Mitochondrial calcium overload is a key determinant in heart failure

Effects of Redox Imbalance on RyR2 Channel in Postischemic HF.

We have previously shown that protein kinase A (PKA) phosphorylation and oxidation of RyR2 channels cause SR Ca²⁺ leak and contribute to HF progression (1, 2). HF-related PKA phosphorylation—in part also attributable to decreased cAMP type 4 phosphodiesterase, PDE4D3, in the RyR2 channel complex (23)—nitrosylation, and oxidation of RyR2 were attenuated in a mouse model (mCAT) with decreased ROS levels obtained via targeted overexpression of human catalase in mitochondria (Fig. 2 A–D).

Open in a separate window

Fig. 2.

Posttranslational modifications of RyR2 complex and HF progression post-MI. (A) Biochemical evaluation of RyR2 macromolecular complex in left ventricular samples from sham and heart failure (HF) mice. To determine channel oxidation, the carbonyl groups in the protein side chains of immunoprecipitated RyR2 were derivatized to 2,4 dinitrophenylhydrazone (2,4 DNPH) by reaction with 2,4 dinitrophenylhydrazine. The 2,4 DNPH signal associated with RyR2 was determined by anti-DNP antibody (specificity for RyR2 was achieved due to immunoprecipitation of the protein). Anti-Cys NO antibody analysis of immunoprecipitated RyR2 was used to measure RyR2 nitrosylation. Quantification of RyR2 oxidation (B), Cys-nitrosylation (C), phosphorylation at Ser²⁸⁰⁸ (D), PDE4D3 (E), and calstabin2 (F) bound to RyR2; note that constitutive phosphorylation of Ser²⁸⁰⁸, mimicked by the aspartate residue substitution in RyR2-S2808D mice, cannot be detected. Data shown represent mean ± SEM from triplicate experiments. *P < 0.05 vs. WT; ^#P < 0.05 vs. RyR2-S2808D, ANOVA, Tukey–Kramer post hoc test; ^!P < 0.05 vs. SHAM, two-tailed t test. (G) Progressive cardiac dysfunction after myocardial infarction (MI) assessed by serial echocardiographic analyses. LVEF, left ventricular ejection fraction. Data are shown as mean ± SEM; *P < 0.05 vs. WT; ^#P < 0.05 vs. RyR2-S2808D; ANOVA repeated measures; n = 16–20 per group. AU, arbitrary units. See also Table S1.

Reduced binding of the RyR2 stabilizing subunit calstabin2 (24) to the channel due to RyR2 oxidation and PKA phosphorylation causes spontaneous diastolic SR Ca²⁺ release contributing to cardiac dysfunction in HF (1, 2). Genetically reducing RyR2 oxidation (mCAT mice) or preventing RyR2 PKA phosphorylation (RyR2-S2808A mice harboring RyR2 channels that cannot be PKA-phosphorylated) improved calstabin2 and PDE4D3 binding to RyR2 (Fig. 2 A–F) and cardiac performance (Fig. 2G and Table S1) after MI. Mitochondrial morphology (Fig. 3 A–I and Fig. S4) and function (Fig. 3 J–M and Table S2) were also improved in mCAT mice.

Open in a separate window

Fig. 3.

Leaky RyR2 channels and mitochondrial dysfunction in heart failure. (A–E) Representative transmission electron micrographs of cardiac mitochondria post myocardial infarction from WT (A), RyR2-S2808A (no leak) (B), RyR2-S2808D (leaky) (C), mCAT (D), mCAT × RyR2-S2808D (E), n = 5 per group. (Magnification: A–E, 15,000×; Insets, 50,000×.) (Scale bars: 500 nm.) Note the diffuse myofibrillar disarray. (F–I) Quantification of ultrastructural mitochondrial alterations depicted in A–E. Mitochondrial size (F) and cristae density (G). Numbers in the bars indicate the number of mitochondria analyzed. Quantification of mitochondrial number per image (H) and percentage of abnormal mitochondria per image at 15,000× (I). Relative number of damaged mitochondria was quantified by blinded observers from 8 to 10 images from different fields. Data are shown as mean ± SEM, *P < 0.05 vs. WT; ^#P < 0.05 vs. RyR2-S2808D, ANOVA, Tukey–Kramer post hoc test. (J) Assessment of the inner mitochondrial membrane potential (Δψ_m); the arrow denotes addition of H₂O₂ (100 μM). FCCP, carbonylcyanide-p-trifluoro-methoxy-phenyl-hydrazone. The * indicates significant difference (P < 0.05, ANOVA repeated measures, Tukey–Kramer post hoc test) of the WT group (n = 8) compared with RyR2-S2808D (n = 7), mCAT × RyR2-S2808D (n = 7), and mCAT (n = 6), or between RyR2-S2808D and mCAT × RyR2-S2808D groups. (K) Mitochondrial DNA (mtDNA)/nuclear DNA (nDNA) copy number and (L) ATP content assessed in left ventricle (n = 8 per group). (M) Measurement of ATP synthesis rates in cardiac mitochondria isolated from failing hearts of the indicated groups. ATP synthesis was driven by complex I (pyruvate/malate, 5 mM) and complex II (succinate 5 mM). The specificity of the measurements was verified using inhibitors (0.5 μM) of respiratory complex, as indicated (n = 3 per group, triplicate measurements per sample). All data are shown as mean ± SEM, *P < 0.05 vs. WT; ^#P < 0.05 vs. RyR2-S2808D, ANOVA, Tukey–Kramer post hoc test.

