SERCA2a upregulation ameliorates cellular alternans induced by metabolic inhibition

Myocyte fluorescence was measured with a custom-made photometry system, which allowed us to excite Ca²⁺ indicators at two distinct wavelengths to simultaneously differentiate two dyes in two different cellular compartments. Our findings demonstrate that impairment of mitochondrial Ca²⁺ cycling and energy production lead to increased susceptibility to alternans. SERCA2a-upregulated mice are more capable of sustaining Ca²⁺ and electrical stability during stress.

Abstract

Cardiac alternans has been associated with the incidence of ventricular tachyarrhythmias and sudden cardiac death. The aim of this study was to investigate the effect of impaired mitochondrial function in the genesis of cellular alternans and to examine whether modulating the sarcoplasmic reticulum (SR) Ca²⁺ ameliorates the level of alternans. Cardiomyocytes isolated from control and doxycyline-induced sarco(endo)plasmic reticulum Ca²⁺-ATPase 2a (SERCA2a)-upregulated mice were loaded with two different Ca²⁺ indicators to selectively measure mitochondrial and cytosolic Ca²⁺ using a custom-made fluorescence photometry system. The degree of alternans was defined as the alternans ratio (AR) [1 − (small Ca²⁺ intensity)/(large Ca²⁺ intensity)]. Blocking of complex I and II, cytochrome-c oxidase, F₀F₁ synthase, α-ketoglutarate dehydrogenase of the electron transport chain, increased alternans in both control and SERCA2a mice (P < 0.01). Changes in AR in SERCA2a-upregulated mice were significantly less pronounced than those observed in control in seven of nine tested conditions (P < 0.04). N-acetyl-l-cysteine (NAC), rescued alternans in myocytes that were previously exposed to an oxidizing agent (P < 0.001). CGP, an antagonist of the mitochondrial Na⁺-Ca²⁺ exchanger, had the most severe effect on AR. Exposure to cyclosporin A, a blocker of the mitochondrial permeability transition pore reduced CGP-induced alternans (P < 0.0001). The major findings of this study are that impairment of mitochondrial Ca²⁺ cycling and energy production leads to a higher amplitude of alternans in both control and SERCA2a-upregulated mice, but changes in SERCA2a-upregulated mice are less severe, indicating that SERCA2a mice are more capable of sustaining electrical stability during stress. This suggests a relationship between sarcoplasmic Ca²⁺ content and mitochondrial dysfunction during alternans, which may potentially help to understand changes in Ca²⁺ signaling in myocytes from diseased hearts, leading to new therapeutic targets.

NEW & NOTEWORTHY

Myocyte fluorescence was measured with a custom-made photometry system, which allowed us to excite Ca²⁺ indicators at two distinct wavelengths to simultaneously differentiate two dyes in two different cellular compartments. Our findings demonstrate that impairment of mitochondrial Ca²⁺ cycling and energy production lead to increased susceptibility to alternans. SERCA2a-upregulated mice are more capable of sustaining Ca²⁺ and electrical stability during stress.

cardiac alternans has been associated with an increased incidence of tachyarrhythmias and sudden cardiac death (41, 46), yet the mechanisms at the cellular level of this multifactorial process remain elusive (30).

The importance of Ca²⁺ as a second messenger in general and as a mediator between electrical excitation and cellular contraction in the heart is well accepted (9), while Ca²⁺ mishandling is a main cause for arrhythmias (28, 30). Mitochondria play a significant role in Ca²⁺ signaling. They provide ATP for excitation/contraction and they serve as a Ca²⁺ buffer by taking up Ca²⁺ via the mitochondrial calcium uniporter (MCU) (40). Also, due to their spatial relation with the ryanodine receptor, mitochondria have been directly implicated in excitation-contraction coupling (12). As a result, it is still highly debated whether mitochondria take up Ca²⁺ on a beat-to-beat basis (22, 35) or whether they integrate it slowly (32, 42).

Prior studies have shown that impairment of mitochondrial function favors the occurrence of alternans (14), and uncoupling of the mitochondrial membrane potential has effects on sarcoplasmic reticulum (SR) Ca²⁺ release (53), while sarco(endo)plasmic reticulum Ca²⁺-ATPase 2a (SERCA2a) upregulation suppressed alternans in computer simulations (34) and experimental studies (10).

To better understand the interplay between cytosolic, SR and mitochondrial Ca²⁺, we have developed a novel system that allows the recording of Ca²⁺ in two organelles simultaneously. We used a modified wide-field fluorescence microscope and were able to simultaneously measure cytosolic or SR Ca²⁺, as well as mitochondrial Ca²⁺ concentration in isolated cardiomyocytes. We examined the hypothesis that impairment of mitochondrial ATP production or perturbation of mitochondrial Ca²⁺ handling will induce alternans. Therefore, the aim of this study was to investigate the effect of impaired mitochondrial function in the genesis of cellular alternans.

METHODS

Experimental setup.

The instrumentation used for myocyte fluorescence measurements was a custom-made fluorescence photometry system from Ionoptix (Milton, MA).

Myocytes were paced using 12-V field-stimulation. Ca²⁺ indicators were excited at two distinct wavelengths to excite two different dyes at the same time. Figure 1A illustrates the light beam inside the hyper switch. Briefly, the light of a xenon lamp was focused onto a galvo-driven mirror. With the help of aluminum mirrors, the light was diverted onto two collecting lenses which sent the light to two different filters. The cube holding the filters was exchangeable, which enabled the use of different wavelengths and, therefore, different dye indicators (Fig. 1A).

Fig. 1.

