Hypersensitivity of excitation-contraction coupling in dystrophic cardiomyocytes

Abstract

Duchenne muscular dystrophy represents a severe inherited disease of striated muscle. It is caused by a mutation of the dystrophin gene and characterized by a progressive loss of skeletal muscle function. Most patients also develop a dystrophic cardiomyopathy, resulting in dilated hypertrophy and heart failure, but the cellular mechanisms leading to the deterioration of cardiac function remain elusive. In the present study, we tested whether defective excitation-contraction (E-C) coupling contributes to impaired cardiac performance. “E-C coupling gain” was determined in cardiomyocytes from control and dystrophin-deficient mdx mice. To this end, L-type Ca²⁺ currents (I_CaL) were measured with the whole cell patch-clamp technique, whereas Ca²⁺ transients were simultaneously recorded with confocal imaging of fluo-3. Initial findings indicated subtle changes of E-C coupling in mdx cells despite matched Ca²⁺ loading of the sarcoplasmic reticulum (SR). However, lowering the extracellular Ca²⁺ concentration, a maneuver used to unmask latent E-C coupling problems, was surprisingly much better tolerated by mdx myocytes, suggesting a hypersensitive E-C coupling mechanism. Challenging the SR Ca²⁺ release by slow elevations of the intracellular Ca²⁺ concentration resulted in Ca²⁺ oscillations after a much shorter delay in mdx cells. This is consistent with an enhanced Ca²⁺ sensitivity of the SR Ca²⁺-release channels [ryanodine receptors (RyRs)]. The hypersensitivity could be normalized by the introduction of reducing agents, indicating that the elevated cellular ROS generation in dystrophy underlies the abnormal RyR sensitivity and hypersensitive E-C coupling. Our data suggest that in dystrophin-deficient cardiomyocytes, E-C coupling is altered due to potentially arrhythmogenic changes in the Ca²⁺ sensitivity of redox-modified RyRs.

duchenne muscular dystrophy (DMD) belongs to the large group of dystrophinopathies, in which cytoskeletal abnormalities and remodeling may lead to heart failure (18, 36). DMD represents the most severe genetic muscular disorder affecting striated muscle. It results from mutations in the dystrophin gene, which encodes for the cytoskeletal protein dystrophin (8). At the cellular level, dystrophin forms a molecular bridge between the F-actin of the cytoskeleton and the extracellular matrix via the dystroglycan protein complex, thus providing mechanical stability to the myocyte. Lack of dystrophin expression disrupts the dystroglycan complex and is thought to confer instability to the cytoskeleton and plasma membrane with a severe impact on cellular signaling. The disease affects 1 in 3,000 boys with the onset of complications in early childhood. Disease symptoms ultimately lead to premature death, mostly due to respiratory failure. Progressive skeletal muscle wasting in DMD is also accompanied by the development of cardiac abnormalities, such as fibrosis with wall motion malfunctions, left ventricular dysfunction, and ventricular arrhythmias, leading to dilated cardiomyopathy and cardiac failure (11). Since therapy progress at the level of skeletal muscle wasting has significantly improved the life expectancy of DMD patients over the years, dystrophic cardiomyopathy becomes more and more a limiting factor in the (symptomatic) treatment of DMD patients.

An established animal model for studying the disease phenotype of DMD is the mdx mouse, where, in analogy to DMD mutations, a point mutation in the dystrophin gene leads to an untranslatable gene (16, 21). Similar to humans, mdx mice show an age-linked disease onset of cardiomyopathy resulting in cardiac dilation, a progressive decrease in fractional shortening, replacement of the functional myocardium by fibrotic tissue, and, thus, conduction defects (5, 31).

A previous study (41) of the mdx mouse showed substantial changes in Ca²⁺ signaling with only slightly elevated levels of resting intracellular Ca²⁺ concentration ([Ca²⁺]_i) in ventricular myocytes. In addition, dystrophic myocytes exhibit an increased susceptibility to mechanical stress-induced Ca²⁺ influx across the sarcolemma, possibly via microruptures, stretch-activated channels, and other pathways as a consequence of the disrupted dystroglycan complex (9, 20, 41, 45). This allows for excessive Ca²⁺ influx via voltage-independent pathways during mechanical stress, which is then synergistically amplified by several cellular mechanisms to generate excessive cytosolic and mitochondrial Ca²⁺ transients and irreversible loss of mitochondrial membrane potential, an early indication of cell death (20). Excessive production of ROS in these cells was identified as a mechanism contributing to this disproportionate Ca²⁺ signal amplification (20, 42). Since ryanodine receptors (RyRs) become more Ca²⁺ sensitive after oxidation (30), the simultaneous activation of Ca²⁺-induced Ca²⁺ release (CICR) and ROS production in dystrophic myocytes could be the mechanism leading to cellular damage. Functional incapability of precise intracellular Ca²⁺ regulation might be an early indicator of changes in Ca²⁺ sensitivity and may have a large impact on excitation-contraction (E-C) coupling and cell survival in the long run.

The goal of the present study was to investigate impairments of E-C coupling in dystrophic cardiomyopathy. In progressive cardiomyopathies, disturbance of the E-C coupling mechanism often contributes to decreased cardiac performance. For example, in several animal models for heart failure, it has been shown that at similar L-type Ca²⁺ current (I_CaL) amplitude, there is a significant reduction in CICR, resulting in smaller Ca²⁺ transients and decreased myocyte contractility (14, 26, 35). A reduction in the ability of the Ca²⁺ current to trigger Ca²⁺ release from the sarcoplasmic reticulum (SR) can indicate a reduced sensitivity of the E-C coupling machinery toward Ca²⁺. Therefore, we expected alterations of E-C coupling in dystrophy to be similar to those observed in other forms of cardiomyopathy, with a reduced reliability of signal transmission between L-type Ca²⁺ channels and RyRs. Opposite to these expectations, our data indicated an abnormal ability of dystrophic myocytes to cope with experimental tests designed to challenge E-C coupling mechanisms, which points to a hypersensitivity of E-C coupling rather than to the reduced efficiency of E-C coupling found in other cardiac pathologies. Our results suggest that the hypersensitive E-C coupling results, at least in part, from an increased probability of RyRs to be activated by Ca²⁺, which may be the consequence of an imbalance in redox regulation in dystrophic myocytes. While such hypersensitivity may initially represent a beneficial mechanism to maintain E-C coupling in the disease, it may also have negative consequences, for example, by increasing the propensity for arrhythmias.

