Spatial and Temporal Inhomogeneities During Ca2+ Release From the Sarcoplasmic Reticulum in Pig Ventricular Myocytes
The [Ca2+]i transient of ventricular myocytes during normal excitation-contraction coupling is the summation of primary Ca2+ release events, which originate at the junction of the sarcoplasmic reticulum (SR) and the T-tubular system. Studies in small mammals have shown a high density of release sites, but little is known of larger mammals. We have studied the spatial distribution of SR Ca2+ release in pig ventricular myocytes using a confocal microscopy. In 69 of 107 cells, large inhomogeneities of Ca2+ release were observed along the longitudinal scan line. Areas where the increase of [Ca2+]i was delayed (time to 50% of peak F/F0 [where F indicates fluorescence intensity, and F0 indicates F at rest] was 26±1 ms in delayed areas versus 11±2 ms in early areas) and smaller (peak F/F0 was 2.27±0.10 for delayed areas versus 2.69±0.13 for early areas; n=13 cells, P<0.05) could be up to 26 μm wide. The sum of all delayed areas could make up to 55% of the line scan. The spatial pattern was constant during steady-state stimulation and was not altered by enhancing Ca2+ channel opening or SR Ca2+ content (Bay K8644, isoproterenol). Imaging of sarcolemmal membranes revealed several areas devoid of T tubules, but SR Ca2+ release channels were homogeneously distributed. In contrast, compared with pig myocytes, mouse myocytes had a very dense T-tubular network, no large inhomogeneities of release, and a faster rate of rise of [Ca2+]i. In conclusion, in pig ventricular myocytes, areas of delayed release are related to regional absence of T tubules but not ryanodine receptors. This lower number of functional couplons contributes to a slower overall rate of rise of [Ca2+]i.
In ventricular cardiac muscle, Ca2+ release from the sarcoplasmic reticulum (SR) is the major source of the transient rise in [Ca2+]i during excitation-contraction coupling. This release is triggered by a local increase in Ca2+ near the Ca2+ release channels of the SR, the ryanodine receptors (RyRs). The major trigger is Ca2+ influx via sarcolemmal L-type Ca2+ channels, dihydropyridine receptors (DHPRs), although Ca2+ influx via Na+-Ca2+ exchange and T-type Ca2+ channels can also contribute (see review1). DHPRs and RyRs are organized in junctional complexes (couplons), which are predominantly found in the T tubules.2,3⇓ After reducing the opening probability of DHPRs and Ca2+ influx, localized Ca2+ release is observed during confocal microscopy as Ca2+ sparks.4,5⇓ Ca2+ sparks represent elementary events during normal excitation-contraction coupling in cardiac muscle, although evidence of smaller units has been reported,6 and more restricted Ca2+ release can be evoked during flash photolysis of caged Ca2+.7 The whole-cell [Ca2+]i transient is the result of a temporal and spatial summation of Ca2+ sparks (see review8). Several studies have shown that in ventricular myocytes, Ca2+ sparks occur at Z lines near T tubules, probably at the junctional complex.6,9,10⇓⇓ The extensive T-tubular system of ventricular myocytes ensures a rapid and homogeneous increase in [Ca2+]i throughout the cell. When the number of T tubules is decreased, Ca2+ release is more inhomogeneous.11 This role of the T tubules is also supported by observations in neonatal ventricular myocytes, atrial myocytes, and Purkinje cells, which lack T tubules. In these cells, Ca2+ release during depolarization occurs first below the sarcolemmal membrane, and the increase of [Ca2+]i in the center of the cell is small and delayed.12–16⇓⇓⇓⇓
So far, studies of spatially resolved SR Ca2+ release in cardiac myocytes have been performed in small mammals, which have a high heart rate. It is not known whether the mechanisms in larger mammals are similar. Pigs have an overall cardiac anatomy and function closely resembling those of humans and are therefore often used for studies of ischemic heart disease (eg, see Schulz et al17). They are also under extensive study for xenotransplantation.18 Therefore, we examined the spatial organization of SR Ca2+ release and T tubules in pig ventricular myocytes in near-physiological conditions.
