Abstract
Inhibition by cardiac glycosides of Na+, K+-ATPase reduces sodium efflux from myocytes and may lead to Na+ and Ca2+ overload and detrimental effects on mechanical function, energy metabolism, and electrical activity. We hypothesized that inhibition of sodium persistent inward current (late INa) would reduce ouabain's effect to cause cellular Na+ loading and its detrimental metabolic (decrease of ATP) and functional (arrhythmias, contracture) effects. Therefore, we determined effects of ouabain on concentrations of intracellular sodium (Na+i) and high-energy phosphates using 23Na and 31P NMR, the amplitude of late INa using the whole-cell patch-clamp technique, and contractility and electrical activity of guinea pig isolated hearts, papillary muscles, and ventricular myocytes in the absence and presence of inhibitors of late INa. Ouabain (1–1.3 μM) increased Na+i and late INa of guinea pig isolated hearts and myocytes by 3.7- and 4.2-fold, respectively. The late INa inhibitors ranolazine and tetrodotoxin significantly reduced ouabain-stimulated increases in Na+i and late INa. Reductions of ATP and phosphocreatine contents and increased diastolic tension in ouabain-treated hearts were also markedly attenuated by ranolazine. Furthermore, the ouabain-induced increase of late INa was also attenuated by the Ca2+-calmodulin-dependent kinase I inhibitors KN-93 [N-[2-[[[3-(4-chlorophenyl)-2-propenyl]methylamino]methyl]phenyl]-N-(2-hydroxyethyl)-4-methoxybenzenesulphonamide] and autocamide-2 related inhibitory peptide, but not by KN-92 [2-[N-(4′-methoxybenzenesulfonyl)]amino-N-(4′-chlorophenyl)-2-propenyl-N-methylbenzylamine phosphate]. We conclude that ouabain-induced Na+ and Ca2+ overload is ameliorated by the inhibition of late INa.
Introduction
Cardiac glycosides inhibit the sarcolemmal Na+, K+-ATPase (sodium pump) and increase intracellular sodium concentration ([Na+]i). The effect of a glycoside to increase [Na+]i may lead to an increase of calcium influx via sodium/calcium exchange (NCX) and an increase in cardiac contractility. However, whereas a small increase of [Na+]i may lead to a positive inotropic effect (Bers et al., 2003), a larger increase may lead to arrhythmias and contractile dysfunction. We hypothesized that reduction of Na+ entry in the presence of the cardiac glycoside ouabain would reduce sodium overloading and its adverse mechanical, metabolic, and electrical consequences. A novel approach to reduce Na+ entry is by reduction of persistent Na+ inward current (late INa) (Ver Donck et al., 1993; Hale et al., 2008).
Late INa is caused by entry of Na+ ions through myocyte Na+ channels that fail to inactivate normally. These channels stay open or reopen during the action potential plateau, when “normal” Na+ channels are inactivated, thereby contributing to intracellular Na+ loading (Makielski and Farley, 2006; Undrovinas and Maltsev, 2008). An increase of late INa caused by impaired Na+ channel inactivation is common in inherited (e.g., SCN5A mutations; Ruan et al., 2009) and acquired (e.g., ischemia, heart failure, remodeling, and oxidative states) pathological conditions (Undrovinas and Maltsev, 2008; for review see Hale et al., 2008). Although small in amplitude relative to peak INa, late INa persists throughout the duration of the cardiac action potential and contributes significantly to Na+ entry in myocytes (Makielski and Farley, 2006). The increase of Na+ entry caused by an enhanced late INa may lead to an increase of [Na+]i. Using a computational model, Noble (2008) found that reduction of late INa attenuated the rise of Na+i caused by repetitive 2-Hz stimulation of a ventricular myocyte. A rise of [Na+]i reduces the reversal potential of NCX and leads to Ca2+ loading of myocardial cells (Bers, 2001; Imahashi et al., 2005). Thus, inhibition of late INa is cardioprotective (Makielski and Farley, 2006; Hale et al., 2008; Sossalla et al., 2008).
