Skip to main content
The Journal of Physiology logoLink to The Journal of Physiology
. 2001 Feb 1;530(Pt 3):417–429. doi: 10.1111/j.1469-7793.2001.0417k.x

Predetermined recruitment of calcium release sites underlies excitation–contraction coupling in rat atrial myocytes

Lauren Mackenzie *, Martin D Bootman *,, Michael J Berridge *,, Peter Lipp *
PMCID: PMC2278433  PMID: 11158273

Abstract

  1. Excitation–contraction coupling (E–C coupling) was studied in isolated fluo-3-loaded rat atrial myocytes at 22 and 37°C using rapid confocal microscopy.

  2. Within a few milliseconds of electrical excitation, spatially discrete subsarcolemmal Ca2+ signals were initiated. Twenty to forty milliseconds after stimulation the spatial overlap of these Ca2+ signals gave a ‘ring’ of elevated Ca2+ around the periphery of the cells. However, this ring was not continuous and substantial Ca2+ gradients were observed.

  3. The discrete subsarcolemmal Ca2+-release sites, which responded in a reproducible sequence to repetitive depolarisations and displayed the highest frequencies of spontaneous Ca2+ sparks in resting cells, were denoted ‘eager sites’.

  4. Immunostaining atrial myocytes for type II ryanodine receptors (RyRs) revealed both subsarcolemmal ‘junctional’ RyRs, and also ‘non-junctional’ RyRs in the central bulk of the cells. A subset of the junctional RyRs comprises the eager sites.

  5. For cells paced in the presence of 1 mM extracellular Ca2+, the response was largely restricted to a subsarcolemmal ‘ring’, while the central bulk of the cell displayed a ∼5-fold lower Ca2+ signal. Under these conditions the non-junctional RyRs were only weakly activated during E–C coupling. However, these channels are functional and the Ca2+ stores were at least partially loaded, since substantial homogeneous Ca2+ signals could be stimulated in the central regions of atrial myocytes by application of 2·5 mM caffeine.

  6. Neither the location nor activation order of the eager sites was affected by increasing the trigger Ca2+ current (by increasing extracellular Ca2+ to 10 mM) or the sarcoplasmic reticulum (SR) Ca2+ load (following 1 min incubation in 10 mM extracellular Ca2+), although with increased SR Ca2+ load, but not greater Ca2+ influx, the delay between the sequential activation of eager sites was reduced. In addition, increasing the trigger Ca2+ current or the SR Ca2+ load changed the spatial pattern of the Ca2+ response, in that the Ca2+ signal propagated more reliably from the subsarcolemmal initiation sites into the centre of the cell. Due to the greater spatial spread of the Ca2+ signals, the averaged global Ca2+ transients increased by ∼500 %.

  7. We conclude that rat atrial myocytes display a predetermined spatiotemporal pattern of Ca2+ signalling during early E–C coupling. A consistent set of eager Ca2+ release sites with a fixed location and activation order on the junctional SR serve to initiate the cellular response. The short latency for activation of these eager sites suggests that they reflect clusters of RyRs closely coupled to voltage-operated Ca2+ channels in the sarcolemma. Furthermore, their propensity to show spontaneous Ca2+ sparks is consistent with an intrinsically higher sensitivity to Ca2+-induced Ca2+ release. While the subsarcolemmal Ca2+ response can be considered as stereotypic, the central bulk of the cell grades its response in direct proportion to cellular Ca2+ load and Ca2+ influx.


In the heart, release of Ca2+ from the sarcoplasmic reticulum (SR) is the key event linking membrane depolarisation and mechanical activity during Excitation–contraction coupling (E–C coupling) (Bers, 1991; Callewaert, 1992). It is generally accepted that Ca2+ influx via voltage-operated Ca2+ channels (VOCCs) is the major source for trigger Ca2+, which subsequently activates ryanodine receptors (RyRs) in the membrane of the SR by a process known as Ca2+-induced Ca2+ release (CICR; Fabiato, 1985). RyRs occur in clusters that give rise to localised Ca2+ release events denoted ‘Ca2+ sparks’ (Cheng et al. 1993; Lipp & Niggli, 1994). Spatiotemporal recruitment of Ca2+ sparks underlies the global Ca2+ signals that subsequently activate myocyte contraction (López-López et al. 1995; for review see Bootman & Berridge, 1995; Lipp & Niggli, 1996; Berridge et al. 1998, 1999, 2000). In ventricular myocytes, a network of tubular membranes oriented transversely to the long axis of the myocyte (‘t-tubules’) conduct the action potential deep within the cells. As a result, Ca2+ spark sites throughout the cell are activated during the action potential ensuring a spatially and temporally homogeneous Ca2+ rise (Cannell et al. 1994; Lipp et al. 1996b). In contrast, atrial myocytes do not possess a t-tubule system (Hüser et al. 1996; Lipp et al. 1996a,b), and the coupling between Ca2+ channels on the sarcolemma and ‘junctional’ RyRs on the SR occurs around the periphery of the cell (Lipp et al. 1990; Berlin, 1995; Lewis Carl et al. 1995; Hüser et al. 1996). Previous studies have indicated that the Ca2+ rise in atrial myocytes occurs initially in the subsarcolemmal region, followed by a variable degree of propagation of the Ca2+ signal into deeper layers of the atrial myocyte (Lipp et al. 1990; Berlin, 1995; Hüser et al. 1996). Since RyRs are present at seemingly equal abundance throughout atrial myocytes (Lewis Carl et al. 1995; Lipp et al. 2000) it is surprising that CICR can largely fail in deeper layers of the cell.