Some preliminary data of this study have been presented in abstract form (39).

MATERIALS AND METHODS

Cell isolation.

Adult C57BL10 mice [wild-type (WT) mice] at the age of 6–12 mo and age-matched dystrophin-deficient mdx mice (C57BL/10ScSn-mdx mice) were used in this study. At this age, mdx mice begin to develop a dilated cardiomyopathy but are not in overt heart failure (31). Animals were provided by Dr. M. Rüegg (University of Basel, Basel, Switzerland) and Dr. U. Rüegg (University of Geneva, Geneva, Switzerland) or were purchased from the Jackson Laboratory. Mouse hearts were rapidly excised after cervical dislocation, immersed in cold Ca²⁺-free saline buffer, and trimmed of excess tissue. Hearts were then cannulated and placed on a Langendorff perfusion apparatus. Enzymatic dissociation of ventricular myocytes was performed as previously described (43). Isolated cells were kept at room temperature until further use. All experiments conformed with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (1996) and were approved by the Institutional Animal Care and Use Committee of the New Jersey Medical School of the University of Medicine and Dentistry of New Jersey and by the State Veterinary Office of Bern, Switzerland.

Electrophysiological experiments.

Membrane currents were measured using whole cell patch-clamp procedures with an Axopatch 200B amplifier (Axon Instruments) controlled by custom-written data-acquisition software developed under LabView (National Instruments). Cells were voltage clamped using low-resistance (1.5–3 MΩ) borosilicate glass micropipettes. The pipette solution contained (in mM) 120 CsAsp, 8 NaCl, 20 tetraethylammonium(TEA)-Cl, 5 MgCl₂, 4 KATP, 5 HEPES, and 0.1 K₅-fluo-3 at pH 7.2 adjusted with CsOH. For experiments with reduced SR Ca²⁺ load, Na⁺ was omitted [to continuously favor Ca²⁺ extrusion via the Na⁺/Ca²⁺ exchanger (NCX)]. The external control solution contained (in mM) 140 NaCl, 5.4 KCl, 5 CsCl, 1.2 CaCl₂, 1.1 MgCl₂, 5 HEPES, and 10 glucose at pH 7.4 adjusted with NaOH. In some experiments, the external solution contained 1.8, 0.5, or 0.25 CaCl₂ instead of 1.2 mM CaCl₂.

Starting from a holding potential of −80 mV, the voltage protocol consisted of an initial step to −40 mV for 2 s. When corrected for the tip potential, this would correspond to a real potential of −52 mV, which inactivates voltage-dependent Na⁺ and T-type Ca²⁺ currents. This inactivation step was followed by 8 pulses of 50 ms to 0 mV at 1 Hz and then by a 400-ms depolarization to various test voltages. The initial preconditioning pulses were always applied in control solution and served to equalize the SR Ca²⁺ load before each test step, which was recorded at different extracellular Ca²⁺ concentrations ([Ca²⁺]_o). Subsequent to the loading pulses, the membrane voltage was kept at −40 mV for 4 s to allow for the complete exchange of extracellular solution by rapid local perfusion (half-life < 0.5 s). The full voltage protocol is shown in Fig. 1A.

Fig. 1.

L-type Ca²⁺ currents (I_CaL), Ca²⁺ transients, and excitation-contraction (E-C) coupling gain under control conditions [1.2 mM extracellular Ca²⁺ concentration ([Ca²⁺]_o)]. A: voltage protocol used for the experiments. The eight prepulses were always applied in control solution, whereas for the test step [Ca²⁺]_o was briefly varied according to the experimental challenge. B and C: representative traces for Ca²⁺ currents (bottom), line-scan images of Ca²⁺-related changes in fluorescence (ΔF/F₀; middle), and normalized cytosolic Ca²⁺ transients (top) elicited by a 400-ms test pulse to 0 mV in control solution in a wild-type (WT) cell (B) and a mdx cell (C). D: voltage dependence of current activation [current-voltage (I–V) curve, bottom traces] and corresponding Ca²⁺ transients (Ca²⁺-V curve, top traces) in WT (black; n = 8–21 cells) and mdx (red; n = 15–30 cells) myocytes. E: E-C coupling gain calculated from the data shown in D in WT (black) and mdx (red) myocytes at different test voltages.

Confocal microscopy.

Changes in [Ca²⁺]_i were simultaneously measured with ionic currents using the fluorescent Ca²⁺ indicator K₅-fluo-3 (100 μM, Biotium) and a laser-scanning confocal microscope (MicroRadiance, Bio-Rad). Fluo-3 was excited with the 488-nm line of an argon ion laser. The emitted light was collected above 500 nm. The acquisition of line-scan images along the cardiomyocytes was triggered by the LabView software 400 ms before the last loading pulse. Line-scan images were recorded for 6 s at a rate of 500 lines/s and captured the last prepulse, the period for solution exchange, and the test step at the end of the voltage protocol.

In experiments with intact ventricular cardiomyocytes (i.e., not dialyzed with a pipette filling solution), intracellular Ca²⁺ signals were recorded by full-frame image acquisition. Cells were loaded with fluo-3 by exposure to the membrane-permeant ester fluo-3 AM (5 μM, Biotium) for 30 min at room temperature followed by at least 10 min of deesterification. Images were acquired using a laser-scanning confocal microscope (MRC 1000, Bio-Rad). Fluo-3 was excited at 488 nm with a solid-state laser (Sapphire, Coherent), and fluorescence detection was performed at 540 ± 15 nm. Cells were continuously superfused with extracellular solution containing (in mM) 140 NaCl, 5.4 KCl, 1.8 CaCl₂, 1.1 MgCl₂, 5 HEPES, and 10 glucose at pH 7.4. Different test solutions contained 70 mM NaCl and 70 mM LiCl, or 140 mM LiCl instead of 140 mM NaCl. The osmolality was 310 mosM for all solutions.

