Chemical uncouplers diacetyl monoxime (DAM) and cytochalasin D (cyto-D) are used to abolish cardiac contractions in optical studies, yet alter intracellular Ca2+ concentration ([Ca2+]i) handling and vulnerability to arrhythmias in a species-dependent manner. The effects of uncouplers were investigated in perfused mouse hearts labeled with rhod-2/AM or 4-[β-[2-(di-n-butylamino)-6-naphthyl]vinyl]pyridinium (di-4-ANEPPS) to map [Ca2+]i transients (emission wavelength = 585 ± 20 nm) and action potentials (APs) (emission wavelength > 610 nm; excitation wavelength = 530 ± 20 nm). Confocal images showed that rhod-2 is primarily in the cytosol. DAM (15 mM) and cyto-D (5 μM) increased AP durations (APD75 = 20.0 ± 3 to 46.6 ± 5 ms and 39.9 ± 8 ms, respectively, n = 4) and refractory periods (45.14 ± 12.1 to 82.5 ± 3.5 ms and 78 ± 4.24 ms, respectively). Cyto-D reduced conduction velocity by 20% within 5 min and DAM by 10% gradually in 1 h (n = 5 each). Uncouplers did not alter the direction and gradient of repolarization, which progressed from apex to base in 15 ± 3 ms. Peak systolic [Ca2+]i increased with cyto-D from 743 ± 47 (n = 8) to 944 ± 17 nM (n = 3, P = 0.01) but decreased with DAM to 398 ± 44 nM (n = 3, P < 0.01). Diastolic [Ca2+]i was higher with cyto-D (544 ± 80 nM, n = 3) and lower with DAM (224 ± 31, n = 3) compared with controls (257 ± 30 nM, n = 3). DAM prolonged [Ca2+]i transients at 75% recovery (54.3 ± 5 to 83.6 ± 1.9 ms), whereas cyto-D had no effect (58.6 ± 1.2 ms; n = 3). Burst pacing routinely elicited long-lasting ventricular tachycardia but not fibrillation. Uncouplers flattened the slope of AP restitution kinetic curves and blocked ventricular tachycardia induced by burst pacing.
- optical action potentials
- action potentials
- intracellular calcium
- restitution kinetics
molecularly engineered mice have been extensively used to genetically alter a specific component of a complex signaling process and to develop models of human diseases. Transgenic mice are used as models for various cardiac diseases and offer an effective strategy to elucidate the mechanisms underlying long QT-related arrhythmias, metabolic diseases, and the pathology of heart failure (29). A limitation of mouse models is the small size of the heart, making it difficult to study changes in contractility, electrophysiology, and vulnerability to arrhythmias in intact hearts. The challenge of measuring changes in cardiac phenotype has been partly overcome by applying optical technique to map electrical activity, but a major technical difficulty in the application of optical techniques to measure action potentials (APs) and intracellular Ca2+ concentration ([Ca2+]i) transients has been the distortion of the signals by movement due to muscle contractions.
Several approaches have been used to reduce movement artifacts: 1) perfusion in Ca2+-free Tyrode solution to abolish contractions, an approach applicable to amphibian hearts (38); 2) design perfusion chambers to mechanically stabilize the heart (18, 37); 3) perfusion with an inhibitor of L-type voltage-gated Ca2+ channels (ICa,L) to reduce [Ca2+]i and force generations (16); and 4) perfusion with a chemical uncoupler of excitation-contraction like diacetyl monoxime (DAM) and cytochalasin D (cyto-D) to block force by a direct inhibition of the contractile filaments (5, 14).
Chemical uncouplers can potentially provide a practical approach to block movement artifacts and have been used to inhibit contractions during optical recordings, particularly for measurements of the recovery phase of the AP and [Ca2+]i transients, which tend to be distorted by movement artifacts. DAM acts as a chemical phosphatase, exerts its biological effects by altering protein phosphorylation, and blocks contractions through the inhibition of myosin ATPase activity such that the rise of intracellular [Ca2+]i elicited by an AP fails to generate force (2, 6). However, DAM also alters repolarization and reduces AP durations (APD) in several species of cardiac tissues, such as cat ventricular muscles (43), dog Purkinje fibers (4), and guinea pig papillary muscle (27). In contrast, DAM prolonged APDs in rat Purkinje fibers (12, 13) and mouse ventricles (3). In sheep and guinea pig ventricular muscles, DAM reduced calcium and potassium conductance (27).
