Although the exact role of hydrogen peroxide (H₂O₂) and other reactive oxygen species (ROS) in pathological processes is still under investigation, increasing evidence strongly suggests a link between ROS and ischemia-reperfusion injury (Dhalla et al., 2000; Turoczi et al., 2003; Paradies et al., 2004), “stunned myocardium”, and the development and progression of heart failure (Bolli and Marbán, 1999; Sawyer et al., 2002; Zeitz et al., 2002; Liu et al., 2005). The myocardial tissue concentration of H₂O₂ is significantly elevated in hearts exposed to ischemia-reperfusion (Slezak et al., 1995) and in the failing heart (Ide et al., 2000). ROS seem to play a major role in the dysregulation of [Na⁺]_i and [Ca²⁺]_i in ischemia/reperfusion injury (Bolli and Marbán, 1999; Zeitz et al., 2002). This effect of ROS is accompanied by cellular electrical instability (e.g., arrhythmias) and contractile dysfunction characterized by a marked increase in diastolic tension (Hara et al., 1993; Zeitz et al., 2002). The cellular Ca²⁺ overload caused by ROS has been proposed to be due to a rise in [Na⁺]_i followed by Ca²⁺ influx via the reverse mode of the Na⁺-Ca²⁺ exchanger (NCX) (Wagner et al., 2003). As to the mechanism of the ROS-induced rise in [Na⁺]_i, there is evidence in support of an increased entry of Na⁺ via noninactivating Na⁺ channels (Ward and Giles, 1997), an enhanced Na⁺-H⁺ exchanger activity (Sabri et al., 1998), and an impaired Na⁺-K⁺-ATPase function (Kim and Akera, 1987). The amplitude of late Na⁺ current (via noninactivating Na⁺ channels) and [Na⁺]_i and [Ca²⁺]_i are significantly increased in myocytes isolated from ischemic (Kihara et al., 1989; Haigney et al., 1994; Huang et al., 2001) or failing hearts (Pogwizd et al., 2003; Valdivia et al., 2005). It has been shown that the increase in [Ca²⁺]_i caused by ROS is attenuated by an NCX inhibitor (Zeitz et al., 2002) and is exacerbated in cells overexpressing NCX (Wagner et al., 2003). Furthermore, hypoxia-induced Na⁺ and Ca²⁺ loading of cardiac myocytes can be reduced by blockade of I_Na (Haigney et al., 1994). A rise of [Ca²⁺]_i caused by hypoxia can also be prevented by inhibition of the Na⁺-Ca²⁺ exchanger (Ziegelstein et al., 1992). Interestingly, verapamil does not attenuate the increase in [Ca²⁺]_i caused by ROS, indicating that the increase in Ca²⁺ entry into myocardial cells is not through L-type Ca²⁺ channels (Zeitz et al., 2002). Thus, the Ca²⁺ overload caused by ROS is probably due to a rise in [Na⁺]_i that in turn leads to an increased exchange of intracellular Na⁺ for extracellular Ca²⁺ via NCX.

The objective of the present study was to determine the contribution of the late Na⁺ current (late I_Na) to the rises in [Na⁺]_i and [Ca²⁺]_i and to the accompanying electrical and contractile dysfunctions caused by H₂O₂. To determine the role of late I_Na, we used low concentrations (<10 μM) of the putative Na⁺ channel blocker tetrodotoxin (TTX), and of ranolazine (Ran), a cardioprotective agent that preferentially inhibits late relative to peak I_Na (Undrovinas et al., 2006). Ranolazine has previously been shown to markedly attenuate, in a concentration-dependent manner, the increase in left ventricular diastolic pressure caused by H₂O₂ in rat isolated, perfused hearts (Matsumura et al., 1998).

Materials and Methods

Chemicals. H₂O₂ was purchased from Fisher (Fairlawn, NJ) and Merck (Darmstadt, Germany). TTX was purchased from Sigma-Aldrich (St. Louis, MO). Ranolazine [(±)-N-(2,6-dimethylphenyl)-(4[2-hydroxy-3-(2-methoxyphenoxy)propyl]-1-piperazine], a piperazine derivative (Matsumura et al., 1998), was synthesized by CV Therapeutics, Inc. (Palo Alto, CA).

Isolation of Ventricular Myocytes. Electrophysiological studies were performed on guinea pig ventricular myocytes at the University of Florida (Gainesville, FL); intracellular Na⁺ and Ca²⁺ measurements were conducted on rabbit ventricular myocytes at Georg-August-University Göttingen (Göttingen, Germany). Use of animals was in accordance with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH publication 85-23, revised 1996) and was approved by the Institutional Animal Care and Use Committee of the University of Florida. Ventricular myocytes were isolated from adult female Hartley guinea pigs and Chinchilla bastard rabbits with standard enzymatic procedures described previously (Wagner et al., 2003; Song et al., 2004).