Although it has been shown that E–C coupling in atrial myocytes is initiated in the subsarcolemmal region, the precise spatiotemporal pattern of the Ca2+ rise has not been investigated so far. In the present study, we used rapid confocal microscopy to probe the development of the Ca2+ signal in the subsarcolemmal and central regions of single electrically paced atrial myocytes. Our data indicate that within the subsarcolemmal and central regions of an atrial myocyte Ca2+ release sites can show markedly different latencies and levels of regenerativity. The consequence is that atrial myocytes have a predetermined microscopic activation sequence of Ca2+ spark sites whereby single cells produce reproducible inhomogeneous Ca2+ increases upon depolarisation.

METHODS

Cell isolation

Atrial myocytes were isolated using established methods (Lipp et al. 2000). Briefly, male Wistar rats (∼200 g) were killed by cervical dislocation after CO2 anaesthesia. The hearts were quickly removed, and perfused with an extracellular solution containing (mM): NaCl, 135; KCl, 5.4; MgCl2, 2; Hepes, 10; glucose, 10; pH 7.35 and 1 mg ml−1 collagenase (Worthington). The atria or ventricular parts were excised and gently shaken in the perfusion solution to release isolated cells.

Confocal Ca2+ imaging

Myocytes were adhered to poly-D-lysinE–Coated (Sigma) coverslips for up to 2 h before use. The cells were loaded with 2 μM fluo-3 AM for 30 min followed by an additional 30 min period for de-esterification, and then subsequently transferred to the stage of a laser scanning confocal microscope (NORAN Oz, Bicester, UK). Fluo-3 was excited at 488 nm and the emitted fluorescence was collected at wavelengths > 505 nm. The confocal slit aperture was set so that the confocal plane was < 1 μm in thickness. Two dimensional confocal images (frame size: 512 × 115 pixels) were acquired at 120 Hz, and line scans were performed at 10 kHz (i.e. 0.1 ms per line). Analysis of the images was conducted off-line using a modified version of NIH Image (NIH, Bethesda, USA). The myocytes were stimulated with 40 V pulses (2 ms duration) using two field electrodes (distance 0.5 cm). The stimulation frequency was set to 1 Hz for all experiments. This resulted in electrically induced Ca2+ transients with a constant amplitude throughout the train of stimulations (10 stimulations per train). Intracellular Ca2+ concentrations were calculated using a self-ratio method as originally described by Minta and co-workers (Minta et al. 1989). The confocal images shown represent self ratios, where every confocal frame of the sequence was normalised by an average resting cell section. All experiments were performed at room temperature (20-22°C) unless stated otherwise. During the experiments, cells were maintained in the same extracellular medium as used for cell isolation, except that collagenase was omitted and 1 or 10 mM CaCl2 was added.

Calculation of time delays

Although the activation sequence of eager sites was reproducible, we attempted to quantify the temporal delay for each site. Since the synchronisation between the confocal microscope and the electrical stimulation was not perfect (we observed jitter in the range of one frame; 8 ms), we deferred to calculating relative delays, whereby all responses were normalised to that ‘eager’ site which consistently activated first. For this, we calculated the relative time each local Ca2+ transient took to reach 10 % of its maximal amplitude by interpolating linearly between the data points. The use of the 10 % value avoided problems with Ca2+ diffusion from out-of-focus release sites.

Immunocytochemistry

Rat atrial and ventricular myocytes adhered to coverslips were fixed with 4 % paraformaldehyde for 1 h at room temperature. The immunocytochemical staining was performed using established methods (Sugiyama et al. 1994; Lipp et al. 2000) with a RyR type II specific monoclonal antibody (V. Sorrentino, Siena, Italy). The subcellular distribution of fluorescein-conjugated secondary antibodies was examined using an UltraView confocal microscope (Perkin Elmer Life Sciences, Cambridge, UK).

RESULTS

Depolarisation of fluo-3-loaded rat atrial myocytes resulted in spatially inhomogeneous Ca2+ transients. A typical example of the spatiotemporal pattern of Ca2+ rise following field stimulation of an atrial myocyte at 22°C and incubated in 1 mM extracellular Ca2+ concentration is illustrated in Fig. 1Aa. The sequences of images show that the Ca2+ signal was more rapid in onset and more substantial in the subsarcolemmal region than in the centre of the cell. However, the Ca2+ rise did not occur with equal timing or amplitude around the circumference of the cell. The earliest signals were observed as eliptical Ca2+ increases at discrete locations (see images at 8 ms in Fig. 1Aa, marked by the black arrowheads in the right hand column). These events presumably reflected Ca2+ sparks at the diadic junctions between junctional SR (containing RyRs) and the plasma membrane (containing VOCCs). At the peak of the response, the lateral diffusion and overlap of these signals resulted in a ‘ring’ of elevated Ca2+ beneath the sarcolemma (e.g. images captured at 32 ms in Fig. 1Aa). However, there were distinct inhomogeneities within the subsarcolemmal Ca2+ rise, such that significant Ca2+ gradients were even observed in regions spaced only a few micrometres apart (e.g. region marked by the red arrowhead in Fig. 1Aa). A more quantitative representation of the Ca2+ concentration gradients observed during field stimulation of the same cell is illustrated by the traces in Fig. 1Ab. The outline of the cell is depicted by the inset image, and coloured circles mark the subcellular regions in which Ca2+ was monitored. The first regions to show Ca2+ rises had a lag time of less than 8 ms and reached maximal amplitude within 24 ms (red circle and corresponding red trace in Fig. 1Ab). Such regions were denoted ‘eager sites’ due to their ability to respond most rapidly. Between the eager sites were regions that exhibited weak regenerative Ca2+ responses (black circle and corresponding black trace in Fig. 1Ab).