A: illustration of the light beam in the Hyper Switch. The light is focused with lenses and mirrors onto two different filters. The filters are mounted on a cube that is exchangeable, allowing the use of different wavelengths. The experimental set up includes the Hyper Switch, microscope, and photomultiplier tubes to collect the emission light in two different applications. Cubes, in the microscope and in front of the photomultiplier tubes, which hold polychroic mirrors, are exchangeable to work with different dyes. Filters in the Hyper Switch can also be changed to allow different applications. B: Fluo-4 AM (for cytosolic Ca²⁺) and x-Rhod-1 AM (for mitochondrial Ca²⁺) representative examples of Ca²⁺ alternans measured simultaneously. C: Rhod-2 AM (for cytosolic Ca²⁺) and Fluo-5N AM (for sarcoplasmic Ca²⁺) representative examples of Ca²⁺ alternans measured simultaneously.

An exchangeable polychroic mirror led the light beam to the myocyte-containing chamber. X-rhod-1- as well as Rhod-2-loaded myocytes were excited at 555 ± 25 nm, and emission signals were collected at 605 ± 55 nm (17, 31). Fluo-4, as well as Fluo-5N-loaded myocytes were excited at 484 ± 15 nm, and the emission signals were collected at 517 ± 30 nm (16, 25) (Fig. 1, B and C). Therefore, the simultaneously measured cytosolic and mitochondrial Ca²⁺ alternans are time aligned. The illumination field was restricted to collect the emission of a single myocyte. The emission light was sorted with the help of corresponding polychroic mirrors to either of three photomultiplier tubes. Cubes holding the polychroic mirrors were exchangeable to work with different dyes.

Doxycycline-induced SERCA2a-upregulated mouse model.

The experimental protocol was approved by the Massachusetts General Hospital Committee on Animal Care, Animal Care and Use Committee.

To probe the role of the interplay of the SR and mitochondrial Ca²⁺ we used the doxycycline (DOX)-induced SERCA2a-upregulated mouse model. The generation of DOX-induced SERCA2a mice has been previously described, in detail (7, 48). To induce expression of SERCA2a, animals received DOX in their drinking water (200 mg/l) for 7 days. The SERCA2a protein expression in this transgenic SERCA2a mouse model is increased by 38% to 45%, as previously reported (48, 49, 51).

Echocardiographic assessment.

Mice received an echocardiogram on the day before DOX was added to their drinking water and on the day they were used for myocyte isolation. The animals were anesthetized with sodium pentobarbital injected intraperitoneally (50 mg/kg). The chest hair was removed with a topical depilatory agent, and ultrasound transmission gel was applied to the transducer and contact area. Animals were imaged in the supine position with a 13-MHz linear probe using an Acuson Sequoia C512 system (Siemens, Mountain View, CA). The heart was first imaged in the two-dimensional mode in the parasternal long-axis view. One-dimensional images were recorded in the parasternal long and short axis with M-mode recordings at the midventricular level to evaluate systolic function. Left ventricular (LV) wall thickness was assessed in the interventricular septum and the posterior wall (LVPW). Then, LV end-systolic dimensions [left ventricular posterior wall, systolic (LVPWs), interventricular septal thickness end-systole (IVSs), left ventricular internal diameter end systole (LVIDs)] were obtained at the time of minimal chamber dimension. From the point of maximal LV diastolic dimension, end-diastolic measurements [left ventricular posterior wall, diastolic (LVPWd), interventricular septal thickness end diastole (IVSd); left ventricular internal diameter end diastole (LVIDd)] were collected.

LV Fractional shorting (FS) was then calculated from LV dimensions derived in M-mode by using the formula: (LVIDd − LVIDs)/LVIDd × 100%.

Isolation of cardiomyocytes.

Cardiomyocytes were isolated as described previously (44). Briefly, mice of either sex were heparinized and anesthetized with isoflurane. The heart was rapidly excised and cooled immediately in cold myocyte isolation buffer supplemented with 0.4 mM EGTA. Myocyte isolation buffer contained (in mM) 130 NaCl, 5.4 KCl, 0.5 MgCl₂, 0.33 NaH₂PO₄, 22 glucose, 50 μU/ml bovine insulin, 25 HEPES (acid free), and 0.3 CaCl₂ adjusted to pH 7.4 with NaOH. The aorta was cannulated, and the heart was then mounted on a Langendorff apparatus and perfused with an enzyme solution consisting of (mg/ml) 0.12 collagenase B (Roche, Minneapolis, MN), 0.12 collagenase D (Roche), 0.08 protease (type XIV; Sigma, St. Louis, MO), and 0.08 trypsin (Sigma-Aldrich) at a constant hydrostatic pressure until the heart was digested. Ca²⁺ concentration was increased up to 0.7 mM during further incubation at 37°C for 15 min in enzyme solution and bovine serum (2 mg/ml; Sigma) was added to the solution. The tissue was softly pipetted to loosen it, filtered through a 200-μm nylon mesh, and then centrifuged at 300 rpm for 3 min.

The myocyte pellet was resuspended in myocyte isolation buffer with 1.2 mM CaCl₂ and 2 mg/ml bovine serum albumin and incubated at 37°C for 10 min to completely stop digestion and then centrifuged at 300 rpm for 3 min. Then, myocytes were transferred to a 1.8 mM Ca²⁺ Tyrode solution that contained (in mM) 140 NaCl, 5.4 KCl, 0.5 MgCl₂, 0.33 NaH₂PO₄, 11 glucose, 5 HEPES, pH adjusted with NaOH to 7.4 and kept at room temperature until use.

Dyes and chemicals.