Open in a separate window

Fig. S4.

Assessment of mitochondrial morphological dynamism in HF. (A) Aspect ratio and (B) form factor were measured as described in SI Materials and Methods in cardiac mitochondria from 6-mo-old WT, RyR2-S2808A, RyR2-S2808D, mCAT, and mCAT × RyR2-S2808D mice, 4 wk after myocardial infarction; n ≥ 5 per group; see also Fig. 3 A–E. All data are shown as mean ± SEM; numbers in the bars indicate the number of mitochondria analyzed.

Oxidative overload in cardiomyocytes originates from multiple sources, including mitochondria, NAD(P)H oxidase, xanthine oxidase, and uncoupled nitric oxide synthase (16, 25, 26). Mitochondrial-derived ROS are elevated during cardiac overload or ischemic stress (4, 26, 27). Mitochondrial membrane potential, Δψ_m, is closely linked to Ca²⁺ levels and to mitochondrial ROS production; indeed, depolarized mitochondria produce more ROS, leading to further organelle depolarization, resulting in a vicious cycle. The decrease in Δψ_m observed in mitochondria from RyR2-S2808D ventricular cardiomyocytes (Fig. S3J) is consistent with a progressive decline in Δψ_m due to increasing [Ca²⁺] in cardiac mitochondria and is most likely due to elevated cytosolic [Ca²⁺] caused by RyR2-mediated SR Ca²⁺ leak (10). Supporting this view, mitochondria exposed to elevated [Ca²⁺] exhibit reduced Δψ_m, due to the large mitochondrial Ca²⁺ current generated during local [Ca²⁺] transients (28).

Ventricular cardiomyocytes harboring constitutively leaky RyR2 channels exhibited a reduction in mitochondrial ATP content and generation (Fig. S3 L and M), consistent with previous observations in failing human hearts (6). Further studies are needed to investigate in detail other systems, including neurohormonal and (epi)genetic mechanisms, endoplasmic reticulum (ER) stress, necrosis/apoptosis, and autophagy, that might participate in the regulation of bioenergetic homeostasis in HF (5, 6, 19, 25, 29).

Distinctive Roles of RyR2 and IP3R2 in the Pathophysiology of Mitochondrial Dysfunction in HF.

To determine the source of SR Ca²⁺ leak that causes mitochondrial overload in failing hearts, we investigated the roles of the two major Ca²⁺ release channels on myocardial SR: RyR2 and IP3R2 (1).

We generated a murine model (IP3R2^CVKO) in which IP3R2 expression was specifically ablated in ventricular cardiomyocytes via Cre/Lox recombination (Fig. S5 A–E). IP3R2^CVKO mice survived to adulthood without alterations in baseline myocardial function, and there was no up-regulation of the other two isoforms of IP3R (IP3R1 and IP3R3) (Fig. S5 F and G). Ca²⁺ sparks, SR Ca²⁺ load (Fig. S6), mitochondrial Ca²⁺ level (Fig. S6C and Fig. S7 A and B), and ROS production (Fig. S6D) were not significantly changed in IP3R2^CVKO ventricular cardiomyocytes evaluated both in sham or post-MI mice. Myocardial mitochondria from IP3R2^CVKO mice were normal (Fig. 4), and there was no major effect on acute HF progression (Fig. S7C and Table S1). Further investigations are warranted to explore the potential role of IP3R2 in ischemia/reperfusion and in long-term ischemic HF, especially given the reported involvement of IP3R2 in advanced stages of HF (30).

Open in a separate window

Fig. 4.