Thapsigargin and ryanodine were from Alamone Laboratories, Mn-cpx3 was from Calbiochem, and mercaptopropionyl-glycine (MPG) was from Sigma-Aldrich. All experiments were performed at room temperature (22°C).

Data analysis.

Electrophysiological data were analyzed using IgorPro software (Wavemetrics). Current inactivation was fitted with a double-exponential function. Confocal images were initially analyzed using Image SXM [free software based on NIH Image (3)] and further processed by IgorPro. Fluorescence changes (ΔF/F₀) during Ca²⁺ transients in line-scan images were fitted with a sigmoidal function to determine peak amplitude and rise time.

Significance was tested by Student's t-test. Values of P < 0.05 were considered significant. Pooled data are given as means ± SE, with n representing the number of cells tested. In the figures and table, significance is indicated by *P < 0.05 and **P < 0.01.

RESULTS

I_CaL, cytosolic Ca²⁺ transients, and E-C coupling gain in WT and dystrophic cardiomyocytes.

During cardiac E-C coupling, depolarization-induced Ca²⁺ influx through L-type Ca²⁺ channels triggers Ca²⁺ release from the SR via CICR, which subsequently initiates the contraction of cardiac muscle cells.

Figure 1 shows the voltage protocol (A), representative traces of I_CaL, and corresponding cytosolic Ca²⁺ signals at peak activation [maximal current amplitude at 0 mV (I_max)] in WT (B) and mdx (C) myocytes. At similar trigger I_max, SR Ca²⁺ release reached comparable amplitudes in both WT (n = 20) and mdx myocytes (n = 16; see also Table 1). The voltage of test steps ranged from −25 to +50 mV. The current-voltage relationship and the associated voltage dependence of cytosolic Ca²⁺ transients under control conditions are shown in Fig. 1D.

Table 1.

Comparison of I_CaL and corresponding intracellular Ca²⁺ transients and E-C coupling gain in WT and mdx cardiomyocytes at different [Ca²⁺]_o and membrane voltages

	WT Cardiomyocytes				mdx Cardiomyocytes
	ICaL, pA/pF	ΔF/F0	Gain, (ΔF/F0)/ICaL	n	ICaL, pA/pF	ΔF/F0	Gain, (ΔF/F0)/ICaL	n
1.2 mM [Ca2+]o
0-mV test voltage	−5.36±0.09	3.36±0.24	0.69±0.07	20	−5.81±0.37	3.83±0.21	0.67±0.04	16
−25-mV test voltage	−0.40±0.06†	0.98±0.12	3.58±0.39	21	−0.83±0.12†	1.39±0.17	2.66±0.38	28
0.5 mM [Ca2+]o
0-mV test voltage	−2.97±0.30	2.61±0.54	0.91±0.20	12	−3.12±0.39	3.65±0.32	1.04±0.08	23
−25-mV test voltage	−0.39±0.07	0.46±0.13	1.98±0.34	15	−0.53±0.09	0.79±0.12	3.04±0.63	25
0.25 mM [Ca2+]o
0-mV test voltage	−1.94±0.37	1.20±0.59*	0.48±0.18*	5	−2.93±0.49	2.90±0.37*	1.04±0.11*	6
−25-mV test voltage	−0.37±0.07	0.27±0.06†	1.20±0.33*	7	−0.47±0.15	0.83±0.11†	3.20±0.60*	9

Values are means ± SE; n, no. of cells. I_CaL, maximum current; ΔF/F₀, Ca²⁺ transients; [Ca²⁺]_o, extracellular Ca²⁺ concentration. Significant differences between datasets of wild-type (WT) and mdx cardiomyocytes are indicated as

^*P < 0.05 and

^†P < 0.01. See the text for statistics for other comparisons.

Analysis of the E-C coupling gain provides an estimate of the coupling fidelity between L-type Ca²⁺ channels and RyRs. It is determined by normalizing the amplitude of the cytosolic Ca²⁺ transient to the corresponding amplitude of I_CaL at each given voltage. When performing this E-C coupling analysis, gain was comparable in dystrophic and WT myocytes, particularly at positive test voltages, as expected from the similarity in Ca²⁺ current amplitudes and Ca²⁺ transients (Fig. 1E). At negative membrane voltages, the E-C coupling gain was always largest, possibly due to higher coupling fidelity between the L-type Ca²⁺ channel and RyR and less redundant Ca²⁺ current (1). However, under control conditions, the gain was only insignificantly reduced in mdx cells at −25 mV (3.58 ± 0.39, n = 21, and 2.66 ± 0.38, n = 28, at −25 mV in WT and mdx cells, respectively).

Thus, in contrast to observations in several other cardiac pathologies, such as hypertrophy and heart failure (4, 14, 28), the efficiency of E-C coupling in dystrophic myocytes did not seem to be affected under control conditions.

I_CaL, cytosolic Ca²⁺ transients, and E-C coupling gain at threshold conditions.

At physiological [Ca²⁺]_o, the E-C coupling machinery is optimized to produce reliable E-C coupling in cardiomyocytes and may therefore be able to compensate for subtle problems resulting from a disease. To detect changes in the functionality of the involved key proteins, the system may be tested under conditions that challenge the efficiency of the E-C coupling mechanism. Lowering the [Ca²⁺]_o has previously been successfully used as an experimental tool to unmask subtle impairments of E-C coupling in a diseased heart (28). Therefore, we repeated the experiments shown in Fig. 1 at lower [Ca²⁺]_o, and the data are shown in Fig. 2.

Fig. 2.