The custom-made fluorescence system described above allowed us to measure the fluorescence from two distinct dyes, at two different wavelengths. For that matter, the myocyte solution was consecutively loaded with two different myocyte-permeable acetoxymethyl (AM) ester derivatives of fluorescent calcium indicators.

To measure [Ca²⁺] in two cellular compartments simultaneously, we used the following combinations of dyes: 1) x-Rhod-1 AM (for mitochondrial Ca²⁺, [Ca²⁺]_m) and Fluo-4 AM (for cytosolic, [Ca²⁺]) or 2) Fluo-5N AM (for sarcoplasmic Ca²⁺, [Ca²⁺]_SR) and Rhod-2 AM (for cytosolic Ca²⁺, [Ca²⁺]_i).

Briefly, to probe the [Ca²⁺]_m, myocytes were loaded with the Ca²⁺ indicator x-Rhod-1 AM (1.4 μM) for 30 min in 1.8 mM Ca²⁺ Tyrode solution at 37°C. After the first dye was washed out, the same myocytes were loaded with Fluo-4 AM (10 μM) for 30 min at room temperature to promote accumulation of the dye in the cytosol. Then, the supernatant was removed and 1.8 mM Ca²⁺ Tyrode solution was added to the myocytes. When [Ca²⁺]_SR was studied, myocytes were loaded with Fluo-5N AM (5 μM) for 2 h at 37°C in 1.8 mM Ca²⁺ Tyrode solution, and after a washout, myocytes were loaded with Rhod-2 AM (1 μM) for 30 min at room temperature, as previously described (15, 18, 43). X-rhod-1 and Fluo-4 were added from a stock solution in DMSO with 5% pluronic. Fluo-5N AM was added from a stock solution in DMSO with 0.25% pluronic. No pluronic was added to Rhod-2 AM solution. After loading of the two dyes, myocytes rested for 30 min for deesterification of the dye in dye-free media before further workup. Once loaded, myocytes were used for experiments for ∼4-6 h. Prior studies have demonstrated that the above described protocols result in localization of Fluo-5N AM in the SR (43), Rhod-2 AM (18), and Fluo-4 AM in the cytosol (37) and x-Rhod-1 AM in the mitochondria (15).

Sodium cyanide, antimycin, FCCP, oligomycin, ruthenium red, cyclosporin A, N-acetyl-l-cysteine (NAC), 2-keto-3-methylvaleric acid (KMVA), DMSO, doxycycline, and pluronic were obtained from Sigma-Aldrich (St. Louis, MO). Fluo-4, x-Rhod-1, Rhod-2, Fluo-5N, and thapsigargin were purchased from Invitrogen (Carlsbad, CA) and CGP 37157 from Tocris (Ellisville, MO). FCCP, antimycin, oligomycin, cyclosporin A, CGP 37157, and thapsigargin were prepared as stock solutions in DMSO. Ruthenium red was prepared as a stock solution in ultra pure water. Sodium cyanide, KMVA, and NAC were directly added to the solution.

Calcium recordings.

Cytosolic calcium and mitochondrial calcium were recorded simultaneously using the fluorescence dyes Fluo-4 AM and x-Rhod-1 AM, as described in Dyes and chemicals. Myocytes were continuously superfused with Tyrode solution containing 1.8 mM calcium. Myocytes were stimulated at that frequency until stable alternans occurred and was not changed until the end of the experiment. Once stable alternans was established, the superfusing solution was switched to Tyrode solution containing a chemical influencing mitochondrial metabolism and calcium cycling (as mentioned in Table 1). All chemicals were supplemented to the superfusing media. In selected experiments, an agonist was added after the alternans developed to investigate whether it can alter the magnitude of alternans.

Table 1.

Alternans ratios before and during superfusion with the presented chemicals

	Control		SERCA2a
Chemical	[Ca2+]c	[Ca2+]m	[Ca2+]c	[Ca2+]m
Baseline	0.17 ± 0.01	0.23 ± 0.03	0.21 ± 0.03	0.25 ± 0.03
Antimycin	0.37 ± 0.03*	0.43 ± 0.03*	0.25 ± 0.03*	0.3 ± 0.03*
(ncontrol = 14, nSERCA2a = 13)
Baseline	0.15 ± 0.01	0.19 ± 0.01	0.22 ± 0.02	0.26 ± 0.02
Sodium cyanide	0.39 ± 0.04*	0.44 ± 0.04*	0.35 ± 0.03*	0.41 ± 0.03*
(ncontrol = 13, nSERCA2a = 10)
Baseline	0.19 ± 0.02	0.23 ± 0.02	0.29 ± 0.03	0.34 ± 0.03
KMVA	0.52 ± 0.03*	0.61 ± 0.03*	0.43 ± 0.04*	0.54 ± 0.03*
(ncontrol = 16, nSERCA2a = 11)
Baseline	0.17 ± 0.01	0.26 ± 0.02	0.23 ± 0.03	0.29 ± 0.03
Oligomycin	0.39 ± 0.03*	0.48 ± 0.03*	0.32 ± 0.04*	0.39 ± 0.04*
(ncontrol = 13, nSERCA2a = 10)
Baseline	0.15 ± 0.01	0.22 ± 0.01	0.19 ± 0.01	0.27 ± 0.01
FCCP	0.48 ± 0.04*	0.57 ± 0.03*	0.42 ± 0.03*	0.52 ± 0.03*
(ncontrol = 12, nSERCA2a = 12)
Baseline	0.24 ± 0.02	0.29 ± 0.02	0.29 ± 0.03	0.34 ± 0.03
Rotenone	0.34 ± 0.02*	0.42 ± 0.03*	0.32 ± 0.03*	0.4 ± 0.03*
(ncontrol = 16, nSERCA2a = 19)
Baseline	0.20 ± 0.02	0.25 ± 0.02	0.29 ± 0.06	0.33 ± 0.06
Ruthenium Red	0.33 ± 0.03*	0.43 ± 0.03*	0.41 ± 0.07*	0.46 ± 0.06*
(ncontrol = 20, nSERCA2a = 7)
Baseline	0.23 ± 0.02	0.28 ± 0.02	0.27 ± 0.04	0.3 ± 0.04
H2O2	0.40 ± 0.02*	0.49 ± 0.03*	0.38 ± 0.03*	0.43 ± 0.03*
(ncontrol = 14, nSERCA2a = 9)
Baseline	0.14 ± 0.02	0.19 ± 0.02	0.25 ± 0.03	0.31 ± 0.04
CGP	0.55 ± 0.02*	0.61 ± 0.02*	0.37 ± 0.03*	0.42 ± 0.02*
(ncontrol = 16, nSERCA2a = 12)

