This Article Abstract Full Text (PDF) All Versions of this Article: 92/1/e1    most recent 01.RES.0000051887.97625.07v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Woo, S.-H. Articles by Morad, M. Search for Related Content PubMed PubMed Citation Articles by Woo, S.-H. Articles by Morad, M. Related Collections Calcium cycling/excitation-contraction coupling Imaging

(Circulation Research. 2003;92:e1.)
© 2003 American Heart Association, Inc.

UltraRapid Communications

Spatiotemporal Characteristics of Junctional and Nonjunctional Focal Ca²⁺ Release in Rat Atrial Myocytes

Sun-Hee Woo, Lars Cleemann, Martin Morad

From the Department of Pharmacology, Georgetown University Medical Center, Washington, DC.

Correspondence to Prof Martin Morad, Dept of Pharmacology, Georgetown University Medical Center, 3900 Reservoir Rd NW, Washington, DC 20057. E-mail moradm{at}georgetown.edu

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Atrial myocytes have two functionally separate Ca²⁺ release sites: those in peripheral sarcoplasmic reticulum (SR) adjacent to the Ca²⁺ channels of surface membrane and those in central SR not associated with Ca²⁺ channels. Recently, we have reported on the gating of these two different Ca²⁺ release sites by Ca²⁺ current. In the present study, we report on the spatiotemporal properties of focal Ca²⁺ releases (sparks) occurring spontaneously in central and peripheral sites of voltage-clamped rat atrial myocytes, using rapid 2-dimensional (2-D) confocal Ca²⁺ imaging. Peripheral and central sparks were similar in size and release time (300 000 Ca²⁺ ions for 12 ms), but significantly larger and longer than ventricular sparks. Both sites were resistant to Cd²⁺ and inhibited by ryanodine. Peripheral sparks were brighter and flattened against surface membrane, had 5-fold higher frequency, 2 times faster diffusion coefficient, and dissipated abruptly. Central sparks, in contrast, occurred less frequently, were elongated along the cellular longitudinal axis, and dissipated slowly. Compound sparks (composed of 2 to 5 unitary focal releases) aligned longitudinally and occurred more frequently at the center. The diversity of peripheral and central sparks with respect to shape, frequency, and speed of spatial development and decay is consistent with regional ultrastructural heterogeneity of SR. The retarded dissipation of central atrial sparks, and high prevalence of compound sparks in cell center may be critical in facilitating the propagation of Ca²⁺ waves in atrial myocytes lacking t-tubular system and provide the atrial myocytes with functional Ca²⁺ signaling diversity. The full text of this article is available at http://www.circresaha.org.

Key Words: atrial myocyte • Ca²⁺ spark • two-dimensional confocal imaging

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Contraction of mammalian ventricular myocytes is controlled by a sequence of events (E-C coupling) that includes Ca²⁺ current (I_Ca)-gated opening of Ca²⁺ release channels (RyRs) and the release of Ca²⁺ from the sarcoplasmic reticulum (SR).^1–5 The close (12 nm) proximity of the RyRs and Ca²⁺ channels (DHPRs) in dyadic junctions of transverse (t-) tubules^6,7 provides for the privileged Ca²⁺ cross signaling between the two sets of proteins.^8,9 Confocal Ca²⁺ imaging of ventricular myocytes reveals the presence of focal Ca²⁺ transients (Ca²⁺ sparks)¹⁰ that are triggered either spontaneously or by I_Ca.^10–13 The unitary property of Ca²⁺ sparks, but not their frequency, appears to be independent of I_Ca and voltage, indicating that they may represent the elementary event underlying cardiac E-C coupling.^5,10–15

In cell types lacking t-tubules (atrial cells, Purkinje cells, avian ventricular cells) RyRs are found not only as peripheral couplings, but also in nonjunctional or corbular SR throughout the central regions of the cell^16–20 with no direct association with DHPRs.¹⁶ The mode of gating and the contribution of this set of RyRs to cytosolic rise of Ca²⁺ remains uncertain. It has been suggested that nonjunctional RyRs might be activated by diffusion of Ca²⁺ from neighboring Ca²⁺ release sites.^20–24 Such a mechanism is likely to be slow and regenerative,²⁵ because it would lack the features that are associated with peripheral or dyadic junctions and the associated negative feedback mechanism.^5,8,26 Using 2-dimensional (2-D) confocal imaging in voltage-clamped rat atrial myocytes dialyzed with different concentration of Ca²⁺ buffer (Fluo-3, EGTA), we have recently shown that the peripheral focal Ca²⁺-release was directly activated by I_Ca, while the magnitude of Ca²⁺ release from the central SR was a cooperative function of peripheral Ca²⁺ release and dependent on a sequential saltatory Ca²⁺ propagating mechanism.²⁴ Although central and peripheral sparks have been previously described^{19,20,23,24,27,28} little is known of their spatiotemporal properties or comparative characteristics. In the present study, we quantify and characterize atrial Ca²⁺ sparks originating from RyRs associated with DHPRs and those not in close proximity of the L-type Ca²⁺ channels using rapid (240 Hz) 2-D confocal imaging. Our study shows that peripheral sparks occur more frequently, develop with higher rate of spatial growth, and dissipate faster, but have similar average Ca²⁺ release flux as those of the central sparks. The central sparks were significantly retarded in their spatial decline, which may facilitate the Ca²⁺ diffusion–dependent Ca²⁺ release mechanism, in the absence of close proximity of Ca²⁺ channels.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Whole-Cell Patch Clamp
The surgical removal of hearts of Wistar WKY rats (250 to 280 g, 32 animals; Charles River Laboratories, Wilmington, Mass) was performed in accordance with institutional and national ethical guidelines. Atrial myocytes were dispersed enzymatically²⁴ and voltage-clamped in the whole-cell configuration²⁹ using glass micropipets (tip resistance 3 M) containing (in mmol/L) 110 CsOH, 110 Aspartic acid, 5 NaCl, 20 TEA-Cl, 20 HEPES, 5 Mg-ATP, 0.2 cAMP, 1 K₅fluo-3 (Molecular Probes Inc.), and 2 EGTA, with the pH adjusted to 7.2 with CsOH. Membrane currents were measured as described previously.²⁴ cAMP (200 µmol/L) was used to prevent run-down of I_Ca during long periods of cell dialysis and to accomplish Ca²⁺ loading of the SR in spite of the added Ca²⁺ buffers.^8,9 Large concentrations of the fluorescent Ca²⁺ indicator fluo-3 in combination with EGTA were used to sharpen the images of focal Ca²⁺ releases by limiting the diffusion distance of free Ca^2+,13,30 so that spatial and temporal resolutions were determined primarily by the less-mobile Ca²⁺-bound fluo-3.³¹ The external solution contained (in mmol/L) 137 NaCl, 5.4 KCl, 2 CaCl₂, 10 HEPES, 1 MgCl₂, 10 Glucose, buffered to pH 7.4 with NaOH. Alternate switching between various external solutions was accomplished rapidly (<50 ms) using an electronically controlled multibarrel puffing system.⁴ All experiments were performed at room temperature (22 to 24°C).