Materials and Methods
Domestic pigs (3 to 6 months old, 30 to 53 kg, n=22) were housed and treated according to the Guide for the Care and Use of Laboratory Animals (National Institutes of Health), and experimental protocols were approved by the in-house ethics committee. Single ventricular myocytes were prepared as previously described19 and detailed in the online data supplement (available at http://www.circresaha.org). Only cells from the central midmyocardial layer were harvested and used for measurements within 24 hours. Ventricular myocytes were isolated from mice (2 to 3 months old) as previously described.20
Cells were studied at 37°C under whole-cell voltage clamp with a pipette solution containing (mmol/L) potassium aspartate 120, KCl 20, potassium HEPES 10, MgATP 5, MgCl2 0.5, NaCl 10, and K5-fluo-3 0.05, pH 7.20. The external solution contained (mmol/L) NaCl 130, KCl 5.4, sodium HEPES 11.8, MgCl2 0.5, CaCl2 1.8, and glucose 10, pH 7.35. Depolarizing steps of 225 ms from −70 to 10 mV were applied at frequencies of 0.5 or 1 Hz.
For imaging of T tubules, cells were incubated with di-8-ANEPPS (5 μmol/L, Molecular Probes) for 5 minutes, followed by washout. Cells from three hearts were used for staining of the RyR with antibodies (anti-RyR2, Affinity BioReagents, clone C3-33), as detailed in the online data supplement.
Confocal [Ca2+]i Measurements
Cells were studied using a Zeiss Axiovert 100M inverted microscope with a ×40/1.3 oil-immersion objective and a Zeiss LSM 510 confocal laser point-scanning system (Zeiss GmbH). Fluo 3 and di-8-ANEPPS were excited with the 488-nm line of a 25-mW argon laser. Pinhole size was set to 1 Airy unit, resulting in an optical slice thickness (Z direction) of 0.9 μm. Pixel width was between 0.21 and 0.36 μm.
To record [Ca2+]i transients, cells were scanned along the longitudinal axis, orthogonal to the Z lines, avoiding scanning through nuclei. The scanning speed and gain settings were adjusted to achieve an adequate contrast-to-noise ratio with a temporal resolution between 1.5 and 3.8 ms per line. Scanning speed, excitation, and amplification settings were kept constant during each experiment. Sequential line scans were stacked over time and are shown as 2D line-scan images.
To study the T-tubular distribution, a stack of adjacent XY images (Z stack) with a spacing of 0.44 μm in the Z direction and a pixel width of 0.11 to 0.20 μm in the XY plane was recorded.
Image analysis was performed with Scion Image and Microsoft Excel software and is detailed in the online data supplement. Intensity values of line-scan images (F) were normalized to the fluorescence intensity at rest (F0). Images were smoothed by a running average of five adjacent pixels along the spatial and temporal axis of the line-scan image. However, to visualize the delayed onset of local Ca2+ release in the spatial intensity profiles of sequentially scanned lines, there was no smoothing along the temporal axis. Onset of the [Ca2+]i transient was identified in the line-scan average, and all temporal data are referred to this time point. Early and delayed [Ca2+]i transients were identified by visual inspection of the intensity distribution along the line at the onset of the transient. A section with an approximate width of 2 μm in the center of these early and delayed line sections was selected, and the average intensity of this section was plotted over time.
Data are shown as mean±SEM. Comparisons of data from early and delayed transients were performed using a 2-tailed unpaired Student t test. The data from one release site in different experimental conditions was compared with a 2-tailed paired Student t test. A value of P<0.05 was taken to indicate significant difference.
An expanded Materials and Methods section can be found in the online data supplement available at http://www.circresaha.org.
Spatial Inhomogeneities in [Ca2+]i Transients During Steady-State Stimulation
Figure 1A shows a typical example of a line-scan image obtained during steady-state stimulation. The increase in [Ca2+]i on depolarization is not simultaneous, with a delayed increase of [Ca2+]i in several areas. Similar large inhomogeneities of the [Ca2+]i transient front could be discriminated in line-scan images of 64% of the cells (n=107 cells). This was not related to the use of the whole-cell patch clamp, inasmuch as it was also observed in unpatched cells during field stimulation (12 of 18 cells, data not shown). These inhomogeneities occurred before the onset of cell shortening and could thus not be explained by motion artifacts.
