Ca²⁺ sparks and Ca²⁺ waves in saponin-permeabilized rat ventricular myocytes

Cell isolation, permeabilization and experimental solutions

Adult Sprague-Dawley rats (200–300 g) were killed by lethal injection of Nembutal (Abbott Laboratories, 100 mg kg⁻¹, i.p.), and single ventricular myocytes were obtained by enzymatic dissociation as described previously (Györke et al. 1997). The cells were loaded with fluo-3 by a 20 min incubation with 5 μm fluo-3 AM (acetoxymethyl ester form, Molecular Probes) at room temperature. The Tyrode solution contained (mm): 140 NaCl, 5.4 KCl, 0.5 MgCl₂, 1–5 CaCl₂, 10 Hepes, 0.25 NaH₂PO₄, 5.6 glucose; pH 7.3. The cardiac myocytes were permeabilized with saponin (0.01 % for 45–60 s) in an ‘internal’ solution containing (mm): 120 potassium aspartate, 3 MgATP (free [Mg²⁺]∼ 1 mm), 0.1 EGTA, 10 phosphocreatine, 5 U ml⁻¹ creatine phosphokinase, and 8 % dextran (40 000, to prevent osmotic swelling of the cells); pH 7.2. The control experimental solution contained (mm): 120 potassium aspartate, 3 MgATP, 0.5 EGTA, 0.114 CaCl₂ (free [Ca²⁺]∼ 100 nm), 10 phosphocreatine, 0.03 fluo-3 potassium salt (TefLabs, Austin, TX, USA) and 5 U ml⁻¹ creatine phosphokinase; pH 7.2. Solutions with different buffering strengths and free [Ca²⁺] were prepared by adding appropriate amounts of K₂EGTA and CaCal₂. The free [Ca²⁺] at given total Ca²⁺, Mg²⁺, ATP and EGTA concentrations was calculated using a computer program (WinmAXC 1.80, Stanford University, CA, USA) and verified by measurements with a spectrofluorometer D-Scan (PTI, Monmouth Junction, NJ, USA) and the Ca²⁺ indicator fura-2 (TefLabs, Austin, TX, USA). The drugs were applied through a gravity-driven perfusion system. All experiments were performed at room temperature (21–23°C). All chemicals except fluo-3 and fura-2 were from Sigma.

Confocal microscope

Experiments were performed as described previously (Györke et al. 1997), using an Olympus laser scanning confocal microscope (LSM-GB200) equipped with an Olympus × 60, 1.4 NA objective. Fluo-3 was excited by light at 488 nm using a 25 mW argon laser with intensity attenuated to 1–3 %. Fluo-3 fluorescence was measured at wavelengths of > 515 nm. Images were acquired in the line-scan mode of the microscope at a rate of 2.1 or 8.3 ms per scan, with the scan line oriented along the longitudinal axis of the cell. An analog recording of fluorescence intensity was digitized into 640 pixels, giving a nominal pixel dimension of 0.41 μm. To reduce cell damage by the laser illumination, the position of the line scan was changed after acquiring three to six images from each particular location. Thus, in a typical cell the measurements could be performed for 10–15 min without significant alterations in Ca²⁺ spark properties.

Ca²⁺ sparks were detected and measured using a computer algorithm similar to that described previously (Song et al. 1997; Cheng et al. 1999). The program defines Ca²⁺ sparks as regions of elevated fluorescence relative to the standard deviation (s.d.) of background noise of the fluorescence image. The performance of the program at different event detection threshold settings was tested by using standard Ca²⁺ sparks of various intensities contaminated with appropriate amounts of random noise (Song et al. 1997). With the detection threshold set at a level of 2.6 ×s.d., the amplitude of events detected with 50 % efficiency was about 1.3F/F_o (where F is the recorded fluorescence intensity and F_o is background fluorescence) while the probability of false events was 1–2 %. The average propagation velocity of Ca²⁺ waves was determined by fitting a linear function to the position of the wave (defined at half-maximal amplitude) in the x–t plane (Lukyanenko et al. 1999). Image processing and analysis were performed using NIH Image (NIH, Bethesda, MD, USA) and IDL software (Research Systems Inc., Boulder, CO, USA).