Two-Dimensional Confocal Ca²⁺ Imaging
Cells were loaded with the Ca²⁺ indicator dye fluo-3 via the patch pipette (see previous section) and were imaged using a Noran Odyssey XL 2-D laser scanning confocal microscopy system (Noran Instruments) as previously described.^13,24,32 Images were scanned at 30 Hz (Figures 1 and 3) or 240 Hz (Figures 2, 4, 5, 6, and 7) depending on the experimental objectives. Data were acquired by the Intervision program in a workstation computer (IRIX-operating system, Indy, Silicon Graphics). The measured point-spread function of the confocal microscope was approximately a truncated cylinder 0.3 µm in radius and 0.8 µm in length. Fluorescence measurement was performed following 6 to 7 minutes after rupture of the membrane with the patch pipette.²⁴ Recording of spontaneous sparks was preceded by a train of ten 100-ms conditioning pulses applied from -90 to -10 mV at 0.1 Hz in 5 mmol/L Ca²⁺-containing external solution. This procedure produced I_Ca-gated Ca²⁺ transients of similar magnitude and kinetics, indicating stable Ca²⁺ loading of the SR in the presence of Ca²⁺ buffers. We confirmed that this approach maintained the Ca²⁺ load of the SR by measuring caffeine-triggered Ca²⁺ release (see Figure 2).

View larger version (99K): [in this window] [in a new window]   Figure 1. Spontaneous Ca2+ sparks in the periphery and center of rat atrial myocytes. A, Sequential 2-D confocal Ca2+ images measured at 30 Hz (at Vh=-80 mV) during periods indicated by box in B. Image Fav shows average fluorescence, revealing the clear outline of the myocyte. Numbered images, 45 to 52, are sequential frames measured differentially as the increase in fluorescence (F) relative to the average fluorescence. In the differential measurements of fluorescence, the outline of the cell is seen as only variations in the noise, but Ca2+ sparks appear clearly (arrowheads). B, Time course of the occurrences of peripheral (PERI) and central (CEN) sparks in cell shown partly in A. C, Average frequency of peripheral and central sparks measured in 22 atrial cells. **P<0.01, significantly different from values measured in the periphery.

View larger version (78K): [in this window] [in a new window]   Figure 2. Spatiotemporal profiles of caffeine-triggered local Ca2+ releases. Numbered sample frames were recorded as total fluorescence (F) at 240 Hz at the corresponding times indicated in the fluorescence tracings (F/Fr) of C. Fr indicates resting average fluorescence measured before the exposure to caffeine. The timing of the exposure to 10 mmol/L caffeine is indicated by arrows in B and C. Simultaneous measurement of the Na+-Ca2+ exchange current was shown in B. C, Inset, Diagram of masks used to measure fluorescence in the periphery (PERI) and center (CEN) of the images partly shown in A to produce the fluorescence tracings (C). Average magnitude of caffeine-triggered local Ca2+ transients measured in the periphery was not significantly different from that measured in the center (P>0.05, n=19).

View larger version (98K): [in this window] [in a new window]   Figure 3. Effects of Cd2+ and ryanodine on the Ca2+ spark occurrence. Top images of atrial myocytes in A and C show the average fluorescence intensity (Fav) obtained from 60 frames recorded at 30 Hz in cells voltage clamped at a holding potential -80 mV. Bottom images in A and C show four consecutive frames recorded at the times indicated by the dotted box in B and D that represent the time courses of spark occurrences in the periphery (PERI) and center (CEN). Left and right images in A and C show frames in the absence and presence of drug interventions, respectively. Notice that the Ca2+ sparks are brief so that at 30 Hz they are seen clearly only in single frames (arrowheads, A and C) and have, at most, a faint afterglow in the following frame.

View larger version (49K): [in this window] [in a new window]   Figure 4. Morphology and parametric measurements of peripheral (A) and central (B) Ca2+ spark. a, Sequential 2-D images of fluo-3 fluorescence recorded at 4.167-ms intervals. b, Local Ca2+ signals, measured at center of mass (red) and at surroundings with increasing distances from the center (gray, inner; light gray, outer), were normalized relative to the resting fluorescence and plotted vs time. c, Normalized local Ca2+ signals at center of release subtracted by that at the inner surrounding, where full-duration was measured at half-maximal amplitude (FDHM). d, Examples of the Gaussian approximations for peripheral (Ad) and central (Bd) sparks in single frame. Straight line (red line) in Ad indicates edge of the cell (see Materials and Methods). e, Time courses of intensity (F/Fo), full-width at half-maximum amplitude (FWHM), and area of Ca2+ spark measured by the Gaussian approximations. Color scales were shown underneath the images in Aa and Ba.

View larger version (32K): [in this window] [in a new window]   Figure 5. Parametric profiles of peripheral and central Ca2+ sparks. Histograms of Ca2+ spark peak amplitude (Fpeak/Fo, A), full-width at half-maximal amplitude (FWHM, B, measured at peak area), peak area (Apeak, C), and full-duration at half-maximal amplitude (FDHM, D) at periphery and center. Total 53 peripheral sparks and 102 central sparks detected in 33 cells were analyzed. Gaussian profile of spark parameters is described by y=aexp(-(x-b)2/(22)), where a indicates magnitude (number of spark), b is mean value of each parameters, and 2 indicates variance of the distribution. All 8 Gaussian distributions in the 4 panels were produced as the best fit with single peak and showed significant goodness of the fits (2 test, P>0.5).

View larger version (27K): [in this window] [in a new window]   Figure 6. Rapid spatial growth and dissipation of peripheral spark. A, Mean of 42 peripheral (PERI) and 75 central (CEN) Ca2+ spark amplitude profiles (normalized to the peak intensity). B, Time courses of mean spark area measured from the 42 peripheral and 75 central sparks (normalized to the peak area). Inset in B compares declining phase of the peripheral and central spark area, showing significantly higher speed of spark dissipation in the periphery. C, Mean of 42 peripheral and 75 central Ca2+ spark width (FWHM) profiles. D, Linear relations of mean FWHM2, calculated by the values in C, to the time. Slope of the linear functions in the periphery (PERI) was significantly larger than that in the center (CEN). Dotted curves in C were obtained by back-calculating the linear fit in D.