Cardiac ablation of IP3R2 does not rescue mitochondrial abnormalities observed in failing hearts. (A–D) Representative transmission electron micrographs of cardiac mitochondria in SHAM conditions (A and B) and post myocardial infarction (C and D) from IP3R2^fl/fl (A and C) and IP3R2^CVKO mice (B and D), n = 5 per group. (Magnification: A–D, 15,000×; Insets, 50,000×.) (Scale bars: 500 nm.) (E–J) Morphometric analysis of mitochondrial ultrastructure. Mitochondrial size (E) and cristae density (F). Quantification of mitochondrial number per image (G) and percentage of abnormal mitochondria per image at 15,000× (H). Evaluation of aspect ratio (I) and format form (J) (see SI Materials and Methods for details). (K) Mitochondrial DNA (mtDNA)/nuclear DNA (nDNA) copy number assessed in left ventricular tissue. (L) Assessment of ATP content in left ventricle (n = 6 per group) and (M) measurement of ATP synthesis rates in isolated mitochondria in sham conditions and 4 wk after coronary artery ligation, as described in Fig. 3M. All data are shown as mean ± SEM, *P < 0.05 vs. SHAM; two-tailed t test. Numbers in the bars indicate the number of mitochondria analyzed.

Open in a separate window

Fig. S5.

Generation of cardiac specific ablation of IP3R2. (A) Schematic illustration of the generation of the IP3R2 floxed mouse (Top) and representative PCR for the genotype detection of WT, IP3R2^fl/fl, and IP3R2KO mice. (B) Determination of IP3R2 mRNA levels by real time quantitative reverse transcription PCR (RT-qPCR) analysis of total RNA from isolated cardiomyocytes, using actin as internal standard; each bar represents mean ± SEM of four independent experiments, in each of which reactions were performed in triplicate using the pooled total RNAs from five mice per group. (C) IP3R2 from isolated ventricular cardiomyocytes was immunoprecipitated and immunoblotted. 2,4 DNPH, 2,4-dinitrophenylhydrazone. (D and E) Quantification of data shown in C. Data shown represent means ± SEM from triplicate experiments. *P < 0.05 vs. IP3R2^fl/fl, two-tailed t test. (F and G) Levels of IP3R isoforms after cardiac-specific ablation of IP3R2. Determination of IP3R1 (F) and IP3R3 (G) mRNA levels by RT-qPCR analysis of total RNA from isolated cardiomyocytes, using actin as internal standard; each bar represents mean ± SEM of four independent experiments, in each of which reactions were performed in triplicate using the pooled total RNAs from five mice/group.

Open in a separate window

Fig. S6.

Effects of cardiac ablation of IP3R2 on Ca²⁺ spark frequency, SR Ca²⁺ load, mitochondrial Ca²⁺ content, and ROS production. (A) Average Ca²⁺ spark frequency, (B) SR Ca²⁺ load, and (C) mitochondrial Ca²⁺ in ventricular cardiomyocytes (n = 20∼22 cells) isolated from at least six mice per group in SHAM and HF conditions. Data are shown as mean ± SEM. (D) Evaluation of mitochondrial ROS generation in ventricular cardiomyocytes using the mitochondria-targeted fluorescent indicator of superoxide production MitoSOX Red. n > 100 ventricular myocytes from at least four mice in each group. AU, arbitrary units. Data are shown as mean ± SEM, *P < 0.05 vs. WT; one-way ANOVA, Tukey–Kramer post hoc test.

Open in a separate window

Fig. S7.

Effects of cardiac ablation of IP3R2 on mitochondrial Ca²⁺ dynamics and HF progression after myocoardial infarction. (A) Mitochondrial Ca²⁺ in response to 3 Hz stimulation was evaluated in cardiomyocytes (n = 25–30) enzymatically isolated from at least seven mice per group in SHAM (solid lines) and HF (dotted lines) conditions. (B) The mitochondrial Ca²⁺ peak. Data are shown as means ± SEM, *P < 0.05 vs. SHAM, two-tailed t test. (C) Progressive cardiac dysfunction after left anterior coronary artery ligation assessed by serial ultrasound in IP3R2^fl/fl and IP3R2^CVKO. LVEF, left ventricular ejection fraction; n = 16–18 per group. See also Table S1.

In Vivo Experiments.