Cyto-D was shown to disrupt F-actin filaments in the cytoskeleton and unexpectedly blocked contractions by disrupting F-actin in myofibrils with little effect on APDs of ventricular rat (40) and canine myocytes (5). Cyto-D was considered to be a better uncoupler than DAM because of its negligible effects on APDs, transmural propagation velocity, and repolarization gradients (5, 44). However, cyto-D was shown to abolish Ca2+-mediated inward rectification of K+ channels (31) and to modulate the kinetics of voltage-gated Na+ current (40) by disrupting the cytoskeleton of cardiac myocytes.
The interpretation of optical data obtained using chemical uncouplers must be carefully reexamined in light of their multiple effects on ionic channels, gap junctions, and intracellular Ca2+ handling in myocytes (12, 13, 21, 27, 28, 30, 39, 42–44). There is also little doubt that the electrophysiological effects of cyto-D and DAM are species dependent (39), yet they have not been extensively studied in murine hearts. Here, we examine the effects of cyto-D and DAM (using the lowest concentrations that block force reliably) on APs and [Ca2+]i transient and on the vulnerability to arrhythmias in mouse hearts.
All animal procedures complied with National Institutes of Health guidelines and were approved by the IACUC of the University of Pittsburgh. FVB mice were anesthetized with pentobarbital (50 mg/kg) and heparinized (35 mg/kg) with an intraperitoneal injection. The heart was rapidly excised, cannulated, placed in a chamber specially designed to immobilize the ventricles, and paced, and a chosen region was imaged on a photodiode array. The perfusate contained (in mM) 112 NaCl, 1.0 KH2PO4, 25.0 NaHCO3, 1.2 MgSO4, 5.0 KCl, 50.0 dextrose, and 1.8 CaCl2, at pH 7.4, and was gassed with 95% O2-5% CO2. Perfusion pressure was adjusted to 60–80 mmHg by controlling the flow rate of a peristaltic pump. The temperature of the bath surrounding the heart was kept at 37°C by continuously monitoring the temperature with a thermistor, which controlled a heating coil located in the back of the chamber via a feedback amplifier. In pilot studies, left ventricular pressure was measured using an intraventricular balloon (23) to ensure that the concentrations of uncoupler effectively blocked contractions. Left ventricular diastolic and systolic pressures were measured, the heart was allowed to reach a stable steady state, and then various concentrations of DAM or cyto-D were added to the perfusate to arrest contractions. The minimum concentrations of DAM and cyto-D that reliably reduced developed pressure by >90% were 15 mM and 5 μM, respectively, after 2–5 min of continuous perfusion. Left peak systolic, end-diastolic pressures, and maximum and minimum first derivative of left ventricular pressure (dP/dt) were measured with a Digi-Med heart performance analyzer.
Hearts were stained with the voltage-sensitive dye 4-[β-[2-(di-n-butylamino)-6-naphthyl]vinyl]pyridinium (di-4-ANEPPS, 10.0 μl of 1 mg/ml dissolved in DMSO) or the calcium indicator dye rhod-2 (0.2 mg in 0.2 ml DMSO). The dyes were delivered as a single one-time bolus through a port in the bubble trap, which served as a compliance chamber and was located proximal to the aortic cannula. [Ca2+]i was calibrated as previously described (11). Briefly, rhod-2 exhibits a >100-fold increase in fluorescence emission (F) at 585 nm upon binding to Ca2+ with excitation wavelength = 530 ± 20 nm. Fmax was determined by adding 100 μM 2,2′-dithiodipyridine and 10 μM A23187 to the port in the bubble trap in perfusate containing 5 mM Ca2+, resulting in a peak F in 2 min. A rapid saturation of [Ca2+]i was achieved because 2,2′-dithiodipyridine elicited rapid release of Ca2+ from the sarcoplasmic reticulum by oxidizing critical sulfhydryl-activating ryanodine receptors, whereas A23187 facilitated Ca2+ entry in heart cells (32). Fmin was then determined by perfusing the hearts with perfusate containing 5 mM EGTA for 20–30 min. The absolute fluorescence intensity recorded during a cardiac AP or a Ca2+ transient is dependent on the optical apparatus, the depth of staining, and the physiological condition of the heart; under the current experimental conditions, the mouse ventricular AP upstroke and Ca2+ transient produced fractional fluorescence change of 10–12% and 30–35%, respectively.