To further characterize these inhomogeneities, we compared the properties of the local [Ca2+]i transients in early and delayed sections of the line scan (Figure 1B). These individual transients showed little beat-to-beat variation in time course and amplitude; consequently, the differences were not canceled out by averaging. On average, the maximal amplitude of the Ca2+ transients in the delayed areas (peak F/F0, 2.27±0.10, 24 areas in 13 cells) was smaller than for the early areas (peak F/F0, 2.69±0.13; P<0.05). Decay of [Ca2+]i was not significantly different (half-time of decay, 203±9 ms in late areas versus 194±8 ms in early areas). In the center of the delayed areas, the time to reach the half-maximal [Ca2+]i amplitude (F50) was significantly longer (26±1 ms in delayed areas versus 11±2 ms in early areas). To establish whether there was a real delay at the onset of the [Ca2+]i transient or just a slower increase, we plotted spatial profiles of the line scan at early time points (Figure 1C). These illustrate that there is indeed a delay. From such line-scan images at 12 ms, we also estimated the size of the delayed areas by measuring the distance between successive peaks that exceeded F50. Mean values were 6.8±1.1 μm (n=33 delayed areas), with an upper limit of 25.5 μm.
Are the Inhomogeneities due to Functionally Silent Release Sites?
We examined whether increasing the probability of Ca2+ release would decrease inhomogeneity. β-Adrenergic stimulation increases the open probability of L-type Ca2+ channels,21 responsiveness of the RyRs,22 and SR Ca2+ content.23 Therefore, we looked at the effect of the β-agonist isoproterenol (3 μmol/L); an example is shown in Figure 2A. The averaged [Ca2+]i transients illustrate the increase in overall Ca2+ release, and we confirmed in separate experiments that SR content was increased (2.5-fold, n=5 cells). Nevertheless, the inhomogeneous pattern of the [Ca2+]i transient was not suppressed. A difficulty of this intervention was that the vigorous contractions often led to loss of the original line-scan position (6 cells). In three cells, including the cell shown in Figure 2A, this was not the case. In all experiments, inhomogeneities were still observed.
Pig ventricular myocytes do not have a pronounced frequency-dependent modulation of the [Ca2+]i transient.19 However, in a number of cells, the first [Ca2+]i transient at the start of stimulation was larger than the steady-state transient. F/F0 of the whole scan-line average was 3.30±0.32 for the first beat versus 2.23±0.16 at steady state (P<0.01, n=6 cells). Because this implies that SR Ca2+ release was larger for the first transient, we examined whether inhomogeneities were less for this first transient (Figure 2B). Despite the increased overall Ca2+ release, the distribution and properties of the delayed areas were similar to those seen with the smaller [Ca2+]i transients at steady state. At steady state, the time to reach F50 was 14±3 ms in early areas versus 49±7 ms in delayed areas. For the first beat, time to F50 was less for both early and delayed areas (9±1 and 32±4 ms, respectively), but the difference between delayed versus early areas to reach F50 was comparable (35±8 ms for steady state versus 23±4 ms for the first beat).
We also examined the effects of Bay K8644 (0.5 to 1 μmol/L), a Ca2+ channel agonist. Bay K8644 increases the open time of Ca2+ channels24 and increases resting spark frequency, although this may be unrelated to increased Ca2+ influx.25 In all cells (n=7), the amplitude of the [Ca2+]i transients increased, but the areas of delayed release remained present, as illustrated in Figure 3.
Distribution of Delayed Areas
We quantified the fraction of the scan line with delayed [Ca2+]i transients from line scans obtained at 12 ms (Figure 4; see online data supplement for details). The fraction of the scan line with delayed transients varied between 7% and 54% (range, n=8 cells). We examined whether the fraction of delayed areas was variable with scan line position by comparing lines of central regions with lines in the periphery, ie, near the external sarcolemma. Different line positions within the cell showed various grades of inhomogeneous Ca2+ increase (Figure 4B), but there was no correlation with position: on average, delayed transients occupied 34±2% of the line scans recorded in the center versus 28±4% of the line scans recorded in the periphery of the cell (P=NS).
