Elevated Na is a dynamic and reversible modulator of mitochondrial metabolism in the heart

Abstract

Elevated intracellular sodium Na_i adversely affects mitochondrial metabolism and is a common feature of heart failure. The reversibility of acute Na induced metabolic changes is evaluated in Langendorff perfused rat hearts using the Na/K ATPase inhibitor ouabain and the myosin-uncoupler para-aminoblebbistatin to maintain constant energetic demand. Elevated Na_i decreases Gibb's free energy of ATP hydrolysis, increases the TCA cycle intermediates succinate and fumarate, decreases ETC activity at Complexes I, II and III, and causes a redox shift of CoQ to CoQH₂, which are all reversed on lowering Na_i to baseline levels. Pseudo hypoxia and stabilization of HIF-1α is observed despite normal tissue oxygenation. Inhibition of mitochondrial Na/Ca-exchange with CGP-37517 or treatment with the mitochondrial ROS scavenger MitoQ prevents the metabolic alterations during Na_i elevation. Elevated Na_i plays a reversible role in the metabolic and functional changes and is a novel therapeutic target to correct metabolic dysfunction in heart failure.

Fig. 1. Intracellular Na elevation and contractility in isolated Langendorff perfused rat hearts transiently exposed to 75 μM ouabain and 150 nM para-aminoblebbistatin (PAB).

a Time-courses of intracellular Na in the absence of PAB and b corresponding left ventricular developed pressure (LVDP) measured before (baseline), during ouabain treatment, and after washout. c Time-courses of intracellular Na and d corresponding left ventricular developed pressure (LVDP) measured before (baseline), during co-infusion of ouabain+PAB, ouabain+CGP + PAB, or PAB alone and after washout. Intracellular Na was measured using ²³Na-TQF NMR spectroscopy. Para-aminoblebbistatin (PAB), used to remove the positive inotropy caused by elevation of Na, was titrated for 5 min following perfusion with ouabain and confirmation of positive inotropy; arrow denotes start of PAB titration in ouabain-treated hearts. White circle = time-matched control; red circle = ouabain treatment; purple diamond = ouabain + CGP blue triangle = PAB treatment only. n = 6 rats per condition; significance determined by two-tailed, unpaired student’s t-test. **P < 0.01. Data plotted as mean ± SEM and source data are provided as a Source data file.

Fig. 2. ³¹P NMR cardiac energetics during intracellular Na elevation and washout.

a ATP, b phosphocreatine (PCr), c inorganic phosphate (Pi) and d free energy (ΔG_ATP) of ATP hydrolysis at the end of the Na elevation (40 min) and at the end of the washout period (60 min) measured with ³¹P NMR spectroscopy. e Representative ³¹P saturation transfer NMR spectra where the arrows (ω₁/2π) denote the frequency of the saturation pulse positioned midway between the α- and γ-ATP resonances (top spectrum, control) or positioned to saturate the γ-ATP peak (bottom spectrum), note the partial saturation of the PCr and Pi peaks due to magnetization transfer. f Reaction scheme showing magnetization transfer in the direction of the coloured arrows. Quantification of the magnetization transfer from PCr to γ-ATP (g) and Pi to γ-ATP (h). Time control = time-matched control; ouab = ouabain; CK = creatine kinase. n = 6 hearts or n = 12 (a, b) in each group. Significance determined by two-tailed, unpaired student’s t-test. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Data plotted as mean ± SEM and source data are provided as a Source data file.

Fig. 3. Cardiac metabolomics during [Na]_i elevation and washout.

a Schematic for ¹³C-labelling of cardiac metabolites arising from 1-¹³C glucose. Orange circles correspond to the first round of labelling. Once ¹³C reaches succinate then it becomes scrambled in the 2 and 3 positions since the molecule is symmetric leading to a second round of labelling denoted by the yellow circles. OAA = oxaloacetic acid, α-KG = α-ketoglutarate. Quantification of b ¹³C lactate and c ¹³C alanine measured using ¹³C NMR spectroscopy. n = 6 hearts per groups. d ¹³C citrate, e ¹³C succinate, f ¹H fumarate, g ¹³C glutamate C2, h ¹³C aspartate, and i ¹³C creatine measured using ¹³C or ¹H NMR spectroscopy at the end of the Na elevation (40 min) and at the end of the washout period (60 min). Time control = time-matched control; ouab = ouabain-treated; CGP+ouab = CGP and ouabain co-treatment; w/o = washout. n = 6 hearts per group. Significance determined by two-tailed, unpaired student’s t-test. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Data plotted as mean ± SEM and source data are provided as a Source data file.