Reduction of late INa can be achieved using either ranolazine or tetrodotoxin (TTX). The antianginal drug ranolazine is a relatively selective late INa inhibitor (Antzelevitch et al., 2004; Hale et al., 2008). Ranolazine reduces late INa with an approximate IC50 value of 6.5 μM (versus an IC50 value of 244 μM for inhibition of peak INa) and causes minimal or no inhibition of L-type Ca2+ channel current, NCX, or sodium proton exchange at therapeutic concentrations (≤ 10 μM) (Antzelevitch et al., 2004; Hale et al., 2008). In other studies it has been shown that ranolazine reduces sea anemone toxin-II- and H2O2-induced late INa in guinea pig and rabbit isolated ventricular myocytes and suppresses early and delayed afterdepolarizations and arrhythmic activity (Song et al., 2004, 2008). Ranolazine attenuates diastolic dysfunction in myocardium isolated from failing human hearts (Sossalla et al., 2008), sea anemone toxin-II-treated and ischemic/reperfused rat hearts (Fraser et al., 2006), and guinea pig hearts exposed to the ischemic metabolite palmitoyl-l-carnitine (Wu Y., et al., 2009). The limitation of ranolazine is that its selectivity for inhibition of late INa relative to human ether-a-go-go-related gene K+ current is only 2-fold (Hale et al., 2008). In contrast, TTX is very selective for Na+ channels relative to other ion channels, but has less selectivity for late relative to peak INa than does ranolazine. Both inhibitors were therefore used in this study to test the hypothesis that a reduction of endogenous late INa will reduce effects of ouabain to cause cellular Na+ loading and metabolic and contractile dysfunction. It also has been reported that a rise in intracellular Ca2+ and phosphorylation of the cardiac Na+ channel by Ca2+-calmodulin-dependent kinase II (CaMKII) can alter Na+ channel inactivation and enhance late INa (Maier and Bers, 2007; Hale et al., 2008; Maltsev et al., 2008; Song et al., 2008; Undrovinas and Maltsev, 2008; Xie et al., 2009). Because ouabain is known to increase intracellular Ca2+, we also determined the effects of inhibition of NCX and CaMKII on late INa and its attendant adverse functional consequences. The findings in this study were that ouabain itself led to an increase of late INa and that in the presence of inhibitors of late INa and CaMKII Na+i accumulation in the presence of ouabain was reduced, energy loss was prevented, and mechanical function was improved.
Materials and Methods
Animals and Isolated Perfused Heart Preparation.
Animal use protocols were approved by the Standing Committee on Animals of Harvard Medical Area and the Institutional Animal Care and Use Committees of CV Therapeutics (now Gilead Sciences, Palo Alto, CA), the University of South Carolina (Columbia, SC), and the University of Florida (Gainesville, FL). Animal use conformed to National Institutes of Health guidelines (National Institutes of Health publication 85-23, revised 1996).
Guinea pigs (Duncan Hartley, 250–350 g, either sex) were anesthetized (180 mg/kg sodium pentobarbital, intraperitoneally), and hearts were isolated and perfused in the isovolumic Langendorff mode at a constant pressure of 60 mm Hg with a modified Krebs-Henseleit buffer (37°C, pH 7.4) containing 118 mM NaCl, 4.8 mM KCl, 1.75 mM CaCl2, 1.2 mM MgSO4, 0.5 mM EDTA, 25 mM NaHCO3, 1.2 mM KH2PO4, 5.5 mM glucose, and 2 mM pyruvate, oxygenated with 95% O2/5% CO2. For experiments in which contractile function was measured, a fluid-filled balloon was inserted into the left ventricle and connected to a physiological pressure transducer (AD Instruments, Colorado Springs, CO). Hearts were stimulated at a rate of 5 Hz during all experiments (SD9 Square Stimulator; Grass Technologies, West Warwick, RI). Data were collected and analyzed as described previously (Shen et al., 2001), using a PowerLab system (Bridge Amp, 8sp interface, Chart 5.Pro software; ADInstruments, Colorado Springs, CO). For experiments in which NMR signals were measured, hearts were isolated and perfused as described above, suspended in a Varian Inova wide-bore spectrometer (Varian Inc., Palo Alto, CA), and paced at a rate of 5 Hz.
23Na and 31P NMR Spectroscopy for Measuring [Na+]i and High-Energy Metabolites in Guinea Pig Isolated Hearts.
For 23Na NMR, 590 free induction decay signals obtained from the Varian Inova spectrometer were acquired at 105.5 MHz and averaged over 2 min (90° pulse, 0.2-s recycle time). To distinguish intracellular from extracellular sodium, the shift reagent sodium thulium (III)1,4,7,10-tetraazacyclododecane-1,4,7,10-tetra(methylene phosphonate) (Na5TmDOTP) (3.5 mM) was added to the Krebs-Henseleit buffer. To determine [Na+]i, the peak areas of 23Na signals were compared with the peak area of a Na+ internal reference standard (Jansen et al., 2003).
For 31P-NMR, 125 free induction decay signals were acquired at 161.4 MHz and averaged over 5 min (60° pulse, 2.4-s recycle time). Cytosolic concentrations of ATP, phosphocreatine (PCr), and Pi were determined according to Shen et al. (2001).
Ouabain, Ranolazine, and Tetrodotoxin Concentrations.
Results of preliminary studies of guinea pig isolated perfused hearts indicated that concentrations of ouabain at 0.5 μM or above 1 μM were associated with either no dysfunction or rapid induction of arrhythmia, respectively. The ouabain concentration of 0.75 μM was therefore used in the majority of experiments. Ouabain induced two effects, a positive inotropic effect and a toxic effect (e.g., arrhythmia, elevated diastolic function).The shift reagent Na5TmDOTP reduced the effect of ouabain, presumably secondary to Ca2+-chelation. Therefore, in 23Na NMR experiments the ouabain concentration was increased to 1.3 μM to achieve an effect equivalent to that seen at 0.75 μM in the absence of Na5TmDOTP. For measurements of contractile function in guinea pig papillary muscle the ouabain concentration was increased to 2 μM, acknowledging that papillary muscle tissue is less ouabain-sensitive than either single cells or perfused hearts.