Transthoracic 2D echocardiography was performed using a 12-MHz probe (VeVo; Visualsonics). M-mode interrogation was performed in the parasternal short-axis view. Left ventricular (LV) end-diastolic and end-systolic dimensions and septal and posterior wall thicknesses were determined and used to calculate fractional shortening and ejection fractions (32). After baseline echocardiography, permanent occlusion of the proximal left anterior descending (LAD) coronary artery was performed in 5-mo-old mice. A small (1.5-cm) left thoracotomy was performed via the fourth intercostal space, and the lungs were retracted to expose the heart. After opening the pericardium, the LAD coronary artery was located and ligated with 8-0 silk suture near its origin between the pulmonary outflow tract and the edge of the left atrium, 2 mm lower than the tip of the left auricle. The ligation was deemed successful when the anterior wall of the LV turned pale. Animals were kept on a heating pad until they recovered. The group of mice undergoing sham ligation had a similar surgical procedure without tightening the suture around the artery. Echocardiography was repeated at 1, 2, and 4 wk post-MI. Serum concentration of troponin I was measured 1 d after coronary artery ligation using a commercially available immunoassay kit to indirectly estimate myocardial infarct size. Infarct area was also assessed in a distinct group of animals using 2-3-5-triphenyltetrazolium chloride and expressed as percentage of total LV area. Blood pressure and cardiac contractility were measured as described (32).

Isolation of Adult Cardiomyocytes and Measurement of Ca²⁺ and Mitochondrial Membrane Potential.

The hearts were excised and washed in Ca²⁺-free Tyrode solution, followed by retrograde perfusion with buffer containing collagenase II (Worthington Biochemical Corporation) at 37 °C for 10 min. The LV was subsequently dissociated into single myocytes, and extracellular Ca²⁺ concentration was progressively increased to reach a final concentration of 1.8 mM. Cardiomyocytes were plated on glass coverslips coated with laminin (Thermo Fisher Scientific). To determine intracellular [Ca²⁺], cardiomyocytes were loaded with 5 μM Fluo-4 acetoxymethyl (AM) ester (Thermo Fisher Scientific) for 20 min and washed three times with dye-free Tyrode solution. Ca²⁺ sparks were recorded using a Zeiss LSM 5 Live confocal microscope in line scan mode. The fluorophore was excited with an argon laser at 488 nm, and emission was recorded at 505–530 nm. Ca²⁺ sparks were quantified using a software algorithm (IDL; ITT Visual Information Solutions) and calculated as sparks/100 μm/s (34–36). Dynamic mitochondrial Ca²⁺ was evaluated in enzymatically isolated cardiomyocytes incubated with Rhod2-AM (2 µm; Thermo Fisher Scientific) for 1 h at 37 °C and then washed for deesterification. After loading, the cells were placed on the stage of a Zeiss LSM 5 Live confocal microscope (63× oil immersion lens) where Rhod2 fluorescence signals were recorded by excitation at 561 nm and measurement of the emitted light at 588 nm. Cardiomyocytes were stimulated by pacing (3 Hz) or pharmacologically triggering Ca²⁺ leak via RyR2 using FK506 (5 μM).

To assess mitochondrial ROS levels, isolated cardiomyocytes were incubated for 20 min with the mitochondria-targeted fluorescent indicator of superoxide production MitoSOX Red (5 μM; Thermo Fisher Scientific) and washed. Using a confocal laser-scanning microscope (Zeiss LSM 5 Live, 40× oil immersion lens), MitoSOX Red fluorescence was excited at 488 nm, and the emitted signal was filtered through a band pass filter (540–625 nm). The scanned parameters were fixed for all scans. For each group, fluorescence intensities of >100 cells randomly selected from several different dishes were examined.

To measure mitochondrial membrane potential (Δψ_m), cardiomyocytes were incubated with the fluorescent indicator tetra-methylrhodamine ethyl ester (TMRE; 30 nM; Thermo Fisher Scientific) for 20 min. TMRE fluorescence was excited at 532 nm, and the emitted signal was collected through a band pass filter (540–625 nm, Zeiss LSM 5 Live, 40× oil immersion lens). To minimize the impact of subcellular variability on Δψ_m, TMRE fluorescence was measured in five different areas in each cell. At the end of each experiment, cells were exposed to the mitochondrial uncoupler carbonylcyanide-p-trifluoro-methoxy-phenyl-hydrazone (FCCP, 300 nM). Images were obtained every 5 min, and fluorescence signals were normalized to the fluorescence measured at the start of the experiment.

Isolation of Cardiac Mitochondria.