View larger version (61K): [in this window] [in a new window]   Figure 7. Compound Ca2+ spark. A and D, Sequential 2-D confocal images of peripheral (A) and central (D) compound spark. Centers of unitary focal Ca2+ releases were identified by the detection kernels (see Materials and Methods) shown in the insets of B and E. B and E, Contrast-enhanced images from the boxed area of raw image (16 ms) in A and 16 ms image in D using each kernel, respectively. Numbered fluorescence signal tracings in C and G were measured from the corresponding centers of focal Ca2+ releases using the unfiltered images of A and D, respectively. F, Location and orientation of the central compound spark (boxed area). Color scales for fluorescence intensity are shown below the images.

Image Analysis
Focal Ca²⁺ releases were identified by a computerized algorithm. This algorithm¹³ screened records based on their signal-to-noise ratio and identified local maxima in the interior of cell by means of a center-minus-surround detection kernel (inset in Figure 7E), which consists of pixels (0.207 µm spacing) approximating a central positively weighted disc (radius 0.6 µm) and a concentric negatively weighted ring (radius 1 to 1.4 µm). To detect local maxima underneath the cell membrane, the detection kernel was modified by excluding the extracellular space from the negatively weighted ring (inset in Figure 7B), and then aligned parallel with the direction of the cell edge. This procedure allowed to resolve sparks positioned close in time and produce records with local maxima that were identified as focal Ca²⁺ releases if they could be followed from frame to frame (at 240 Hz; often increasing in intensity, then spreading and fading). The focal Ca²⁺ releases that had one stationary center for their growth and decay were then subjected to Gaussian approximations,³² which allowed detailed morphometry and routine measurements of the amplitude, width, and equivalent area of sparks originating from the peripheral and central regions. Individual focal Ca²⁺ release signal in single frames were approximated in a restricted area (30x30 pixels) by a Gaussian function defined in space^32,37 by its center (x_o,y_o), the central increase in fluorescence (F), and its standard deviation (), giving a total fluorescence: equation

where F_o is the background fluorescence measured in the restricted area. The fitted values of were converted to full-width at half maximum amplitude (FWHM=2.3548). The size of the spark calculated as the equivalent area of background fluorescence, A, was derived from the following equation:equation

For peripheral Ca²⁺ sparks, the Gaussian distribution was approximated by the data obtained within a straight line defining the edge of the cell (Figure 4Ad), and the effective area was calculated accordingly (A_corr) using the error function (Erf) to compensate for the tail of the Gaussian distribution extending outside of the edge of the cell:equation

where l is the distance of the center of the Gaussian distribution to the edge. Central sparks were elongated (see Figure 4Ba) and could be approximated better by the following expression:equation

where the local coordinates are rotated by the angle [X=xcos()-ysin(), Y=xsin()+ycos()] to be aligned with long and short axes where the standard deviations (d and /d) ares, respectively, expanded and compressed by the factor d (eccentricity, d²-1).³⁸ We measured the angle of the longitudinal direction of the spark, , relative to the longitudinal axis of the cell, _cell, in such a way that the absolute difference, |-_cell|, is in the rage from 0° to 90°. This means that 0° corresponds to the exact alignment of the long axis of sparks in the direction of the cell, 90° corresponds to the exact alignment in the transverse direction of the cell, and 45° corresponds to random orientation . The expressions for the FWHM and area (Equation 2) of the spark remain unchanged in terms of the mean standard deviation equation

We have routinely measured the quality of the fit by calculating the standard deviation (SD, root-mean-square) between the raw data and the fitted Gaussian distribution. We compared these SD values to an estimate of the noise (individual pixels as compared with the mean their 8 nearest neighbors) and found that they agree within 20% in 85% of analyzed sparks.

To calculate frequency of sparks, central and peripheral areas were estimated by considering the major part of cell image excluding both ends of the cell, recorded at 30 Hz (see Figures 1A, 3A, and 3C), to be a square. The area up to 1.5 µm immediately underneath the cell membrane was denoted as the peripheral domain (peripheral area=2x1.5 µmximage length, µm). The central area was then measured as a difference between the whole image area and the peripheral area.

Only cells with little leak current and clear edges were included in the final analysis of the results. Statistical comparisons were performed using Student’s t test. Differences were considered to be statistically significant to a level of P<0.05. Numerical results are given as mean±SEM.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Spontaneous Focal Ca²⁺ Releases in the Periphery and Center
Figure 1A shows 2-dimensional confocal images of spontaneous Ca²⁺ sparks in a rat atrial myocyte voltage clamped at -80 mV. Imaging in extended periods (2 seconds) at 30 frames/s were used to monitor the major part of the cell and compensate for the sparsity of spontaneously occurring sparks. Spontaneous Ca²⁺ sparks (arrowheads) were detected in both the periphery (images 45 and 46) and center (images 47 and 49) of myocyte. At imaging speeds of 30 Hz each, Ca²⁺ spark was clearly seen for only a single frame. In 22 cells (12 hearts) examined, we found spontaneous sparks occurrence at rates of 2.5±0.9/s in the peripheral (220±19 µm²) and 2.6±0.6/s in the central regions (1008±261 µm²). Thus, the frequency of occurrence of sparks, normalized for the myocyte area (sparks/[10³ µm² second]), was significantly higher (5 times; P<0.01) in the periphery (11.2±1.25) compared with the center (2.44±0.48) (Figure 1C).

To examine if the regional spark frequency differences was caused by differential Ca²⁺ loading of the SR,^10,33 local Ca²⁺ transients were measured in response to "puffs" of caffeine (10 mmol/L) after the conditioning pulses (see Materials and Methods). Confocal images recorded at 240 Hz show that the shape (rise and decay times) and magnitude of caffeine-induced Ca²⁺ release was not significantly different in the periphery and center of the cell (Figures 2A, 2C, and 2D). Quantification of Ca²⁺ release in the periphery and center suggests an increase in fluorescence (F/F_r) of 3.49±0.39 in the periphery and 3.32±0.43 in the center (n=19, P>0.05; Figure 2D), and a velocity of release (time-to-peak) of 216±25 ms in the periphery versus 220±31 ms in the center (n=19, P>0.05). The rise in Ca²⁺ concentrations, as expected, activated only a brief and small Na⁺-Ca²⁺ exchange current (V_h=-80 mV),³⁴ in part because cytosolic Ca²⁺ is well buffered by 2 mmol/L EGTA and 1 mmol/L fluo-3 (Figure 2B).⁹ The data suggest that SR Ca²⁺ loading is uniform in different regions of the myocytes.