Ranolazine is a relatively selective late INa inhibitor. The ranolazine concentrations used in this study (3, 5, and 10 μM) are in the mid to high therapeutic range and are known to significantly reduce the late INa (Antzelevitch et al., 2004; Hale et al., 2008).
TTX reduces peak and late INa with IC50 values of 6.0 ± 0.2 and 0.5 ± 0.1 μM (Wu L. et al., 2009), respectively. At concentrations of 0.5 to 1 μM, TTX is a relatively selective late INa inhibitor.
Papillary Muscle Preparation and Tension Measurement.
Guinea pigs were anesthetized, and hearts were quickly removed and placed in an ice-cold Tyrode's solution containing 136 mM NaCl, 2.8 mM KCl, 1 mM CaCl2, 1.5 mM MgCl2, 0.3 mM NaH2PO4, 10 mM HEPES, 10 mM glucose, and NaOH to adjust the pH to 7.4. The right ventricular papillary muscle was dissected free, mounted in a 36.5 ± 0.5°C tissue bath in Tyrode's solution equilibrated with 100% 02, and electrically paced at a rate of 1 Hz. Muscle strips were equilibrated for 30 min with two changes of bathing solution and stretched stepwise with a micromanipulator to a rest length at which stimulated developed tension was maximal. Contractile force was measured isometrically using a force displacement transducer (TRI 201; LSi Letica Scientific Instruments, Barcelona, Spain) and digitized using a PowerLab system (ADInstruments).
Isolation of Ventricular Myocytes and Electrophysiological Recordings.
Single guinea pig ventricular myocytes were isolated using standard enzymatic procedures as described previously (Song et al., 2004). Transmembrane Na+ currents were measured with an Axopatch-200 amplifier, a Digidata-1440 digitizer, and pClamp-10 software (Molecular Devices, Sunnyvale, CA), using the whole-cell patch-clamp technique. The recording pipettes had a resistance of 2 to 3 MΩ when filled with a solution containing 120 mM Cs-aspartate, 20 mM CsCl, 1 mM MgSO4, 4 mM Na2ATP, 0.1 mM Na3GTP, and 10 mM HEPES, pH 7.2, and the series resistance was compensated by approximately 85%. Late INa was activated using 300-ms voltage-clamp pulses from −90 to −50 mV at a frequency of 0.16 Hz. Transmembrane current during the last 100 ms of depolarizing pulse was integrated and expressed as nanocoulombs (nC) or picocoulombs (pC). Cell membrane capacitance was minimized using the amplifier, and values of capacitance compensation in picofarads (pF) were used to normalize the integrated current to the magnitude of the membrane capacitative current (pC/pF). During experiments, myocytes were superfused with a bath solution (36°C) containing 135 mM NaCl, 4.6 mM CsCl, 1.8 mM CaCl2, 1.1 mM MgSO4, 0.01 mM nitrendipine, 0.1 mM BaCl, 10 mM glucose, and 10 mM HEPES, pH 7.4. Barium was present in the bath solution to reduce potential contamination of late INa by IK1.
In selected experiments with isolated myocytes, KN-93 [N-[2-[[[3-(4-chlorophenyl)-2-propenyl]methylamino]methyl]phenyl]-N-(2-hydroxyethyl)-4-methoxybenzenesulphonamide], KN-92 [2-[N-(4′-methoxybenzenesulfonyl)]amino-N-(4′-chlorophenyl)-2-propenyl-N-methylbenzylamine phosphate], or EGTA were included in the recording pipette solution to avoid the CaMKII-independent effects of KN-93 that are reported to occur when the compound is applied extracellularly (Rezazadeh et al., 2006). Ouabain, TTX, and ranolazine were applied extracellularly via the bath solution.
Chemicals.
Ranolazine was provided by CV Therapeutics, and KN-92, KN-93, and myristoylated autocamtide-2 related inhibitory peptide (AIP) were obtained from Calbiochem (La Jolla, CA). The shift reagent Na5TmDOTP was purchased from Macrocyclics (Dallas, TX). All other drugs and reagents were obtained from Sigma-Aldrich (St. Louis, MO).
Statistics.
Results are expressed as mean ± S.E.M. Data were analyzed by one-way analysis of variance (Prism 5.01; GraphPad Software, Inc., San Diego, CA) or analysis of variance with repeated measures (Statistica 8.0; StatSoft, Tulsa, OK), followed by a post hoc test (e.g., Tukey's test) when significant differences were observed. A p value < 0.05 was considered to indicate a significant difference.
Results
Changes in Contractile Function of the Isolated Heart during Ouabain-Induced Sodium Pump Inhibition in the Absence and Presence of Ranolazine or TTX.