Figure 1. E–C coupling in rat atrial myocytes is characterised by inhomogeneous Ca2+ signals.

Figure 1

A and B show the development of the depolarisation-induced Ca2+ responses at 20 and 37°C, respectively. The left columns in Aa and Ba show self-ratio colour-coded images of the entire cell, and the right columns show the part of the cell bounded by the dashed box at a higher magnification. Cool colours (e.g. blue and green) indicate low Ca2+ concentrations and warm colours (e.g. red and yellow) denote higher Ca2+ concentrations. The times at which the images were taken are indicated between the columns, and the time at which the electrical pulse was applied is shown by the open arrow. Three identified ‘eager’ Ca2+ spark sites are marked by black arrowheads in Aa and Ba, and a failure site is denoted by the red arrowhead in Aa. The traces in Ab depict the time course of Ca2+ changes at the three subcellular sites (∼1 μm in diameter) shown on the inset cell image. The same cellular regions were analysed for the traces in Bb. The surface plots in Ac depict the spatial and temporal development of the Ca2+ signal. The Ca2+ concentration is encoded by both the colour and the height of the surface. Ad and Ae illustrate that the limited Ca2+ response in the central region of the cell was consistently observed during a train of depolarisations (at 1 Hz). The red and black coloured traces in Ae show the Ca2+ responses observed at the correspondingly coloured regions marked in Ad. The timing of the electrical depolarisations is indicated by the arrowheads in Ae.

Such ‘failure sites’, although immediately adjacent to eager sites, gave only slowly developing Ca2+ signals with ∼25 % of the amplitude of neighbouring regions. The centre of the cells typically gave a low-amplitude Ca2+ rise of ∼200 nM (green circle and corresponding green trace in Fig. 1Ab). The spatially heterogeneous Ca2+ rise is also visible in the surface plots in Fig. 1Ac. The maximum response at 32 ms consisted of a discontinuous ring of peaks, with the centre of the cell showing a ‘valley’ of low Ca2+ concentration. Such Ca2+ gradients were typical for the majority of atrial myocytes paced in 1 mM Ca2+-containing solution (n >100), and were repetitively observed during a 1 Hz train of depolarisations (Fig. 1Ad and e). Since Ca2+ homeostatic and signalling mechanisms are steeply temperature dependent, we investigated the effect of increasing the temperature on the subcellular properties of Ca2+ signalling observed during E–C coupling.

The data depicted in Fig. 1A were obtained at 22°C, and the corresponding response of the same cell to depolarisation at 37°C is illustrated in Fig. 1B. Although the response was larger and more rapid in onset, and the ‘ring’ of elevated Ca2+ was more continuous, the Ca2+ rise was still restricted to the subsarcolemmal region (Fig. 1B a). Furthermore, analysis of the Ca2+ signals at the same subcellular regions as in Fig. 1A revealed that the eager sites were generally at the same locations for both temperatures (Fig. 1B a, marked by the black arrowheads in the right hand column).

Since the eager Ca2+ release sites were commonly activated within the 8 ms taken to acquire the first image after depolarisation (Fig. 1), we used confocal line scanning to obtain a better temporal resolution of the Ca2+ rise. Scanning along a subsarcolemmal region (Fig. 2Aa) revealed the rapid onset of eager sites at discrete regions within the cell (Fig. 2Ab). Between the eager sites, the Ca2+ signal rose more slowly (compare black and grey traces in Fig. 2Ac). Scanning through the centre of the same cell shows the slow development and low amplitude of the Ca2+ signal deeper within the cell (Fig. 2Ad-f).

Figure 2. Spatial and temporal stability of eager Ca2+ spark sites.

Figure 2

A depicts the development of Ca2+ signals in the subsarcolemmal (Aa-c) and central (Ad-f) regions of the same atrial myocyte using confocal line scanning. The positions of the scanned lines are depicted on the cell images in Aa and Ad. For Ab and Ae, time runs from left to right and the spatial dimension of the scanned line is vertical (the scanned line was turned 90 deg anticlockwise). The development of the Ca2+ signals along the regions indicated by the black and grey arrows is illustrated by the correspondingly coloured traces in Ac and Af (the bars to the right of the arrows indicate the segment that was averaged along the scanned line). The asterisk in Ae indicates a Ca2+ spark that occurred at the right boundary of the cell, while the left end of the cell (lower end of the line scan image) indicates a failure to elicit SR Ca2+ release. Due to the modest contraction of the cell, both Ab and Ae show movement artifacts visible as curvature of the edges of the plots. The dashed vertical lines in Ab and Ae indicate the time of field stimulation. Traces in panel A were representative for six cells from three rat hearts. B illustrates the reproducible recruitment of eager Ca2+ sparks sites with a train of six successive depolarisations. The images in Ba1-6 depict individual line scans for each depolarisation. The dashed vertical lines mark the time of stimulation. The traces in Bb indicate the time course of the Ca2+ responses with each depolarisation at the regions marked by the black and grey arrows in Ba1. Similar results were obtained in five atrial cells from three hearts.