Measurements of Transmembrane Potential and Current. Guinea pig ventricular myocytes were placed into a recording chamber that was bathed with prewarmed (35-36°C) Tyrode's solution, without or with drug(s), at a rate of 2 ml/min. The Tyrode's solution contained 135 mM NaCl, 4.6 mM KCl, 1.8 mM CaCl₂, 1.1 mM MgSO₄, 10 mM glucose, and 10 mM HEPES, pH 7.4. The temperature of the bathing media in a given experiment did not vary more than 0.3°C. Transmembrane voltages and currents were recorded from quiescent, rod-shaped myocytes with clear striations, using borosilicate glass pipettes (1- to 3-MΩ resistance when filled) in a whole-cell configuration of the patch-clamp technique. An Axopatch-200 amplifier, a DigiData interface, and a computer with pCLAMP software (Axon Instruments, Union City, CA) were used to amplify, store, and analyze the recorded signals. The electrode capacitance and whole-cell capacitance currents were maximally compensated with the amplifier. The series resistance was compensated by 60 to 80%. The liquid junction potential between pipette and bath medium was calculated with pCLAMP software and corrected online.

When measuring action potentials, recording pipettes were filled with a solution containing 120 mM potassium aspartate, 20 mM KCl, 1 mM MgSO₄, 4 mM Na₂ATP, 0.1 mM Na₃GTP, and 10 mM HEPES, pH 7.2. For inducing action potentials, a 5-ms depolarizing pulse was applied at a frequency of 0.16 Hz. The duration of the action potential was measured from onset of upstroke to 50% of repolarization (APD₅₀).

To elicit late I_Na, myocytes were voltage-clamped at a holding potential of -90 mV, and a 300-ms depolarizing pulse to -30 mV was applied at a frequency of 0.16 Hz. In experiments to determine the effect of ranolazine on late I_Na, K⁺ and Ca²⁺ were omitted from the bath and pipette solutions to reduce contamination of I_Na by K⁺ and Ca²⁺ currents. In these experiments, recording pipettes were filled with 120 mM cesium aspartate, 20 mM CsCl, 1 mM MgSO₄, 4 mM Na₂ATP, 0.1 mM Na₃GTP, and 10 mM HEPES, pH 7.2. The magnitude of late I_Na was determined by integration of the current over the last 50 ms of the -30-mV clamp pulse, using the integration (area) feature of the pCLAMP program.

Measurement of Cell Contraction. Twitch shortenings of guinea pig ventricular myocytes were elicited by field stimulation from a Grass S88 (Quincy, MA) stimulator. The amplitude of twitch shortening was determined by a video-motion detector (Crescent Electronics, Logan, UT) and was recorded on a chart recorder (2200S; Gould, Cleveland, OH). In this study, a twitch shortening denotes a normal systolic contraction, whereas an early aftercontraction denotes an additional contraction that occurs during relaxation and is triggered by events following the preceding normal contraction. The amplitude of twitch shortening was measured from maximal cell relaxation to peak contraction, and the rate of relaxation was calculated by dividing the amplitude (micrometers) with the time (seconds) required from peak contraction to maximal relaxation.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1.

Ranolazine (10 μM; A) and 10 μM TTX (B) attenuated 200 μM H₂O₂-induced late sodium current of guinea pig ventricular myocytes. Sodium current was elicited by 300-ms voltage-clamp pulses from -90 to -30 mV. Myocytes were sequentially treated with no drug (control) (a), H₂O₂ (b), and H₂O₂ plus either ranolazine or TTX (c).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 2.

Attenuation by 10 μM Ran and 2 μM TTX of 200 μMH₂O₂-induced prolongation of action potential duration of guinea pig ventricular myocytes. A and B, superimposed action potentials recorded from a single cell in the presence of no drug (control) (a), H₂O₂ (b), and H₂O₂ plus Ran (c) (A) or TTX (B). C and D, summary of data from experiments similar to those shown in A and B, respectively. * and **, p < 0.001 versus control and H₂O₂ alone, respectively.