To examine if the higher spark frequency in the cell periphery was related to the activity of Ca²⁺ channels, preferentially associated with junctional release sites,³⁵ cadmium (Cd²⁺) was used to block the Ca²⁺ channels. Figure 3A compares sequential images of a myocyte clamped at -80 mV before and after exposure to 50 µmol/L Cd²⁺. Full suppression of I_Ca (activated at 0 mV) failed to alter the spontaneous Ca²⁺ spark frequency in the periphery or center of the cells (Figure 3A and 3B). Spark frequency (events/[10³ µm² second]) was 11.8±1.2 in the periphery and 2.4±0.2 in the center in control, and 13.2±1.6 in the periphery and 2.8±0.2 in the center in the presence of Cd²⁺ (P>0.05, n=6). On the other hand, 10 µmol/L ryanodine (known to block the RyRs by causing long-lasting subconductance states)³⁶ completely suppressed spontaneous sparks occurrences in 1 minute (n=9). At times <30 seconds after the application of ryanodine, when a submaximal inhibitory effect was achieved, the regional ratio of spark occurrences of peripheral and central release remained constant (data not shown). These results are consistent with previous reports that the spontaneous Ca²⁺ sparks in resting cardiac cells are generated by RyRs independent of Ca²⁺ channel activity.¹⁰

2-D Spatiotemporal Properties of Peripheral and Central Spark
Figure 4 compares the unitary properties of representative spontaneous Ca²⁺ sparks measured at 240 Hz at the periphery (Figure 4A) and center (Figure 4B) of an atrial cell. As expected, peripheral sparks were flattened against the surface membrane (Figure 4Aa, dashed line). Such releases were first analyzed by measuring the fluorescence intensity at the spark center versus the surrounding areas with increasing distances from the center, excluding the space outside of the membrane (Figure 4Ab, inset). Compared with the normalized peak fluorescence at the center of the mass (red, Figure 4Ab, F/F_o=2.8), the more distant signals from the center were reduced and delayed in amplitude (Figure 4Ab, gray and light gray), consistent with Ca²⁺ diffusion. Figure 4Ac shows the difference between the normalized signals of the center of the mass and the area immediately adjacent to it (inner surround). The duration of Ca²⁺ spark was measured as the time at half-maximal amplitude (FDHM, 15 ms) of the normalized center-minus-surround signal, with temporal resolution of 2 ms. Figure 4B shows the spatial and temporal profiles of a Ca²⁺ spark monitored in the cell interior. Ca²⁺ signal intensity (F/F_o) was highest at the center of the spark and decreased gradually with distance in the surrounding rings (Figure 4Bb) in a manner similar to peripheral sparks (Figure 4Ab). The FDHM of central spark, estimated by the similar method (center-minus-surround, Figure 4Ac), was 22 ms (Figure 4Bc).

2-D spark images were also analyzed in greater detail by approximating the individual focal Ca²⁺ release signal in single frames by a spatial Gaussian function^32,37 (see Materials and Methods). For peripheral Ca²⁺ sparks, the Gaussian distribution was approximated by the data obtained within a straight line defining the edge of the cell (arrowheads, Figure 4Ad). We used this method (Equation 3) rather than eccentricity (Equation 4) to account for the flattening of the spark against the membrane. Figure 4Ae illustrates the time courses of normalized amplitude (F/F_o), width (FWHM), and equivalent area (Area) of the peripheral Ca²⁺ spark measured using Gaussian analysis. The amplitude was measured as peak fluorescence of Gaussian curve (F) normalized relative to average resting background fluorescence (F_o) in the restricted area (30x30 pixels; Figures 4Ad and 4Bd). Spark amplitude reached peak value of 1.75 within 8 ms and then decayed to 50% of its peak value in 20 ms. The peak amplitudes of sparks (F/F_o) in the cell periphery measured by the Gaussian approximations were usually somewhat smaller than the peak Ca²⁺ signals obtained using pixel measurement at the center of release (compare Figures 4Ab and 4Ae). This difference arises primarily because the F_o used for normalization of the pixel measurements is so close to the edge that it is somewhat reduced as compared with the value that pertains to the Gaussian measured over an area extending away from the edge of the cell. The time course of spark width revealed a gradual enlargement over 4 to 12 ms that succeeds an initial rapid expansion phase (<4 ms) and is then followed by a maintained constant phase (<12 ms). The effective spark area (A_corr, µm²) increased progressively to a peak value of 5.3 µm² in 12 ms, where the spark width was 1.75 µm. This rising phase appeared to be terminated abruptly and followed by a rapid decline.

Central sparks often appeared elongated with roughly elliptical isofluorescence curves (Equation 4; Figures 4Ba and 4Bd). The eccentricity (d²-1),³⁸ the degree of deviation of the spark from circular symmetry, measured in 36 central sparks, showed eccentricity >0.5 (0.5 to 2) in 23, and of 0.25 to 0.5 in 11 sparks. Only 2 sparks had eccentricity <0.25. The long axes () of each spark with eccentricity >0.5, when compared with the direction of longitudinal axis of the cell (_cell), showed that the long axis of the sparks was predominantly oriented in the longitudinal cellular axis (|_cell-|=21±3.4°). This means that sparks typically had orientations in the range from -21° to +21° relative to the axis of the cell, which is significantly different from random orientation (|_cell-|=45°; see Materials and Methods). Figure 4Be shows time courses of central spark parameters measured by the Gaussian approximations. Peak value of 2.0 was achieved for spark brightness (F/F_o) in 8 ms. The spread of the fluorescence signal in space had an initial rapid expanding phase of 4 ms, followed by a slower growth. Area of the spark reached its peak value of 7 within 12 ms, when FWHM value was 1.8 µm. Interestingly the effective area of the central spark declined slowly compared with the peripheral spark (compare bottom panels in Figures 4Ae and 4Be).

The profiles of unitary characteristics of 53 peripheral and 102 central spontaneous Ca²⁺ sparks in 33 cells are quantified in Figure 5. Peripheral sparks showed larger fluorescence amplitudes with rightward shift of the histogram compared with central sparks (Figure 5A). Average peak amplitude (F_peak/F_o) of peripheral sparks (1.19±0.03) was significantly higher than that of the central sparks (0.93±0.01; P<0.001). The distribution of peripheral FWHM (measured at the time when the area reached its peak) was similarly shifted to the right (Figure 5B), with an average FWHM of 2.76±0.20 µm for peripheral sparks, significantly larger than the width of the central sparks (2.26±0.10 µm; P<0.05). Profiles for peak area of the sparks revealed broader distribution of the central spark areas between 1 and 9 µm² compared with narrower distribution 2 to 7 µm² in the periphery (Figure 5C), but the difference in the mean values was insignificant (periphery, 4.5±0.2; center, 4.4±0.2; P>0.05). In this context it should be recalled that the effective area was derived not only from its amplitude and FWHM (Equation 2), but was corrected by cutting the tail of the Gaussian distribution extending outside the cell (Equation 3). There was considerable variation in duration (FDHM) of individual peripheral or central Ca²⁺ sparks, producing a wide distribution histogram (Figure 5D). Mean duration (ms) of the peripheral sparks (23.1±1.82) was not, however, significantly different from that of central sparks (21.5±2.21; P>0.05).