The locations of eager and failure sites were the same for successive depolarisations, and were independent of the orientation of the cell relative to the stimulating electrodes. The line scan images in Fig. 2Ba depict the initiation of six sequential Ca2+ signals. Several eager sites can be distinguished in the line-scan images, with similar patterns of Ca2+ increase arising with each depolarisation. Although the same eager sites responded, one noticeable difference between them was that some sites had a consistent time course whilst others varied in their rise times. The traces in Fig. 2Bb depict the development of the Ca2+ signal at the locations marked by arrows in Fig. 2Ba1. One of the regions (black traces in Fig. 2Bb) showed little variation in its time course (< 1 ms difference for time to 10 % peak in six trials), whilst at another site (grey traces in Fig. 2Bb) the latency before the rapid upstroke of the Ca2+ was more variable (∼4-5 ms difference for time to 10 % peak in six trials). Despite this variability, the activation order of the eager sites was on average the same with each depolarisation.

The consistent order of eager site recruitment is shown quantitatively in Fig. 3. The locations of 23 eager sites identified in an individual atrial myocyte are depicted on the cell image in Fig. 3A. The traces in Fig. 3B show the Ca2+ rise in three of the identified sites. To calculate the relative activation order of the 23 eager sites, the traces were expanded and the relative times at which each site reached 10 % of its maximal eventual amplitude was estimated from the curves (see Methods), as illustrated in Fig. 3C. Using time to 10 % of maximal amplitude minimises potential problems caused by diffusion of Ca2+ from neighbouring sites or motion artifacts during the contraction of the cell. Two-dimensional imaging, and not faster line scanning, was used to examine the activation order since eager sites around the circumference of a cell could be monitored. The first of the 23 eager sites to respond was always the site ‘S’ shown in Fig. 3A. The graph in Fig. 3D depicts the relative times, with reference to site S, taken for the Ca2+ signals at each eager site to reach 10 % of their maximal amplitude. The data indicate that these eager sites responded in a consistent order, and that there was a ∼10 ms delay in the development of the Ca2+ signal at different sites. In additional experiments, we observed that the order of activation for the eager sites did not depend on the orientation of the applied electrical field. Reversing the field direction did not alter the activation sequence (data not shown). We tested whether the preferential activation of the eager sites was due to an intrinsically higher sensitivity to CICR. For this, we compared the eagerness of sites (i.e. relative latencies) and their ability to produce spontaneous Ca2+ sparks during rest. The correlation between eager sites and Ca2+ spark frequency is illustrated in Fig. 4. For the atrial myocyte shown, 14 subsarcolemmal Ca2+ release sites were analysed (Fig. 4Aa). These sites were found to have a reproducible order of activation (Fig. 4Ab; relative time taken to reach 10 % of peak amplitude; mean value of five consecutive stimulations) with a difference of up to 8 ms. Following termination of electrical pacing, a similar order was found when the frequency of spontaneous Ca2+ sparks was plotted (Fig. 4Ac). These data indicate that the most eager sites with the least lag times also displayed the highest frequency of spontaneous Ca2+ release events.

Figure 3. Reproducible activation order of eager sites.

Figure 3

The relative activation times of the 23 eager sites depicted in A were analysed. Example traces from three of the eager sites are shown in B. To calculate the activation order of the eager sites, the time taken for the signals to reach 10 % of maximal amplitude (demarked by the horizontal dashed lines) was calculated as illustrated in C. The average delay of each eager site to respond after site ‘S’ is plotted in D (mean ±s.e.m. of 6 stimulations). Similar results were found in four additional myocytes from two hearts.

Figure 4. Properties of eager sites in rat atrial myocytes.

Figure 4

Aa depicts a single atrial myocyte in which 14 discrete eager sites were identified. The relative activation order of these sites was calculated as described in the text, and is plotted in Ab. The bar graph in Ac shows the frequency of spontaneous Ca2+ sparks arising at the eager sites during 15 s in which the cell was not electrically stimulated. To assess Ca2+ spark frequency, the fluorescence images (collected at 120 Hz) were individually inspected for the appearance of new Ca2+ spark events. The relative activation order and degree of spontaneous activity were analysed in four additional myocytes from two rat hearts with similar results. B illustrates that the activation order of eager sites was not a result of different diffusion times for Ca2+ release sites out of focus into the confocal plane. The cell image in Ba depicts the 11 identified eager sites. The relative order of activation of these sites at three confocal planes differing axially by 0.5 μm is shown in Bb. A similar behaviour was found in three additional myocytes from two rat hearts.

Since localised Ca2+ release events can appear to have slower onset if they are out of focus, we were concerned that our observation of eager sites and their predetermined activation order reflected Ca2+ release at different focal planes within the myocytes. To address this issue, the activation order of 11 Ca2+ release sites within a single atrial myocyte (Fig. 4Ba) was monitored at three focal planes differing by 0.5 μm. The sequence in which the Ca2+ release sites were activated was similar at each plane, although the absolute values varied slightly (Fig. 4Bb). These data indicate that the observed order of activation of the Ca2+ signals was not due to different positions of the Ca2+release sites relative to the confocal image plane.