I_CaL, Ca²⁺ transients, and E-C coupling gain challenged by low (0.5 mM) [Ca²⁺]_o. Please note that the prepulses to load the sarcoplasmic reticulum (SR) were done in control solution (1.2 mM [Ca²⁺]_o). A: voltage dependence of current (top lines) and corresponding Ca²⁺ transients (bottom lines) in WT (black) and mdx (red) cells. B: voltage dependence of the E-C coupling gain from the data shown in A in WT (black; n = 4–17 cells) and mdx (red; n = 9–25 cells) myocytes. C: statistical comparison of the E-C coupling gain in 1.2 and 0.5 mM Ca²⁺ in WT (black; n = 15 cells) and mdx (red; n = 25 cells) cells.

First, I_CaL and intracellular Ca²⁺ transients were simultaneously measured during test steps at 0.5 mM [Ca²⁺]_o. Please note that the SR Ca²⁺-loading procedure (as in Fig. 1A) was performed before each test pulse by applying the prepulses in control solution containing 1.2 mM [Ca²⁺]_o to assure a SR Ca²⁺ load equal to that under control conditions.

As expected upon reduction of [Ca²⁺]_o, Ca²⁺ current was reduced in both WT and dystrophic myocytes. The Ca²⁺ transients were also smaller. However, contrary to our expectations, mdx myocytes were much less sensitive to reductions of [Ca²⁺]_o than WT cells. The E-C coupling gain was calculated for each current-transient pair, and the averages are shown in Fig. 2B as a function of the test voltage. The phenomenon was most pronounced at negative test potentials but was consistently seen at all voltages tested. Whereas in WT cells the E-C coupling gain at −25 mV decreased by 45% in low [Ca²⁺]_o (to 1.98 ± 0.34, n = 15), it increased by 14% in mdx cells (to 3.04 ± 0.63, n = 25; Fig. 2, B and C and Table 1). In other words, in mdx cardiomyocytes, the SR Ca²⁺ release was less impaired by reductions of [Ca²⁺]_o, and thus the trigger signal for CICR, than in WT cells. Averaged amplitudes of I_CaL and Ca²⁺ transients at two different test voltages (0 and −25 mV) are shown in Table 1. Current-voltage relationship and the associated voltage dependence of cytosolic Ca²⁺ transients are shown in Fig. 2A.

Because of this surprising finding, we challenged the E-C coupling mechanism even more by further lowering [Ca²⁺]_o to near critical levels for functional coupling (to 0.25 mM). Figure 3 shows representative Ca²⁺ current traces, line scans of Ca²⁺-dependent changes in fluorescence, and line profiles of cytosolic Ca²⁺ transients at test voltages of 0 mV (A and B) and −25 mV (C and D) for WT and mdx myocytes. Table 1 shows averaged amplitudes of I_CaL and Ca²⁺ transients obtained in this group of experiments. At both test potentials, the amplitude of the Ca²⁺ transient was significantly larger in mdx cells than in WT cells in 0.25 mM [Ca²⁺]_o at similar trigger I_CaL.

Fig. 3.

I_CaL, Ca²⁺ transients, and E-C coupling gain in 0.25 mM [Ca²⁺]_o elicited with the same voltage protocol as shown in Fig. 1. A–D: representative traces of Ca²⁺ currents (bottom), line-scan images (middle), and normalized cytosolic Ca²⁺ transients (top) elicited by test pulses to 0 mV (A and B) and −25 mV (C and D) in WT (A and C) and mdx (B and D) myocytes. Insets in C and D show enlarged Ca²⁺ signals at −25 mV. E: statistical comparison of the E-C coupling gain in 1.2 mM (from Fig. 2) and 0.25 mM Ca²⁺ at −25 mV for WT (black; n = 7 cells) and mdx (red; n = 9 cells) cells. *P < 0.05.

The differences in CICR between WT and mdx cells are emphasized when the values at 0.25 mM [Ca²⁺]_o are compared with the corresponding controls at 1.2 mM [Ca²⁺]_o. Figure 3E shows the E-C coupling gain in WT and mdx myocytes at −25 mV in 0.25 mM [Ca²⁺]_o. Whereas in WT cells the gain decreased by 66% (n = 7) in 0.25 mM [Ca²⁺]_o, it increased by 20% (n = 9) in mdx myocytes compared with control [Ca²⁺]_o.

Thus, in contrast to WT cells, lowering of [Ca²⁺]_o unmasked an unusual resistance to reductions in triggering I_CaL in mdx myocytes and resulted in an abnormal increase in the E-C coupling gain. Again, this reflects a behavior that is opposite to the results from previous studies of altered E-C coupling in cardiac diseases, such as heart failure (4, 14, 28). This observation could, for example, be explained by a hypersensitivity of E-C coupling and RyRs toward Ca²⁺ triggers in mdx cardiomyocytes.

Initial support for this possibility comes from the analysis of the dependence of half-maximal SR Ca²⁺ release on [Ca²⁺]_o. We determined Ca²⁺ transient amplitudes during test steps to −25 mV in 1.8, 1.2, 0.5, and 0.25 mM [Ca²⁺]_o. The half-maximal Ca²⁺ release was observed at 0.69 mM [Ca²⁺]_o in WT, whereas it was shifted to lower Ca²⁺ concentrations (to 0.41 mM) in mdx myocytes.

RyR Ca²⁺ sensitivity.