Exposure to 0.75 μM ouabain for 60 min led to a transient increase of left ventricular systolic pressure (LVSP) by 55 ± 5% (n = 13), followed by arrhythmic activity and episodes of cardiac standstill (i.e., absence of a contractile response during continuous electrical pacing at 5 Hz) alternating with periods of rhythmic contraction in 11 of 13 hearts tested (Fig. 1A). A marked elevation of left ventricular end diastolic pressure (LVEDP) and a decrease of left ventricular developed pressure (LVDevP) were observed (Fig. 1A). The late INa inhibitors ranolazine and TTX reduced the occurrence of episodes of cardiac standstill and the rise of LVEDP caused by ouabain. Of eight hearts treated with 0.75 μM ouabain + 3 μM ranolazine, four hearts showed episodes of cardiac standstill including elevated LVEDP (Fig. 1B), whereas the remaining four hearts maintained enhanced but irregular contraction. The responses of hearts that were exposed to ouabain in the presence of 0.5 μM TTX were comparable with those exposed to ouabain in the presence of 3 μM ranolazine. When the concentration of ranolazine was increased to 5 μM, hearts treated with 0.75 μM ouabain (n = 6; Fig. 1C) did not have episodes of cardiac standstill, although biphasic contractions were occasionally observed and LVEDP was slightly, but not significantly, elevated. In hearts treated with either 10 μM ranolazine (n = 8; Fig. 1D) or 1 μM TTX (n = 6; not shown) throughout the 60-min duration of ouabain exposure, neither episodes of cardiac standstill or biphasic contractions nor changes in LVEDP were observed. Ranolazine (Fig. 1, B–D) and TTX (data not shown) alone caused slight concentration-dependent decreases of LVSP but did not inhibit the positive inotropic response to ouabain (Fig. 1). LVSP decreased by 8 and 16% to 17% (n = 6–8 each) during treatment with ranolazine alone (3, 5, or 10 μM, respectively; p < 0.05) and by 9 and 18% (n = 5–6) during treatment with TTX alone (0.5 and 1 μM, respectively; p < 0.04). The values of rate pressure product in the presence of 0.75 μM ouabain with or without ranolazine or TTX increased significantly and were not different from each other (Tables 1 and 2). Hearts exposed to ouabain in the presence of 3, 5, or 10 μM ranolazine or 1 μM TTX also showed better recovery of contractile function (higher developed pressure) after drug washout than hearts treated with ouabain alone (Fig. 1). The effects of drug treatment on values of contractile parameters are summarized in Tables 1 and 2.
Changes in Contractile Function in Papillary Muscle Preparations during Ouabain-Induced Sodium Pump Inhibition in the Absence and Presence of Ranolazine.
Contractile function of the guinea pig papillary muscle was measured to confirm results of experiments performed using the isolated heart. A concentration of 2 μM ouabain increased developed tension of papillary muscles by almost 4-fold after 20 min, from 35.4 ± 7.7 (n = 8) to 133.9 ± 37.0 mg (Fig. 2, A and C). The developed tension of papillary muscles declined with time during a 60-min exposure to ouabain, and episodes of tachyarrhythmia were often observed (Fig. 2, A and C). Diastolic tension of papillary muscle strips also increased significantly by 15% after a 1-h ouabain treatment (Fig. 2E). Ranolazine attenuated the ouabain-induced contractile/electrical dysfunction in guinea pig papillary muscle preparations (Fig. 2, B, D, and E). Ranolazine (10 μM) alone had no significant effect on developed tension and did not seem to decrease the effect of ouabain to increase developed tension (Fig. 2, B and D). In muscles pretreated with 10 μM ranolazine, the addition of 2 μM ouabain also caused a 4-fold increase in contractile force from 51.0 ± 6.7 to 202.9 ± 44.0 mg, but tachyarrhythmias were not observed and diastolic force was not significantly increased compared with control.
Changes in [Na+]i during Sodium Pump Inhibition in the Absence and Presence of Ranolazine or TTX.
[Na+]i of the guinea pig isolated, perfused heart in the absence of drug was 6.9 ± 0.6 mM (n = 9; Fig. 3A), as determined by 23Na NMR spectroscopy in the presence of the shift reagent Na5TmDOTP. After perfusion of the heart with ranolazine (10 μM) for 30 min, [Na+]i was unchanged, 6.5 ± 0.4 mM (n = 5, p > 0.1 versus control; Fig. 3B). Upon exposure of the heart to 1.3 μM ouabain, [Na+]i increased rapidly by 3.7-fold at 60 min to reach a plateau level of 25.1 ± 1.2 mM (n = 9, p < 0.001 versus control; Fig. 3C). After washout of ouabain for 20 min, [Na+]i was 10.9 ± 1.2 mM (n = 8, p < 0.001 versus plateau level, p < 0.05 versus control), indicating that the ouabain effect was at least partially reversible. The 1.3 μM ouabain-induced increase of [Na+]i could be attenuated by treatment of hearts with either ranolazine (10 μM) or TTX (1 μM) for 10 min before and during the exposure to ouabain (Fig. 3, C and D). During treatment of hearts with 1.3 μM ouabain in the presence of either 10 μM ranolazine (n = 9) or 1 μM TTX (n = 5), values of [Na+]i reached plateau concentrations of 15.6 ± 0.1 or 10.5 ± 0.1 mM, respectively (p < 0.001 ranolazine or TTX versus ouabain alone). The decrease in the ouabain-induced rise of [Na+]i by ranolazine and TTX was concentration-dependent (Fig. 3D).
Changes in Energy-Related Phosphates during Sodium Pump Inhibition in the Absence and Presence of Ranolazine.