We examined whether increasing the SR Ca2+ load or Ca2+ influx determined the position and activation order of the eager sites. The SR Ca2+ load was increased by bathing atrial myocytes in extracellular medium containing 10 mM Ca2+ concentration for 1 min, a procedure that typically enhances Ca2+ influx and store loading. The experimental protocol is schematically illustrated in Fig. 5Aa. Essentially, the responses of cells paced at 1 Hz in 1 mM Ca2+-containing medium were compared before and after incubation in 10 mM Ca2+. In the example shown in Fig. 5A, the global amplitude of the electrically evoked Ca2+ transients was increased by ∼500 % following the 10 mM Ca2+ treatment (Fig. 5Ab). This was essentially due to propagation of the Ca2+ signal throughout the cell, as reported by Hüser et al. (1996), in contrast to the largely subsarcolemmal response before altering SR Ca2+ load (data not shown). The activation order of nine eager sites identified in this cell (depicted in Fig. 5Ac) is shown in Fig. 5Ad. Incubation of the cells in 10 mM Ca2+ did not alter the location or activation order of the eager sites, although it did reduce the time interval between the sequential activation of the sites. The increased store load did not result in recruitment of extra eager sites in the observed confocal plane (data not shown). We confirmed that the 1 min incubation in 10 mM Ca2+ enhanced the Ca2+ load of the SR by directly activating the RyRs with caffeine. The experimental protocol for investigating the SR Ca2+ loading is depicted in Fig. 5Ba, which indicates that cells were superfused with caffeine either before or after the 10 mM Ca2+ treatment. The traces in Fig. 5Bb and c show that following the incubation in 10 mM Ca2+, there was a substantial increase of SR Ca2+.

Figure 5. Enhanced SR Ca2+ load and Ca2+ influx did not change the activation order of eager sites.

Figure 5

Aa depicts the experimental protocol used to increase the SR Ca2+ load. The arrowheads indicate the times at which electrical stimulation was given (1 Hz frequency). Ab illustrates Ca2+ transients before (black squares) and after (grey diamonds) the 1 min incubation in medium containing 10 mM Ca2+. The arrowheads below the traces indicate the times of electrical stimulation. The white circles on the cell images in Ac depict the position of nine identified eager sites within the cell. The activation order of these sites under control conditions (black squares) and after the 1 min incubation in medium containing 10 mM Ca2+ (grey diamonds) is shown in Ad. For both the control and 10 mM Ca2+ conditions, the relative delays of the eager sites during four transients was averaged (marked by corresponding symbols above the traces in Ab). A similar behaviour was found in two additional myocytes. B depicts the typical increase in SR Ca2+ load caused by the 1 min incubation in 10 mM Ca2+ containing medium. Ba depicts the experimental protocol used. The Ca2+ responses evoked by superfusion with 5 mM caffeine before and after the 10 mM Ca2+ treatment are shown in Bb and Bc, respectively. C illustrates the effect of acutely increasing the extracellular Ca2+ concentration whilst the cells were electrically paced (the arrowheads beneath the trace in Ca indicate when depolarising pulses were applied; 1 Hz frequency). The white circles on the cell images in Cb depict the position of 10 identified eager sites within the cell. The activation order of these sites under control conditions (black squares) and after extracellular Ca2+ to 10 mM (grey diamonds) is shown in Cc. For both the control and 10 mM Ca2+ conditions, the relative delays of the eager sites during four transients was averaged (marked by corresponding symbols above the traces in Ca).

Elevating the extracellular Ca2+ concentration from 1 to 10 mM during pacing significantly increased the amplitude of electrically evoked Ca2+ transients (Fig. 5Ca). Despite this acute change of extracellular Ca2+ concentration, the activation order of the ten eager sites depicted in Fig. 5C b was unaffected (Fig. 5Cc). Similar to the situation where cells were incubated with 10 mM Ca2+ for 1 min (Fig. 5A), acutely increasing the extracellular Ca2+ concentration also changed the response from a largely subsarcolemmal ‘ring’ of Ca2+ to a globally regenerative response (data not shown). The data presented in Fig. 5 indicate that the location and activation order of eager Ca2+ release sites was unaffected by increasing either SR Ca2+ load or voltage-activated Ca2+ entry. Apart from decreasing the time interval between the activation of eager sites by an enhanced SR Ca2+ load (Fig. 5A), increasing extracellular Ca2+ had no effect on the early events of atrial myocyte E–C coupling.