Time Courses of Peripheral and Central Sparks
Figure 6 compares the mean time courses of amplitude, area, and width of 75 central and 42 peripheral sparks (29 cells), which showed single focal Ca²⁺ release site (see Materials and Methods) for 30 ms after their activation. The time-to-peak of central ("CEN") and peripheral ("PERI") Ca²⁺ spark amplitudes were similar (8 ms; Figure 6A). There was also no difference in the decay time constant () of the spark amplitude between peripheral (12.0±1.1 ms) and central sparks (14.1±1.2 ms; P>0.05; Figure 6A), but there was significant difference in the time course of their area (Figure 6B). Peripheral sparks dissipated immediately after reaching their peak with a clear break point. In sharp contrast, the area of central sparks reached its peak gradually and was maintained for 16 ms at its peak value, before its dissipation (Ca²⁺ removal in space). Efficacy of spark dissipation was evaluated as the slope (µm²/ms) of the declining phase of the area (Figure 6B, inset). The mean declining slope of the peripheral sparks (-0.106±0.006) was significantly steeper than that of central sparks (-0.033±0.009; P<0.001).

To investigate whether the difference in the regional spark dissipation was caused by Ca²⁺ diffusion during the focal Ca²⁺ release, we evaluated diffusion coefficients of focal Ca²⁺ release using the signal averaged time courses of spark width in the cell periphery and center (Figure 6C). As predicted from Fick’s diffusion law, FWHM was expected to increase proportional with the square root of time, as shown by the linear fit of a straight line when plotting FWHM² against time (Figure 6D). The dotted curves in Figure 6C were obtained by back-calculating the linear functions in Figure 6D. The slopes of the linear fits (periphery, 0.50±0.022, n=42; center, 0.27±0.012, n=75; P<0.001; Figure 6D) could then be used to estimate the apparent diffusion coefficient, D³⁹: C/C_o=exp(-r²/4Dt), where C/C_o is the Ca²⁺ concentration at any given time normalized to the Ca²⁺ concentration at the center of the spark at the same time. Then, D can be derived from, D=(slope)/(-16ln(C/C_o)). In this approximation (dotted curve in Figure 6C), the FWHM at 0 ms gives a rough estimate of lateral point-spread-function (optical blurring of confocal microscopy, 0.7 to 1 µm) of the spark. Apparent diffusion coefficient (D, µm²/s) for the fluo-3-bound Ca²⁺ during spatial growth of spark was 25 at center and 45 at periphery, suggesting 1.8 times faster diffusion coefficients of the peripheral focal Ca²⁺ releases. The spatial spread of peripheral sparks terminated at 12 ms as the spark width reached a value of approximately 2.5 µm (Figure 6C, top panel), whereas the central sparks progressively expanded for 25 ms as the FWHM reached about 2.7 µm (Figure 6C, top panel). These results suggest that the peripheral sparks initially grow significantly faster in space, but that diffusion alone is not sufficient to account for their time course.

Quantification of Focal Ca²⁺ Releases
The size of Ca²⁺ spark was estimated in terms of the number of Ca²⁺ ions required to produce a spark. The following assumptions were made in this estimation: (1) the dissociation constant (K_d) of the dye is substantially larger than the background Ca²⁺ concentration, [Ca²⁺]_bg, (2) the total concentration of the fluo-3 within the cell, [fluo]_t, is the same as that of pipette solution, and (3) the point-spread-function in the vertical direction, h, is large enough to cover most of the spark. Considering these assumptions the amount of Ca²⁺ (S, mole) associated with a spark can be calculated from the following: equation

where A (or A_corr for peripheral sparks), the mean area of sparks (µm²), and h ( 0.8 µm) together determine the equivalent volume, [fluo]_t/K_d is the buffering capacity ([1 mmol/L]/[0.4 µmol/L] 2500), and [Ca²⁺]_bg is 60 nmol/L.⁹ Computer simulations of Ca²⁺ binding by fluo-3 and EGTA suggested that, in these experiments, the time resolution of measurements of Ca²⁺ release depends mainly on diffusion of Ca²⁺ bound fluo-3.^13,31 Thus, mean amounts of focal Ca²⁺ releases were similar in the periphery, 336 000±30 000 Ca²⁺ ions, and in the center, 318 000±30 000 Ca²⁺ ions, for 12 ms release time. Therefore, the flux of Ca²⁺ release associated with a spark was 4.5x10^-17 moles/s, corresponding to ionic currents of 8.7 pA.

Compound Ca²⁺ Spark
We occasionally found localized Ca²⁺ release composed of two to five unitary events occurring more or less synchronously, "compound spark," occupying normally more than two sarcomeres. A total of 19 compound sparks were detected in 8 myocytes out of 33 cells, 14 of which were observed in the cell interior and 5 in the periphery. Considering the 5 times large central region the occurrence of compound sparks normalized for area was 2 times higher in the cell interior. Figure 7A and 7D display the morphology of representative compound sparks monitored in the periphery and center of the same atrial myocyte. Figure 7B and 7E show contrast-enhanced images of the 16-ms images in Figure 7A and 7D, respectively, using center-minus-surround kernel (inset),¹³ representing well-resolved focal maxima of the Ca²⁺ releases (numbered sites). The compound sparks occurring at the periphery were mostly composed of well-defined two or three smaller unitary release events (Figure 7B), whereas the central compound sparks had two to five unitary events (Figure 7E) distributed in a longitudinal direction of the cell (80%; see Figure 7F). The center-to-center separations between the subunits of peripheral and central compound sparks were 1.4 to 2 µm. The time courses of fluorescence, measured from the centers of each unitary release composing a compound spark, revealed similar rise and decay of the signals (Figure 7C and 7G).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
The rapid 2-D confocal imaging and analysis of Ca²⁺ sparks in atrial myocytes provides the first 2-dimensional comparison of peripheral and central Ca²⁺ sparks. We report three rather novel findings: (1) although the frequency of spontaneous unitary sparks was significantly higher in the periphery, the compound sparks appeared more prevalent in cell center; (2) peripheral and central sparks were similar in size and release time (300 000 Ca²⁺ ions in 12 ms), but were significantly larger than ventricular sparks (100 000 Ca²⁺ ions in 7 ms);^13,31 and (3) peripheral sparks were flattened against the surface membrane, grew faster in space, and dissipated abruptly and rapidly, as compared with central sparks, which were elongated along the cellular longitudinal axis and grew slower (2 times) in space and were retarded in their dissipation. The diversity in unitary functional characteristics of atrial myocytes appears to be consistent with their ultrastructural regional differences^16,17 and may provide the atrial myocyte with its unique E-C coupling characteristics.