The subcellular distribution of rhod-2 was examined by confocal microscopy in perfused mouse hearts loaded with rhod-2 to discriminate between mitochondrial versus cytosolic loading. The perfused heart was mounted horizontally in a chamber with a Sylgard bottom carved in the shape of the heart and a 3-mm diameter glass window (0.2 mm thick) on the bottom. The chamber was placed on the stage of the inverted microscope such that the objective viewed the left ventricular epicardium. To acquire confocal images, KCl (20 mM) was added to the perfusate to arrest the heart. An argon laser excited the epicardial surface, and fluorescence was collected through a confocal aperture by a photomultiplier under manual gain and black-level control. Confocal images were recorded from hearts loaded with rhod-2 and then after perfusion with 20 μM digitonin and 2 μM free Ca2+ to permeabilize cell membranes and release rhod-2 trapped in the cytosol while retaining rhod-2 trapped in mitochondria and other subcellular organelles. (15)
Optical apparatus, computer interface, and analysis.
The optical apparatus, computer interface, and analysis of APs have been previously described (3). Briefly, light from a 100-W tungsten-halogen lamp was collimated, passed through a 530 ± 20-nm interference filter, and focused on the epicardial surface of the left ventricle of the mouse heart. The fluorescence from dye bound to the heart was passed through a cut-off filter (>610 nm) for di-4-ANEPPs or an interference filter (585 ± 20 nm) for rhod-2. An image of the heart was focused on a 12 × 12-element photodiode array of which 124 diodes were simultaneously monitored. Each diode recorded the summed electrical activity from a 312 × 312-μm2 region of the ventricle with a depth of 70 μm. Image magnification was ×4.5, and a 4 × 4-mm2 tissue area was viewed by the array. The photocurrent from each diode was passed through a current-to-voltage converter (50 MΩ feedback resistor), AC or DC coupled, amplified (1, 50, 200, or 100×), digitized at 2,000 frames/s at a 12-bit resolution (DAP 3200e/214 Microstar Laboratories), and stored in computer memory.
APDs were determined from measurements of the time point of maximum upstroke velocity (dF/dt)max minus the time point at which the downstroke recovered to 75% back to baseline, i.e., APD75. APs with signal-to-noise ratios of <10 or excessive movement artifact were not included in the analysis. Conduction velocities were calculated as previously described (36). The duration of [Ca2+]i transients was determined from the maximum first derivative of the [Ca2+]i upstroke to the time point of 75% recovery of [Ca2+]i to its original baseline. The rise time of the Ca2+ transient was taken as the time to peak from the minimum to the maximum of the [Ca2+]i upstroke (3). [Ca2+]i was calibrated from hearts loaded with rhod-2 AM using the equation: [Ca2+] = Kd·[(F − Fmin)/(Fmax − F)], where the dissociation constant (Kd) is 710 nM, Fmin is the rhod-2 fluorescence when all the dye is in the free form or in zero Ca2+, and Fmax is the rhod-2 fluorescence when all the dye is bound to Ca2+ or in a Ca2+-saturated solution, as previously described (11).