Effects of permeabilization

Figure 1 illustrates the main steps of a permeabilization experiment in a single isolated rat ventricular myocyte pre-loaded with fluo-3 AM. The series of line-scan images in B were acquired before permeabilization (a), after permeabilization with 0.01 % saponin for 1 min in an internal solution containing no dye (b) and after addition to the internal solution of 30 μm fluo-3 potassium salt in the presence of 0.1 mm (c) or 0.5 mm EGTA (d). Photographic wide field images of the myocyte before and after permeabilization are illustrated at the top (A). As described previously (Cheng et al. 1993, 1996; Lukyanenko et al. 1996), the intact cell spontaneously exhibited sporadic Ca²⁺ sparks and occasional Ca²⁺ waves. Permeabilization was confirmed by the disappearance of all the fluorescence signals in the bathing solution containing no dye and re-emergence of the signals after introduction of the free acid form of the dye into the bathing solution. Note that the overall appearance of the cell and the Ca²⁺ signals before and after permeabilization are very similar. Figure 1C shows surface plots of Ca²⁺ sparks in intact and permeabilized cells in the presence of 0.1 or 0.5 mm EGTA. Table 1 summarizes spark statistics for the same conditions. As can be seen, permeabilization had no significant impact on the frequency, amplitude, width or length of the events (for all groups, P > 0.05). The similarities in the spatio-temporal properties of Ca²⁺ sparks at 0.1 and 0.5 mm EGTA are likely to be due to the slow rate of Ca²⁺ binding by EGTA. The overall similarities between Ca²⁺ sparks and waves in intact and permeabilized cells suggest that our permeabilization procedure does not significantly alter the Ca²⁺ signalling mechanisms in the cell.

Effects of EGTA

Theoretical studies predict that soluble intracellular Ca²⁺ buffers must have a strong effect on interaction between release sites by lowering the rate of effective diffusion of Ca²⁺ (Keizer et al. 1998). We experimentally tested the effects of intracellular Ca²⁺ buffering on Ca²⁺ waves in permeabilized cardiac myocytes. In these experiments Ca²⁺ buffering strength was varied by addition of different concentrations of EGTA to the bathing solution, while adjusting basal Ca²⁺ to a constant value of 100 nm. In internal solutions with low Ca²⁺ buffering strength (0.05 mm EGTA), Ca²⁺ waves arose relatively frequently (0.06 ± 0.01 s⁻¹, n= 21) and propagated typically through the entire cell with a velocity of about 70 μm s⁻¹ (66 ± 6 μm s⁻¹, n= 21; Fig. 2A). Elevation of EGTA concentration to 0.1 mm resulted in decreased frequency and propagation velocity of waves (0.03 ± 0.01 s⁻¹ and 53 ± 1 μm s⁻¹, respectively, n= 12; Fig. 2B). Further increase in [EGTA] resulted in fragmentation of Ca²⁺ waves into abortive responses (Fig. 2C and D). Images obtained at high [EGTA] also clearly show that Ca²⁺ waves arise from sequential activation of discrete release events, revealing the saltatory nature of wave propagation. Propagating Ca²⁺ waves were completely abolished at [EGTA] > 0.3 mm. The fast Ca²⁺ buffer BAPTA prevented Ca²⁺ waves even at lower concentrations (∼0.1 mm, not shown). These results indicate that Ca²⁺ buffering has a profound influence on Ca²⁺ waves.

Effects of [Ca²⁺]