One of the consequences of Na+ and Ca2+ overload is a mismatch of energy supply and demand. Therefore, we measured changes of energy-related phosphates with 31P NMR spectroscopy in ouabain-treated guinea pig isolated, perfused hearts in the absence and presence of ranolazine. Under control conditions [ATP], [PCr], and [Pi] were 10 ± 0.1, 18 ± 0.5, and 3 ± 0.2 mM, respectively (n = 18 each; Fig. 4), and intracellular pH (pHi) was 7.15 ± 0.01 (n = 18). Exposure of hearts to ranolazine alone for 10 min (n = 5) did not alter either the concentrations of phosphates or pHi. After exposure to 0.75 μM ouabain for 60 min, [ATP] and [PCr] declined by 53 ± 7 and 49 ± 5%, respectively, [Pi] increased by 3.6 ± 1-fold (from 3 ± 0.2 to 10.4 ± 1.3 mM; all n = 5), and pHi declined to 7.07 ± 0.01 (Fig. 4A). Values of pHi and [Pi] recovered fully or partially during a 20-min washout period; pHi returned to 7.15 (control) and [Pi] decreased from 10.4 ± 1.3 to 6.6 ± 0.8 mM (p < 0.04; Fig. 4A). In hearts treated with 0.75 μM ouabain in the presence of 10 μM ranolazine, [ATP] and [PCr] did not change significantly after 60-min ouabain treatment (Fig. 4A). The value of [Pi] increased slightly, but not significantly, from 3.7 ± 0.3 to 4.8 ± 0.2 mM in hearts exposed to ouabain in the presence of ranolazine (p > 0.05 versus control). Ranolazine (10 μM; Fig. 4A) also attenuated the ouabain-induced decrease of pHi (7.13 ± 0.01 versus 7.07 ± 0.01; p < 0.05). In summary, inhibition of late INa effectively prevented or reduced the ouabain-induced decreases in high-energy phosphates and pHi and the increase in [Pi].
Ouabain-Induced Late INa.
To determine whether ouabain has an effect on sodium channels we measured the amplitude of late INa in guinea pig isolated ventricular myocytes exposed to ouabain in the absence and presence of ranolazine. The amplitude of late INa was increased by the exposure of cells to ouabain (1 μM). After a 3- to 5-min exposure of myocytes to ouabain, the integrated late INa was increased by 4.2-fold from 23.5 ± 4.9 to 99.6 ± 15.2 pC/pF (n = 8, p < 0.001; Fig. 5, A–C). Ranolazine (10 μM) applied to cells in the continuous presence of ouabain reduced late INa by 69 ± 9%, from 99.6 ± 15.2 to 50.6 ± 13.6 pC/pF (n = 8, p < 0.001; Fig. 5, A and C). In some experiments, after washout of ranolazine, cells were exposed to TTX (3 μM, n = 6; Fig. 5B). Ouabain-induced late current was completely inhibited by 3 μM TTX, to 21.2 ± 7.9 pC/pF (p < 0.001), indicating that this current was a Na+ channel current (e.g., NaV1.5).
The ouabain-induced increase of intracellular Na+ may lead to Ca2+ uptake and activation of CaMKII. To examine the hypothesis that a Ca2+-dependent, CaMKII-mediated mechanism may underlie the effect of ouabain to increase late INa, cells were incubated with ouabain when either the CaMKII inhibitor KN-93 (10 μM) or the Ca2+ chelator EGTA (1 mM) was dialyzed into them by inclusion in the patch pipette solution. KN-92 (10 μM), an inactive analog of KN-93, was used as a control. It has previously been shown that KN-93 applied intracellularly selectively blocks ion channels (Rezazadeh et al., 2006).
Ouabain alone (1 μM, n = 6) caused a time-dependent increase of late INa by 318 ± 74% from 21 ± 2 to 84 ± 12 pC/pF (p = 0.003) in 5 to 10 min (Fig. 6, A–C). In comparison, at the end of a 10-min exposure to ouabain in the presence of intracellular KN-93 late INa was increased by only 76 ± 35% (from 21 ± 2 to 33 ± 6 pC/pF; n = 7, p = 0.003 versus ouabain alone). In contrast, in the presence of the inactive analog KN-92 late INa at the end of a 10-min exposure to ouabain was increased by 273 ± 39% (from 20 ± 1 to 72 ± 7 pC/pF; n = 6, p > 0.05 versus ouabain alone, and p < 0.01 versus KN-93). The intracellular application of 1 mM EGTA (a Ca2+-chelating agent) via the patch pipette before a 10-min exposure of isolated myocytes to ouabain also attenuated the ouabain-induced increase of late INa: late INa increased by only 33 ± 28% (from 23 ± 3 to 31 ± 8 pC/pF; n = 6, p < 0.001 versus ouabain alone; Fig. 6D). A similar intracellular application of 1 mM EGTA has been shown to reduce the effect of Ca2+ to induce delayed afterdepolarizations in myocytes (Song et al., 2008), indicating that this application of EGTA is effective to attenuate an action mediated by a rise of intracellular Ca2+, presumably by reduction of [Ca2+]i itself.
Changes in Cardiac Contractility, High-Energy Phosphates, and [Na+]i during Sodium Pump Inhibition in the Presence of CaMKII Inhibitors.