APD was characterized as a function of basic cycle length using basic pacing at S1-S1 intervals of 40–200 ms, in increments of 20 ms. On the basis of the measurements of spatial dispersion of repolarization, single premature stimuli were delivered at decreasing coupling intervals at the base or apex of the heart. The heart was paced at a basic interval or cycle length (e.g., S1-S1 = 200 ms) for 10 beats to obtain a stable APD. Every tenth beat, an extra impulse S2 was applied to interrupt the basic cycle length. The S1-S2 interval was gradually decreased in 1- or 2-ms steps (particularly in the steep zone of the restitution curve) until S2 failed to capture an AP. The refractory period at that site was defined as the shortest S1-S2 interval that elicited a propagating AP. Ventricular arrhythmias in the murine heart were induced by applying a train of stimuli, burst pacing. Burst pacing consisted of 10 electrical impulses, 1 ms in duration, with 15-ms interpulse interval at three times threshold voltage.
Data are presented as means ± SE and changes in APDs recorded under different conditions were compared by paired or unpaired Student’s t-test as appropriate. The results were considered significant for P < 0.05.
Effects of DAM and cyto-D on mouse AP.
A digital picture of a perfused mouse heart is shown mounted in a chamber to abate movement artifacts without chemical uncouplers, with examples of four simultaneously recorded APs from different sites on the heart (Fig. 1A). The left epicardium faces the optical apparatus, and a silhouette of the array is superimposed on the heart to identify the region of optical recordings. In Fig. 1, B–D, the silhouette of the array is illustrated with the AP recorded by each diode drawn in its respective location from hearts perfused with control solutions (Fig. 1B), 15 mM DAM (Fig. 1C), or 5 μM cyto-D (Fig. 1D). Below each panel, a trace of AP recordings from a diode at the center of the field of view is shown at a fast sweep speed. Upon the addition of the chemical uncouplers, there was an immediate (within a minute) and marked prolongation of APDs with APD at 75% recovery to baseline (APD75) increasing from 20.0 ± 3 ms in controls (n = 8) to 46.6 ± 5 ms in DAM (n = 4) and 39.9 ms in cyto-D (n = 4) (Table 1). Both uncouplers increased the refractory periods of the epicardial APs, which would in principle increase the wavelength of reentrant circuits (Table 1). The chemical uncouplers caused marked changes in the shape and time course of APs. In particular, note the spike and dome appearance of APs in hearts treated with cyto-D.
In control mouse hearts, APDs are shorter at the apex than at the base of the left ventricle, producing a gradient of repolarization of 10.4 ± 4.1 ms (3). Gradients of APDs were within experimental error similar in controls and hearts treated with chemical uncouplers. In DAM and cyto-D, APDs increased progressively in going from apex to base with APD75 increasing from 63 ± 4 to 69 ± 5 ms with DAM and from 45 ± 5 to 50 ± 8 ms with cyto-D.
Effects of uncouplers on restitution kinetics and conduction velocity.
Abrupt changes in heart rate or the firing of a premature impulse can produce dynamic heterogeneities of AP amplitudes (APA) and durations. In mouse hearts, APDs are considerably shorter than in other mammalian hearts, show only a slight variation as a function of rate (Fig. 2A), and tend to have flat APD restitution curves. In the presence of DAM and cyto-D, APDs varied steeply as a function of cycle length for long cycle lengths; however, for short cycle lengths that are 10–15 ms above the refractory period, the curve remained flat and close to zero (Fig. 2A). In Fig. 2B, the restitution kinetics curve of the AP amplitude is compared before and after the addition of DAM or cyto-D by plotting the APA as a function of S1-S2 intervals. With cyto-D and DAM, the shortest possible S1-S2 intervals were considerably longer than in controls because the uncouplers increased APDs and refractory periods. As a result, the steep phases of the restitution kinetics curves at S1-S2 <75 ms were abolished. From the analysis of AP recordings from eight sites per heart, DAM and cyto-D increased the refractory periods from 45.14 ± 2.1 (n = 8 hearts) to 82.5 ± 3.5 (n = 4) and 78 ± 4.24 ms (n = 4), respectively. At the longer S1-S2 intervals (>75 ms), the chemical uncouplers had a slightly steeper restitution kinetics curves compared with controls. Cyto-D reduced conduction velocity from 0.55 ± 0.03 m/s in controls to 0.47 ± 0.08 m/s (n = 4), which was statistically significant (P < 0.01, ANOVA), and there was a tendency by DAM to reduce velocity from 0.58 ± 0.06 to 0.54 ± 0.04 m/s DAM (n = 5) that did not reach statistical significance (Fig. 2C).