To investigate whether mdx myocytes indeed exhibit higher sensitivity toward changes in [Ca²⁺]_i, we provoked moderate and slow Ca²⁺ influx into cardiomyocytes via the activation of NCX in its Ca²⁺ influx (reverse) mode. Lowering the Na⁺ concentration in the extracellular medium by substitution with Li⁺ forced the NCX into the Ca²⁺ influx mode. Figure 4, A and B, shows representative intracellular Ca²⁺ signals in WT and mdx cells in response to a stepwise reduction of external Na⁺ to 70 or 0 mM. Ca²⁺ influx via NCX triggered substantial and reversible Ca²⁺ oscillations (Fig. 4, A and B). Whereas at 70 mM extracellular Na⁺ the frequency of oscillations was still comparable in the two types of myocytes (0.27 ± 0.05 Hz, n = 19, and 0.29 ± 0.02 Hz, n = 9, in WT and mdx cells, respectively), it was significantly higher upon complete removal of Na⁺ in mdx myocytes (0.39 ± 0.04 Hz, n = 19, and 0.68 ± 0.08 Hz, n = 8; Fig. 4C). Equally important, Ca²⁺ oscillations at reduced Na⁺ occurred significantly earlier in mdx cells (after 23.1 ± 4.1 vs. 8.87 ± 1.69 s in 70 mM Na⁺ and 13.0 ± 2.1 vs. 5.1 ± 0.4 s in 0 mM Na⁺ in WT and mdx cells, respectively; Fig. 4D). While the higher frequency of Ca²⁺ oscillations can have several reasons, the shorter delay is a hallmark of increased RyR Ca²⁺ sensitivity (19).

Fig. 4.

Ca²⁺ oscillations in WT and mdx cells. A and B: representative cytosolic Ca²⁺ signals in a WT cardiomyocyte (A) and a mdx (B) cardiomyocyte upon rapid reduction to 70 mM Na⁺ or complete removal of Na⁺ from the external solution (0 mM Na⁺). C and D: averaged frequency of oscillations (C) and time delay to their initiation (D) in WT (solid bars; n = 19) and mdx (shaded bars; n = 8) cardiomyocytes. *P < 0.05; **P < 0.01.

To verify that the above observations are linked to a putatively increased Ca²⁺ sensitivity of RyRs in mdx myocytes, but not to changes in NCX activity itself, we estimated the influx of Ca²⁺ via NCX while eliminating SR function. The latter was achieved by the incubation of cells with thapsigargin (1 μM) and ryanodine (10 μM). Figure 5A shows a representative trace of slow changes in [Ca²⁺]_i elicited by the rapid removal of external Na⁺ in a WT cardiomyocyte. Figure 5B shows the results of similar experiments in all WT (n = 11) and mdx (n = 5) cells studied with this protocol. There were no significant differences detected in the NCX-elicited Ca²⁺ influx in WT and mdx cells. Ca²⁺ influx via NCX in 70 mM Na⁺ raised cytosolic Ca²⁺ levels by 0.37 ± 0.05 and 0.33 ± 0.18 ΔF/F₀ in WT and mdx cells, respectively. The increase was larger upon complete removal of Na⁺ but was not significantly different in the two groups. Cytosolic [Ca²⁺]_i rose by 1.65 ± 0.28 and 1.13 ± 0.71 ΔF/F₀ in WT and mdx myocytes. Taken together, these findings indicate an increased Ca²⁺ sensitivity of RyRs in mdx cells that may result in the dystrophy-related changes of E-C coupling.

Fig. 5.

Activity of the Na⁺/Ca²⁺ exchanger (NCX) in WT and mdx myocytes. A: Ca²⁺ influx into a WT cardiomyocyte upon rapid reduction to 70 mM Na⁺ or complete removal of Na⁺ from the external solution (0 mM Na⁺) with eliminated SR function (1 μM thapsigargin and 10 μM ryanodine). B: averaged intracellular Ca²⁺ responses to a stepwise reduction or removal of external Na⁺ in WT (solid bars; n = 11) and mdx (shaded bars; n = 5) cells.

Estimations of SR Ca²⁺ load.

When E-C coupling gain values are analyzed and compared among various experimental series or preparations, several confounding factors need to be considered. Importantly, the E-C coupling gain depends on the SR Ca²⁺ content (34). It has been shown that an elevated SR Ca²⁺ concentration ([Ca²⁺]_SR) increases the open probability of RyR channels at any given cytosolic Ca²⁺ level (17). Moreover, the increase is nonlinear and much more pronounced at high [Ca²⁺]_SR. This was the reason that we carried out most of our experiments at low SR Ca²⁺ loads to reduce interference resulting from small alterations of SR Ca²⁺ load. This is of particular relevance for the present experiments, because the SR Ca²⁺ content has been reported to be elevated in mdx cardiomyocytes (41). To minimize alterations of our gain estimates that could result from differences in SR Ca²⁺ load, we applied a voltage-clamp Ca²⁺ preloading protocol throughout the experiments (see Fig. 1A). Furthermore, to keep the SR Ca²⁺ content in the lower range, where the dependence of Ca²⁺ release on SR Ca²⁺ load is much less steep (34), we generally omitted Na⁺ from the pipette solution. This maneuver favors Ca²⁺ extrusion via NCX and reduced the SR Ca²⁺ load by ∼30%.

We compared SR Ca²⁺ content (and SR Ca²⁺ release) in WT and mdx cells under experimental conditions of low intracellular Na⁺ concentration ([Na⁺]_i) using several complementary approaches: analysis of caffeine-induced intracellular Ca²⁺ transients and NCX currents and analysis of the inactivation kinetics of I_CaL (Fig. 6). Figure 6A shows representative intracellular Ca²⁺ transients elicited by 10 mM caffeine in WT and mdx myocytes in patch-clamped cells at −40 mV. Figure 6B shows the averaged amplitudes of Ca²⁺ transients as estimated from caffeine-induced Ca²⁺ transients (2.88 ± 0.32 ΔF/F₀ in WT cells and 3.49 ± 0.33 ΔF/F₀ in mdx myocytes, respectively, n = 10 and 13). In the presence of 8 mM [Na⁺]_i, SR Ca²⁺ content increased to 4.61 ± 0.68 ΔF/F₀ in WT cells and 4.72 ± 0.38 ΔF/F₀ in mdx cells, respectively (n = 8 in both groups; not shown). Thus, SR Ca²⁺ content was not significantly different between WT and mdx cells at either [Na⁺]_i (P = 0.89 and P = 0.21 at 8 and 0 mM, respectively).