Data are presented as means ± SE. Cardiomyocytes were loaded with two different dyes Fluo-4 AM [Ca²⁺]_c and x-Rhod-1 AM [Ca²⁺]_m, respectively.

^*P < 0.05 before vs. after treatment with the chemical.

The degree of alternans was defined as the alternans ratio = 1-S/L, where S/L represent the amplitude of the small and large Ca²⁺ signal, respectively. An alternans ratio of 1/0 represents the most/least significant alternans, respectively. The background was subtracted from all fluorescence signals.

Statistical analysis.

Data are presented as means ± SE for the indicated number (n) of cells. Statistical comparisons were performed for statistical significance (P < 0.05) using Student's t-test for paired and unpaired data. The experimental data were analyzed with IonWizard 6.0 (IonOptix, Milton, MA) and Origin (OriginLab, Wellesley Hills, MA).

RESULTS

Echocardiographic assessment of the SERCA2a mouse model.

Control mice had a mean FS of 46%, while SERCA2a-upregulated mice showed a higher FS of 54%, compared with control (P < 0.0001).

SR calcium content of control and SERCA2a-upregulated mice.

Myocytes were paced at 2 Hz for 30 beats. Then a rapid application of caffeine (10 mM) was applied to assess the SR Ca²⁺ content. Each caffeine induced Ca²⁺ transient was normalized to the mean value of five previous (before application of caffeine) Ca²⁺ transients.

Figure 2B illustrates representative [Ca²⁺]_SR recordings (Fluo-4 AM) from control and SERCA2a-upregulated mice after rapid application of caffeine. Myocytes from SERCA2a-upregulated mice had a significantly higher [Ca²⁺]_SR (P < 0.0001), compared with myocytes from control mice, suggesting that upregulation of SERCA2a led to higher [Ca²⁺]_SR in those mice (Fig. 2A).

Fig. 2.

A: comparison of calcium intensity after rapid application of caffeine in control (n = 17) and SERCA2a mice (n = 20); data are normalized. B: representative [Ca²⁺]_c recording after an impulse of caffeine in myocytes isolated from control and SERCA2a mice loaded with Fluo-4 AM. *P < 0.05 normal vs. SERCA2a mice; data are presented as means ± SE.

Role of inhibition of ATP synthesis and mitochondrial calcium metabolism on alternans.

Mitochondria play a central role in energy metabolism of the myocyte. The electron transport chain in the mitochondria creates an electrochemical gradient that is the driving force for the F₀F₁-synthase that catalyzes the synthesis of ATP. Impairment of ATP synthesis may, therefore, have effects at the whole myocyte level. To investigate the role of TCA cycle in the genesis of cardiac alternans, we investigated the effect of antimycin, rotenone, and sodium cyanide on the alternans ratio (AR).

Blocking complex III with antimycin (5 μg/ml) increased the [Ca²⁺]_c and [Ca²⁺]_m AR from 0.17 ± 0.01 AU to 0.37 ± 0.03 AU (P < 0.0001) and from 0.23 ± 0.03 AU to 0.43 ± 0.03 AU (P < 0.0001), respectively, in control mice (Fig. 3A). In SERCA2a-upregulated mice, the [Ca²⁺]_c and [Ca²⁺]_m AR increased from 0.21 ± 0.03 AU to 0.25 ± 0.03 AU (P < 0.0001) and from 0.25 ± 0.03 AU to 0.3 ± 0.03 AU (P < 0.0001), respectively.

Fig. 3.

Comparison of alternans ratio in control and SERCA2a mice before and after the exposure of chemicals that influence mitochondrial energy. Alternans were elicited with rapid pacing, and myocytes were then superfused with the chemical. Myocytes were incubated under conditions designed selectively to enhance the accumulation of x-Rhod-1 AM in the mitochondria, and Fluo-4 AM in the cytosol respectively. Antimycin (A), rotenone (B), sodium cyanide (C), KMVA (D), oligomycin (E), and FCCP (F). *P < 0.05; data are presented as means ± SE.

Rotenone (5 μM), an inhibitor of complex I of electron chain, enhanced [Ca²⁺]_c and [Ca²⁺]_m AR from 0.24 ± 0.02 AU to 0.34 ± 0.02 AU (P < 0.0001) and from 0.29 ± 0.02 AU to 0.42 ± 0.03 AU (P < 0.0001), respectively. Rotenone-induced alternans were also less severe in SERCA2a-upregulated mice; [Ca²⁺]_c and [Ca²⁺]_m AR increased from 0.29 ± 0.03 AU to 0.32 ± 0.03 AU (P < 0.0001) and from 0.34 ± 0.03 AU to 0.4 ± 0.03 AU (P < 0.0001), respectively (Fig. 3B).