Effects of uncouplers on [Ca2+]i transients.
Hearts were loaded with rhod-2/AM, and the left ventricles were imaged on the array to record [Ca2+]i transients from multiple sites. A symbolic map of the array and the [Ca2+]i transients recorded by each diode are shown in their respective locations (Fig. 3A). The shape and time course of [Ca2+]i transients are shown for four diodes at faster sweep speeds (Fig. 3B) from a heart paced at 200 ms cycle length. Figure 3C compares Ca2+ transients from a control heart and in the presence of DAM or cyto-D where all three signals were calibrated in terms of free cytosolic Ca2+, as described in methods (Fig. 3C). DAM decreased diastolic and systolic [Ca2+]i, whereas cyto-D increased both (Table 2). DAM produced a statistically significant prolongation of the duration of [Ca2+]i transients, whereas cyto-D had negligible effects (Table 2).
Distribution of rhod-2 in mouse ventricular myocytes.
The intracellular distribution of rhod-2 in perfused murine hearts was examined by confocal microscopy from hearts loaded with rhod-2 AM using identical conditions that were used to load the hearts with rhod-2 to map [Ca2+]i transients. Confocal images of the rhod-2 distribution in a cell on the epicardium of the perfused heart revealed a pattern of cytosolic milieu without the punctate appearance of mitochondrial loading (Fig. 4A, n = 4). The loading procedure resulted in a similar distribution of dye throughout the epicardium with no apparent hot spots of high rhod-2 fluorescence, which would be expected if the dye was accumulated in subcellular organelles with higher concentrations of hydrolyzed dye and/or high [Ca2+]i (i.e., in the sarcoplasmic reticulum network or the mitochondria). To verify that rhod-2 was not trapped in a subcellular compartment, the heart was perfused with digitonin for 10 min to increase the permeability of the cellular plasma membrane cells without compromising the integrity of subcellular organelles. Low Ca2+ (2 μM) was used in the perfusate containing digitonin to maintain a high level of rhod-2 fluorescence and a normal mitochondrial potential (15). As shown in Fig. 4B, perfusion with digitonin resulted in an extensive loss of rhod-2 from all regions of the cells with similar observations made throughout the epicardium, indicating that the dye was not trapped in subcellular compartments (n = 4 hearts).
Antiarrhythmic actions of uncouplers.
While the chemical uncouplers prolonged APDs and refractory periods and eliminated the steep phase of restitution kinetic curves without significantly changing gradients of refractoriness, they decreased conduction velocity. Because these changes in the myocardial substrate can potentially reduce the vulnerability to arrhythmias, we tested for the propensity to arrhythmias by applying burst stimulation in attempts to elicit ventricular tachycardia (VT). The incidence of arrhythmias in control hearts was compared with that in hearts perfused with a chemical uncoupler. In most control hearts (n = 8/10), one or two bursts (10 pulses/burst) were sufficient to elicit an immediate monomorphic VT. In a few hearts, several bursts were required to elicit VT (n = 2/10). Most VTs were long lasting, 30 s to an hour (n = 6/10 hearts) (Fig. 5A) or spontaneously returned to sinus rhythm in 15–20 min (n = 4/10). In contrast, burst pacing of isolated hearts perfused with DAM (Fig. 5B) or cyto-D (Fig. 5C) triggered the firing of APs, but the hearts became quiescent immediately after the end of the burst (n = 5 for cyto-D and n = 5 for DAM). In control hearts, patterns of activation during VTs were highly reproducible exhibiting stable frequencies ranging from 10 to 19 Hz, (n = 6) for the duration of the arrhythmia (Fig. 6).