Murine cardiac mitochondria were isolated using previous established procedures (11, 13, 37). In brief, mice were euthanized, and the heart was excised. The LV was minced on ice, resuspended, and homogenized in 1.5 mL of buffer (10 mM Tris, 250 mM sucrose, 1× protease inhibitor mixture, pH 7.4) supplemented with 1 mM EDTA. Homogenates were centrifuged at 1,000 × g for 5 min at 4 °C (pellet discarded and supernatant recentrifuged at 500 × g). The supernatant was centrifuged at 10,000 × g for 15 min to pellet the mitochondria. The pellet was then resuspended in the homogenization buffer (without EDTA) and used for experiments within 1 h. All steps were performed at 4 °C. ATP synthesis rates in isolated cardiac mitochondria were determined using a bioluminescence kit (Sigma-Aldrich), according to the manufacturer’s instructions. Briefly, 10 μg of cardiac mitochondria were dissolved in 50 μL of buffer (10 mM Hepes, 125 mM KCl, 5 mM MgCl₂, 2 mM K₂HPO₄, pH 7.42) to determine complex I (5 mM pyruvate/malate) or complex II (5 mM succinate) driven ATP synthesis. To determine the rates of nonmitochondrial ATP production, measurements with substrates were repeated in the presence (0.5 μM) of inhibitors of respiratory complex: rotenone (complex I), antimycin (complex III), and oligomycin (complex IV). To avoid the reverse electron transfer effect, succinate-driven ATP synthesis was assessed in the presence of rotenone (0.5 μM). ATP content in the LV was determined as described (37). The endogenous mitochondrial Ca²⁺ content was measured in cardiac mitochondria isolated as reported above, except that all buffers were EGTA/EDTA-free to avoid chelating Ca²⁺. Isolated mitochondrial pellets were repeatedly washed in buffers, resuspended in 0.6 M HCl, and sonicated (2 × 10 s at 40% of maximal power output). Absolute Ca²⁺ content (expressed as nmoles/mg protein) was determined using the o-cresolphthalein complexone assay (Cayman Chemical), according to the manufacturer’s instructions.

Real-Time Quantitative Reverse Transcription PCR.

Total RNA was isolated using TRIzol reagent (Thermo Fisher Scientific) in combination with the RNeasy Mini kit (Qiagen) followed by DNase treatment (38, 39), and cDNA was synthesized by means of a Thermo-Script RT-PCR System (Thermo Fisher Scientific), following the manufacturer’s instructions. After reverse transcription, real-time quantitative PCR was performed in triplicate using the SYBR Green RT-PCR master mix kit and quantified by built-in SYBR Green Analysis (Thermo Fisher Scientific) (39, 40). To determine mitochondrial and nuclear DNA copy number (mtDNA and nDNA, respectively), a quantitative RT-qPCR was performed on 30 ng of total left ventricular DNA in each qPCR reaction using primer pairs for mitochondrial and genomic loci. Two different primer pairs were used to quantify and confirm relative mitochondrial (mt)/nuclear (n) DNA ratio: 12S and 16S for mtDNA, H19 and 18S for nDNA. Samples were measured in triplicate, and results were confirmed by at least three independent experiments. All of the primer sequences (Sigma-Aldrich) for gene analysis are listed in Table S2.

Immunoprecipitation/Immunoblot Analysis.

Left ventricular samples were incubated with specific antibodies and protein A Sepharose beads (Amersham Pharmacia Biotech Inc.) at 4 °C for 2.5 h, and the beads were then washed. Immunoprecipitates were size-fractionated on SDS/PAGE gels (6% for RyR2 and IP3R2, 15% for calstabin2) and transferred onto nitrocellulose membranes for 2 h at 200 mA. To determine channel oxidation, the carbonyl groups in the protein side chains in the immunoprecipitated material were derivatized to 2,4-dinitrophenylhydrazone (2,4 DNPH) by reaction with 2,4-dinitrophenylhydrazine. The 2,4 DNPH signal was determined using a specific antibody, according to the manufacturer’s instructions (Millipore). Anti-S-nitrosocysteine (anti-CysNO) antibody was purchased from Sigma-Aldrich. All immunoblots were developed with the Odyssey system (LI-COR Biosciences), using infrared-labeled anti-mouse and anti-rabbit IgG (1:10,000 dilution) secondary antibodies (40). The intensity of the bands was quantified by using the LI-COR Image Studio Software (LI-COR Biosciences).

Ethical Approval.

All studies were performed according to protocols approved by the Institutional Animal Care and Use Committee (IACUC) of Columbia University.

We are grateful to Bi-Xing Chen, Qi Yuan, and Jingyi Yang (Columbia University) for technical support, Alain Lacampagne and Jeremy Fauconnier (INSERM and University of Montpellier) for critical discussion and helpful assistance, and Peter S. Rabinovitch (University of Washington) for providing the mCAT mouse founders. This work was supported by American Heart Association Grants 13POST16810041 and 15SDG25300007 (to G.S.), NIH Grant R01HL061503, and the Leducq Foundation (to A.R.M.).