Blocking the cytochrome-c oxidase (complex IV) with sodium cyanide (1 mM), increased the [Ca²⁺]_c and [Ca²⁺]_m AR, respectively, from 0.15 ± 0.01 AU to 0.39 ± 0.04 AU (P = 0.0001) and from 0.19 ± 0.01 AU to 0.44 ± 0.04 AU (P = 0.0001). In SERCA2a-upregulated mice, changes in alternans ratio with these chemicals were not as pronounced as in control mice (Fig. 3C): [Ca²⁺]_c increased from 0.25 ± 0.03 AU to 0.37 ± 0.03 AU (P = 0.00017) and [Ca²⁺]_m from 0.31 ± 0.04 AU to 0.42 ± 0.02 AU (P < 0.0001).

We also evaluated the effect of KMVA (5 mM)—a blocker of the α-ketoglutarate dehydrogenase of the TCA cycle that results in a diminished ATP supply—on alternans. Superfusion with KMVA caused a change in AR [Ca²⁺]_m from 0.23 ± 0.02 AU to 0.61 ± 0.03 AU (P = 0.0001) and [Ca²⁺]_c AR from 0.19 ± 0.02 AU to 0.52 ± 0.03 AU (P < 0.0001). In myocytes from SERCA2a-upregulated mice, KMVA also induced an increase in alternans (Fig. 3D). AR in the cytosol changed from 0.29 ± 0.03 AU to 0.43 ± 0.04 AU (P < 0.0001) and in the mitochondria from 0.34 ± 0.03 AU to 0.54 ± 0.03 AU (P < 0.0001).

We also inhibited ATP synthesis by exposing cardiomyocytes to oligomycin (1 μg/ml), an inhibitor of the F₀F₁ synthase. The [Ca²⁺]_c and [Ca²⁺]_m AR increased from 0.17 ± 0.01 AU to 0.39 ± 0.03 AU (P < 0.0001) and from 0.26 ± 0.02 AU to 0.48 ± 0.03 AU (P < 0.0001), respectively. [Ca²⁺]_c and [Ca²⁺]_m AR in myocytes from SERCA2a-upregulated mice after treatment with oligomycin also increased from 0.23 ± 0.03 AU to 0.32 ± 0.04 AU (P = 0.002) and from 0.29 ± 0.03 AU to 0.39 ± 0.04 AU (P = 0.002), respectively (Fig. 3E).

Finally, we did not just directly inhibit ATP synthesis, but we also treated the myocytes with the uncoupling agent FCCP (0.1 μM), which abolishes the proton gradient across the inner mitochondrial membrane and indirectly leads to a diminished ATP production. We have observed that [Ca²⁺]_c and [Ca²⁺]_m increased significantly from 0.15 ± 0.01 AU to 0.48 ± 0.04 AU (P < 0.0001) and from 0.22 ± 0.01 AU to 0.57 ± 0.03 AU (P < 0.0001) in control mice, respectively (Fig. 3F). [Ca²⁺]_c and [Ca²⁺]_m AR also increased in SERCA2a-upregulated mice from 0.19 ± 0.01 AU to 0.42 ± 0.03 AU (P < 0.0001) and from 0.27 ± 0.01 AU to 0.52 ± 0.03 AU (P < 0.0001), respectively.

Overall, we have observed that SERCA2a-upregulated mice exhibited a smaller increase of the cytosolic and mitochondrial AR across all perturbations of mitochondrial function.

Role of mitochondrial calcium uptake on alternans.

We then sought to examine whether impairment of the mitochondrial Ca²⁺ uptake and extrusion mechanisms may be involved in the genesis of cellular alternans. Mitochondria take up Ca²⁺ via the MCU, which is driven by the electrical gradient. Once in the mitochondria, Ca²⁺ is sequestered through the mitochondrial sodium calcium exchanger and possibly through the mitochondrial transition pore. We, therefore, hypothesized that blocking those pathways may affect the AR.

Superfusing myocytes from control mice with ruthenium red (10 μM), an MCU inhibitor, led to a significant increase of [Ca²⁺]_c and [Ca²⁺]_m AR from 0.20 ± 0.02 AU to 0.33 ± 0.03 AU (P < 0.0002) and from 0.25 ± 0.02 AU to 0.43 ± 0.03 AU (P < 0.0001), respectively. The same experiment was repeated with myocytes from SERCA2a-upregulated mice, in which the [Ca²⁺]_c and [Ca²⁺]_m AR increased from 0.33 ± 0.06 AU to 0.46 ± 0.06 AU (P < 0.01) and from 0.29 ± 0.06 AU to 0.41 ± 0.07 AU (P < 0.01), respectively (Fig. 4).

Fig. 4.

Comparison of alternans ratio in control and SERCA2a mice before and after the exposure of ruthenium red in myocytes loaded with Fluo-4 AM and x-Rhod-1 AM. *P < 0.05; data are presented as means ± SE.

Ability to suppress alternans.

CGP, an antagonist of the mitochondrial Na⁺-Ca²⁺ exchanger, blocks calcium extrusion from mitochondria, resulting in an accumulation of Ca²⁺ in the mitochondria, opening of the mitochondrial transition pore and loss of mitochondrial function. In myocytes from control mice, CGP (5 μM) exhibited a significant effect on AR (Fig. 5A). [Ca²⁺]_c and [Ca²⁺]_m AR increased from 0.14 ± 0.02 AU to 0.55 ± 0.02 AU (P < 0.0001), and from 0.19 ± 0.02 AU to 0.61 ± 0.02 AU (P < 0.0001), respectively. Specifically, in myocytes from SERCA2a-upregulated mice, CGP increased [Ca²⁺]_c and [Ca²⁺]_m AR from 0.25 ± 0.03 AU to 0.37 ± 0.03 AU (P < 0.0002) and from 0.31 ± 0.04 AU to 0.42 ± 0.02 AU (P < 0.0001), respectively.