Fig. 6.

Estimation of luminal SR Ca²⁺ content. A: caffeine-induced cytosolic Ca²⁺ transients in a WT myocyte (black) and a mdx myocyte (red). Confocal line-scan images and average fluorescence signals are shown. B: averaged amplitudes of caffeine-induced Ca²⁺ transients in both types of cells. C: NCX current (top) and integrated NCX current (bottom) activated by caffeine-induced Ca²⁺ transients in a WT myocyte (left) and a mdx myocyte (right). D: averaged integrated NCX currents in WT (n = 6) and mdx (n = 8) myocytes. E: representative traces of I_CaL during a test step from −40 mV to maximal current at 0 mV under control conditions in WT (black line) and mdx (red line) cells. F: fast (τ₁) and slow (τ₂) time constants of current inactivation at 0 mV at different [Ca²⁺]_o in WT (black) and mdx (red) cells. At 1.2 mM [Ca²⁺]_o, τ₁ and τ₂ were 14.5 ± 1.1 and 78.9 ± 6.1 ms (n = 42) in WT cells and 18.8 ± 2.0 and 87.7 ± 5.5 ms (n = 57) in mdx cells, respectively.

Integrated NCX currents activated by caffeine-induced Ca²⁺ transients are more reliable estimates of SR Ca²⁺ content because they do not depend on the particular time course of the caffeine-induced Ca²⁺ release. These estimates were even more similar in both types of myocytes than those derived from fluorescence transients (Fig. 6, C and D). Figure 6C shows representative traces of NCX current in WT and mdx myocytes (top left and top right traces, respectively) and the corresponding current integrals (bottom left and bottom right traces). The averaged data of integrated NCX currents are shown in Fig. 6D. They were 0.41 ± 0.09 pC/pF (n = 6) in WT cells and 0.43 ± 0.07 pC/pF (n = 8) in mdx cells.

Yet another, although more indirect, way to estimate luminal Ca²⁺ content and SR Ca²⁺ release is to compare the inactivation kinetics of I_CaL. In cardiac myocytes, a massive release of Ca²⁺ from the SR prominently contributes to the inactivation of Ca²⁺ currents via a Ca²⁺-dependent mechanism (33). Membrane currents were measured during a voltage step from −40 to 0 mV. As before, the test depolarization was preceded by a series of preconditioning pulses to optimize SR Ca²⁺ load. Figure 6E shows representative traces of I_max in a WT cardiomyocyte and a mdx cardiomyocyte under control conditions (1.2 mM [Ca²⁺]_o). The fast and slow time constants of current inactivation were not significantly different in WT and mdx cells studied at various [Ca²⁺]_o (Fig. 6F).

Taken together, these results indicates that there were no significant differences in SR Ca²⁺ content between WT and mdx cells under our experimental conditions. Therefore, luminal Ca²⁺ is unlikely to explain the observed enhanced Ca²⁺ sensitivity of RyRs and hypersensitive E-C coupling in mdx myocytes.

E-C coupling gain in the presence of ROS scavengers.

There is increasing evidence of oxidative stress in the dystrophic heart, which is caused by an imbalance in the production of ROS and the cell's innate capability to inactivate them (20, 42). Permanent exposure of the cells to increased levels of ROS may severely damage several components of the cell, resulting in cell death. In a recent study (20), we found abnormally elevated ROS levels in dystrophic cardiomyocytes both at rest and after mechanical stress.

The RyR contains a number of free cysteine residues, which are prone to redox modifications. In the case of oxidation of the cytosolic environment, free and neighbored thiol groups may form disulfide bonds, thereby altering the secondary structure of the RyR and increasing its sensitivity to Ca²⁺ activation (10). We speculated that the increased intracellular levels of ROS observed previously may be responsible for the abnormal Ca²⁺ sensitivity of RyRs and consequently lead to unusually reliable E-C coupling in dystrophic myocytes. Therefore, we tested whether a reduction of cytosolic ROS levels would normalize E-C coupling in dystrophic myocytes. For these experiments, we used two approaches: 1) the addition of a synthetic compound (Mn-cpx3) that acts as a natural SOD mimetic, which helps to maintain a reducing environment within the cell; and 2) the addition of a general antioxidant (MPG), which directly interacts with the oxidized protein and reduces it.

Preincubation of mdx cells with either 10 μM Mn-cpx3 or 800 μM MPG for 30 min normalized SR Ca²⁺ release during E-C coupling under conditions of reduced [Ca²⁺]_o and eliminated the hypersensitivity of E-C coupling. Figure 7 shows the averaged E-C coupling gain in dystrophic cardiomyocytes preconditioned with Mn-cpx3 and MPG, respectively, and studied in 0.5 and 0.25 mM [Ca²⁺]_o. For better comparison, these data are shown together with some that are replotted from Figs. 2C and 3E. The E-C coupling gain of mdx myocytes pretreated with Mn-cpx3 or MPG was no longer significantly different from that in WT cells. Thus, reduction of intracellular levels of ROS and oxidative stress eliminated the hypersensitivity of RyRs toward Ca²⁺ and restored normal E-C coupling gain in dystrophic myocytes. Since RyRs are redox sensitive even at normal cytosolic redox potentials (10), treatment of WT myocytes with MPG also led to a reduction of E-C coupling gain, as expected.

Fig. 7.

Reduction of oxidative stress prevents the hypersensitivity of E-C coupling gain in mdx cardiomyocytes. E-C coupling gain data for WT and mdx are replotted from Figs. 2C and 3E for better comparison. Preincubation of mdx cells with 800 μM mercaptopropionyl-glycine (MPG; 30 min, n = 3) or with 10 μM Mn-cpx3 (30 min, n = 7) reduced the E-C coupling gain to levels similar to those in WT cells kept in control conditions (dashed lines). WT myocytes pretreated with 800 μM MPG also led to the expected reduction of the E-C coupling gain. *P < 0.05 and **P < 0.01.

