Effects of Electrical Shocks on Cai2+ and Vm in Myocyte Cultures
Changes in intracellular calcium concentration (ΔCai2+) induced by electrical shocks may play an important role in defibrillation, but high-resolution ΔCai2+ measurements in a multicellular cardiac tissue and their relationship to corresponding Vm changes (ΔVm) are lacking. Here, we measured shock-induced ΔCai2+ and ΔVm in geometrically defined myocyte cultures. Cell strands (width=0.8 mm) were double-stained with Vm-sensitive dye RH-237 and a low-affinity Cai2+-sensitive dye Fluo-4FF. Shocks (E≈5 to 40 V/cm) were applied during the action potential plateau. Shocks caused transient Cai2+ decrease at sites of both negative and positive ΔVm. Similar Cai2+ changes were observed in an ionic model of adult rat myocytes. Simulations showed that the Cai2+ decrease at sites of ΔV+m was caused by the outward flow of ICaL and troponin binding; at sites of ΔV−m it was caused by inactivation of ICaL combined with extrusion by Na–Ca exchanger and troponin binding. The important role of ICaL was supported by experiments in which application of nifedipine eliminated Cai2+ decrease at ΔV+m sites. Largest ΔCai2+ were observed during shocks of ≈10 V/cm causing simple monophasic ΔVm. Shocks stronger than ≈20 V/cm caused smaller ΔCai2+ and postshock elevation of diastolic Cai2+. This was paralleled with occurrence of biphasic negative ΔVm that indicated membrane electroporation. Thus, these data indicate that shocks transiently decrease Cai2+ at sites of both ΔV−m and ΔV+m. Outward flow of ICaL plays an important role in Cai2+ decrease in the ΔV+m areas. Very strong shocks caused smaller negative ΔCai2+ and postshock elevation of diastolic Cai2+, likely caused by membrane electroporation.
Calcium ions play crucial roles in regulation of cardiac excitation and contractility, and they may be an important determinant of the tissue response to defibrillation shocks. The interaction between electrical field and Cai2+ may affect the outcome of a defibrillation attempt in several ways. First, it was suggested that very strong shocks cause calcium overload, which can lead to abnormal impulse generation, re-induction of rapid arrhythmias,1 and defibrillation failure.2 Second, it was reported that relatively weak shocks with an energy below the defibrillation threshold applied during fibrillation can prevent the loss of cardiac contractility often observed after successful defibrillation, so-called pulseless electrical activity syndrome.3,4 In a related study, it was reported that cardiac contractility is enhanced when shocks are applied during the absolute refractory period.5 Two alternative mechanisms were proposed to explain these effects: stimulation of intracardiac sympathetic nerves by the shock4 or an increase of peak Cai2+ concentration caused by a direct effect of the shock on myocytes.6
The direct assessment of the mechanisms of shock-Cai2+ interaction and its role in defibrillation requires measurements of shock-induced Cai2+ and Vm changes with high spatial and temporal resolution. Previously, shock-induced Cai2+ changes were measured in single myocytes7,8 and at a single point in whole hearts.6 No spatially resolved data on shock-induced Cai2+ changes and colocalized Vm changes in multicellular cardiac tissue are currently available. Such data are especially important because of the known complexity of shock-induced ΔVm in the heart. It is well established that shocks produce highly nonuniform patterns of ΔVm with areas of positive, negative, or negligible polarizations present in different parts of cardiac tissue,9,10 suggesting that Cai2+ changes may also be nonuniform. In addition, shocks produce different types of Vm responses that depend on the shock strength and the tissue geometry. With the exception of very weak shocks, shocks applied during the action potential (AP) plateau produce nonlinear ΔVm of 2 main types.11 Shocks of moderate strength induce asymmetric ΔVm, with negative ΔVm being much larger than positive ΔVm.10,12,13 Stronger shocks can induce ΔVm of another type, which is characterized by a nonmonotonic behavior of negative ΔVm in which strong hyperpolarization is followed by a return of Vm to more positive levels.11,14 In cell cultures, the transition from one nonlinear ΔVm type to another with increasing shock strength was paralleled by the occurrence of postshock arrhythmias originating from the area of shock-induced negative polarization1 as well as with membrane electroporation in the same areas.15 It can be hypothesized that these 2 types of nonlinear ΔVm are associated with different Cai2+ changes as well.