Fig. 5.

A: alternans ratio of control and SERCA2a myocytes under baseline conditions, during exposure of CGP and during application with cyclosporin A (CsA) added to the superfusing solution containing CGP. B: alternans ratio of control and SERCA2a myocytes under baseline conditions, during exposure of H₂O₂, and during application of the reducing agent N-Acetyl-l-cysteine (NAC) added to the superfusing solution containing H₂O₂. Myocytes were loaded with Fluo-4 AM and x-Rhod-1 AM simultaneously. *P < 0.05; data are presented as means ± SE.

We next attempted to examine whether the observed CGP-induced changes of [Ca²⁺]_c and [Ca²⁺]_m ARs were reversible. Cyclosporin A inhibits the opening of the mitochondrial permeability transition pore and, therefore, preserves membrane potential and mitochondrial function. Cyclosporin A (10 μM) led to less pronounced alternans in both, mitochondrial and cytosolic signals (Fig. 5A). [Ca²⁺]_c and [Ca²⁺]_m AR decreased from 0.55 ± 0.03 AU to 0.24 ± 0.02 AU (P < 0.0001) and from 0.60812 ± 0.03 AU to 0.19 ± 0.02 AU (P < 0.0001). In myocytes from SERCA2a-upregulated mice, the [Ca²⁺]_c and [Ca²⁺]_m AR decreased from 0.37 ± 0.03 AU to 0.18 ± 0.01 AU (P < 0.0002) and from 0.42 ± 0.02 AU to 0.16 ± 0.02 AU (P < 0.0001), respectively (Fig. 5A).

We furthermore examined whether the oxidizing agent H₂O₂ has an effect on [Ca²⁺]_c and [Ca²⁺]_m AR (Fig. 5B). We again established alternans by rapid pacing and then supplemented the superfusing solution with H₂O₂ (0.1 mM). This caused a change in [Ca²⁺]_c AR from 0.23 ± 0.02 AU to 0.40 ± 0.02 AU (P < 0.0003) and in [Ca²⁺]_m AR from 0.28 ± 0.02 AU to 0.49 ± 0.03 AU (P < 0.0001). The [Ca²⁺]_c AR in myocytes from SERCA2a-upregulated mice increased from 0.27 ± 0.04 AU to 0.38 ± AU (P < 0.001), whereas the [Ca²⁺]_m from 0.3 ± 0.04 AU to 0.43 ± 0.03 AU AR increased (P < 0.001).

Moreover, we hypothesized that N-Acetyl-l-cysteine (NAC), which has an antioxidant effect and reduces free radicals, rescues alternans. Therefore, following exposure to H₂O₂, additional exposure of the myocytes to NAC (2 mM) caused a change of AR in control and SERCA2a-upregulated mice (Fig. 5B). [Ca²⁺]_c AR decreased to 0.22 ± 0.01 AU (P < 0.0001) and [Ca²⁺]_m AR to 0.17 ± 0.02 AU (P < 0.0001). In SERCA2a-upregulated mice, [Ca²⁺]_c AR decreased to 0.24 ± 0.04 AU (P < 0.01) and [Ca²⁺]_m AR to 0.21 ± 0.04 AU (P < 0.002). The observed changes were higher in the mitochondrial AR than in the cytosolic AR for both, control, and SERCA2a-upregulated mice (p_control = 0.0131, p_SERCA2a = 0.0195).

Restitution curve.

To further investigate the role of [Ca²⁺]_c and [Ca²⁺]_m, restitution properties and its relationship to alternans, myocytes from control and SERCA2a-upregulated mice were stimulated, according to the following S1–S2 protocol: 30 pulses were delivered at 2 Hz (S1) followed by a single S2 that ranged from 2 to 8.5 Hz. Myocytes were loaded with Rhod-2, and either Fluo-5N or Fluo-4, as described above.

Figures 6, A–F, show summary results of the restitution curves of cytosolic, mitochondrial, and SR Ca²⁺ for control and SERCA2a mice. The time constant τ was determined by fitting a single exponential to each curve: τ was smaller in SERCA2a-upregulated mice (Table 3) than in control mice (p_{Rhod-2 AM} = 0.0488, p_{x-Rhod-1 AM} = 0.2778, p_{Fluo-5N AM} = 0.0105).

Fig. 6.

Control and SERCA2a mice were paced according to an S1–S2 protocol. Myocytes were stimulated at 2 Hz (S1) followed by a single S2 that ranged from 2 Hz to 8.5 Hz. Restitution curves are presented for each cellular compartment, for control (A–C) (n_{Rhod-2 AM} = 12, n_{x-Rhod-1 AM} = 6, n_{Fluo-5N AM} = 8) and SERCA2a mice (D–F) (n_Rhod-2,AM = 14, n_{x-Rhod-1 AM} = 6, n_{Fluo-5N AM} = 7). Time constants (τ) were calculated for each experiment, and these data are presented in Table 3.

Table 3.