Fig. 4. Effect of [Na]_i elevation on mitochondrial oxidative phosphorylation.

Oxygen consumption was measured in isolated rat cardiomyocytes using the Seahorse XFe24 platform. a Oxygen consumption rate (OCR) measured in intact cardiomyocytes treated with 50 mM ouabain, 100 mM ouabain, or 100 mM ouabain + CGP plotted as % change relative baseline. b–d Oxygen consumption rate (OCR) measured in permeabilised cardiomyocytes supplemented with different buffer Na concentrations, b 10 vs 15 mM, c 10 vs 20 mM or d 10 vs 30 mM. Also shown is 30 mM + CGP treatment. Control Na concentration was 10 mM. Data normalized to baseline. For (a–d), n = 4 rats, N = 4 technical replicates per rat per condition. Significance determined by nested (hierarchical) one-way ANOVA. *P < 0.05. e–g Electron transport chain activities of e Complex I, f Complex II and g Complex III, measured at the end of the Na elevation (40 min) and at the end of the washout period (60 min) for time-matched control (white data points) vs ouabain treated (high Na, filled red data points) vs washout (open red data points); inhib = Complex inhibitor rotenone for e, malonate for f and antimycin A for g. n = 6 hearts per group; n = 4 hearts for inhibitor. Significance determined by two-tailed, unpaired student’s t-test. Significance determined by nested (hierarchical) one-way ANOVA. *P < 0.05, **P < 0.01, ***P < 0.001. Data plotted as mean ± SEM and source data are provided as a Source data file.

Fig. 5. Effect of [Ca]_i elevation cardiac function and metabolomics.

a Time-course of left ventricular developed pressure (LVDP) in high Ca buffer (3.5 mM CaCl₂, dark blue filled symbols) and high Ca buffer titrated with para-aminoblebbistatin (3.5 mM CaCl₂ + PAB, light blue filled symbols). b Intracellular Na measured by ²³Na-TQF NMR is unchanged during perfusion with 3.5 mM CaCl₂ + PAB. c ATP, d phosphocreatine (PCr), e inorganic phosphate (Pi) and f free energy (ΔG_ATP) of ATP hydrolysis measured after perfusion with 3.5 mM CaCl₂ + PAB showing unaltered cardiac energetics. Metabolite concentration measured for control and high Ca buffer + PAB: g lactate, h alanine, i succinate, j fumarate and k glutamate using ¹H NMR spectroscopy. n = 7 hearts per groups. Significance determined by two-tailed, unpaired student’s t-test. ****P < 0.0001. Data plotted as mean ± SEM and source data are provided as a Source data file.

Fig. 6. [Na]_i elevation reduces CoQ pool and causes ROS production.

Percentage of reduced CoQH₂ over total CoQ (CoQ+CoQH₂) pool for a CoQ₉ and b CoQ₁₀ in myocardium measured at the end of the Na elevation (40 min) and at the end of the washout period (60 min). n = 5 rats per group (n = 6 for ouabain+PAB + CGP). c Time course of ROS production in isolated mouse ventricular cardiomyocytes measured as an increase in the fluorescence of the ROS reporter MitoSOX at 570 nm in response to increasing concentration of ouabain. d Rate of change of superoxide accumulation measured in MitoSOX-loaded mouse cardiomyocytes subject to Na elevation via ouabain and ouabain washout; values normalized to baseline. N = 30 cells from n = 3 mice per group; N = 5 cells from n = 1 mouse for 0.1 mM ouabain group. e–k MitoQ rats were fed 500 μM of MitoQ in drinking water for 2 weeks and their cardiac metabolism compared against sham rats on normal drinking water. e Time-course of left ventricular developed pressure (LVDP) in the hearts of sham vs MitoQ hearts with ouabain-induced Na elevation and washout period. f ATP and g ΔG_ATP during Na elevation in sham and MitoQ-treated hearts. Concentration of h ¹³C lactate, i ¹³C alanine, j ¹³C succinate, and k ¹H fumarate in sham vs MitoQ hearts during Na elevation and washout (w/o). n = 5 (for sham) and n = 6 hearts per group. Significance determined by nested (hierarchical) one-way ANOVA for d two-tailed, unpaired student’s t-test for (a, b) and (e–g) two-way ANOVA with post-hoc Sidak’s multiple comparisons test (h–k); *P < 0.05, **P < 0.01, ****P < 0.0001. Source data are provided as a Source data file.