The data described above illustrate that atrial myocytes display spatially heterogeneous Ca2+ signals upon depolarisation. Stereotypic responses are observed in the subsarcolemmal region, with variable amplitude Ca2+ transients in the central bulk of the cell. Within the subsarcolemmal region where the stereotypic Ca2+ rise occurs, there are substantial differences in the responsiveness of Ca2+ release sites, leading to an inhomogeneous subsarcolemmal Ca2+ signal. The mechanisms causing the Ca2+ release sites within an atrial myocyte to behave so distinctly are unclear. One plausible explanation would be a variable distribution of RyRs. Isolated atrial myocytes probed with a type II RyR-specific antibody generally displayed consistent distributions of immunoreactivity that did not allude to obvious reasons why Ca2+ release should fail at certain positions (Fig. 6A). Two regularly arranged populations of atrial RyRs were observed: junctional RyRs that underlie the initial subsarcolemmal Ca2+ elevation during E–C coupling and the non-junctional RyRs aligned along the SR within the centre of cells where the response is SR Ca2+-load dependent (Fig. 6Ab). The only location where atrial myocyte RyRs appeared to be less regularly spaced was in a perinuclear region that was sometimes devoid of immunostaining (Fig. 6A; marked PR). A more detailed inspection of the immunostaining reveals that within both junctional and non-junctional RyR populations the channel distribution was punctate. Furthermore, between the junctional and non-junctional RyRs, there was a clear gap in the immunostaining of 0.5-1 μm that extended completely around the cell (marked by red lines in Fig. 6Ab). Ventricular myocytes also displayed regularly arranged RyRs (Fig. 6B). However, in contrast to atrial cells, the RyRs are all ‘junctional’ since they align the t-tubules that project into the cells (Fig. 6B b).

Figure 6. Immunolocalisation of type II RyRs in atrial and ventricular myocytes.

Figure 6

A and B depict type II RyR immunostaining of single atrial and ventricular myocytes, respectively. In Aa, PR denotes the perinuclear region. For Aa and Ba, N indicates the position of the nuclei. The regions bounded by dashed box (b) in Aa and Ba are expanded in Ab and Bb, to show more clearly the punctate RyR staining and the separation of junctional and non-junctional RyRs in atrial cells.

To examine the functionality of the RyRs identified by immunostaining, caffeine (2.5 mM) was superfused onto atrial myocytes to directly activate the channels (Fig. 7). Irrespective of the external Ca2+ concentration or the SR Ca2+ load, application of caffeine evoked homogeneous Ca2+ signals that developed equally rapidly within the subsarcolemmal and central regions of the cells. Figure 7 depicts the response of a cell paced at 1 Hz in 1 mM Ca2+-containing medium where electrical pacing evoked typical Ca2+ signals with subsarcolemmal gradients and a poorly regenerative response in the central region of the cell (Fig. 7B). In contrast to the electrically evoked Ca2+ signals, with caffeine there was no evidence of subsarcolemmal gradients similar to those observed during pacing. Furthermore, the central bulk always responded with a Ca2+ signal of similar amplitude to that observed in the subsarcolemmal region (Fig. 7C and D).

Figure 7. Caffeine, but not electrical depolarisation, evokes spatially homogeneous Ca2+ signals.

Figure 7

Comparison of the spatial pattern of Ca2+ signals evoked by electrical depolarisation and caffeine. The cytosolic Ca2+ concentration was monitored at the two sites shown on the inset cell image in A. Electrical depolarisation (at 1 Hz frequency in 1 mM Ca2+ containing medium) evoked typically heterogeneous Ca2+ responses, with a larger Ca2+ increase in the subsarcolemmal regions (black trace in B) than in the centre of the cell (grey trace in B). The arrowheads denote the timing of the electrical pulses. C shows that application of caffeine to the same cell shortly after termination of the electrical pulses evoked Ca2+ signals with similar amplitudes and kinetics in both regions. The spatially homogeneous Ca2+ response to caffeine is illustrated by the surface plot in D, which shows the caffeine-induced Ca2+ response at the maximal amplitude. The asterisks in B represent spontaneous Ca2+ sparks.

DISCUSSION

The present study shows that within the Ca2+ response of atrial myocytes to electrical stimulation there is a consistent microscopic sequence of Ca2+ spark activation, along with regions that repeatedly fail to respond (Figs 1 and 2). Thus the spatially heterogeneous Ca2+ transient in an atrial myocyte is not stochastic, but is actually constructed from a predetermined recruitment of eager Ca2+ spark sites. Previous studies have reported that atrial myocyte E–C coupling is characterised by spatially heterogeneous Ca2+ signals (Berlin, 1995; Hüser et al. 1996). Due to the lack of t-tubules in atrial myocytes (Hüser et al. 1996), junctional couplings between VOCCs and RyRs only occur around the periphery of the cells (Sommer & Jennings, 1986; Lewis Carl et al. 1995; Lipp et al. 2000). For this reason, the Ca2+ rise is expected to be more rapid in subsarcolemmal regions (Figs 1 and 2; Berlin, 1995; Hüser et al. 1996). Our data indicate that the RyRs at the junctional couplings differ significantly in their ‘eagerness’ to activate during E–C coupling. The mechanisms underlying the eagerness of some Ca2+ release sites and the lack of responsiveness of others are unclear. Our immunostaining (Fig. 6A) and that of others (Lewis Carl et al. 1995) suggests that both RyRs and VOCCs are evenly distributed around the periphery of atrial myocytes. There are no apparent regions devoid of either RyRs (Fig. 6A) or VOCCs (Lewis Carl et al. 1995) that could explain the failure sites (Fig. 1) observed during Ca2+ responses. Nor are there any sites with apparently more RyRs or VOCCs that might underlie the eager sites. In addition, the occurrence of eager sites, their subcellular localisation and activation order also appeared to be independent of the SR Ca2+ load or Ca2+ influx. An alternative explanation is therefore that the functional coupling between RyRs and VOCCs at diadic junctions varies significantly, such that the Ca2+ influx signal is unable to trigger a regenerative release in some regions. Furthermore, the observation that spontaneous Ca2+ sparks occur most frequently at eager sites (Fig. 4A) suggests that there may be some intrinsic difference in RyR activity or sensitivity to CICR at those locations.