Time constants (τ) calculated from restitution curves of an S1–S2 protocol in control and SERCA2a-upregulated mice

Indicator	τcontrol, ms	τSERCA2a, ms
Fluo-5N AM	178 ± 22.3	71 ± 14.4
x-Rhod-1 AM	152 ± 26.8	70.4 ± 22.4
Rhod-2 AM	202 ± 87.2	45.7 ± 6.8

Data are presented as means ± SE. The time constant τ was determined by fitting a single exponential to each curve: τ was smaller in SERCA2a upregulated mice than in control mice (control mice: p_{Rhod-2 AM} = 0.0488, p_{x-Rhod-1 AM} = 0.2778, p_{Fluo-5N AM} = 0.0105; n_{Rhod-2 AM} = 12, n_{x-Rhod-1 AM} = 6, n_{Fluo-5N AM} = 8; and SERCA2a mice: n_{Rhod-2 AM} = 14, n_{x-Rhod-1 AM} = 6, n_{Fluo-5N AM} = 7).

The smaller time constants in myocytes from SERCA2a mice suggest that myocytes from SERCA2a-upregulated mice exhibit faster restitution capacity (more rapid recovery), which is likely to be one of the factors that constitute them as less prone to alternans and more capable of sustaining a more regular Ca²⁺ homeostasis under stress.

Comparison of cytosolic and mitochondrial alternans.

Overall, we have observed that modulating mitochondrial metabolism has a significant effect on pacing-induced alternans in all tested conditions (P < 0.01). Furthermore, agonists, such as cyclosporin A and NAC can partially reduce (P < 0.01) the amplitude of alternans (Table 2).

Table 2.

Alternans ratio before and during superfusion with the presented chemicals

	Control		SERCA2a
Chemical	[Ca2+]c	[Ca2+]m	[Ca2+]c	[Ca2+]m
Baseline	0.23 ± 0.02	0.28 ± 0.02	0.27 ± 0.04	0.3 ± 0.04
H2O2	0.40 ± 0.02*	0.49 ± 0.03*	0.38 ± 0.03*	0.43 ± 0.03*
(ncontrol = 14, nSERCA2a = 9)
H2O2	0.40 ± 0.02	0.49 ± 0.03	0.38 ± 0.03	0.43 ± 0.03
H2O2 + NAC	0.22 ± 0.01*	0.17 ± 0.02*	0.24 ± 0.04	0.21 ± 0.04
(ncontrol = 14, nSERCA2a = 9)
Baseline	0.14 ± 0.02	0.19 ± 0.02	0.25 ± 0.03	0.31 ± 0.04
CGP	0.55 ± 0.02*	0.61 ± 0.02*	0.37 ± 0.03*	0.42 ± 0.02*
(ncontrol = 16, nSERCA2a = 12)
CGP	0.55 ± 0.03	0.61 ± 0.03	0.37 ± 0.03	0.42 ± 0.02
CGP + CsA	0.24 ± 0.02*	0.19 ± 0.02*	0.18 ± 0.01	0.16 ± 0.02
(ncontrol = 10, nSERCA2a = 8)

Data are presented as means ± SE. Myocytes were loaded with two different dyes, Fluo-4 AM [Ca²⁺]_c and x-Rhod-1 AM [Ca²⁺]_m, and alternans were elicited by rapid stimulation. Agonist N-Acetyl-l-cysteine (NAC) and cyclosporin A (CsA) were added to examine whether the observed changes are reversible.

^*P < 0.05 before vs. after treatment with the chemical.

SERCA2a upregulation results in smaller AR in SERCA2a-upregulated mice than in control ones: Unpaired two sided t-test between [Ca²⁺]_c in control and SERCA2a mice revealed a significant difference in the induced change in AR (Δcytosol, in control vs. SERCA2a mice) when superfusing with antimycin (P < 0.0001), sodium cyanide (P = 0.0430), KMVA (P < 0.0001), oligomycin (P = 0.0016), rotenone (P = 0.0006), and CGP (P < 0.0001). [Ca²⁺]_m, changes in AR (Δmitochondria, in control vs. SERCA2a mice) were also significantly smaller in SERCA2a-upregulated mice when superfusing with KMVA (P < 0.0001), oligomycin (P = 0.0054), rotenone (P = 0.0030), and CGP (P < 0.0001) compared with control mice.

DISCUSSION

This study presents the first description that cytosolic Ca²⁺ alternans can be paralleled by mitochondrial Ca²⁺ alternans. The major findings of this study are that the impairment of mitochondrial Ca²⁺ cycling and energy production lead to a higher alternans ratio. Any pathway affecting mitochondrial metabolism investigated in this study had a profound effect on AR. Specifically, we have found 1) that impairment of mitochondrial function leads to a higher degree of alternans in both the cytosol and mitochondria and is more pronounced in the mitochondria than in the cytosol, 2) that the alternans magnitude can be suppressed to a certain degree in the mitochondria as well as in the cytosol, and 3) that myocytes from SERCA2a-upregulated mice are less susceptible to alternans than those in control mice.

Mitochondria contribute to the genesis of cardiac alternans.

In our study, modulating mitochondrial function, Ca²⁺ homeostasis and ATP production have always been accompanied by a change of the AR (P < 0.01). Mitochondria are known to respond to Ca²⁺ and to contribute actively to the regulation of spatial and temporal patterns of intracellular Ca²⁺ signaling (6). A recent study has shown that limiting the interaction of mitochondria with the SR during reperfusion can protect cardiomyocytes against lethal reperfusion injury through reduction of mitochondrial Ca²⁺ (36). Mitochondrial dysfunction is known to be involved in many cardiac disease states (20, 29); the present study demonstrates that it may be also involved in the genesis of cardiac alternans. Prior studies have shown that changes of the mitochondrial function can alter cytosolic Ca²⁺ signaling, consequently leading to abnormal Ca²⁺ handling and the occurrence of alternans (14). Our work supports the idea that there is a close relationship between cytosolic and mitochondrial Ca²⁺.