Optical mapping has become an established technique to investigate mechanisms underlying cardiac arrhythmias, metabolic diseases, and the pathology of heart failure by providing accurate recordings of cardiac APs and [Ca2+]i transients at high spatial and temporal resolution. However, muscle contractions produce movement artifacts that can distort optical signals, raising concerns regarding their validity, particularly of the recovery phase of APs and [Ca2+]i transients. Mechanical immobilization of the heart has been extensively used to abate movement artifacts relative to the voltage-dependent component of the optical signal (18, 37). Another approach has been to block force and movement artifacts by reducing [Ca2+]i with Ca2+ free Tyrode solution (38) or blocking L-type Ca2+ channels (16), which alters the substrate being investigated. Alternatively, pharmacological agents were used as chemical uncouplers of contractions, which ideally eliminate force generation without affecting the shape, time course, or propagation of the cardiac AP or of [Ca2+]i transients. DAM and cyto-D have been extensively used as chemical uncouplers in cardiomyocytes, because they interfere with the contractile apparatus at the level of the myofibrils presumably without significantly changing [Ca2+]i (5, 14) Unfortunately, these uncouplers are far from ideal and are found to alter the myocardial substrate through changes intracellular Ca2+ handling, ion channels, and AP characteristics in a species-dependent manner.
Multiple effects of DAM.
DAM was first introduced as a nucleophilic agent with phophatase-like activity that acts by removing the phosphate groups required for protein activation. Its phosphatase activity appears to be nonspecific having numerous effects at the cellular level by altering the properties of ion channels, gap junctions, and other intracellular processes regulated by phosphorylation. In cat ventricular muscle, DAM was found to inhibit contractions and depress the plateau phase of the AP (43). DAM has been shown to alter delayed rectifier K+ current and L-type Ca2+ current in certain species. DAM prolongs APD in mouse and rat hearts but shortens APD in rabbit and guinea pig hearts, yet depresses contraction in all these species. Because DAM has opposite effects on APDs in different species, the depression of contractions cannot be explained simply by an inhibition of L-type Ca2+ currents. It has been suggested that the lengthening effect of DAM on APD results mainly from the simultaneous reduction of both the slow inward calcium current (Iso) and the transient outward current (Ito), two antagonistic currents with unequal influences on AP plateau development.(13) In guinea pig myocytes, DAM reduced voltage-gated Ca2+ currents and the inward and delayed rectifying K+ currents (28).
In rat ventricular myocytes, DAM produced a rapid, dose-dependent, and reversible blockade of gap junctional conductance (4) that potentially reduced conduction velocity in heart muscle. Interestingly, DAM reduced gap junction conductance in neonatal rat cardiomyocytes without changing the phosphorylation state of connexin 43, the main gap junctional protein in rat hearts (17). The reduced conductance of connexin 43 without a change in protein phosphorylation state suggested that DAM acted at associated regulatory proteins that determine the functional state of gap junctions (17) or acted by an entirely different mechanism. Our current results on mouse hearts are consistent with these previous observations because DAM tended to decrease conduction velocity in mouse ventricles and prolonged APDs, consistent with observations in rat heart APs.
DAM was also shown to reduce the L-type Ca2+ current in rabbit, rat, and guinea pig ventricular myocytes. In rat heart trabeculae, DAM at 10–20 mM decreased peak systolic [Ca2+]i with no significant changes in the time course of [Ca2+]i transients (1). However, in murine hearts, we found a substantial prolongation in the duration of [Ca2+]i transients induced by DAM. Others have shown an increased sensitivity of mouse hearts to changes in extracellular Ca2+ concentration within the physiological range compared with the rat (7), which could account for the different effects of DAM on [Ca2+]i transient in mice and rats. In the current experiments, the rise time of [Ca2+]i transients measured at 200-ms cycle length were increased by DAM from 14.1 ± 1.2 to 25.0 ± 3.31 ms (n = 5). It is interesting to note that the rise times of [Ca2+]i in mouse hearts (14.1 ± 1.2 ms) were considerably shorter than in guinea pig hearts (25.65 ± 5 ms) (11).
Actions of cyto-D.