The purpose of the present study was to use high-resolution optical mapping of Vm and Cai2+ to determine spatio-temporal Cai2+changes during and after shocks, as well as the relationship between these ΔCai2+ and different types of ΔVm in multicellular cardiac tissue. Because shock-induced ΔVm strongly depend on the tissue geometry,11,16 experiments were performed in geometrically defined cultured cell monolayers. To elucidate the ionic mechanisms of Cai2+ changes, shock effects were also investigated in a computer model of rat ventricular myocardium.
Growth substrates containing linear strands (width=0.8 mm) were prepared as described previously.17 Neonatal rat myocytes were obtained from 1- to 2-day-old Wistar rats and cultured according to the previously published procedures.11 Cell cultures were incubated in UltraCulture medium (BioWhittaker) supplemented with 20 μg/mL vitamin B12, antibiotics, 0.1 mmol/L of bromodeoxyuridine, and 0.5 μmol/L of epinephrine at 37°C in a humidified atmosphere containing 4% CO2. Culture medium was exchanged on the next day after preparation and every second day thereafter. Experiments were performed after 4 to 6 days in culture.
Optical Mapping of Vm and Cai2+
Vm and Cai2+ changes were measured using a previously described optical mapping technique18 with some modifications. Cells were first stained with 5 μmol/L of a Cai2+-sensitive dye (see later) by incubating with a dye solution for 30 to 45 minutes. After that, cultures were transferred into experimental chamber and superfused with Hanks solution (Life Technologies) having a pH of 7.4 and a temperature of 36°C. Cells were stained for 5 minutes with 2.5 μmol/L of Vm-sensitive RH-237 (Molecular Probes). To reduce leakage of the Cai2+ dye, 1 mmol/L of probenecid was added to the staining and perfusion solutions.18
Previous Cai2+measurements using dye Fluo-3 demonstrated very long durations of Cai2+ transients (CaD).18 We hypothesized that such long CaD were caused by the high affinity of Fluo-3 to Ca2+ ions. Because this can affect measurements of shock-induced ΔCai2+, we examined dyes with different affinities to Ca2+ ions from 2 major groups. The first group included Fluo-3 analogs: Fluo-4, Fluo-5F, Fluo-4FF, and Fluo-5N (Molecular Probes) that have nominal dissociation constants (Kd) of 0.345, 2.3, 40, and 90 μmol/L, respectively. The second group included Rhod-2 and Rhod-FF, with Kd of 0.57 and 17 μmol/L, respectively. Based on these measurements (see Results), the dye Fluo-4FF was selected for measurement of shock-induced ΔCai2+ and ΔVm in double-stained preparations.
ΔCai2+ and ΔVm were measured sequentially using different sets of optical filters. For Vm measurements, fluorescence was excited at 560/55 nm and emitted light was measured at >650 nm. For Cai2+ measurements using Fluo dyes, the respective wavelength ranges were 480/40 nm and 530/50 nm. With Rhod dyes, they were 530/40 nm and 580/40 nm. Tests described previously18 showed that there was no optical cross-talk between RH-237 and Fluo dyes. Fluorescence was measured using a 16×16-photodiode array (Hamamatsu) and a microscopic mapping system11 at a sampling rate of 2 to 5 kHz/channel and spatial resolution of 110 or 55 μm/diode.
Cells were paced at a cycle length of 500 ms. Rectangular uniform-field shocks with a duration of 10 ms were delivered via 2 platinum plate electrodes. The field strength (E) was measured using a bipolar electrode. Delivery of shocks was synchronized with stimulation pulses. The delay between AP upstrokes and shocks was 10 to 20 ms. At each mapping location, 3 to 4 shocks of different strengths selected from the values of ≈5, 10, 20, 30, and 40 V/cm were applied. A shock of the same strength was applied twice to measure ΔVm and ΔCai2+. For shocks of 20 V/cm or stronger, a 3-minute interval was allowed after shocks for tissue recovery.1 Before shock application, control (no shocks) Vm and Cai2+ recordings were made. To check for data reproducibility, at the end of each series, measurements with the initial shock strength were repeated. In some cases, shocks of the same strength were applied repeatedly. Both ΔVm and ΔCai2+ measurements were highly reproducible. Changing the shock polarity resulted in symmetric reversal of Vm and Cai2+ responses.