Distribution of T Tubules and SR Ca2+Release Channels
Tubular structures were visualized with di-8-ANEPPS. In Figure 5, typical examples of a pig myocyte and of a mouse ventricular myocyte are shown. T tubules are identifiable as transverse lines with a regular spacing of ≈2 μm. This pattern is very homogeneous in the mouse but not the pig myocyte. Three-dimensional analysis of the T-tubular network from the Z stack showed that in the pig myocyte many regions are devoid of T tubules. The values for T-tubular signal density in pig myocytes were ≈50% of the values obtained for mouse myocytes (20±2% versus 45±2% of XY area, respectively; n=6 cells and n=60 images from each species; P<0.001; see online data supplement for methods).
Another approach for evaluating the T-tubular network is measuring the membrane area in relation to cell volume (S/V).26 In the pig myocytes in the present study, capacities were 93±4 pF (n=62) for an average cell length of 140±2 μm and cell width of 23±1 μm (n=188), values that are comparable to our previous simultaneous measurements of cell size and capacity.19 Mouse myocytes had a larger average cell capacity of 130±4 pF (n=74) for somewhat smaller cell size (cell length 117±3 μm, cell width 23±1 μm), again comparable to data previously obtained in this laboratory.27 In a smaller number of cells, we measured cell depth from the Z-stack images, and these values were not different for mice and pigs (22 mouse myocytes, 17±1 μm; 17 pig myocytes, 17±1 μm). Although these measurements do not allow for precise volume calculation as described in Satoh et al,26 we can extrapolate and determine that the S/V for the mouse is larger than for the pig, consistent with a lower T-tubular density in pig myocytes.
We also examined the distribution of RyRs by antibody labeling in fixed cells (n=3 hearts, n=18 cells), as shown in Figure 6A. We found that in contrast to the T tubules, RyRs were present throughout the cell. We tested whether these RyRs were functional by examining the response to a fast application of 10 mmol/L caffeine (Figure 6B). The onset of the caffeine-induced release was not simultaneous at all sites, as has been reported by others,28,29⇓ which could be due to the limiting diffusion time of caffeine into the cell.30 Nevertheless, release was observed along the entire scan line in all cells (n=6). We compared time to F50 in the presence of caffeine at early and delayed sites (n=18 of each). As shown in Figure 1, during the depolarizing step, time to F50 was significantly longer for delayed areas. When release was evoked by caffeine, time to F50 decreased for the delayed areas, whereas it increased for the early areas. As a result, time to F50 in the presence of caffeine was no longer significantly different for early and delayed areas. These data suggest that functional RyRs are present in both early and delayed areas.
Is the [Ca2+]i Transient Front in Mouse Myocytes More Homogeneous?
In mouse cells, we never observed the large areas of delayed release as shown in Figure 1. We did see smaller inhomogeneities during the rise of [Ca2+]i, which tended to be more variable from one pulse to the next, as described by Bridge et al.31 In Figure 7, spatial profiles during the rise of [Ca2+]i are shown for a pig myocyte (panel A) and a mouse myocyte (panel B). From these profiles, we calculated the percentage of the scan line with substantial Ca2+ increase (defined as F above F50) as a function of time (Figure 7C).
In scan lines of pig myocytes with an inhomogeneous Ca2+ transient front, F above F50 was detectable in >50% of the scan line after 16±1 ms (n=6 cells). For comparison, in six mouse cells, this value was 6±1 ms. These data indicate that the inhomogeneities in pig myocytes will contribute to a slower overall rise of [Ca2+]i. Measurements of the time to peak of the whole-cell spatially averaged [Ca2+]i transient obtained in a nonconfocal fluorescence setup support this notion. In similar experimental conditions, the time to peak [Ca2+]i for pig myocytes (60±4 ms, n=10) was significantly longer than that for mouse myocytes (32±1 ms, n=14; P<0.05).
To exclude the possibility that a slower rise of [Ca2+]i was related to a intrinsically lower density of functional Ca2+ channels, we also compared current density in a separate set of experiments (K+-free solutions, nifedipine-sensitive current). Peak L-type Ca2+ current was −4.3±0.7 pA/pF for mouse myocytes (n=7) and −5.3±0.6 pA/pF for pig myocytes (n=9, P=NS).
Early and Delayed Release in Relation to T Tubules
We examined whether delayed and early release would colocalize with low and high T-tubular density, respectively. Figure 8A shows a line-scan image at baseline and after the cell was superfused briefly with di-8-ANEPPS. Early and delayed release areas are indicated by the arrows. In Figure 8B, selected single line-scan records at 12 and 90 ms are shown, and the local [Ca2+]i transients corresponding to the areas are indicated by arrows. This example illustrates that delayed Ca2+ increase is associated with a low density of T-tubular structures, as was observed in six other cells.