The finding that not only late INa inhibitors but also CaMKII inhibitors reduced ouabain-induced late INa in myocytes suggested that CaMKII inhibitors may also improve function in the isolated guinea pig heart exposed to ouabain. KN-93 (1.8 μM) alone decreased LVSP by 24 ± 1.1% (p < 0.03, n = 5; Fig. 1E), whereas AIP (0.3 μM, n = 2) alone had no measurable effect on cardiac contractility. Exposure of hearts to ouabain in the presence of the CaMKII inhibitors KN-93 or AIP resulted in an increase of contractile amplitude (Table 2). In contrast to hearts treated with ouabain alone, however, hearts exposed to ouabain in combination with CaMKII inhibitors experienced neither episodes of cardiac standstill nor an elevation of LVEDP (Table 2). Thus, inhibition of either late INa or CaMKII caused similar reductions of both late INa and electrical/contractile dysfunction in the presence of ouabain.
The CaMKII inhibitors KN-93 and AIP also attenuated the effects of ouabain on high-energy phosphates, [Pi], and pHi. Exposure of hearts to KN-93 (1.8 μM, n = 3–6) or AIP (0.3 μM, n = 2) alone for 10 min did not alter the concentrations of phosphates or pHi (data not shown). During exposure to 0.75 μM ouabain in the presence of either 1.8 μM KN-93 or 0.3 μM AIP, [ATP] and [PCr] did not change significantly from baseline after 60 min (not shown). The values of [Pi] increased slightly, but not significantly, from 2.5 ± 0.2 to 4.5 ± 0.8 mM and from 2.5 ± 0.2 mM to 3.9 ± 0.03 mM (p > 0.05 versus control), and the values of pHi decreased from 7.15 ± 0.01 to 7.13 ± 0.01 and 7.14 ± 0.01 in hearts treated with ouabain in the presence of KN-93 or AIP, respectively (p < 0.05 versus ouabain alone).
Finally, the effect of KN-93 on [Na+]i in hearts exposed to ouabain was determined. KN-93 (2 μM) alone did not significantly alter [Na+]i. The concentrations of intracellular sodium in the absence and presence of KN-93 were 7.3 ± 0.4 and 6.9 ± 0.6 mM, respectively (p > 0.05, n = 4). The increase of [Na+]i in hearts exposed to 1.3 μM ouabain was significantly reduced in the presence of KN-93 from 25.1 ± 1.2 mM in hearts treated with ouabain alone to 18.9 ± 1.8 mM in hearts treated with ouabain in the presence of 2 μM KN-93 (p < 0.05). In summary, the deleterious effects of ouabain on cardiac function (contractility, energy metabolism, intracellular sodium) were diminished by either CaMKII or late INa inhibitors.
To exclude the possibility that either ranolazine or the CaMKII inhibitors KN-93 or AIP had a direct effect on the sodium pump, three different concentrations of each inhibitor were tested in a Na+, K+-ATPase activity assay (Chassande et al., 1988) by measuring the 86Rb+ uptake of A7r5 cells in the presence of ouabain with or without ranolazine or CaMKII inhibitor. The activity of Na+, K+-ATPase was inhibited 77% by 1 mM ouabain in the absence (control) of either inhibitor. Values of 86Rb+ uptake were 91 ± 11, 98 ± 2.5, and 93 ± 8.3% of control (activity in presence of ouabain) in the presence of 3, 10, and 30 μM ranolazine, respectively. Neither KN-93 nor myristoylated AIP had significant effects on Na+, K+-ATPase activity in this assay. Values of 86Rb+ uptake were 90 ± 4, 92 ± 14, and 87 ± 20% of control in the presence of KN-93 (0.2, 2, and 5 μM, respectively) and 98 ± 10, 89 ± 12, and 106 ± 11% of control in the presence of AIP (0.03, 0.3, and 0.8 μM, respectively).
Discussion
The results presented here suggest that a reduction of late INa attenuates sodium accumulation and metabolic, contractile, and electrical dysfunction induced by the cardiac glycoside ouabain in the guinea pig isolated perfused heart and papillary muscle. Ouabain markedly increased [Na+]i and [H+]i and decreased [ATP] and [PCr] in the heart. Ranolazine (10 μM) and TTX (1 μM) at concentrations reported to inhibit late INa (Song et al., 2008) significantly reduced the rise in [Na+]i and attenuated the losses of [ATP] and [PCr] and the decrease of pHi that were observed in the presence of ouabain alone. Ranolazine (5 and 10 μM), TTX (1 μM), KN-93 (1.8 μM), and AIP (0.3 μM) all prevented the rise of LVEDP and reduced occurrences of cardiac standstill caused by ouabain in the isolated perfused heart, and ranolazine attenuated the increase of diastolic tension of isolated guinea pig papillary muscles during ouabain treatment.
Changes in [Na+]i during Sodium Pump Inhibition in the Absence and Presence of Ranolazine, TTX, and KN-93.