A shock-induced ΔVm was measured as the difference between a linear fit of the plateau phase and the magnitude of the shock response 9 ms after the shock onset and normalized by the action potential amplitude.16 To measure a shock-induced change in Cai2+, 2 recordings of Cai2+ transients taken with and without a shock were normalized according to their levels at the moment preceding the shock. This normalization compensated for changes in optical signals caused by dye washout and photobleaching. The ΔCai2+ was calculated at the shock end (t=9 ms) as the difference between the 2 signals and expressed as a percentage of the amplitude of the control Cai2+ transient (% ACa).
An activation time was measured at the level of 50% change of the AP upstroke amplitude. The onset of Cai2+ transients was measured at the level of 30% change. The difference between the 2 times was taken as the Vm−Cai2+ delay. The durations of AP and Cai2+ transients were measured as time intervals between their respective onsets and as 50% or 80% levels of signal recovery.
Data were expressed as mean±SD. Differences were compared using the 2-tailed t test or the 1-way ANOVA test. Results were considered statistically significant if P<0.05.
Computer Modeling of Shock Effects
Computer simulations were performed in a 1-dimensional cable with ionic model of adult endocardial rat ventricular myocytes.19 The Ca-handling system included ICaL, Na–Ca exchange (NCX), sarcoplasmic reticulum (SR), sarcolemmal pump current (ICaP), and buffering by troponin (Tpn). SR and Tpn buffering were slightly modified according to a previous publication.20 Full description of ionic model and cable parameters is presented in the online data supplement at http://circres.ahajournals.org. Cable equations were solved numerically using previously described procedures.21 The cable was stimulated at all nodes simultaneously. After 15 ms of the stimulus, a current was injected at one cable end and the same current was withdrawn from the other end. Such current injection is equivalent to application of extracellular uniform-field shocks.
Selection of Cai2+-Sensitive Dye
Figure 1 illustrates the effects of dye affinity on optical measurements of Cai2+ transients. Calcium transients (Figure 1A) measured with high-affinity dyes Fluo-4 and Fluo-5F were much longer than those measured with low-affinity Fluo-4FF. The average CaD80 (Figure 1B) measured with Fluo-4 and Fluo-5F (243±29 ms, n=6, and 220±15, n=7, number of strands) was ≈80% longer than with Fluo-4FF (132±10 ms, n=6; P<0.001). A further reduction of dye affinity by using Fluo-5N did not reduce CaD80 (141±11 ms, n=7) in comparison to Fluo-4FF. In all series of measurements, AP durations were not significantly different (Figure 1B, right panel). Decreasing the dye affinity caused reduction of the signal-to-noise ratio (Figure 1A).
Contrary to the recovery phase of Cai2+ transients, the parameters of the rising phase were not dependent on the dye choice. As shown in Figure 1C, the rise-times of Cai2+ transients measured using high-affinity and low-affinity dyes Fluo-4 and Fluo-4FF dyes were 16.3±1.0 ms and 17.5±1.3 ms (n=6, NS), respectively. The respective delays between AP and Cai2+ upstrokes were 3.1±0.1 and 3.1±0.3 (n=6, NS).
Similar data were obtained using Rhod dyes. The low-affinity Rhod-FF resulted in markedly shorter Cai2+ transients than the high-affinity Rhod-2 (Figure 1D). The CaD80 values measured by these 2 dyes (Figure 1E) were similar to those measured by Fluo-4FF and Fluo-4, respectively.
These data indicate that the dye affinity plays an important role in optical measurements of Cai2+. Long Cai2+ transients measured with high-affinity dyes can be explained by dye saturation (see Discussion). Therefore, in conditions in which in situ dye calibration compensating for this saturation is not available, low-affinity dyes are more appropriate for the faithful reproduction of Cai2+ time course. Because Fluo-4FF exhibited a much higher signal-to-noise ratio than Fluo-5N, it was selected for measurements of shock-induced ΔCai2+ and ΔVm in double-stained preparations described later.
Shock-Induced Cai2+ Changes and Monophasic Asymmetric ΔVm
It was previously shown that depending on the field strength, shocks applied during AP plateau produce nonlinear ΔVm of 2 main types, monophasic asymmetric ΔVm and biphasic ΔVm.10–13 Figure 2 illustrates ΔVm of the first type and accompanying Cai2+ changes induced by a 10-V/cm shock. The shock caused negative and positive ΔVm on the opposite strand edges with a gradual transition of ΔVm amplitude in between (Figure 2A). The ΔVm distribution was strongly asymmetric with maximal ΔV−m (trace 1) being ≈2.5-times larger than maximal ΔV+m (trace 7). At the strand edges and at most other locations, polarizations had simple monophasic shapes. A more complex ΔVm shape was observed at the boundary between areas of positive and negative polarizations (trace 5), reflecting electrotonic interactions with areas of ΔV−m and ΔV+m. After the shock, Vm returned to levels similar to those in the control.
As shown in Figure 2B, a shock of the same strength reduced Cai2+ at sites of both negative (traces 1 to 3) and positive polarizations (traces 6, 7). At the intermediate location where ΔVm was small (trace 5), ΔCai2+ was negligible. Cai2+ reduction was larger at sites of negative, rather than at positive, ΔVm, with ΔCai2+ magnitude being −25% and −11% ACa at sites of maximal ΔV−m (trace 1) and ΔV+m (trace 7), respectively.
After the shock, Cai2+ continued to decrease for ≈4 ms at sites of negative ΔVm, where ΔCai2+ could reach −40%ACa (trace 1). After that, Cai2+ began to rise and, at most locations, Cai2+ returned to the control level soon after the shock (Figure 2C). A slight postshock increase of peak Cai2+ amplitude was observed at some locations (trace 6). Similarly, the duration of Cai2+ transients could be slightly prolonged (traces 6, 7). The diastolic level of Cai2+ was not changed.
Shock-Induced ΔCai2+ and Biphasic ΔV−m
Figure 3 illustrates Cai2+ changes during a shock (E=29 V/cm) that produced another nonlinear ΔVm type. In this case, the negative ΔVm waveforms (Figure 3A, traces 1 to 3) were biphasic, exhibiting a strong positive shift after the initial large hyperpolarization. At the same time, the positive ΔVm at the cathodal strand edge (trace 7) was not different from the weaker shock.
Similar to the effect of the weaker shock, Cai2+ was transiently reduced at sites of both negative and positive ΔVm (Figure 3B), but these changes were smaller than ΔCai2+ during the weaker shock. At sites of maximal ΔV−m and ΔV+m, the decreases of Cai2+ were 16% and 5% ACa, respectively. Another difference was that the duration of Cai2+ transients was somewhat prolonged in the areas of both positive and negative ΔVm. A small elevation of the diastolic Cai2+ level was observed, more prominent at the anodal edge of the strand where it was ≈12% ACa 300 ms after the rising phase (Figure 3C, trace 1). In addition, the shock induced an extra beat, with a coupling interval of ≈360 ms.
Shock-induced Vm and Cai2+ changes were measured in a total of 19 cell strands from 15 cell monolayers. Each shock strength was examined in 7 to 10 strands. In all strands, Vm and Cai2+ responses of the 2 types described were observed. Thus, all shocks caused Cai2+ decreases at sites of both ΔV+m and ΔV−m (Figure 4B). The ΔCai2+magnitude was always larger at ΔV−m sites than at ΔV+m sites. The largest −ΔCai2+ (≈16% ACA) were observed at a shock strength of 10 V/cm, and they decreased with increasing shock strength (P<0.05, 40-V/cm group versus 10-V/cm group).
The average duration of Cai2+ transients was increased in a strength-dependent fashion with mean ΔCaD50 at the strand edges reaching 25% to 30% at 40-V/cm shocks (Figure 4D). The diastolic Cai2+ measured 300 ms after the Cai2+ rising phase was not significantly affected by shocks with strength of 10 V/cm or less (Figure 4C). Increasing the shock strength to 20 V/cm caused elevation of diastolic Cai2+, but the magnitude of this elevation was very small (≈3% to 4% ACa). Measurements of diastolic Cai2+ changes by shocks stronger than 20 V/cm were precluded in the majority of cases because of occurrences of extra beats with relatively short coupling intervals. In 1 case, when a 30-V/cm shock induced an extra beat with a coupling interval longer than 300 ms, a more substantial elevation of diastolic Cai2+ level was registered, with ΔCai2+dia measuring 12% and 7% ACa at the anodal and cathodal edges, respectively.
Figure 5 illustrates effects of shocks in a computer model. A shock caused ΔVm and ΔCai2+ that were qualitatively similar to those observed in experiments. Thus, shock-induced polarizations were asymmetric with ΔV−m>ΔV+m and Cai2+ was decreased at sites of both ΔV−m and ΔV+m (Figure 5A). ICaL recordings showed that shocks caused rapid inactivation of this current in the ΔV−m area (trace 1), which contributed to Cai2+ decrease at this location. In the ΔV+m region, ICaL remained active, but it changed its direction from inward to outward (trace 2), causing removal of Ca2+ ions from the intracellular space.
To examine the roles of various Cai2+ handling pathways in shock-induced Cai2+ decrease, these pathways were selectively inhibited during the shock. As shown in Figure 5B, Cai2+ decrease in ΔV−m area (site 1) became smaller when NCX or Tpn buffering was disabled (green and blue traces). Inhibition of SR (both release and uptake), ICaL, or ICaP did not reduce ΔCai2+. In the area of ΔV+m (site 2), Cai2+ decrease became smaller after disabling of ICaL or Tpn. Blocking other pathways (NCX, SR, ICaP) produced no or minor effects.
The effect of ICaL inhibition on ΔCai2+ was more pronounced when ICaL was partially inhibited during the whole simulation. With ICaL reduced by 60%, a shock caused practically no change in Cai2+ in the ΔV+m area (Figure 5C). In these conditions, ΔV+m and ΔV−m became equal. Thus, block of ICaL resulted in elimination of ΔVm asymmetry.
With increasing shock strength, Cai2+decrease became larger in the ΔV−m area (Figure 5D). In the ΔV+m area, ΔCai2+ dependence was nonmonotonic, with ΔCai2+ becoming positive at stronger shocks. This ΔCai2+ reversal was eliminated by blocking NCX (not shown), indicating that positive ΔCai2+ was caused by inflow of Ca2+ via reversed NCX at large ΔV+m. Within the range of ΔV+m observed experimentally (<≈100 mV), however, shock-induced ΔCai2+ were negative (Figure 5E).
Effects of Nifedipine, Caffeine, and Thapsigargin on Shock-Induced ΔCai2+
The roles of ICaL and SR in ΔCai2+ were investigated in cell cultures using drug application. Figure 6 illustrates the effects of 1 μmol/L nifedipine on ΔCai2+. Nifedipine completely eliminated the shock-induced Cai2+ decrease at the cathodal side (Figure 6A, traces 1). This effect was reversible (not shown). Similar results were obtained in 7 monolayers. As shown in Figure 6B, nifedipine caused radical reduction of the average cathodal ΔCai2+ (P<0.05 from control), which became not different from zero (NS); the average anodal ΔCai2+ was reduced by ≈37% (P<0.05). In 2 monolayers, ΔVm were measured together with ΔCai2+. In accordance with a previous publication,15 nifedipine reduced ratio of ΔV−m/ΔV+m caused by an increase of ΔV+m (Figure 6C).
Figure 7 demonstrates the effects of caffeine (10 mmol/L) and thapsigargin (1 μmol/L) on ΔCai2+. Although both drugs caused slowing of rising and recovery phases of Cai2+ transients, they did not eliminate shock-induced Cai2+ decrease (Figure 7A and 7B). The average ΔCai2+ was not strongly affected by both drugs (Figure 7C and 7D).
Measurements of Shock-Induced ΔCai2+ With High-Affinity Dyes
As shown in Figure 1, dye affinity affects measurements of Cai2+ transient duration. To examine whether it also affects measurements of shock-induced ΔCai2+, these measurements were repeated using a high-affinity dye Fluo-4. As shown in Figure 8A, use of this dye resulted in measurements of much smaller negative ΔCai2+ than those measured with Fluo-4FF. A 9-V/cm shock caused only ≈5% decrease of Cai2+ at the anodal strand edge (trace 1) and no Cai2+ change at the cathodal edge (trace 2). In addition, Fluo-4 resulted in measurements of larger postshock Cai2+ changes. Thus, the 29-V/cm shock caused elevation of the anodal diastolic Cai2+ level by ≈65% ACa (Figure 8B). Qualitatively similar results were observed in a total of 4 cell monolayers stained with Fluo-4.
This study presents high-resolution measurements of Cai2+ and Vm changes caused by electrical shocks in cultured cell monolayers. The main findings are: (1) shocks induced transient decreases of Cai2+ in areas of both ΔV−m and ΔV+m; computer simulations and experiments with channel blockers indicate that ICaL played an important role in Cai2+ decrease; (2) the magnitude of ΔCai2+ had a nonmonotonic dependence on shock strength, decreasing at stronger shocks; this was paralleled with the occurrence of biphasic negative ΔVm; and (3) optical measurements of Cai2+ changes were strongly dependent on the affinity of the Cai2+-sensitive dye.
Role of Dye Affinity in Cai2+ Measurements
It is well-recognized that dye affinity plays an important role in Cai2+ measurements.22 Although there is no “gold” standard for validation of fluorescent dyes, it is generally agreed that low-affinity dyes are more appropriate for dynamic Cai2+ imaging.22 High-affinity dyes may misrepresent Cai2+ transients because of several factors, including (1) saturation at high Ca2+ concentrations generated by cardiac cells and (2) slow dye dissociation from Cai2+ ions. Without in situ calibration compensating for dye saturation, using high-affinity dyes may result in “clipping” of the upper portion of Cai2+ transients, which can explain long measured Cai2+ transient durations (see online supplement for detailed explanation). This mechanism may also exaggerate small Cai2+ changes near the resting level, which can explain the large elevation of diastolic Cai2+ level measured with high-affinity dyes. The fact that these dyes may underestimate Cai2+ changes near the peak of the Cai2+ transient contributes to the apparent small shock-induced ΔCai2+.
Shock-Induced ΔCai2+ and Relationship With ΔVm
This is the first study in which spatio-temporal Cai2+ changes were measured and directly related with colocalized Vm changes during electrical shocks in multicellular cardiac tissue. It was found that shocks of all strengths caused transient decreases of Cai2+ in the areas of both negative and positive polarizations. Qualitatively similar results were obtained in a computer model of rat myocytes in which shocks induced asymmetric ΔVm and Cai2+ decrease at sites of ΔV−m and ΔV+m within the range of ΔVm observed experimentally. The model allowed evaluating roles of different Cai2+-handling pathways in shock-induced ΔCai2+. Simulations showed that in ΔV−m areas, Cai2+ reduction was caused by inactivation of ICaL combined with extrusion of Ca2+ ions by NCX and troponin binding. Unexpectedly, SR played no role in Cai2+ decrease. This could be because of the fact that SR activity is relatively low in this model; it is responsible for only ≈30% of Cai2+ transient amplitude.
In regions of ΔV+m, ICaL remained active but it changed its direction from inward to outward when Vm exceeded the ICaL reversal potential causing a reduction of Cai2+ at these locations. This mechanism accounted for a significant portion of shock-induced Cai2+ decrease with the remaining portion being caused by troponin binding. NCX did not play a role in Cai2+ decrease in this area because NCX in this model reverses at potentials above ≈0 mV.
Experiments in cell cultures supported the important role of ICaL in ΔCai2+. Thus, inhibition of ICaL by nifedipine caused elimination of Cai2+ decrease at the cathodal strand side indicating that the reversed flow of ICaL contributed to Cai2+ decrease in this area in control conditions. In accordance with the model, disabling of SR by caffeine and thapsigargin did not affect ΔCai2+. This may suggest that, similar to the model, SR played a negligible role in ΔCai2+ in cell cultures, which is consistent with the relatively low level of SR development in neonatal rat myocytes.23,24 Alternatively, the lack of drug effects may be explained by their inhibition of both Cai2+ release and uptake. This may affect amplitudes of Cai2+ transients and ΔCai2+ to the same extent, leaving relative ΔCai2+ unchanged.
The fact that Cai2+ was decreased during shocks at the majority of locations is in apparent contradiction with previous studies in whole hearts reporting an increase of peak Cai2+ in response to shocks applied during the absolute refractory period.6 Also, no Cai2+ decrease during shocks was reported in a study in isolated cells.7 These discrepancies may be caused by a combination of several factors. First, previous studies used high-affinity dyes that, because of their saturation and slow dissociation, may underestimate negative ΔCai2+ near peak of Cai2+ transient. Second, these studies did not assess regional differences in Cai2+ changes; according to the present work, ΔCai2+ are nonuniform and may remain unchanged during shocks in areas where ΔVm are small. Therefore, the definitive resolution of the discrepancy between this and the previous studies requires spatially and temporally resolved measurements of shock-induced ΔCai2+ and ΔVm in whole hearts or whole tissue preparations using low-affinity Cai2+ dyes.
It was previously reported that shocks with a strength below the defibrillation threshold applied during the absolute refractory period or during fibrillation can cause a positive inotropic effect, alleviating postshock “pulseless electrical activity.”3–5 Two alternative mechanisms were proposed to explain this effect. First, shocks could have a direct effect on myocytes by increasing their peak Cai2+. Second, shocks could affect myocytes indirectly via stimulation of sympathetic intracardiac nerves. The shock strength used in these studies was below the defibrillation threshold, which is estimated at ≈5 V/cm.25 According to the present study, shocks of such strength should not increase Cai2+ directly. Instead, the main effect of 5-V/cm shocks was a decrease of Cai2+ during the shocks at sites of both positive and negative ΔVm (Figure 4). There was also a small increase of the Cai2+ transient duration (by ≈10%), but such an effect is unlikely to cause an increase in peak force. Therefore, these data do not support the hypothesis relating inotropic shock effect with the direct increase of Cai2+. More likely, it is caused by parasympathetic nerve stimulation.
Strong shocks (E ≥20 V/cm) that caused biphasic ΔVm produced smaller Cai2+decreases than the weaker shocks that caused monophasic ΔVm. The stronger shocks also often caused a small but statistically significant postshock elevation of Cai2+, which was more prominent in the areas of negative ΔVm (Figure 4). Shocks of such strength were shown to cause membrane electroporation.15 Therefore, both Cai2+ changes can be explained by shock-induced electroporation causing entry of Ca2+ ions via membrane pores. This interpretation is also supported by simulations in a model without membrane electroporation in which increasing shock strength caused larger negative ΔCai2+ at sites of ΔV−m. At sites of ΔV+m, ΔCai2+ dependence was nonmonotonic, with negative ΔCai2+ first increasing and then decreasing because of Ca2+ entry via reverse NCX. Because this effect was associated with relatively large positive ΔVm, which were not observed experimentally, it is unlikely that this mechanism played a large role in ΔCai2+ in cell cultures.
Shock-induced Cai2+ overload has been implicated in the generation of postshock focal arrhythmias1 via activation of calcium-dependent inward currents.26 The threshold for these arrhythmias in 0.8-mm-wide cultured cell strands is ≈20 V/cm.1 Although such shocks caused a statistically significant increase of diastolic Cai2+, the change was too small (≈4%) to make a definitive conclusion about possible contribution of such Cai2+ changes to arrhythmia generation. It is more likely that Cai2+ elevation plays a role in arrhythmias induced by stronger shocks.
Postshock Cai2+ elevation was previously described in a study on isolated myocytes.7 In that work, Cai2+ elevation comparable to the amplitude of Cai2+ transients (≈100%ACa) was observed after shocks with strengths of 30 V/cm and higher. Such Cai2+ changes are much larger than the diastolic Cai2+ changes observed here at comparable shock strengths (Figure 4C). This difference may be partially attributed to the high affinity of Fura-2 used in these experiments. When dye Fluo-4 with a similar affinity was used here, significant postshock elevation of Cai2+ was observed after 30-V/cm shocks, also (Figure 7). This and the arguments presented indicate that previous measurements might have substantially overestimated shock-induced elevation of diastolic Cai2+.
Cai2+ changes were measured in strands of only one width. Because Cai2+ changes are mediated by Vm changes, shock-induced ΔCai2+ in strands of other widths can be estimated using the dependence of ΔVm magnitude on the strand width published previously.11 There are species- and age-dependent differences in ionic currents and handling of intracellular calcium23,24,27 that may affect shock-induced Cai2+ changes. On the qualitative level, however, calcium and other ionic currents have similar voltage dependence and kinetic properties in different cell types, indicating that the effects of shocks on Cai2+ should be qualitatively similar as well.
This work was supported by NIH Grants HL67748 and HL67961 and AHA Grant 0255025B. We thank Reuben Collins for help with preparation of cell cultures.
Original received October 21, 2003; resubmission received January 29, 2004; revised resubmission received May 5, 2004; accepted May 7, 2004.
Fast VG, Cheek ER. Optical mapping of arrhythmias induced by strong electrical shocks in myocyte cultures. Circ Res. 2002; 90: 664–670.
Gold JH, Schuder JC, Stoeckle H, Granberg TA, Hamdani SZ, Rychlewski JM. Transthoracic ventricular defibrillation in the 100 kg calf with unidirectional rectangular pulses. Circulation. 1977; 56: 745–750.
Rosborough J, Deno D. Electrical therapy for pulseless electrical activity. Pacing Clin. Electrophysiol. 2000; 23 Part II: 591.
Walcott GP, Melnick SB, Killingsworth CR, Smith WM, Ideker RE. Effects of burst stimulation during ventricular fibrillation on cardiac function after defibrillation. Am J Physiol. 2003; 285: H766–H774.
Mohri S, He KL, Dickstein M, Mika Y, Shimizu J, Shemer I, Yi GH, Wang J, Ben-Haim S, Burkhoff D. Cardiac contractility modulation by electric currents applied during the refractory period. Am J Physiol. 2002; 282: H1642–H1647.
Mohri S, Shimizu J, Mika Y, Shemer I, Wang J, Ben-Haim S, Burkhoff D. Electric currents applied during refractory period enhance contractility and systolic calcium in the ferret heart. Am J Physiol. 2003; 284: H1119–H1123.
Sharma V, Tung L. Effects of uniform electric fields on intracellular calcium transients in single cardiac cells. Am J Physiol. 2002; 282: H72–H79.
Efimov IR, Aguel F, Cheng Y, Wollenzier B, Trayanova N. Virtual electrode polarization in the far field: implications for external defibrillation. Am J Physiol. 2000; 279: H1055–H1070.
Fast VG, Sharifov OF, Cheek ER, Newton JC, Ideker RE. Intramural virtual electrodes during defibrillation shocks in left ventricular wall assessed by optical mapping of membrane potential. Circulation. 2002; 106: 1007–1014.
Fast VG, Rohr S, Ideker RE. Non-linear changes of transmembrane potential caused by defibrillation shocks in strands of cultured myocytes. Am J Physiol. 2000; 278: H688–H697.
Cheek ER, Fast VG. Nonlinear changes of transmembrane potential during electrical shocks: role of membrane electroporation. Circ Res. 2004; 94: 208–214.
Cheek ER, Ideker RE, Fast VG. Nonlinear changes of transmembrane potential during defibrillation shocks: role of calcium current. Circ Res. 2000; 87: 453–459.
Winslow RL, Rice J, Jafri S, Marban E, O’Rourke B. Mechanisms of altered excitation-contraction coupling in canine tachycardia-induced heart failure, II: model studies. Circ Res. 1999; 84: 571–586.
Pollard AE, Cascio WE, Fast VG, Knisley SB. Modulation of triggered activity by uncoupling in the ischemic border. A model study with phase 1b-like conditions. Cardiovasc Res. 2002; 56: 381–392.
Takahashi A, Camacho P, Lechleiter JD, Herman B. Measurement of intracellular calcium. Physiol Rev. 1999; 79: 1089–10125.
Wibo M, Bravo G, Godfraind T. Postnatal maturation of excitation-contraction coupling in rat ventricle in relation to the subcellular localization and surface density of 1,4-dihydropyridine and ryanodine receptors. Circ Res. 1991; 68: 662–673.
Schlotthauer K, Bers DM. Sarcoplasmic reticulum Ca2+ release causes myocyte depolarization - Underlying mechanism and threshold for triggered action potentials. Circ Res. 2000; 87: 774–780.