DISCUSSION

Patients suffering from muscular dystrophy develop early cardiomyopathy with reduced cardiac force. Therefore, we expected that the contractility of dystrophic cardiomyocytes would be diminished. In particular, we anticipated that E-C coupling would be impaired. Surprisingly, our study revealed an extremely reliable E-C coupling mechanism in dystrophic cardiomyocytes, which is even more resistant to experimental challenges than in WT cells. This unique feature has not been previously observed in studies of other cardiac pathologies and may be due to an increased sensitivity of RyRs to Ca²⁺ resulting from redox modification of the channel proteins.

E-C coupling in dystrophic cardiomyocytes.

E-C coupling comprises the early steps that link electrical excitation of the cardiac muscle to mechanical contraction. The quality of the E-C coupling mechanism can be assessed by the E-C coupling gain, which is a measure for the effectiveness of a cardiomyocyte to amplify the Ca²⁺ influx via L-type Ca²⁺ channels by triggered Ca²⁺ release from the SR. Amplification of the cardiac Ca²⁺ signal is well controlled and not only depends on Ca²⁺ influx via L-type Ca²⁺ channels but also on the complex regulation of RyRs and accessory proteins as well as on SR Ca²⁺ content, ultrastructural features, and the metabolic state of the cell. Therefore, the amplification mechanisms are likely to be affected in various diseases. So far, only a few studies have specifically investigated this phenomenon by focusing on the actual E-C coupling mechanism. They found that the E-C coupling gain is significantly decreased in several animal models of heart failure or dilated cardiomyopathies (for examples, see Refs. 4, 12, 14, 26, 28, and 35). The reduction in E-C coupling gain could be, at least partially, responsible for the decreased cardiac contractility in these pathologies (but see Ref. 38).

In our present study, there was no obvious deficiency found when E-C coupling was analyzed under control conditions. Since E-C coupling presumably incorporates a high degree of reliability, this mechanism may have a large margin of safety, and small shifts in its sensitivity may not lead to obvious signaling disturbances. Thus, we needed experimental conditions to unmask even subtle alterations of E-C coupling. Our method to discover such changes was to challenge E-C coupling by reducing [Ca²⁺]_o. This procedure has already proven useful in a previous study by McCall et al. (28), who showed that reducing the trigger signal for CICR by lowering [Ca²⁺]_o can be used to reveal slight E-C coupling problems in heart failure. Whereas this test was well tolerated by control cells, in their study it lead to a pronounced and disproportionate decrease in contractile function in myocytes from failing hearts at similar trigger Ca²⁺ currents.

In our experiments, we used this method to reduce the safety margin for functional E-C coupling, so that the system does not operate above saturation any longer. With this approach, changes in the coupling between I_CaL and RyR Ca²⁺ release are easier to uncover. When compared with the study on heart failure mentioned above (28), the dystrophic cardiomyocytes examined here behaved in the exact opposite way. Acutely lowering [Ca²⁺]_o reduced Ca²⁺ release from the SR to a lesser extent in mdx cardiomyocytes than in WT cardiomyocytes. This resulted in an increase of E-C coupling gain, not only relative to that in WT cells but also compared with the gain at control [Ca²⁺]_o. At this point, it is important to mention again that all experiments have been performed at matching SR Ca²⁺ load. Using our voltage protocol, conditioning pulses were always applied in control [Ca²⁺]_o, and only the test step that directly followed the loading pulses was applied in different reduced [Ca²⁺]_o.

The discovery of Ca²⁺ hypersensitivity at reduced [Ca²⁺]_o does not mean that this hypersensitivity is not present under normal conditions. It may just be masked by the already redundant activation of RyRs during E-C coupling and is therefore best identified at smaller trigger I_CaL (1). Thus, the physiological or pathophysiological relevance of the hypersensitivity is not limited to depolarizations to −25 and 0 mV but present for the entire range of membrane potentials that are covered during the cardiac action potential (see Fig. 2B). Therefore, our data indicate a hypersensitive E-C coupling mechanism in dystrophic cardiomyocytes, which are more resistant to this specific experimental challenge.

Altered Ca²⁺ sensitivity in dystrophic cardiomyocytes.

Increased E-C coupling gain actually means that Ca²⁺ is more easily released from the SR at similar trigger I_CaL. Since the properties of I_CaL in mdx myocytes showed no difference compared with WT myocytes, the critical changes of E-C coupling seemed to have occurred on the intracellular level, downstream of I_CaL. Most likely, a change in RyR Ca²⁺ sensitivity has occurred. Several mechanisms are known to modify the actual or apparent Ca²⁺ sensitivity of RyRs. They include, but are not limited to, ultrastructural changes in the dyadic junction, changes in luminal Ca²⁺ concentration or [Ca²⁺]_i, oxidative stress, and/or nitrosative stress.

At the level of the microarchitecture, it is conceivable that loss of the cytoskeletal protein dystrophin causes disarrangements in the dyadic cleft and disruption of the coupling between the L-type Ca²⁺ channel and RyR. However, one would expect that such structural changes lead to a reduction in the apparent Ca²⁺ sensitivity of the RyR and to a decrease in the E-C coupling gain. Such loss of function has been reported for a cardiac myocyte hypertrophy model (13). Pressure overload of the cardiomyocytes resulted in the disruption of the cytoskeleton, a significant decrease in E-C coupling efficiency, and a reduction in cardiac output. In contrast, a significant gain in the sensitivity of E-C coupling was observed in dystrophic cardiomyocytes.

An increase in [Ca²⁺]_SR can also have profound consequences on RyR function and intracellular Ca²⁺ signaling (23), leading to elevated RyR Ca²⁺ sensitivity and even to spontaneous SR Ca²⁺-release events like Ca²⁺ waves (22). Potential alterations in the expression and function of several major SR-related proteins, such as RyRs, SR Ca²⁺ pumps, and calsequestrin, in mdx myocytes might modify SR Ca²⁺ refilling, Ca²⁺ leak, and, finally, SR Ca²⁺ content (27, 32, 41). Loss of dystrophin reduces the mechanical stability of the myocyte and may favor the formation of microruptures and activation of other Ca²⁺ influx pathways, thus leading to uncontrolled Ca²⁺ entry into the cells. The resulting slight elevation of resting [Ca²⁺]_i found in mdx cardiomyocytes could, in principle, also add to a putative SR Ca²⁺ overload (20, 41).