The principle of ‘eager’ Ca2+ release sites was initially suggested for skeletal muscle E–C coupling (Blatter et al. 1996), although it has not been widely reported. In smooth muscle cells, Ca2+ spark sites with a significantly higher rate of spontaneous activation have been observed, but it is unclear whether they also play a prominent role during electrical or hormonal stimulation of the cells (for review see Bolton et al. 1999). Similarly, in line-scan images of spontaneous Ca2+ sparks from unstimulated ventricular myocytes, it appears that only a few of the many Ca2+ spark sites show repeated activity, consistent with an intrinsic difference in RyR activation similar to that seen in the present study (Parker & Wier, 1997). However, for ventricular cardiomyocytes and skeletal muscle, it is more generally accepted that homogeneous Ca2+ rises occur during E–C coupling (for a review see Lipp & Niggli, 1996). Therefore in contrast to our observations with atrial myocytes, the potentially eager sites in ventricular and skeletal muscle are not apparent during E–C coupling.

Whereas the subsarcolemmal response can be considered as stereotypic, since it is unaffected by variations of temperature, SR Ca2+ load or Ca2+ influx, the central bulk of the atrial cell shows a response that is enhanced by greater SR Ca2+ load and Ca2+ influx (Fig. 5). The lack of central response from the cells (Fig. 1; see also Hüser et al. 1996), is surprising since the RyRs are regularly arranged throughout the major part of an atrial cell (Fig. 5; Lewis Carl et al. 1995; Lipp et al. 2000) with only the perinuclear regions showing less abundant RyR immunofluorescence. Furthermore, these RyRs within the central regions of atrial myocytes are functional and even under conditions of 1 Hz stimulation and 1 mM extracellular Ca2+ concentration the Ca2+ stores are substantially replete, since the RyR agonist caffeine causes a substantial homogeneous Ca2+ rise (Fig. 7).

Although there are no t-tubules to conduct the depolarisation deep into atrial myocytes, the subsarcolemmal Ca2+ rise is significant and could be expected to trigger CICR in deeper layers, so that a regenerative Ca2+ wave would sweep into the cells and cause a global Ca2+ increase (see Berlin, 1995; Hüser et al. 1996; Lipp et al. 1996a,b). Our data and those of Hüser et al. (1996) therefore indicate that with low SR Ca2+ loads the Ca2+ influx occurring during E–C coupling in atrial myocytes is only sufficient to trigger subsarcolemmal RyRs. The ∼1 μM Ca2+ rise that follows is unable to activate significant CICR from RyRs that are less than a few micrometres away. To what degree the 1 μm gap between the junctional and non-junctional RyRs (see Fig. 6A), which was also described earlier by Lewis Carl et al. (1995), might be responsible for that failure remains unclear and needs further attention in the future.

In summary, atrial myocyte E–C coupling appears to occur as a predetermined sequence of Ca2+ spark recruitment. The sequential activation of eager sites and the lack of response from the failure sites produce a Ca2+ signal that is stereotypic for successive depolarisations in a single atrial myocyte. This predetermined nature of the atrial Ca2+ response might reflect variations in the microarchitecture of atrial diadic junctions or the intrinsic activation properties of RyRs.

Acknowledgments

This work was supported by the BBSRC, and by the MRC (grant no. G9808149). L.M. is an MRC student. M.D.B. is supported by a Royal Society University Research Fellowship.