In the present study, we provide a first description of spatially resolved Ca2+ release from the SR in pig ventricular myocytes in near-physiological conditions. We find substantial spatial and temporal inhomogeneities. Areas with delayed [Ca2+]i transients, which do not reflect variability of release, are apparently related to a lower T-tubular density. This pattern of release contributes to the overall slower rise in [Ca2+]i in pig ventricular myocytes.
Relation to Previous Findings
Many data on the properties of elementary release were obtained after Ca2+ influx was reduced (eg, see Cannell et al4 and Lopez-Lopez et al5), or Ca2+ buffers were added to restrict Ca2+ diffusion (eg, see Cleemann et al10 and Song et al32). Such studies have provided essential information on the gating of RyRs by Ca2+ influx through L-type Ca2+ channels and are at the basis of our current understanding of the local control of excitation-contraction coupling (eg, see reviews1,8⇓). Spatial properties of [Ca2+]i transients in more physiological conditions have been described in a few studies. Cannell et al have examined spatial nonuniformities of the [Ca2+]i transient that occur at the onset of [Ca2+]i transients of rat ventricular cells.33 They distinguished two types of inhomogeneities, one being time independent and probably related to structural inhomogeneities and another being a more variable type of inhomogeneity. The latter type was further studied and seemed most likely related to the stochastic activation of SR release channels. The properties and sources of the time-independent inhomogeneities were not further explored in that study. Other studies have reported inhomogeneities resulting from the variable activity of release units (eg, see Bridge et al31) and gradients between adjacent release units,9 but to our knowledge, there are no further reports on inhomogeneities in ventricular myocytes such as those observed by us. Predominantly, the data presented from rat, guinea pig, and mouse ventricular myocytes show a pattern of release marked by distances of ≈2 μm between foci, consistent with the presence of release units at nearly every Z line.6,9,10,31⇓⇓⇓ The widths of the line sections with delayed Ca2+ increase that we measured in pig ventricular myocytes varied greatly but could exceed by far the single sarcomere size.
Source of Large Inhomogeneities in Pig Ventricular Myocytes
The inhomogeneities that we observed during the upstroke of the Ca2+ transient are unlikely to be due to couplons with low activity. Indeed, delayed areas remained unchanged from beat to beat even in long stimulation protocols over several minutes, and every intervention aimed at increasing SR Ca2+ release probability failed to induce an early increase of [Ca2+]i in the delayed areas. This is in contrast to the synchronization of Ca2+ release induced by β-adrenergic stimulation observed in a rabbit model of heart failure.34 Our data do not exclude the existence of functional variability of release in pig ventricular myocytes. Indeed, we could induce variability in release site activity with increased inhomogeneity, eg, by lowering external [Ca2+] or by varying the amplitude of the depolarizing step (data not shown). The time-independent large inhomogeneities that we observed in near-physiological conditions seem more likely to be related to structural inhomogeneities. With di-8-ANEPPS staining, the distribution of T tubules had an irregular appearance, with several regions of low density. This contrasted strongly with the very homogeneous T-tubular pattern in mouse cells. This difference cannot be explained by a more convoluted course of T tubules in pig cells, inasmuch as we quantified density in successive Z sections. The lower T-tubular density in pig myocytes is also supported by a lower surface area estimated from membrane capacitance measurements. In di-8-ANEPPS–labeled cells, delayed areas corresponded to low T-tubular density, and early Ca2+ release preferentially occurred at the line sections with the highest density of T-tubular structures. This latter relation did not hold in all cells studied, because sometimes we could observe early release though T tubules could not be clearly identified. This most likely resulted from the presence of out-of-plane structures. Indeed, opening the pinhole and increasing the Z optical slice thickness to ≥5 μm could show T-tubular structures above and below the plane. In contrast to the T tubules, RyR staining was evenly distributed throughout the cell. These imaging data are supported by the observation that during rapid application of caffeine, release was observed along the entire scan line.