The concentration of Na+ in resting heart cells of many mammals is in the range of 4 to 8 mM (Bers et al., 2003). In this study using 23Na NMR spectroscopy, [Na+]i was found to be ∼7 mM in guinea pig isolated hearts paced at 5 Hz, consistent with literature reports (Jelicks and Siri, 1995; Hotta et al., 1998). Treatment of hearts with 10 μM ranolazine for up to 30 min or 1 μM TTX for 10 min did not significantly change [Na+]i (Fig. 3, B and C). This finding suggests that physiological late INa is either a small contributor to sodium entry in the beating isolated heart or a decrease of Na+ influx via late INa does not lead to reduction of [Na+]i because the reserve capacity of the Na+, K+-ATPase to extrude Na+ from the cell is not normally exceeded (Akera and Ng, 1991). In this study, ouabain (1.3 μM, in the presence of the NMR shift reagent and Ca2+ chelator Na5TmDOTP) led to a 3.7-fold increase in [Na+]i (Fig. 3C). Ranolazine (10 μM) and TTX (1 μM) as well as KN-93 (2 μM) significantly attenuated the increase of [Na+]i caused by ouabain (Fig. 3D), suggesting that an enhancement of persistent Na+ current (late INa) by ouabain (Fig. 5), was a factor contributing to the increase of [Na+]i.
Ouabain-Induced Late INa.
A novel finding in this study is that ouabain increased late INa in guinea pig isolated ventricular myocytes. Ranolazine and TTX as well as intracellular applications of the CaMKII inhibitor KN-93 or the Ca2+-chelator EGTA all reduced late INa in the presence of ouabain (Figs. 5 and 6) and attenuated the ouabain-induced increase of [Na+]i (Fig. 3). These findings suggest that an increased late INa contributes to elevation of [Na+]i in the whole heart (Makielski and Farley, 2006; Noble, 2008) and elevation of [Ca2+]i and/or activity of CaMKII are potential causes of the increase of late INa that occurs during exposure of cardiac myocytes to ouabain. This interpretation is supported by results of previous studies showing that glycosides increased both [Na+]i and [Ca2+]i in the heart and activated CaMKII (Sapia et al., 2010) and that Ca2+/calmodulin/CaMKII may directly regulate the function of the cardiac Na+ channel to increase late INa (Wagner et al., 2006; Maltsev et al., 2008; Bers and Grandi, 2009; Aiba et al., 2010 and references therein). An increase of late INa itself leads to Ca2+ overload (Maier and Bers, 2007; Xie et al., 2009) to close a positive feedback loop between increases of Ca2+ and late INa. Furthermore, it has been reported that ouabain can stimulate reactive oxygen species production by the Na+, K+ ATPase (Xie et al., 1999; Liu et al., 2000), and an increase of reactive oxygen species is reported to activate CaMKII by both Ca2+-dependent and -independent pathways (Palomque et al., 2009; Xie et al., 2009). Thus, there are several potential mechanisms by which the ouabain-induced increase of intracellular Na+ may lead to Ca2+ dysregulation and altered cell function.
We suggest that the effect of ouabain on cardiac Na+/Ca2+ homeostasis and cardiac function has at least two components: first, the rise of [Na]i caused by decreased Na+ efflux due to inhibition of Na+, K+-ATPase; second, the rise of [Na]i caused by an enhanced late INa, which leads to a further increase of Na+ influx. In the present work we sought to diminish the latter component in three ways: use of a late INa inhibitor, inhibition of CaMKII, and reduction of Ca2+ overload with EGTA. Each of these interventions (e.g., ranolazine, TTX, KN-93, and EGTA) reduced late INa. However, we are not able to distinguish how much of the increase in intracellular sodium comes from the late sodium current versus the sodium pump inhibition because inhibition of the sodium pump and increase of late INa probably act synergistically. More importantly, each of the interventions also reduced Na+ accumulation, loss of [ATP] and [PCr], and electrical and mechanical dysfunction caused by ouabain. These findings suggest that late INa plays a role in glycoside-induced cardiac dysfunction, and that either a direct (by TTX or ranolazine) or indirect inhibition of late INa is cardioprotective when [Na+]i is elevated as a result of glycoside-induced inhibition of the Na+, K+-ATPase.
In addition to ranolazine and TTX, the putative late INa inhibitor R56865 [N-[1-[4-(4-fluorophenoxy)-butyl]-4-piperidinyl]-N-methyl-2- benzothiazolamine] is reported to reduce Na+ and Ca2+ overload and improve electrical and mechanical function (e.g., reductions of arrhythmic activity and contracture) during exposure of cardiac tissues to cardiac glycosides (Ver Donck et al., 1993; Watano et al., 1999). Inhibition of NCX with KB-R7943 (2-[2-[4-(4-nitrobenzyloxy)phenyl]ethyl]isothiourea) in isolated guinea pig atria exposed to ouabain has also been shown to reduce Ca2+ overload pathology (Watano et al., 1999). The mitochondrial NCX inhibitor CGP-37157 [7-chloro-5-(2-chlorophenyl)-1,5-dihydro-4,1-benzothiazepin-2(3H)-one] also improved electrical and mechanical function in isolated guinea pig hearts as well as energy metabolism in myocytes during concomitant exposure to ouabain and isoproterenol (Liu et al., 2010). Taken together, these results indicate that strategies to prevent a pathological increase in late INa, and/or its downstream effects, may be cardioprotective.