Faster Expansion and Dissipation of Peripheral Spark
The dominant process (75%) that determines the decline of Ca²⁺ spark is thought to be Ca²⁺ diffusion.⁴⁰ In the present study, we found that the diffusion of released Ca²⁺ was about 2 times faster in the periphery than the center (D=45 versus 25 µm²/s). The faster Ca²⁺ diffusion in the peripheral sites may arise from the proximity of these sites to the cell membrane acting as a diffusion barrier, providing also for the flattening of the spark on one side (Figure 4A). Because the amount of Ca²⁺ released is similar in both the peripheral and central sparks (300 000 ions), the limited diffusion space in the peripheral junctions can account for the apparent faster Ca²⁺ diffusion constant. In fact, the surface membrane may serve as a reflective boundary to diffusion, which would effectively double the strength of the Ca²⁺ source, thereby increasing their apparent peak amplitudes and their rapid spatial expansion and dissipation. The exogenous Ca²⁺ buffers introduced in the myocytes are not likely to be responsible for the differential regulation of the time courses of peripheral and central sparks as Ca²⁺ buffering would be equally effective throughout the cytosol.^13,31,32

The enhancement of the activity of SR Ca²⁺-ATPase has been reported to accelerate the decay of spark amplitude and reduce its width.^40,41 It is possible therefore that the larger peak local concentration of Ca²⁺ at the peripheral sites could have activated more rapid Ca²⁺ removal by SR Ca²⁺-ATPase in its close vicinity. In our study, however, the decay time constant of spark amplitude was not found to be significantly different between the two regions (Figure 6A). It is likely that the routine use of 200 µmol/L cAMP in the patch pipette masks such a mechanism. We also considered the contribution of Na⁺-Ca²⁺ exchanger to the rapid dissipation of the peripheral sparks, as high levels of exchanger appear to be expressed in the interdyadic junctional membranes,^30,42 making the exchanger favor its Ca²⁺ efflux mode following the release of Ca²⁺. Nevertheless, the role of Na⁺-Ca²⁺ exchanger in the decay kinetics of Ca²⁺ spark amplitude remains somewhat controversial as inhibition of the exchanger in 0 Na⁺- and 0 Ca²⁺-external solutions failed to alter the kinetics of decay of spark amplitude.^40,43 We also failed to find a significant difference in the decay time constant of amplitude of sparks at cell periphery and center, thus minimizing the role of Na⁺-Ca²⁺ exchanger in determining the kinetics of the spark amplitude (Figure 6A).

Eccentricity of the Central Sparks
We often found that central sparks were elongated in longitudinal direction (Figure 4Ba). The eccentricity of spark has been previously reported for ventricular myocytes (18% longer longitudinally)⁴⁴ and skeletal muscle.^38,45 In skeletal muscle the eccentricity of spark appears to be aligned with the Z planes,³⁸ especially when caffeine was used to sensitize the RyRs. We find that eccentricity in spark appearance is also present during the growing phase of the central spark (Figure 4Ba). In addition, the elongated sparks were somewhat symmetric in the Gaussian approximation, and the eccentricities were constant over time and not correlated with amplitude or width (data not shown). These observations suggest that the eccentricity is not caused by anisotropy in the diffusion or binding properties of Ca²⁺ pump or SR membranes, etc,^41,44 but rather by the Ca²⁺ release from a spatially extended source, such as corbular SR with multiple channels.³⁸ The elongation of central sparks in the longitudinal direction may be critical in often observed longitudinal orientation of spontaneous Ca²⁺ propagation waves in the cell, also seen for the central compound sparks (Figure 7D).

Size and Frequency of Atrial Ca²⁺ Spark
The amount of Ca²⁺ released during activation of peripheral and central sparks was similar (300 000 Ca²⁺ ions for 12 ms; equivalent to 8.7 pA). There were, however, significant variations in the duration (FDHM, Figure 5D) and area of sparks, with prominent variations mostly in the peak area of the central sparks compared with peripheral sparks (Figure 5C). This is consistent with the existing structural evidence on variation in the number of RyRs (feet) in the junctional and nonjunctional domains of the SR.⁷ In addition, both our data and the structural data support the idea that Ca²⁺ sparks result from the opening of multiple Ca²⁺ release channels clustered within discrete SR domains. Interestingly, the amount of the focal Ca²⁺ release in rat atrial myocytes was significantly larger than that measured in rat ventricular cells under similar conditions (100 000 Ca²⁺ ions for 7 ms, 4.6 pA).^13,31 Although the larger Ca²⁺ spark finding is consistent with the atrial ultrastructural data,¹⁶ the higher extracellular Ca²⁺ (5 mmol/L) used in the present study could have contributed to the larger Ca²⁺ release. This possibility was evaluated in some experiment by using the same [Ca²⁺]_o. Consistently, we found atrial sparks to be significantly larger in duration and size than ventricular sparks in the rat heart (unpublished data, 2001).

The frequency of occurrence of spontaneous spark was significantly higher in the cell periphery than in its center (Figure 1C), as also reported for cat atrial myocytes.²³ Because the Ca²⁺ content of the SR in peripheral and central domains are similar (Figure 2), and both sites are resistant to Cd²⁺ but sensitive to ryanodine (Figure 3), one possible explanation for the higher frequency of spontaneous sparks is that there is a higher density of dyadic couplings in the periphery as suggested from ultrastructural studies in rat, rabbit, and mouse.^16,46,47 It has been also recently suggested that the higher peripheral spark frequency arises from the spatial differences in the RyR sensitivity to Ca²⁺. Such "eager sites,"¹⁹ found primarily in subsarcolemmal regions, were suggested to have the highest frequency of spontaneous activity, not associated with SR Ca²⁺ load or Ca²⁺ influx.¹⁹

Interestingly, in sharp contrast to the frequency of unitary sparks, the compound sparks appear to be more frequent in the cell interior. It is possible that the retarded dissipation of unitary central focal Ca²⁺ releases may facilitate the activation of neighboring release sites, leading to recruitment of the larger number of unitary events associated with the compound sparks (Figure 7D).