Last potential sources of structural inhomogeneities are mitochondria. Because we used the salt form of fluo 3, mitochondria were not loaded. Even in cells with a high dye concentration, we could not discern “empty” areas in an XY image of the resting cell, suggesting that individual mitochondria are small and that they are homogeneously distributed. Consistent with this, mitochondria visualized with a MitoTracker (Molecular Probes) were present throughout the cell. This distribution was the same in mouse cells as in pig cells, making it less likely that mitochondria are responsible for the presence of delayed release areas in pig myocytes.
Taken together, these data suggest that the number of functional couplons is smaller in pig myocytes than in mouse myocytes because the sarcolemmal component is less developed. It is at present not fully clear whether the rise in [Ca2+]i in the delayed areas represents actual release from the local RyRs triggered by diffusion from the neighboring areas or simple diffusion without triggered release. The observation that the amplitude was smaller than that in early areas and the fact that the rise time of the Ca2+ signal in the delayed areas is slightly faster with caffeine than with depolarization would argue for the latter hypothesis.
Implications and Perspectives
Time-independent inhomogeneities, not related to stochastic variation of SR Ca2+ release but most likely related to a lower T-tubular density, may be specific to the pig ventricle or may be more general to larger mammals. So far, such inhomogeneities have not been described in normal myocytes of rats, rabbits, or guinea pigs, and we did not observe similar inhomogeneities in mouse cells. For larger mammals, such as dogs, or for human cells, no data are presently available. Images of T tubules in dog ventricular myocytes, published by He et al,35 do not show large areas of rarefaction, but a direct comparison with our findings is difficult. Currently, there are no data on T-tubular density in normal human myocytes.
In the dog with pacing-induced heart failure, T-tubular density was decreased, and this was proposed to be implicated in the reduced efficiency of Ca2+ release with heart failure.35 In the failing human heart, such a decrease of T-tubular density could not yet be established.36 The alterations of T-tubular density during early development12 and, in particular, culture conditions11 suggest that this is a cellular property under tight control, which can potentially be part of a remodeling process.
A lower T-tubular density has important functional consequences. The presence of delayed areas of release leads to an overall slowing of the upstroke of [Ca2+]i with substantial Ca2+ gradients. With RyRs “uncoupled” from Ca2+ channels, control of release is altered and could give rise to spontaneous release, as seen in Purkinje cells.15 A lower T-tubular density will also lead to an overall reduction of S/V with consequences for concentration changes in relation to membrane influx, even if channel density on the membranes is comparable.
Our data support the importance of T tubules, but they also underscore that in normal myocytes, their density is not necessarily as high as that seen, for example, in rat37 or mouse (this study) myocytes (see also histology1).
In conclusion, in pig ventricular myocytes, Ca2+ release from the SR is inhomogeneous. Areas of delayed release are related to regional absence of T tubules but not RyRs. This lower number of functional couplons contributes to a slower overall rate of rise of [Ca2+]i. Our findings further emphasize the importance of T tubules for synchronized and spatially homogeneous Ca2+ release.
This study was supported by the Fonds voor Wetenschappelijk Onderzoek-Vlaanderen (FWO), the Flanders Fund for Scientific Research (K.R.S., K.M.), and the “Herz- und Kreizlaufzentrum, Essen” (F.R.H.). The authors wish to thank Dr Tania Stankovicova and Prof Gerd Heusch for their contributions. The expert help of Dr Lothar Blatter with image acquisition and analysis is greatly appreciated. We thank Patricia Holemans for technical assistance and David Decoux for animal care.
↵*Both authors contributed equally to this study.
Original received June 28, 2002; revision received October 28, 2002; accepted October 28, 2002.
- ↵Bers DM. Excitation-Contraction Coupling and Cardiac Contractile Force. Boston, Mass: Kluwer Academic Publishers; 2001.
- ↵Cannell MB, Cheng H, Lederer WJ. The control of calcium release in heart muscle. Science. 1995; 268: 1045–1049.
- ↵Lopez-Lopez JR, Shacklock PS, Balke CW, Wier WG. Local calcium transients triggered by single L-type calcium channel currents in cardiac cells. Science. 1995; 268: 1042–1045.
- ↵Wier WG, Balke CW. Ca2+ release mechanisms, Ca2+ sparks, and local control of excitation-contraction coupling in normal heart muscle. Circ Res. 1999; 85: 770–776.