Changes in High-Energy-Related Phosphates and Contractile Performance during Sodium Pump Inhibition in the Absence and Presence of Ranolazine, KN-93, and AIP.
By inducing increases of [Na+]i and [Ca2+]i, ouabain had a positive inotropic effect in the guinea pig isolated heart and papillary muscle preparations. This effect was transient and was followed by mechanical, electrical, and metabolic dysfunction, including a rise of LVEDP, a decrease in left ventricular systolic function, episodes of cardiac standstill (contracture, inexcitability), and a pronounced loss of approximately 50% of ATP and PCr. Sodium-induced Ca2+ overload is known to lead to a mismatch of energy demand and supply in the heart (Hotta et al., 1998; O'Rourke and Maack, 2007). Energy demand increases caused by activation of myosin ATPase, sarcoplasmatic reticulum Ca2+ ATPase, the sarcolemmal Ca2+ ATPase, and the sodium pump. ATP synthesis may be reduced because of Na+ and Ca2+ overload (Balaban, 2002; O'Rourke and Maack, 2007) as the activity of Ca2+-dependent Krebs cycle dehydrogenases (pyruvate, isocitrate, and α-ketoglutarate) is reduced when the mitochondrial Ca2+ level falls in response to increased mitochondrial NCX driven by elevation of intracellular Na+ (Maack et al., 2006; Kohlhaas et al., 2010). The mismatch of energy demand and supply results in decreases in [ATP] and [PCr], increases in [ADP], [Pi], and cellular acidosis, and ultimately in a decrease of the free energy available from hydrolysis of ATP.
Limitations of this study are that we have not measured the level of intracellular Ca2+, and the CaMKII inhibitors KN-93 and AIP may not have been completely selective. Off-target effects of KN-93 include inhibition of L-type calcium and potassium channels (Xie et al., 2009 and references therein), which may reduce Ca2+ overload and arrhythmia. Furthermore, an effect of AIP to prolong the duration of the myocyte action potential (presumably a nonspecific peptide effect; Xie et al., 2009) may counteract any potential shortening of action potential duration caused by ouabain (Lee and Klaus, 1971). Any of the above may potentially reduce ouabain toxicity independently of CaMKII inhibition.
In summary, our results show that ouabain stimulated an increase of late INa in guinea pig myocytes, and direct and indirect inhibition of late INa attenuated the ouabain-induced Na+ overload and metabolic, electrical, and mechanical dysfunction in the guinea pig isolated heart and papillary muscle. The effects of an enhanced late INa to cause Ca2+ overload and both electrical and mechanical dysfunction of the heart suggest that late INa has a key pathophysiological role in the heart. Therefore, inhibition of late INa in the diseased heart and during digitalis therapy may be of clinical relevance.
Authorship Contributions
Participated in research design: Hoyer, Song, Wang, Phan, Ingwall, Belardinelli, and Shryock.
Conducted experiments: Hoyer, Song, Wang, Phan, and Balschi.
Performed data analysis: Hoyer, Song, and Wang.
Wrote or contributed to the writing of the manuscript: Hoyer, Song, Wang, Ingwall, Belardinelli, and Shryock.
Footnotes
This project was supported by CV Therapeutics, a recent acquisition of Gilead Sciences, Inc., Palo Alto, CA. K.H., D.P., L.B., and J.C.S. are employees of Gilead Sciences, and J.B., J.S.I., Y.S., and D.W. have received funding from CV Therapeutics.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
doi:10.1124/jpet.110.176776.
-
ABBREVIATIONS:
- [Na+]i
- intracellular sodium concentration
- AIP
- autocamtide-2 related inhibitory peptide
- CaMKII
- Ca2+-calmodulin-dependent kinase II
- late INa
- late sodium current (persistent sodium current)
- LVDevP
- left ventricular developed pressure
- LVEDP
- left ventricular end diastolic pressure
- LVSP
- left ventricular systolic pressure
- Na5TmDOTP
- sodium thulium (III) 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetra(methylene phosphonate)
- NCX
- sodium/calcium exchange
- PCr
- phosphocreatine
- pHi
- intracellular pH
- TTX
- tetrodotoxin
- pC
- picocoulomb
- pF
- picofarad
- KN-93
- N-[2-[[[3-(4-chlorophenyl)-2-propenyl]methylamino]methyl]phenyl]-N-(2-hydroxyethyl)-4-methoxybenzenesulphonamide
- KN-92
- 2-[N-(4′-methoxybenzenesulfonyl)]amino-N-(4′-chlorophenyl)-2-propenyl-N-methylbenzylamine phosphate
- R56865
- N-[1-[4-(4-fluorophenoxy)-butyl]-4-piperidinyl]-N-methyl-2-benzothiazolamine
- KB-R7943
- 2-[2-[4-(4-nitrobenzyloxy)phenyl]ethyl]isothiourea
- CGP-37157
- 7-chloro-5-(2-chlorophenyl)-1,5-dihydro-4,1-benzothiazepin-2(3H)-one.
- Received November 12, 2010.
- Accepted February 14, 2011.
- Copyright © 2011 by The American Society for Pharmacology and Experimental Therapeutics