Physiological Implication
The present findings suggest that junctional and nonjunctional atrial Ca²⁺ sparks differ significantly with respect to their spatial and temporal characteristics and frequency. Highly frequent peripheral sparks may be related to large number of dyadic couplings or the "eager sites," reported in rat atrial myocytes.¹⁹ The 2 times faster diffusion and larger amplitude of focal Ca²⁺ releases in the peripheral junctions may be critical in activation of central release sites not associated with the Ca²⁺ channels. It is likely that the focal Ca²⁺ releases in the periphery may efficiently regulate Ca²⁺-dependent membrane conductance,^21,48 and/or inositol 1,4,5-trisphosphate receptors associated with peripheral SR in response to hormonal interventions.²⁸ The diversity of peripheral and central sparks with respect to shape, frequency, and speed of spatial development and decay is consistent with regional ultrastructural differences of SR. The significantly larger and longer atrial sparks, the retarded dissipation of central atrial sparks, and high prevalence of compound sparks in cell center maybe critical in facilitating the propagation of Ca²⁺ waves in atrial myocytes lacking t-tubular system. Such Ca²⁺ propagation waves in the cell interior have been observed during Ca²⁺ current activation of atrial myocytes lacking t-tubules.^19,20,24

	Acknowledgments

 
This work was supported by the National Institutes of Health grant (HL-16152 to M.M.) and American Heart Association (0225383U, to S.-H.W.). The authors would like to thank Dr John. C. Pezzullo for helpful discussion and help on statistical evaluation of imaging data and Alexandru N. Mihai for his help for algorithm for image analysis.