References

  1. Berlin J. Spatiotemporal changes of Ca2+ during electrically evoked contractions in atrial and ventricular cells. American Journal of Physiology. 1995;38:H1165–1170. doi: 10.1152/ajpheart.1995.269.3.H1165. [DOI] [PubMed] [Google Scholar]
  2. Berridge MJ, Bootman MD, Lipp P. Calcium - a life and death signal. Nature. 1998;395:645–648. doi: 10.1038/27094. [DOI] [PubMed] [Google Scholar]
  3. Berridge MJ, Lipp P, Bootman MD. Calcium signalling. Current Biology. 1999;9:157–159. doi: 10.1016/s0960-9822(99)80101-8. [DOI] [PubMed] [Google Scholar]
  4. Berridge MJ, Lipp P, Bootman MD. The versatility and universality of calcium signalling. Nature Reviews Molecular Cell Biology. 2000;1:11–21. doi: 10.1038/35036035. [DOI] [PubMed] [Google Scholar]
  5. Bers DM. Excitation–contraction Coupling and Cardiac Contractile Force. Dordrecht: Kluver Academic Publishers; 1991. [Google Scholar]
  6. Blatter LA, Tsugorka A, Shirokova N, Rios E. Eager triads in skeletal muscle: Heterogeneous distribution of voltage-elicited Ca2+ release revealed by confocal microscopy. Biophysical Journal. 1996;70:5. (abstract) [Google Scholar]
  7. Bolton TB, Prestwich SA, Zholos AV, Gordienko DV. Excitation–contraction coupling in gastrointestinal and other smooth muscles. Annual Review of Physiology. 1999;61:85–115. doi: 10.1146/annurev.physiol.61.1.85. [DOI] [PubMed] [Google Scholar]
  8. Bootman MD, Berridge MJ. The elemental principles of calcium signalling. Cell. 1995;83:675–678. doi: 10.1016/0092-8674(95)90179-5. [DOI] [PubMed] [Google Scholar]
  9. Callewaert G. Excitation–contraction coupling in mammalian cardiac cells. Cardiovascular Research. 1992;26:923–932. doi: 10.1093/cvr/26.10.923. [DOI] [PubMed] [Google Scholar]
  10. Cannell M, Cheng H, Lederer WJ. Spatial non-uniformities in [Ca2+]i during Excitation–contraction coupling in cardiac myocytes. Biophysical Journal. 1994;67:1942–1956. doi: 10.1016/S0006-3495(94)80677-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Cheng H, Lederer WJ, Cannell MB. Calcium sparks: elementary events underlying Excitation–contraction coupling in heart cells. Science. 1993;262:740–744. doi: 10.1126/science.8235594. [DOI] [PubMed] [Google Scholar]
  12. Fabiato A. Time and calcium dependence of activation and inactivation of calcium-induced release of calcium from the sarcoplasmic reticulum of a skinned canine cardiac Purkinje cell. Journal of General Physiology. 1985;85:247–289. doi: 10.1085/jgp.85.2.247. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Hüser J, Lipsius SL, Blatter LA. Calcium gradients during Excitation–contraction coupling in cat atrial myocytes. The Journal of Physiology. 1996;494:641–651. doi: 10.1113/jphysiol.1996.sp021521. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Lewis Carl S, Felix K, Casswell AH, Brandt NR, Ball WJ, Vaghy PL, Meissner G, Ferguson DG. Immunolocalization of sarcolemmal dihydropyridine receptor and sarcoplasmic reticular triadin and ryanodine receptor in rabbit ventricle and atrium. Journal of Cell Biology. 1995;129:673–682. doi: 10.1083/jcb.129.3.673. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Lipp P, Hüser J, Pott L, Niggli E. Spatially non-uniform Ca2+ signals induced by the reduction of transverse tubules in citrate-loaded guinea-pig ventricular myocytes in culture. The Journal of Physiology. 1996a;497:589–597. doi: 10.1113/jphysiol.1996.sp021792. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Lipp P, Hüser J, Pott L, Niggli E. Subcellular properties of triggered Ca2+ waves in isolated citrate-loaded guinea-pig atrial myocytes characterised by ratiometric confocal microscopy. Journal of Physiology. 1996b;497:599–610. doi: 10.1113/jphysiol.1996.sp021793. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Lipp P, Laine M, Tovey SC, Burrell K, Li W, Berridge MJ, Bootman MD. Functional IP3 receptors that may modulate Excitation–contraction coupling in the heart. Current Biology. 2000;10:939–942. doi: 10.1016/s0960-9822(00)00624-2. [DOI] [PubMed] [Google Scholar]
  18. Lipp P, Niggli E. Modulation of Ca2+ release in neonatal cultured rat cardiac myocytes: Insight from subcellular release patterns revealed by confocal microscopy. Circulation Research. 1994;74:979–990. doi: 10.1161/01.res.74.5.979. [DOI] [PubMed] [Google Scholar]
  19. Lipp P, Niggli E. A hierarchical concept of cellular and subcellular Ca2+-signalling. Progress in Biophysics and Molecular Biology. 1996;65:265–296. doi: 10.1016/s0079-6107(96)00014-4. [DOI] [PubMed] [Google Scholar]
  20. Lipp P, Pott L, Callewaert G, Carmeliet E. Simultaneous recording of Indo-1 fluorescence and Na+/Ca2+ exchange current reveals two components of Ca2+-release from sarcoplasmic reticulum of cardiac atrial myocytes. FEBS Letters. 1990;275:181–184. doi: 10.1016/0014-5793(90)81467-3. [DOI] [PubMed] [Google Scholar]
  21. López-López J, Shacklock P, Balke C, Wier W. Local calcium transients triggered by single L-type calcium-channel currents in cardiac cells. Science. 1995;268:1042–1045. doi: 10.1126/science.7754383. [DOI] [PubMed] [Google Scholar]
  22. Minta A, Kao JPY, Tsien RY. Fluorescent indicators for cytosolic calcium based on rhodamine and fluorescein chromophores. Journal of Biological Chemistry. 1989;264:8171–8178. [PubMed] [Google Scholar]
  23. Parker I, Wier WG. Variability in frequency and characteristics of Ca2+ sparks at different release sites in rat ventricular myocytes. The Journal of Physiology. 1997;505:337–344. doi: 10.1111/j.1469-7793.1997.337bb.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Sommer JR, Jennings RB. The ultrastructure of cardiac muscle. In: Fozzard HA, Haber E, Jennings RB, Katz AM, Morgan HE, editors. The Heart and Cardiovascular System. New York: Raven Press; 1986. pp. 61–100. [Google Scholar]
  25. Sugiyama T, Furuya A, Monkawa T, Yamamotohino M, Satoh S, Ohmori K, Miyawaki A, Hanai N, Mikoshiba K, Hasegawa M. Monoclonal-antibodies distinctively recognizing the subtypes of inositol 1,4,5-trisphosphate receptor - application to the studies on inflammatory cells. FEBS Letters. 1994;354:149–154. doi: 10.1016/0014-5793(94)01099-4. [DOI] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES