| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Department of Biomedical Engineering, Case Western Reserve University, Cleveland, Ohio 44106; and 2Institut Non Linéaire de Nice, Centre National de la Recherche Scientifique, 06560 Valbonne, France
Submitted 24 July 2003 ; accepted in final form 25 September 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Vm) response induced by a strong external electrical field in cardiac cells. We investigated the contribution of electroporation to Vm transients during high-intensity shocks using optical mapping. Rectangular and ramp stimuli (10–20 ms) of different polarities and intensities were applied to the rabbit heart epicardium during the plateau phase of the action potential (AP). Vm were optically recorded under a custom 6-mm-diameter electrode using a voltage-sensitive dye. A gradual increase of cathodal and well as anodal stimulus strength was associated with 1) saturation and subsequent reduction of Vm; 2) postshock diastolic resting potential (RP) elevation; and 3) postshock AP amplitude (APA) reduction. Weak stimuli induced a monotonic Vm response and did not affect the RP level. Strong shocks produced a nonmonotonic Vm response and caused RP elevation and a reduction of postshock APA. The maximum positive and maximum negative Vm were recorded at 170 ± 20 mA/cm2 for cathodal stimuli and at 240 ± 30 mA/cm2 for anodal stimuli, respectively (means ± SE, n = 8, P = 0.003). RP elevation reached 10% of APA at a stimulus strength of 320 ± 40 mA/cm2 for both polarities. Strong ramp stimuli (20 ms, 600 mA/cm2) induced a nonmonotonic Vm response, reaching the same largest positive and negative values as for rectangular shocks. The transition from monotonic to nonmonotonic morphology correlates with RP elevation and APA reduction, which is consistent with cell membrane electroporation. Strong shocks resulted in propidium iodide uptake, suggesting sarcolemma electroporation. In conclusion, electroporation is a likely explanation of the saturation and nonmonotonic nature of cellular responses reported for strong electric stimuli.
defibrillation; arrhythmia; electrophysiology; cardiac tissue electrical damage
Vm) in single cells (12, 18), cell culture strands (9), and heart tissue (7, 10). However, reports on strong shocks of defibrillation strength disagree on the morphology and amplitude of shock-induced Vm responses.
When a stimulus is applied to a single cell during the early plateau phase of the action potential (AP), the optical recordings show depolarization at the cathodal end and hyperpolarization at the anodal end of the cell (14, 22). During a progressive increase in stimulus intensity, hyperpolarization (or, more accurately, negative polarization) Vm first gradually increase in amplitude but soon start to decay, causing an elevation of the cell average potential (18). Similar effects were observed in the narrow strands of cultured rat myocytes (9). Such a decrease of hyperpolarization Vm responses with shock amplitude was attributed to activation of an unknown hyperpolarization-activated channel(s) rather than to electroporation of the cell membrane, because there was no membrane-impermeable dye uptake even with 50 V/cm shocks (8). However, other data demonstrated such an uptake at similar electrical field strengths (11, 13). In contrast to negative Vm responses, positive polarization was found to gradually increase with shock strength, saturating below 100% of the AP amplitude (APA) (9). Whole heart studies have revealed a different type of asymmetry for the positive and negative polarizations during strong shocks. Neunlist and Tung (17) presented measurements of the epicardial cellular response recorded from a 150-µm-diameter area of stimulus application, showing a hyperpolarization overshoot during anodal stimulation (21) similar to that found in cell strands (8, 9). Fast et al. (10) demonstrated in a slab preparation that areas having positive polarization during weak shocks showed negative polarization for a shock strength >34 V/cm. However, the small size of the stimulated area in the Neunlist and Tung (17) study could affect the measurements due to virtual electrode phenomena (16), leading to the development of positive and negative polarizations at nearby locations. Results in the slab preparation (16) could be affected by the interruption of the muscle fibers by the transmural tissue cut. In the present study, we sought to determine epicardial Vm responses during high-density electrical current stimuli of both polarities applied at the large area of the left ventricle. We also investigated whether changes in Vm responses are associated with electroporation. We hypothesized that changes in Vm response morphology during strong shocks are caused by electroporation.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We designed the electrode shown in Fig. 1A to achieve high-current density stimulation over a relatively large area of the epicardium without electrochemical electrode-saline interface artifacts (bubbles) in the optical recordings. A transparent plastic box filled with Tyrode solution with a 6-mm-round hole was gently pressed against the left ventricular free wall (see Fig. 1B) to seal the opening. The 16 x 16 photodiode array optical recording system (2) was focused on the 4 x 4-mm area of epicardial surface inside the hole. A 20 x 20-mm silver wire loop was positioned inside the box 10 mm from the heart surface around the opening, allowing unobstructed optical recordings from the field of view. Stimulating current was injected inside the box through this wire. The reference electrode was a 9-mm-diameter Ag-AgCl disk placed in the bath away from the heart (see Fig. 1). The current was allowed to exit the box only through the hole directly into the epicardium, producing homogeneous transmembrane polarization in the field of view. Figure 2 shows a representative example of optical recordings from all 256 channels. To minimize possible effects of heterogeneous virtual electrode polarization near the edges of the hole, we analyzed recordings only from the center of the mapped area. The average current density of the stimulus was determined as the ratio between the applied current and total cross-sectional area of the hole. The total impedance between the test and reference electrodes was 200–300 
.
|
|
The heart was paced at a cycle length of 300 ms through a bipolar electrode placed at the apex. Rectangular and ramp stimuli of both polarities and with a duration of 10 or 20 ms were generated by the bipolar operational power supply/amplifier (Kepco) during the plateau phase of the AP (20–40 ms after the upstroke). To calibrate the transmembrane voltage optical signal, we assumed that the APA was 100 mV and the resting potential was –85 mV (5). Applied shock voltage and current waveforms were recorded simultaneously with optical maps. We used two sampling rates: a sampling rate of 1.5 kHz was used during acquisition of all 256 optical channels, and a sampling rate of 5 kHz was used during recordings of only 4 optical channels from the center of the field of view.
To detect sarcolemma electroporation during the shocks, we perfused the heart with 30 µM of propidium iodide (PI; a membrane-impermeable nucleic acid stain) and recorded the increase of its fluorescence after shock application for 10–20 min as a marker of electroporated areas. We then washed out the PI for 30 min, cryosectioned the heart from the base to apex with 1-mm steps in 20-µmthick slices, and analyzed the sections with an epifluorescent microscope (Nikon Eclipse600FN). To avoid possible spectra overlapping, we did not use voltage-sensitive dye staining in these experiments at all or used RH-237, which has a different emission spectrum than PI.
Student's t-test for paired data was used to compare strength of stimuli. Values of P < 0.05 were considered significant. All quantitative data are expressed as means ± SD unless otherwise specified.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Figure 3 shows one representative example (of 8 preparations) of optical recordings of the transmembrane potential obtained from one channel in the middle of the field of view. For moderate strength (light gray traces), shock-induced transients (Vm) were monotonic and gradually increased with the increase of the stimuli. However, with a further increase in shock strength, Vm became nonmonotonic. First, they rapidly increased (phase 1) and then saturated and relatively slowly decreased (phase 2). These observations agree with observations in cell culture and isolated cells (9, 18) for hyperpolarizing responses but are different for depolarizing ones. Interestingly, with a gradual increase of stimulus strength, the amplitude of the voltage transient decreases, whereas the slope of the second phase increases. The appearance of a second phase in Vm was accompanied by another transition on a different time scale (Fig. 3B)–appearance of shock-induced diastolic depolarization, suggesting an onset of electroporation.
|
In each of eight preparations for every shock strength applied, we measured (see Fig. 4, inset) the following: 1) the largest deviation of optical potential from the preshock level during the stimulus application (Vmax); 2) the final deviation at the end of the shock (Ve); 3) the postshock diastolic potential (DP) elevation; and 4) the postshock APA (Vamp). The plot in Fig. 4 shows how these characteristics depend on the shock strength in another representative preparation. Weak rectangular stimuli induced monotonic Vm (in this case, Ve follows Vmax) and did not affect DP. Strong rectangular shocks produced nonmonotonic Vm (as a result, Ve deviates from Vmax), caused DP elevation, and reduced postshock APA. Figure 5A shows the average data and SDs for the largest positive and negative values of Vm during shocks (Vmax) and the postshock DP elevation for eight hearts.
|
|
To characterize the onset of nonmonotonic Vm transients and DP elevation, we determined the following from each individual plot (see Fig. 4): 1) the shock strengths causing a DP elevation of 10% of APA (IDP10%); and 2) the shock strengths inducing the maximum depolarization or hyperpolarization (IVmax). Figure 5B shows the average and SDs for IVmax and IDP10% for cathodal and anodal stimuli. A paired Student t-test comparison of anodal versus cathodal stimuli gave P = 0.003 for IVmax and P = 0.94 for IDP10%. The summary data were as follows: 1) DP elevation reached 10% of APA at an average stimulus strength of 320 ± 40 mA/cm2 for both polarities (mean ± SE, n = 8); and 2) the maximum positive Vm was recorded at an average stimulus strength of 170 ± 20 mA/cm2 for cathodal polarity, and maximum negative Vm was recorded at an average stimulus strength of 240 ± 30 mA/cm2 for anodal polarity (means ± SE, n = 8).
As in previous studies, observed largest positive and negative values of polarization were about twice smaller than expected for electroporation (100 vs. 300–400 mV) (17, 20). One of the possible explanations proposed previously was the insufficient time resolution of recording systems, i.e., the observer could fail to notice the initial high amplitude transient Vm at the onset of rectangular stimuli because this large Vm would immediately disappear due to cell membrane electroporation (6, 17). To overcome possible bandwidth limitations, we applied ascending and descending voltage ramps instead of a rectangular pulse stimulus. Strong 20-ms 600 mA/cm2 ramp stimuli induced Vm reaching the same largest positive and negative values that were observed during a series of rectangular shocks in all eight experiments.
Figure 6 shows typical Vm transients during 20-ms voltage ramps superimposed with the responses to the rectangular pulses. The transient decays start at similar Vm levels as with the rectangular pulses despite the fact that the ramp stimulation voltage was still rising. The slow rate of shock voltage increase guaranteed that we could reliably record Vm at all moments. This example proves that potential metrological problems (signal undersampling or amplifier bandwidth limitations) cannot explain why the maximum values of Vm recorded during electroporating rectangular shocks in previous studies and in our experiments were far less than is assumed to be required for electroporation (17, 20).
|
To test our hypothesis that our stimuli could electroporate the sarcolemma, we applied strong shocks to the heart perfused with 30 µM PI (n = 2). Figure 7, top, shows the initial increase of PI fluorescence after the beginning of perfusion recorded inside the stimulating hole (solid line) and 3 mm outside the hole (dashed line). After a strong cathodal shock (1,600 mA/cm2, 20 ms; arrow in Fig. 7), there was an accelerated accumulation of the fluorophore in the tissue inside the hole but not outside the hole. Similar results were obtained for anodal shocks. The average fluorescence increase during 10 min after the shocks in this experiment was 81 ± 19% inside and 13 ± 1% outside (means ± SD, n = 2) the stimulated area. After a single shock was applied, the heart was cryosectioned. At the electroporated area, PI penetrated inside the cells and bound to the nuclei. Figure 7, bottom, shows the fluorescent images made with x4 and x40 lens for a 20-µm slice sectioned throughout the stimulated area. The electroporated region is clearly demarcated by the PI-stained nuclei (bright specks in the high-resolution image of dye staining).
|
To relate the changes in optical potentials to the PI uptake, we dual-stained the hearts (n = 2) with PI and the voltage-sensitive dye RH-237. RH-237 and PI fluorescence were collected at the same spot beneath the electrode. Figure 8 shows that two consequent 300 mA/cm2 shocks of both polarities caused neither a diastolic optical potential elevation (black traces) nor PI fluorescence increase (laser trace after black arrow). Two consequent 700 mA/cm2 shocks of both polarities caused DP elevation (gray traces) and the onset of PI fluorescence increase (laser trace after gray arrow). The same phenomena were observed in a second preparation. After all four stimuli were applied, the heart was cryosectioned. Histological evaluation of the same tissue slice cut through the center of stimulated area (Fig. 8, right) showed that PI staining was localized at the thin layer of cell nuclei near the epicardium in the area adjacent to the electrode, confirming the occurrence of electroporation. Comparing the thickness of the damaged area in Figs. 7 and 8, one can see that lowering stimuli from 1,600 to 700 mA/cm2 resulted in a tremendous reduction of the electroporated area depth, from 
2 to 0.2 mm. We did not observe PI uptake in the cryosections where there was no effect of the shock on the PI fluorescence intensity from the epicardium. This means that the electroporation threshold in this experiment was between 700 and 300 mA/cm2 when determined by either DP elevation or the PI uptake method.
|
The similar relatively slow time course of PI uptake was observed in a cell culture (19), where the dye influx continued 10 min after the application of an electric pulse.
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The correlation between anodal (negative) Vm and diastolic Vm elevation was recently reported by Fast and Cheek (8) in myocyte cultures. The same result was shown before by Neunlist and Tung (17) and Cheng et al. (6). Fast and Cheek (8) also suggested that electroporation was a likely mechanism of nonmonotonic Vm, but, because Lucifer yellow dye uptake was not observed at a shock strength of 50 V/cm, which was above the 30 V/cm threshold of nonmonotonic negative Vm, this conclusion was not definitive. However, Gillis et al. (11) did observe an accumulation of this dye in the areas close to the anode and cathode after shocks with a field strength of 
22 V/cm in a experimental setup similar to the one used by Fast and Cheek (8), which is puzzling. Interestingly, Fast and Cheek (8) did not observe nonmonotonic Vm at the cathodal end of the cell strand even at highest shock strengths. In the present study, we showed PI dye uptake at a shock strength of 700 mA/cm2 (35 V/cm), when we detected nonmonotonic Vm, and no PI uptake at 300 mA/cm2 (15 V/cm), when Vm was monotonic. We do not have an explanation for the absence of positive Vm decay in the Fast and Cheek study (8). The lack of detectable dye uptake in their work could be related to the small exposure time and to the lesser sensitivity of the Lucifer yellow technique. PI has 20- to 30-fold increases in fluorescence after being bound to nucleic acids. Our optical recordings of negative Vm responses to high-intensity stimuli are in agreement with results reported by other investigators in strands of cultured myocytes (3, 9), single cells (4), and the frog heart (17) for hyperpolarizing stimuli. The positive Vm response in our study was different. We did not observe a plateau or an increase in depolarization transients during cathodal stimuli for the same stimulus strengths that caused decayed hyperpolarization responses in their studies. Both polarities had the same injury threshold judging by DP elevation, which fits well with Knisley and Grant (15) data showing that cell injury is independent of the intrinsic transmembrane potential. This suggests that electroporation is responsible for saturation and decay of Vm.
Recent data (10) showed that the initial positive polarization in virtual cathode areas in wedge preparation is changing to hyperpolarizing responses as stimuli strengths increase to 30 V/cm and above, similar to the behavior of the middle of a single myocyte in the Sharma and Tung study (18). Such observations have been reported previously by Cheng et al. (6) and Zhou et al. (23), who detected hyperpolarization transients near the cathodal shock electrode. If this phenomenon takes place during epicardial stimulation as well, it can explain why depolarization saturation is observed at lower shock current densities than hyperpolarization saturation.
We observed, similar to previous investigators, a smaller maximum hyperpolarization response than could be expected to be high enough for electroporation. Among the possible reasons for this were 1) a "dog bone" virtual polarization near the pacing electrode (17), which could attenuate the response due to optical averaging over areas of opposite polarizations; or 2) insufficient temporal resolution of the optical mapping system (17), which could underestimate the true instantaneous transmembrane voltage produced by a square pulse. Our results for polarization transients recorded during 20-ms ramp waveform stimulation (no temporal resolution limitations) over the 6-mm-diameter area of epicardium (opposite polarization is located 3 mm away from the center recording point) reject such explanations.
In our four experiments with PI, we did not detect a PI fluorescence increase during the shock. This suggests that the amount of PI molecules that penetrated through the electroporation holes during the 20-ms stimulus is undetectable in our setup. This also explains why we did not observe a difference in PI uptake for shocks of different polarities despite the charge of the PI molecule. The major amount of PI enters the cells when the external electrical field is already turned off. This is why fluorescence is continuously rising during dye perfusion in our experiments as it did in a cell culture study (19). This means that, in our experiments, electroporated cells were repaired within minutes rather than seconds. We suggest that that DP elevation might be a more sensitive indicator of electroporation than PI uptake because DP elevation can be detected within 1 s after shock application.
Limitations. First measurement of the cellular response directly at the place of stimulation was performed by Neunlist and Tung (17) for a 150-µm pipette with a single-channel fiber fluorimeter. In our study, with a 6-mm-diameter stimulating hole, we were able to create a larger stimulating field and avoid the possible interference from virtual electrode polarization. However, we cannot exclude that transient decays were in some part related to electrotonic interference from the areas of opposite polarization around the stimulating hole. We performed control experiments with a 14-mm-diameter hole and found that the nonmonotonic nature of the responses in the central area was preserved. The use of 2,3-butanedione monoxime could affect the observed transients due to partial ion channel blockage.
In conclusion, changes in the morphology of transmembrane polarization transients during anodal and cathodal shocks from monotonic to nonmonotonic responses are associated with elevation of the resting potential, postshock APA reduction, and PI uptake, implying the occurrence of electroporation. Electroporation changes in transmembrane potential traces are present for hyperpolarized as well as depolarized stimuli of a similar strength.
|
ACKNOWLEDGMENTS |
|---|
This study was supported by National Heart, Lung, and Blood Institute Grant HL-67322 and a grant from the Whitaker Foundation.
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  V. P. Nikolski and I. R. Efimov Electroporation of the heart Europace, January 1, 2005; 7(s2): S146 - S154. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Ashihara and N. A. Trayanova Cell and tissue responses to electric shocks Europace, January 1, 2005; 7(s2): S155 - S165. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Ashihara and N. A. Trayanova Asymmetry in Membrane Responses to Electric Shocks: Insights from Bidomain Simulations Biophys. J., October 1, 2004; 87(4): 2271 - 2282. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. T. Sambelashvili, V. P. Nikolski, and I. R. Efimov Virtual electrode theory explains pacing threshold increase caused by cardiac tissue damage Am J Physiol Heart Circ Physiol, June 1, 2004; 286(6): H2183 - H2194. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