- ↵Isenberg G, Etter EF, Wendt-Gallitelli MF, Schiefer A, Carrington WA, Tuft RA, Fay FS. Intrasarcomere [Ca2+] gradients in ventricular myocytes revealed by high speed digital imaging microscopy. Proc Natl Acad Sci U S A. 1996; 93: 5413–5418.
- ↵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.
- ↵Haddock PS, Coetzee WA, Cho E, Porter L, Katoh H, Bers DM, Jafri MS, Artman M. Subcellular [Ca2+]i gradients during excitation-contraction coupling in newborn rabbit ventricular myocytes. Circ Res. 1999; 85: 415–427.
- ↵Boyden PA, Pu J, Pinto J, Keurs HE. Ca2+ transients and Ca2+ waves in Purkinje cells: role in action potential initiation. Circ Res. 2000; 86: 448–455.
- ↵Schulz R, Post H, Sakka S, Wallbridge DR, Heusch G. Intraischemic preconditioning: increased tolerance to sustained low-flow ischemia by a brief episode of no-flow ischemia without intermittent reperfusion. Circ Res. 1995; 76: 942–950.
- ↵Lai L, Kolber-Simonds D, Park KW, Cheong HT, Greenstein JL, Im GS, Samuel M, Bonk A, Rieke A, Day BN, Murphy CN, Carter DB, Hawley RJ, Prather RS. Production of α-1,3-galactosyltransferase knockout pigs by nuclear transfer cloning. Science. 2002; 295: 1089–1092.
- ↵Stankovicova T, Szilard M, De Scheerder I, Sipido KR. M cells and transmural heterogeneity of action potential configuration in myocytes from the left ventricular wall of the pig heart. Cardiovasc Res. 2000; 45: 952–960.
- ↵Yue DT, Herzig S, Marban E. β-Adrenergic stimulation of calcium channels occurs by potentiation of high-activity gating modes. Proc Natl Acad Sci U S A. 1990; 87: 753–757.
- ↵Valdivia HH, Kaplan JH, Ellis-Davies GC, Lederer WJ. Rapid adaptation of cardiac ryanodine receptors: modulation by Mg2+ and phosphorylation. Science. 1995; 267: 1997–2000.
- ↵Brown AM, Kunze DL, Yatani A. The agonist effect of dihydropyridines on Ca channels. Nature. 1984; 311: 571–573.
- ↵Katoh H, Schlotthauer K, Bers DM. Transmission of information from cardiac dihydropyridine receptor to ryanodine receptor: evidence from BayK 8644 effects on resting Ca2+ sparks. Circ Res. 2000; 87: 106–111.
- ↵Ver Heyen M, Heymans S, Antoons G, Reed T, Periasamy M, Awede B, Lebacq J, Vangheluwe P, Dewerchin M, Collen D, Sipido K, Carmeliet P, Wuytack F. Replacement of the muscle-specific sarcoplasmic reticulum Ca2+-ATPase isoform SERCA2a by the nonmuscle SERCA2b homologue causes mild concentric hypertrophy and impairs contraction-relaxation of the heart. Circ Res. 2001; 89: 838–846.
- ↵Ritter M, Su Z, Spitzer KW, Ishida H, Barry WH. Caffeine-induced Ca2+ sparks in mouse ventricular myocytes. Am J Physiol. 2000; 278: H666–H669.
- ↵Litwin SE, Zhang D, Bridge JH. Dyssynchronous Ca2+ sparks in myocytes from infarcted hearts. Circ Res. 2000; 87: 1040–1047.
- ↵He J, Conklin MW, Foell JD, Wolff MR, Haworth RA, Coronado R, Kamp TJ. Reduction in density of transverse tubules and L-type Ca2+ channels in canine tachycardia-induced heart failure. Cardiovasc Res. 2001; 49: 298–307.
- ↵Ohler A, Houser SR, Tomaselli GF, O’Rourke B. Transverse tubules are unchanged in myocytes from failing human hearts. Biophys J. 2002; 80: 590a. Abstract.
- ↵Soeller C, Cannell MB. Examination of the transverse tubular system in living cardiac rat myocytes by 2-photon microscopy and digital image-processing techniques. Circ Res. 1999; 84: 266–275.