Ca²⁺ is the principal regulator of the activity of ryanodine receptors (RyRs) in the heart. Although the Ca²⁺ dependence of RyR activation and inactivation has been described in in vitro experiments (Coronado et al. 1994), the Ca²⁺ dependency of RyRs in situ remains uncertain because of the difficulties in measuring and controlling [Ca²⁺] in the diadic cleft during E-C coupling. We took advantage of our permeabilized myocyte preparation to investigate the relationship between Ca²⁺ and the activity of Ca²⁺ release sites. Figure 3A illustrates the effects of increasing [Ca²⁺] from 50 nm to 100, 150 or 200 nm in the presence of 0.1 mm EGTA. Elevating [Ca²⁺] resulted in increased frequency of Ca²⁺ waves. In addition, at elevated [Ca²⁺] multi-focal Ca²⁺ waves arising simultaneously from several independent sites became evident. Quantitative assessments of spark properties at high [Ca²⁺] were impaired by the presence of Ca²⁺ waves and elevated background fluorescence. Increases in background fluorescence also limited the Ca²⁺ concentrations in the experimental solution to a rather limited range because of saturation of the photomultiplier at high [Ca²⁺]_i. Therefore, we employed bathing solutions containing high [EGTA] to abolish spontaneous Ca²⁺ waves. We also corrected the images for changes in background fluorescence at different [Ca²⁺]_i by adjusting the fluorescence of the bathing solution outside the permeabilized cells to the same level. Figure 3B shows representative images from such an experiment recorded during successive increases of [Ca²⁺] from 100 to 150, 250 and 400 nm and a subsequent return to 100 nm. The effects of [Ca²⁺] on the frequency and amplitude of Ca²⁺ sparks from the same experiment are documented in Fig. 3C. Increasing [Ca²⁺] in the range 100–400 nm resulted in a gradual increase in frequency of sparks. Changing back to the original solution with 100 nm Ca²⁺ resulted in restoration of spark frequency to the control level, thus indicating no significant signs of a run-down in sparking activity. The changes in the frequency of sparks were accompanied only by insignificant alterations in the amplitude of sparks (10 % at 150 nm[Ca²⁺], grey bars). The results of five experiments are summarized in Fig. 3D, which plots the frequency of Ca²⁺ sparks as a function of [Ca²⁺]. It is unlikely that the amplitude of Ca²⁺ sparks was saturated at elevated [Ca²⁺]. The K_D of fluo-3 in the myoplasm has been estimated to be near 1 μm (Harkins et al. 1993). Ca²⁺ sparks with an amplitude of 200–300 nm would rise above a background [Ca²⁺] of 400 nm only to a level of 600–700 nm, which still would be in the linear range of the indicator.

The observed potentiation of Ca²⁺ sparks could be attributed also to a possible increase in the SR Ca²⁺ content (Fabiato, 1992; Orchard et al. 1998) affecting the activity of the Ca²⁺ release channels at luminal sites (Györke & Györke, 1998). We assessed the potential role of this mechanism, by using thapsigargin to prevent accumulation of extra Ca²⁺ into the SR upon elevation of cytosolic Ca²⁺. Representative images from such an experiment are shown in Fig. 4A. The effects of [Ca²⁺] on spark frequency in the same experiment are quantified in Fig. 4B. We applied 10 μm thapsigargin for 1 min before elevating [Ca²⁺] from 100 nm to 250 nm. At this concentration and exposure time thapsigargin inhibits SR Ca²⁺ uptake without causing a significant loss in the SR Ca²⁺ content (Bassani et al. 1993; Lukyanenko et al. 1999). Under these experimental conditions increasing [Ca²⁺]_i resulted in an about 6-fold (from 7.1 ± 0.9 to 39.8 ± 1.9 event s⁻¹ (100 μm)⁻¹, n= 5) increase in sparking frequency, which is similar to that observed in experiments without thapsigargin. Following more than 2–3 min of continuous exposure to thapsigargin, the frequency of sparks gradually decreased, apparently due to a loss of Ca²⁺ from the SR (Bassani et al. 1993; Lukyanenko et al. 1999). Taken together, these results suggest that under our experimental conditions, the increase in sparking activity cannot be attributed to increased SR Ca²⁺ load.

Effects of calmodulin and cADPR

We tested the effects on release site activity of certain membrane-impermeable putative regulators of the SR Ca²⁺ release such as cyclic adenosine diphosphate-ribose (cADPR, MW 541.3) and calmodulin (MW 16 680), which would be difficult to study in cells with an intact sarcolemma. Figure 5 illustrates representative images of Ca²⁺ sparks before and after exposure of two permeabilized cells to calmodulin (A, 5 μm) and cADPR (C, 5 μm), while Table 2 summarizes the effects of these drugs on Ca²⁺ spark properties. Both drugs caused dramatic increases in the frequency of Ca²⁺ sparks with no significant changes in the amplitude and spatio-temporal properties of the events. The effects of cADPR developed over a period of 2–6 min, and they were readily reversible. The onset of potentiation of Ca²⁺ sparks by calmodulin was considerably slower, 10–15 min. In addition, the cells showed no significant recovery upon reverting to the control solution during a period of 5–10 min. The slow onset and lack of reversibility of calmodulin effects could be attributed to the slow diffusion rate of this high molecular weight agent as well as to the fact that the effects are likely to involve long-lasting biochemical changes (phosphorylation of SR proteins). Incubation of the cells with calmodulin also resulted in a significant increase in the amplitude of caffeine-induced Ca²⁺ transients (31.2 ± 4.4 %, n= 5; Fig. 5B), suggesting that the effects of this agent on Ca²⁺ sparks might be mediated by changes in the SR Ca²⁺ load. These results further illustrate our ability to manipulate the environment of the release sites in ways not possible in intact cells.