The exact molecular basis for cardiac contraction failure induced by cyto-D remains unknown. Cytochalasins selectively bind to rapidly polymerizing and depolymerizing actin filaments and disrupt F-actin of the cytoskeleton by binding to the net polymerizing end of actin filament. Cyto-D has been the agent of choice to disrupt cytoskeletal actin, but in skinned rat myocytes, cyto-D interacts with sarcomeric actin and shifts the force versus pCa curve to lower pCa values (8). The direct interaction of cyto-D with sarcomeric actin can account for the inhibition of cardiac contractions, and the interaction with cytoskeletal actin may account for the various effects on ion channels. Other experiments suggest that cyto-D (10 μM) does not directly bind to cytoskeletal actin but reduces the activation of an actin-depolymerizing factor (cofilin) that binds to actin, leading to the depolymerization of F-actin and the subsequent reduction of the Ca2+ current (ICa-L) (35). Similarly, cyto-D was found to reduce the Na+ current through a decrease in open probability and perhaps by slowing the inactivation rate (40). The interaction of cyto-D (10 μM) with the cytoskeleton of guinea pig ventricular myocytes has been implicated as the mechanism for reducing the inward and delayed rectifying K+ currents (31) and for the acceleration of the rundown of ATP-sensitive K+ channels (19). Jalife et al. (22) recorded optical APs from murine hearts and reported a prolongation of APDs by cyto-D and at higher concentrations of cyto-D (80 μM) reported a “hump” on the plateau phase. In conclusion, the interplay of all these effects of cyto-D on ion channels must be considered to explain the changes in AP and [Ca2+]i transients that are elicited by cyto-D.
Restitution kinetics and ventricular arrhythmias.
The restitution kinetics of APs were measured by pacing the heart at a basic cycle length (S1-S1 = 200 ms) for 10 beats and then applying a premature stimulus at varying S1-S2 intervals. The short duration of the mouse heart AP made it difficult to detect small decreases in APDs of the premature beats, particularly during the steep slope of the restitution curve. We therefore plotted the restitution of APA that depends on the restitution of inward currents (INa and ICa) and indirectly on K+ repolarizing currents. We have previously shown that APA decreases with decreasing S1-S2 interval in mice (3).
DAM and cyto-D prolonged APDs in mouse ventricles as well as the APD versus cycle length and AP amplitude restitution curve. The changes in APD as a function of cycle length are most likely mediated by rate-dependent changes in Ca2+ and K+ conductance (10). DAM and cyto-D caused marked changes in the restitution curve of AP amplitude; both increased refractoriness and made the curve steeper at long cycle lengths and flatter at short cycle lengths. Their effects on refractoriness may be due to their actions on both depolarizing and repolarizing currents, which in the mouse are dominated by voltage-gated sodium and calcium channels (INa and ICa-L) and by the transient outward currents (Ito,f and Ito,s) more related to the delayed rectifier potassium current as previously described in other species.
Electrical restitution has been suggested to play an important role in the initiation and maintenance of arrhythmias (24). The rationale behind the restitution kinetics hypothesis is that a dynamic change in APD (i.e., long to short APD) causes the subsequent wavefront to encounter refractory myocardium resulting in unidirectional conduction block and wave breakup, which promotes the initiation and maintenance of VF (24). Theoretical and experimental studies have proposed that the slope of the restitution curve can serve as an index of vulnerability to arrhythmia (9, 20, 25, 33, 34). If the slope is >1, a small perturbation in diastolic interval produces a larger change in APD, which becomes amplified upon iteration. Eventually, diastolic interval reaches a value shorter than the refractory period resulting in local conduction block, wave break, and turbulence.
Hearts treated with cyto-D or DAM had flatter restitution curves and had a reduced vulnerability to arrhythmias, which is consistent with the restitution kinetics hypothesis. Indeed, burst pacing consistently elicited ventricular arrhythmias in control mice, but in hearts treated with an uncoupler, all attempts to induce an arrhythmia failed (n = 5/5 for DAM and n = 5/5 for DAM). The antiarrhythmic actions of chemical uncouplers in murine hearts could also be due to the lengthening of reentrant wavelengths as a result of prolonged APDs and slower conduction velocity. In swine hearts, Lee et al. (26) showed that DAM converts ventricular fibrillation (VF) to tachycardia, whereas cyto-D did not alter the organization or dynamics of fibrillation.
Can mouse ventricles fibrillate?