Fig. 7. Acute [Na]_i elevation causes pseudohypoxia in hearts.

a HIF-1α protein expression measured by Western blotting and normalised to GAPDH loading control for control hearts and hearts treated with ouabain (high Na) for 20 min and 50 min. b Quantification of band density using ImageJ for control and ouabain treated hearts. n = 4 hearts per group. Data plotted as mean ± SEM. Significance determined by two-tailed, unpaired student’s t-test; *P > 0.05. c Characterisation of tissue hypoxia by monitoring the cardiac accumulation of the hypoxia selective radiotracer ⁶⁴Cu-CTS by real-time γ-detection. Radioactivity measured in counts per second (CPS). Numbered arrows denote time-point and number of injections of radiotracer. 0% O₂ used as positive control for hypoxia. n = 4 hearts for ouabain-treated group, n = 2 for time-matched control (time control) and positive control (hypoxia). d Baseline corrected ⁶⁴Cu retention in time-matched control, ouabain treated and hypoxic hearts calculated as percentage of injected peak. Data plotted as mean ± SEM and source data are provided as a Source data file.

Fig. 8. Proposed sequence of events leading to altered mitochondrial metabolism by elevated intracellular Na.

a Elevated cytoplasmic Na leads to activation of mitochondrial NCLX leading to decreased Ca in the mitochondrial matrix and reduced activity of Ca-sensitive mitochondrial enzymes including pyruvate dehydrogenase (PDH), isocitrate dehydrogenase (IDH), α-ketoglutarate dehydrogenase (KDH) and Complex III and V (ATP synthase). Ca-sensitive enzymes are indicated in red while Ca-insensitive enzymes are blue. The normal flow of electrons through the ETC complexes are accepted by CoQ to produce CoQH₂ at Complexes I and II. CoQH₂ is reoxidised back to CoQ at Complex III allowing electron transport to continue at Complexes I and II. b Under conditions of high cytoplasmic Na, matrix Ca will be decreased leading to reduced activity of Complex III, decreased recycling of CoQ and a build-up of CoQH₂. Excess electrons in the CoQH₂ pool can lead to ROS generation via forward electron transport (FET) or reverse electron transport (RET) at Complexes I and II. There are two possible mechanisms for the build-up of succinate, (i) reduced CoQ availability will decrease activity of Complex II resulting in decreased oxidation of succinate to fumarate, or (ii) RET at Complex II could reduce fumarate back to succinate in the reverse direction. The build-up of succinate will inhibit proly hydroxylases (PHDs) causing decreased degradation of HIF1α. c Pre-treatment in vivo with MitoQ loads the mitochondria with an exogenous antioxidant that is an analogue of CoQ. This could buffer the CoQ pool at Complex II by reducing MitoQ to MitoQH₂, limiting the build-up of succinate. MitoQH₂ is recycled to MitoQ by scavenging ROS, preventing the ROS-induced functional deficit on ouabain washout. Note: MitoQ can access the active site of Complex II but not Complexes I and III. Additional abbreviations: aconitase (A); citrate synthase (CS); cytochrome C (Cyt C); ubiquinone (CoQ); ubiquinol (CoQH₂); fumarase (F); malate dehydrogenase (MDH); mitochondrial pyruvate carrier (MPC); mitochondrial Na/Ca exchanger (NCLX); succinyl CoA synthetase (SCS); succinate dehydrogenase (SDH). Basic layout redrawn from Williams et al. (2015). Panel d links the observations made in this study to this model.