Protonophores (i.e., FCCP) lead to disconnection of the structural organization of mitochondria as a tubular network (19). We demonstrate that this morphologic alteration influences mitochondrial Ca²⁺ homeostasis, as well as cytosolic Ca²⁺. Furthermore, while mitochondria produce and can scavenge reactive oxyen species (ROS) within a physiological range (3), elevated ROS favor mPTP opening (38) and promotes arrhythmogenesis (24). In our study, alternans increased under conditions that favor ROS production, whereas mPTP inhibition using NAC reduced the level of alternans.

Beat-to-beat changes in mitochondrial calcium.

Our study addresses the kinetics of mitochondrial Ca²⁺ transport, which remains a controversial issue. Several studies suggest that [Ca²⁺]_m cannot change on a beat-to-beat basis (13, 21), while others support the idea that beat-to-beat changes can be seen also in mitochondria, similar to those observed in the cytosol (8, 22, 26, 50).

Our results confirm previous work suggesting that impairment of mitochondrial function affects Ca²⁺ homeostasis leading to proarrhythmic alternans. Florea et al. (14) have demonstrated that impairment of mitochondrial function leads to a higher ratio of cytosolic alternans. Our work extends these observations by demonstrating for the first time that alternans can be also seen in the mitochondria, through simultaneous cytosolic and mitochondrial Ca²⁺ measurements. Since there is a growing acceptance of the idea that mitochondria play various roles in Ca²⁺ homeostasis and pathologic mechanisms, this study introduces a novel approach to extend these investigations.

Furthermore, these results provide a framework for understanding how changes in mitochondrial metabolism may affect whole myocyte (patho)physiology. Our observations demonstrate a relationship between alternans and mitochondrial dysfunction and, therefore, indicate that altered mitochondrial metabolism, affects mitochondrial Ca²⁺ handling and may be responsible for the genesis of cardiac alternans. Therefore, our findings underscore the importance of understanding basic mitochondrial Ca²⁺ regulation, under similar conditions in the diseased heart (1, 2, 29, 33, 47) to further probe the mechanisms of cardiac alternans.

Myocytes from SERCA2a-upregulated mice are less susceptible to alternans.

Ca²⁺ handling is known to be impaired in heart failure (4, 5, 23). The present study shows that myocytes from SERCA2a-upregulated mice are less susceptible to alternans than myocytes from control mice following disruption of mitochondrial metabolism, which may provide an explanation for why previous studies have shown a reduced risk to ventricular arrhythmias when SERCA2a is elevated (11, 39).

There is increasing evidence that fractional calcium release, sarcoplasmic reticulum load, and cytosolic calcium sequestration play a pivotal role in Ca²⁺ instability and alternans (30). This study supports the idea that SR Ca²⁺ cycling plays an important role in the genesis of cellular alternans and is in line with prior work demonstrating the influence of SR Ca²⁺ release on intracellular Ca²⁺ restitution (27, 52). Nivala and Qu (34) observed in a computer model that increased SERCA2a from its control value suppressed alternans, which is in agreement with experimental studies suggesting that overexpressing SERCA2a suppresses Ca²⁺ alternans (10). On the other hand, experimental studies in atrial myocytes by Shkryl et al. (45) showed that beat-to-beat alternation of the restitution kinetics of SR Ca²⁺ release seems to be a major contributor for the occurrence of Ca²⁺ alternans. In addition, our data also suggest that perturbed mitochondrial function results in SR Ca²⁺ cycling abnormalities and alternans and, therefore, electric instability.

The findings of this work demonstrate a relationship between alternans, SR Ca²⁺ content, and mitochondrial dysfunction, which may help us understand changes in Ca²⁺ signaling in myocytes from HF patients, leading to new therapeutic targets. In conclusion, this study proposes that impairment of mitochondrial Ca²⁺ cycling and energy production lead to increased susceptibility to alternans in both control and SERCA2a-upregulated mice. However, the magnitude of these changes in SERCA2a-upregulated mice is less pronounced, indicating that SERCA2a-upregulated mice are more capable of sustaining electrical stability during metabolic stress.

Study limitations.

Although we always turned off the excitation light between recordings and changed the myocyte containing solution in the chamber after each recording, it is possible that myocytes may be affected by photo-bleaching. However, given these data presented in this manuscript pertain to paired comparisons (before/after metabolic inhibition), we expect the effect to be minimal. Another possible limitation may be due to potential interaction between the excitation/emission wavelengths of the fluorescent dyes. However, we have taken care that the excitation/emission filter cubes (see methods) are designed such that there is no spectral overlap, and we have used previously evaluated protocols that have demonstrated preferential dye localization at the respected organelles.

GRANTS

This work was supported by an American Heart Association (AHA) Scientist Development Grant (no. 0635127N), AHA Grant-in-Aid (no. 14GRNT20400001), and by National Institute of Aging Grant 1R21AG035128 to A. A. Armoundas. Support for this work was also provided by Stiftung Charité, Deutscher Akademischer Austauschdienst and Ärztefinanzzentrum Berlin to V. Stary.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

Author contributions: V.S., D.P., M.S.-C., and W.H.D. performed experiments; V.S. and D.P. analyzed data; V.S., D.P., M.S.-C., and A.A.A. interpreted results of experiments; V.S. prepared figures; V.S. and A.A.A. drafted manuscript; V.S., D.P., M.S.-C., and A.A.A. edited and revised manuscript; V.S. and A.A.A. approved final version of manuscript; A.A.A. conception and design of research.