Vaidya et al. (41) challenged the critical mass hypothesis for fibrillation by showing that VT (n = 4/5) and VF (n = 4/7) could be induced by burst pacing the small heart of a mouse. In contrast, the present experiments failed to elicit VF even in mice that were not treated with a chemical uncoupler. Here, VTs were readily obtained by burst pacing, were long lasting, and did not progress to VF. It should be noted, however, that the stimulation protocol used by Vaidya et al. was significantly different from the current protocol, because multiple bursts (20 stimuli each) of different cycle length (shorter than that needed for 1:1 capture), of various stimulation strengths, and at various location were needed to elicit VT or VF (41). In addition, VF was only observed in hearts that were not perfused with DAM, and the numbers of bursts needed to elicit VF and the duration of VF episodes was not reported (41).
Distribution of rhod-2 in mouse hearts.
This study validates the measurement of cytosolic [Ca2+]i in perfused mouse hearts with rhod-2. A criticism against the use of rhod-2 to measure cytosolic [Ca2+]i is that the dye has a positive charge, which results in dye accumulation in the mitochondria (due to their negative potential inside −120 to −180 mV). The mitochondrial content could raise the background fluorescence and give errors in diastolic and systolic [Ca2+]i. We have previously shown in perfused rabbit hearts that rhod-2 can be used to selectively measure intracellular cytosolic calcium with negligible contributions from dye loaded in mitochondria, sarcoplasmic reticulum, or nuclei. In addition, rhod-2 loaded in endothelial cells was minimal, making it the best dye to measure cytosolic [Ca2+]i transients in intact hearts (15). We further examined the distribution of rhod-2 in isolated guinea pig hearts and showed that the dye was remarkably selective for the cytosol and did not load in mitochondria compared with fura-2 and fluo-3 (15). Here, we examined the distribution of rhod-2 in mouse hearts by loading perfused hearts with rhod-2, immobilizing the heart with 20 mM KCl, and placing the immobilized heart on a confocal microscope. As shown in Fig. 4A, epicardial cells from the perfused heart were effectively and rapidly loaded with rhod-2 within 5–10 min after passing a bolus of dye with the coronary perfusate. Rhod-2 fluorescence did not exhibit a punctate appearance of mitochondrial loading that is typically seen with tetramethylrhodamine ethyl ester (a voltage-sensitive dye used to measure mitochondrial potential) (15). The subsequent perfusion of the same hearts with digitonin revealed that rhod-2 became freely permeable across the plasma membrane and within 5–10 min diffused out of the cells and was washed out by the perfusion. No significant dye was found trapped in subcellular organelles (Fig. 4B). Similar results were obtained with confocal images of mouse, guinea pig, and rabbit myocytes. Thus reports of mitochondrial dye accumulation appear to be highly dependent on staining conditions (temperature and time).
Reversibility of chemical uncouplers.
Most studies observed an effective recovery of contractions after washing out DAM but not after washing out cyto-D. In mouse hearts perfused with DAM or cyto-D (at concentrations that decreased left ventricular pressure by ≥90%), the subsequent washout of the uncouplers partially reversed the suppression of developed pressure by 50% of the original pressure with cyto-D and 70% with DAM. Other studies in mouse myocytes and canine trabeculae showed that the block of force by cyto-D could not be reversed by extensive washing (5, 22). Our partial recovery of force after treatment with cyto-D in the present experiments was most likely due to the lower concentrations used to block contractions.
In summary, DAM and cyto-D inhibit contraction but are far from being ideal uncouplers and should be used with caution in studies of arrhythmia mechanisms because of marked effects on APs, [Ca2+]i transients, APDs, conduction velocity, and refractoriness.
This study was supported by National Heart, Lung, and Blood Institute Grants HL-59614, HL-57929, and HL-69097 (to G. Salama), HL-70250 (to B. London), and Postdoctoral Fellowships from the Western Pennsylvania Affiliate of the American Heart Association (to L. Baker, R. Wolk, A. Shah, and B.-R. Choi).
The authors thank William Hughes of our departmental machine shop for the construction of optical components and the mouse chamber and Greg J. Szekeres of our departmental electronic shop for building the computer interface.
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.
- Copyright © 2004 by the American Physiological Society