Our data also show that slow increases in cytosolic Ca²⁺ via NCX operating in its Ca²⁺ influx mode trigger Ca²⁺ oscillations significantly earlier in mdx cells than in WT cells. This rapid response might be an indicator for higher diastolic [Ca²⁺]_i, increased SR Ca²⁺ content, or abnormally high sensitivity of RyRs toward changes in [Ca²⁺]_i. However, minimally raised diastolic Ca²⁺ levels are unlikely to account for the observed differences in E-C coupling in patch-clamped cells, where diastolic [Ca²⁺]_i is also determined by the pipette solution and, thus, is similar in both WT and mdx myocytes. In addition, our voltage protocols were specifically designed to guarantee equal SR Ca²⁺ load before each test step. The repetitive application of preconditioning pulses in control [Ca²⁺]_o ensured similar SR loading conditions throughout all experiments and for the test step at various [Ca²⁺]_o. The latter was confirmed by estimates of SR Ca²⁺ content. A comparison of the caffeine-sensitive Ca²⁺ content of the SR revealed no significant differences, nor were there any differences in NCX activity in WT and dystrophic myocytes. Taken together, these observations suggest that the hypersensitivity of E-C coupling in mdx myocytes is unlikely to result from changes in [Ca²⁺]_i and luminal [Ca²⁺]_SR but rather is related to changes of RyR function.

Oxidative and nitrosative stress are known to have marked consequences on the activity of the macromolecular RyR complex. In particular, ROS/reactive nitrogen species have been described as potent modulators of RyR gating properties and Ca²⁺ sensitivity resulting in RyR hyperactivity and increased SR Ca²⁺ leak without the need of elevated SR Ca²⁺ load (2, 7, 15, 44). We and others (20, 42) have recently reported an increased basal ROS production in the dystrophic heart. In addition, we (20) have previously shown that ROS generated by NAD(P)H oxidase and mitochondria are causally involved in the excessive Ca²⁺ signaling response to mechanical stress observed in mdx cardiomyocytes. Recently, it was found that oxidative stress during overt heart failure can lead to RyR oxidation and elevated Ca²⁺ sensitivity, thus contributing to the elevated SR Ca²⁺ leak (37). Abnormal levels of ROS in dystrophic cardiomyocytes might therefore contribute to the enhanced Ca²⁺ sensitivity of RyRs and to the increased E-C coupling gain when the cardiomyopathy has not yet evolved into heart failure. Our experiments using reducing agents support the latter possibility. We (20) have previously shown that a number of different ROS reducing substances (MnTBAP, Mn-cpx3, tiron, and apocynin) effectively normalized the stress sensitivity of dystrophic cardiomyocytes. In the present study, we used Mn-cpx3, which acts as a SOD mimetic, and MPG, which directly reduces the proteins. Reducing the cytosolic ROS concentration and oxidative stress by these means eliminated the hypersensitive E-C coupling in mdx cells. Although through different mechanisms, both tested substances more or less led to the normalization of E-C coupling gain in dystrophic myocytes, suggesting that protein oxidation may indeed be involved. As expected, treatment of WT myocytes with the reducing agent MPG led to some reduction in E-C coupling gain, since the RyR open probability is also sensitive to changes in cytosolic redox potentials, even in normal animals (10).

Under normal and healthy conditions, RyR hypersensitivity by itself may not lead to any dramatic changes in E-C coupling. However, if we extrapolate our findings to conditions of mechanical stress and undue Ca²⁺ accumulation (due to microruptures and other voltage-independent Ca²⁺ influx pathways), as it is suspected for dystrophic myocytes, one can imagine that excess cytosolic Ca²⁺ may easily activate hypersensitive RyRs and initiate spontaneous, arrhythmogenic Ca²⁺ release from the SR–and even more so during β-adrenergic stimulation (29). Related to this notion, it is believed that in the presence of RyR hypersensitivity the combination of two (or more) proarrhythmogenic mechanisms dramatically increases the susceptibility for arrhythmias (40). In an animal model harboring a gain of function mutation in the cardiac RyR, such a “double-hit” scenario has been observed directly (6).

Similarly, hypersensitivity of RyRs caused by several recently discovered RyR mutations is known to have profound arrhythmogenic potential. In these patients, catecholaminergic polymorphic ventricular tachycardias (CPVTs) can be triggered by physical or emotional stress and cause sudden cardiac arrest (24). Thus, we speculate that the frequent CPVTs observed in dystrophic patients may also be linked to RyR hypersensitivity, at least in part (25). The fact that RyR Ca²⁺ hypersensitivity and E-C coupling gain can be normalized in dystrophic cells by antioxidants may open new perspectives for the therapeutical treatment of DMD patients.

Conclusions.

The key result of our study is the discovery of a RyR hypersensitivity for Ca²⁺ due to redox modifications in dystrophic cardiomyopathy, which is not only responsible for excessive stress responses but also changes the signal transduction linking L-type Ca²⁺ channels to RyRs during E-C coupling.

GRANTS

This work was supported by the Swiss Foundation for Research on Muscle Diseases (to E. Niggli and N. Shirokova), Swiss National Science Foundation Grant 31-109693.05 (to E. Niggli), the Muscle Dystrophy Association (to N. Shirokova), National Institute of Arthritis and Musculoskeletal and Skin Diseases Grant AR-053933 (to N. Shirokova), the American Heart Association (to N. Shirokova), and the Sigrist Foundation (to N. Shirokova and E. Niggli).

DISCLOSURES

No conflicts of interest are declared by the author(s).