Received August 16, 2002; revision received November 27, 2002; accepted December 2, 2002.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Beuckelmann DJ, Wier WG. Mechanism of release of calcium from sarcoplasmic reticulum of guinea-pig cardiac cell. J Physiol. 1988; 405: 233–255.[Abstract/Free Full Text]
2. Näbauer M, Callewaert G, Cleemann L, Morad M. Regulation of calcium release is gated by calcium current, not gating charge, in cardiac myocytes. Science. 1989; 244: 800–803.[Abstract/Free Full Text]
3. Niggli E, Lederer WJ. Voltage-independent calcium release in heart muscle. Science. 1990; 250: 565–568.[Abstract/Free Full Text]
4. Cleemann L, Morad M. Role of Ca²⁺ channel in cardiac excitation-contraction coupling in the rat: evidence from Ca²⁺ transients and contraction. J Physiol. 1991; 432: 283–312.[Abstract/Free Full Text]
5. Morad M. Cardiac Ca²⁺ signaling. In: Morad M, Kostyuk P, eds. Calcium Signaling. Washington, DC: IOS press; 2001: 331: 3–13.
6. Sun XH, Protasi F, Takahashi M, Takeshima H, Ferguson DG, Franzine-Armstrong C. Molecular architecture of membrane involved in excitation-contraction coupling of cardiac muscle. J Cell Biol. 1995; 129: 659–671.[Abstract/Free Full Text]
7. Franzine-Armstrong C, Protasi F, Ramesh V. Shape, size, and distribution of Ca²⁺ release units and couplons in skeletal and cardiac muscles. Biophys J. 1999; 77: 1528–1539.[Abstract/Free Full Text]
8. Sham JSK, Cleemann L, Morad M. Functional coupling of Ca²⁺ channels and ryanodine receptors in cardiac myocytes. Proc Natl Acad Sci U S A. 1995; 92: 121–125.[Abstract/Free Full Text]
9. Adachi-Akahane S, Cleemann L, Morad M. Cross-signaling between L-type Ca²⁺ channels and ryanodine receptors in rat ventricular myocytes. J Gen Physiol. 1996; 108: 435–454.[Abstract/Free Full Text]
10. Cheng H, Lederer WJ, Cannell MB. Calcium sparks: elementary events underlying excitation-contraction coupling in heart muscle. Science. 1993; 262: 740–744.[Abstract/Free Full Text]
11. Wier WG, Egan TM, López-López JR, Balke CW. Local control of excitation-contraction coupling in rat heart cells. J Physiol. 1994; 474: 463–471.[Abstract/Free Full Text]
12. Cannell MB, Cheng H, Lederer WJ. Spatial non-uniformities in [Ca²⁺]_i during excitation-contraction coupling in cardiac myocytes. Biophys J. 1994; 67: 1942–1956.[Abstract/Free Full Text]
13. Cleemann L, Wang W, Morad M. Two-dimensional confocal images of organization, density, and gating of focal Ca²⁺ release sites in rat cardiac myocytes. Proc Natl Acad Sci U S A. 1998; 95: 10984–10989.[Abstract/Free Full Text]
14. Shacklock PS, Wier WG, Balke CW. Local Ca²⁺ transients (Ca²⁺ sparks) originate at transverse tubules in rat heart cells. J Physiol. 1995; 487: 601–608.[Medline] [Order article via Infotrieve]
15. Parker I, Zang WJ, Wier WG. Ca²⁺ sparks involving multiple Ca²⁺ release sites along Z-lines in rat heart cells. J Physiol. 1996; 497: 31–38.[Medline] [Order article via Infotrieve]
16. Carl LS, Felix K, Caswell 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. J Cell Biol. 1995; 129: 673–682.[Abstract/Free Full Text]
17. McNutt NS, Fawcett DW. The ultrastructure of the cat myocardium, II: atrial muscle. J Cell Biol. 1969; 42: 46–67.[Abstract/Free Full Text]
18. Hatem SN, Bénardeau A, Rucker-Martin C, Marty I, Chamisso P, Villaz M, Mercardier J-J. Different compartments of sarcoplasmic reticulum participate in the excitation-contraction coupling process in human atrial myocytes. Circ Res. 1997; 80: 345–353.[Abstract/Free Full Text]
19. Mackenzie L, Bootman MD, Berridge MJ, Lipp P. Predetermined recruitment of calcium release sites underlies excitation-contraction coupling in rat atrial myocytes. J Physiol. 2001; 530: 417–429.[Abstract/Free Full Text]
20. Kockskämper J, Sheehan KA, Bare DJ, Lipisius SL, Mignery GA, Blatter LA. Activation and propagation of Ca²⁺ release during excitation-contraction coupling in atrial myocytes. Biophys J. 2001; 81: 2590–2605.[Abstract/Free Full Text]
21. Lipp P, Pott L, Callewaert G, Carmeliet E. Simultaneous recording of Indo-1 fluorescence and Na⁺/Ca²⁺ exchanger current reveals two components of Ca²⁺-release from sarcoplasmic reticulum of cardiac atrial myocytes. FEBS Lett. 1990; 275: 181–184.[CrossRef][Medline] [Order article via Infotrieve]
22. Berlin JR. Spatiotemporal changes of Ca²⁺ during electrically evoked contractions in atrial and ventricular cells. Am J Physiol. 1995; 267: H1165–H1170.
23. Hüser J, Lipsius SL, Blatter LA. Calcium gradients during excitation-contraction coupling in cat atrial myocytes. J Physiol. 1996; 494: 641–651.[Medline] [Order article via Infotrieve]
24. Woo SH, Cleemann L, Morad M. Ca²⁺ current-gated focal and local Ca²⁺ release in rat atrial myocytes: evidence from rapid 2-D confocal imaging. J Physiol. 2002; 543: 439–453.[Abstract/Free Full Text]
25. Berridge MJ. Elementary and global aspects of calcium signaling. J Physiol. 1997; 499: 291–306.[Medline] [Order article via Infotrieve]
26. Soldatov NM, Oz M, O’Brien KA, Abernethy DR, Morad M. Molecular determinants of L-type Ca²⁺ channel inactivation. J Biol Chem. 1998; 273: 957–963.[Abstract/Free Full Text]
27. Blatter LA, Hüser J, Ríos E. Sarcoplasmic reticulum Ca²⁺ release flux underlying Ca²⁺ sparks in cardiac muscle. Proc Natl Acad Sci U S A. 1997; 94: 4176–4181.[Abstract/Free Full Text]
28. Mackenzie L, Bootman M, Laine M, Berridge MJ, Thuring J, Holmes A, LI W-H, Lipp P. The role of inositol 1,4,5-trisphosphate receptors in Ca²⁺ signalling and the generation of arrhythmias in rat atrial myocytes. J Physiol. 2002; 541: 395–409.[Abstract/Free Full Text]
29. Hamill OP, Marty A, Neher E, Sakmann B, Sigworth FJ. Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pflügers Arch. 1981; 319: 85–100.
30. Langer GA, Peskoff A. Calcium concentration and movement in the diadic cleft space of the cardiac ventricular cell. Biophys J. 1996; 70: 1162–1182.
31. Cleemann L, Wang W, Morad M. Calcium, and cellular metabolism.In. Sotelo JR, Bnesh JC, eds. Transport and Regulation New York, NY: Plenum; 1997; 57–65.
32. Wang W, Cleemann L, Jones LR, Morad M. Modulation of focal and global Ca²⁺ release in calsequestrin-overexpressing mouse cardiomyocytes. J Physiol. 2000; 524: 399–414.[Abstract/Free Full Text]
33. Györke I, Györke S. Regulation of the cardiac ryanodine receptor channel by luminal Ca²⁺ involves luminal Ca²⁺ sensing sites. Biophys J. 1998; 75: 2801–2810.[Abstract/Free Full Text]
34. Callewaert G, Cleemann L, Morad M. Caffeine-induced Ca²⁺ release activates Ca²⁺ extrusion via Na⁺-Ca²⁺ exchanger in cardiac myocytes. Am J Physiol. 1989; 268: 1045–1049.
35. Sheehan KA, Blatter L. Initiation mechanism and frequency of spontaneous Ca²⁺ sparks in atrial myocytes. Biophys J. 2002; 82: 281Abstract.
36. Meissner G, El-Hashem A. Ryanodine as a functional probe of the skeletal muscle sarcoplasmic reticulum Ca²⁺ release channel. Mol Cell Biochem. 1992; 114: 119–123.[CrossRef][Medline] [Order article via Infotrieve]
37. Shirokova N, Ríos E. Small event Ca²⁺ release: a probable precursor of Ca²⁺ sparks in frog skeletal muscle. J Physiol. 1997; 502: 3–11.[CrossRef][Medline] [Order article via Infotrieve]
38. Brum G, González A, Rengifo J, Shirocova N, Ríos E. Fast imaging in two dimensions resolves extensive sources of Ca²⁺ sparks in frog skeletal muscle. J Physiol. 2000; 528: 419–433.[Abstract/Free Full Text]
39. Crank J. The Mathematics of Diffusion. Oxford, UK: Clarendon Press; 1975: 28–29.
40. Gómez AM, Cheng H, Lederer WJ, Bers DM. Ca²⁺ diffusion and sarcoplasmic reticulum transport both contribute to [Ca²⁺]_i decline during Ca²⁺ sparks in rat ventricular myocytes. J Physiol. 1996; 496: 575–581.[Medline] [Order article via Infotrieve]
41. Smith GD, Keizer JE, Stern MD, Lederer WJ, Cheng H. A simple numerical model of calcium spark formation and detection in cardiac myocytes. Biophys J. 1998; 75: 15–32.[Abstract/Free Full Text]
42. Scriven DRL, Dan P, Moore DW. Distribution of proteins implicated in excitation-contraction coupling in rat ventricular myocytes. Biophys J. 2000; 79: 2682–2691.[Abstract/Free Full Text]
43. Goldhaber JI, Lamp ST, Walter DO, Garfinkel A, Fukumoto GH, Weiss JN. Local regulation of the threshold for calcium sparks in rat ventricular myocytes: role of sodium-calcium exchanger. J Physiol. 1999; 520: 431–438.[Abstract/Free Full Text]
44. Cheng H, Lederer MR, Xiao RP, Gómez AM, Zhou YY, Ziman B, Spurgeon H, Lakatta EG, Lederer WJ. Excitation-contraction coupling in heart: new insights from Ca²⁺-sparks. Cell Calcium. 1996; 20: 129–140.[CrossRef][Medline] [Order article via Infotrieve]
45. González A, Kirsch WG, Shirokova N, Pizarro G, Stern MD, Ríos E. The spark and its ember: Separately gated local components of Ca²⁺ release in skeletal muscle. J Gen Physiol. 2000; 115: 139–158.[Abstract/Free Full Text]
46. Sommer JR, Waugh RA. The ultrastructure of the mammalian cardiac muscle cell-with special emphasis on the tubular membrane system: a review. Am J Pathol. 1976; 82: 192–232.[Medline] [Order article via Infotrieve]
47. Bossen EH, Sommer JR, Waugh RA. Comparative stereology of mouse atria. Tissue Cell. 1981; 13: 71–77.[CrossRef][Medline] [Order article via Infotrieve]
48. Lipsius SL, Hüser J, Blatter LA. Intracellular Ca²⁺ release sparks atrial pacemaker activity. News Physiol Sci. 2001; 16: 101–106.[Abstract/Free Full Text]

This article has been cited by other articles:

				P. Sasse, J. Zhang, L. Cleemann, M. Morad, J. Hescheler, and B. K. Fleischmann Intracellular Ca2+ Oscillations, a Potential Pacemaking Mechanism in Early Embryonic Heart Cells J. Gen. Physiol., July 30, 2007; 130(2): 133 - 144. [Abstract] [Full Text] [PDF]

				M.-A. Bray, N. A. Geisse, and K. K. Parker Multidimensional Detection and Analysis of Ca2+ Sparks in Cardiac Myocytes Biophys. J., June 15, 2007; 92(12): 4433 - 4443. [Abstract] [Full Text] [PDF]