentalized glycolytic ATP production and consumption occurs in the triads (i.e. the microcompartment enclosed by the t-tubular and SR membrane of the terminal cisternae; Han et al. 1992). In addition, the glycolytic enzyme glyceraldehyde 3-phosphate dehydrogenase was found in the junctional foot complex, which physically couples the SR Ca²⁺ release channel to the DHPR in the t-tubular membrane (Brandt et al. 1990). In addition, Pierce & Philipson (1985) demonstrated that glycolytic enzymes bind to myocardial sarcolemmal and SR membranes.

We propose to extend this scheme by introducing the hypothesis that compartmentalized glycolytic ATP formation in the microenvironment of the RyR channel plays a crucial role in the generation of cardiac alternans. RyR channels are clustered into functional units within morphologically discrete compartments of the SR. Release from one channel in the cluster rapidly activates neighbouring channels. The number of active channels within the cluster is determined by the phosphorylation state of the RyRs, which in turn depends critically on the ATP available to protein kinase(s). Reducing the flux through glycolysis will reduce the local phosphorylation potential ([ATP]/[ADP][P_i]) in the immediate vicinity of the RyRs. Consequently, the rate of phosphorylation of RyRs is slowed and the equilibrium between phosphorylated and non-phosphorylated channels is shifted towards the non-phosphorylated state. In the intact heart, electrical and mechanical alternans are most frequently observed during acute myocardial ischaemia, a condition that is likely to affect glycolytic metabolism through restricted substrate availability. As a result, the phosphorylation reactions are slowed and the availability of active RyR channels on a beat-to-beat basis is reduced. Furthermore, ATP itself is an important modulator of RyR release channels. Thus, alternans may result from alternating changes in the gain or efficiency of E-C coupling due to metabolic alterations and impairment of phosphorylation processes in the microdomain of the SR calcium release channel. There is indeed evidence that glycolytic metabolites directly affect the behaviour of the RyR. Kermode et al. (1998) have shown that intermediates of the glycolytic pathway can activate the cardiac RyR. Furthermore, we have shown (Zima et al. 2002) that pyruvate significantly reduces the open probability of the RyR release channel (single channel recording from RyRs incorporated into the lipid bilayer) and the occurrence of Ca²⁺ sparks, both findings strongly suggesting that glycolysis (or its metabolites) have a profound effect on SR Ca²⁺ release. In addition, one might speculate that the same metabolic factors (i.e. ATP, phosphorylation etc.) may also modulate IP₃-dependent Ca²⁺ signalling and thereby contribute to the development of alternans in atrial myocytes.

Outlook

Atrial dysrhythmias, especially atrial fibrillation, are among the most common clinical cardiac problems, representing a major cause of morbidity and mortality (see e.g. Gersh, 1995; Hart & Halperin, 1999, 2001; Nattel et al. 2000; Nattel, 2002). Despite the magnitude of the problem, little is known about the cellular mechanisms underlying atrial dysrhythmias. There is strong evidence that alternans correlates with the occurrence of certain forms of cardiac arrhythmias (e.g. Dilly & Lab, 1988; Konta et al. 1990), and so the study of the subcellular regulatory mechanisms of Ca²⁺ alternans may generate novel insights into the prevention and treatment of atrial dysrhythmias.