Measurement of Intracellular Sodium and Calcium. Rabbit ventricular myocytes on laminin-coated recording chambers were loaded with either 10 μM SBFI-acetoxymethyl ester for 2 h or 10 μM Indo1-acetoxymethyl ester for 30 min, in the presence of 0.02% (w/v) Pluronic acid (Molecular Probes, Eugene, OR). The chambers were mounted on the stage of an inverted microscope (Nikon Eclipse TE2000-U) and superfused with normal Tyrode's solution (37°C) containing 140 mM NaCl, 4 mM KCl, 1 mM MgCl₂, 5 mM HEPES, 10 mM glucose, and 2 mM CaCl₂, pH 7.4. Myocytes were continuously paced at 0.5 Hz using electrical field stimulation (10-20 V). Intracellular SBFI was alternatively excited at 340 and 380 nm, and emitted epifluorescence was monitored at 510 nm for both wavelengths (F₃₄₀ and F₃₈₀). Intracellular Indo1 was excited at 360 nm, and emitted epifluorescence was measured at 405 and 485 nm. For both dyes, fluorescence emission was recorded using IonWizard software (IonOptix Corporation, Boston, MA). After washing out the external dye for 10 min, background fluorescence was determined for each excitation or emission wavelength. The ratios of F₃₄₀/F₃₈₀ and F₄₀₅/F₄₈₅, respectively, were then calculated from the background-subtracted emission intensities at each wavelength and converted to [Na⁺]_i and [Ca²⁺]_i with calibration curves. In situ calibration of SBFI was accomplished by exposing myocytes to bath solutions containing 0, 5, 10, and 20 mM Na⁺ in the presence of 10 μM gramicidin D, 80 μM monensin, and 50 μM strophanthidin (Maier et al., 1997). The solutions with various concentrations of Na⁺ were prepared from two stock solutions of equal ionic strength containing 140 mM NaCl, 10 mM HEPES, and 1 mM EGTA or 140 mM KCl, 10 mM HEPES, and 1 mM EGTA, pH was adjusted to 7.2 with Tris base. In situ calibration of Indo1 was accomplished using the equation [Ca²⁺]_i = K_D · β[(R - R_min)/(R_max - R)] as described previously (Bassani et al., 1995).

Statistical Analysis. Data are expressed as mean ± S.E.M. Values of “n” indicate the number of cells studied. The Student's t test, one-way repeated measures analysis of variance followed by Student-Newman-Keuls test, and two-way analysis of variance were applied where it was appropriate. A difference with a p value <0.05 was considered statistically significant.

Results

Ranolazine and TTX Inhibited H₂O₂-Induced Late Sodium Current of Guinea Pig Ventricular Myocytes. Late I_Na in the absence of drug was a small inward current (Fig. 1, A and B). Incubation of cells in 200 μM H₂O₂ caused an increase of late I_Na from -3.419 ± 0.392 to -6.215 ± 0.471 nC (Fig. 1A). Ranolazine (10 μM), an inhibitor of late I_Na, reduced the current to -5.072 ± 0.440 nC in the continued presence of H₂O₂ (n = 10; p < 0.01; Fig. 1A), a 51 ± 9% decrease of H₂O₂-induced late I_Na. To confirm that the H₂O₂-induced late current was indeed a Na⁺ current, the specific Na⁺ channel blocker 10 μM TTX was applied in the presence of H₂O₂. Late I_Na was increased by 200 μM H₂O₂ from -0.114 ± 0.538 to -4.423 ± 1.384 nC and was decreased by TTX to -0.476 ± 0.850 nC (n = 5; p < 0.01; Fig. 1B), a decrease of 91 ± 5%.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 3.

Inhibition by 10 μM ranolazine of 200 μMH₂O₂-induced early EADs of a guinea pig ventricular myocyte. The myocyte was sequentially treated with no drug (control) (a), H₂O₂ (b), H₂O₂ plus ranolazine (c), and H₂O₂ alone (d) (to wash out ranolazine). Each panel shows five superimposed, consecutive action potentials. Arrows indicate EADs.

Ranolazine and TTX Attenuated H₂O₂-Induced Action Potential Prolongation and Early Afterdepolarizations in Guinea Pig Ventricular Myocytes. An increase of late I_Na caused by H₂O₂ would cause prolongation of APD and early afterdepolarizations (EADs), which may explain, at least in part, the arrhythmogenic effect of H₂O₂ on cardiac myocytes. Alternatively, attenuation of late I_Na may antagonize these effects of H₂O₂. Therefore, the interactions between H₂O₂ and both ranolazine and TTX on action potentials were examined. For these experiments, equally effective concentrations of TTX (2 μM) and ranolazine (10 μM) were used. At these concentrations, neither drug alone had a significant effect on APD (data not shown). The earliest response to H₂O₂ (200 μM) was a progressive prolongation of APD. Prolonged treatment with H₂O₂ led to development of EADs, followed by spontaneous activity and membrane depolarization (data not shown). Both ranolazine (10 μM; Fig. 2A) and TTX (2 μM; Fig. 2B) attenuated an H₂O₂-induced prolongation of APD. The APD at 50% of repolarization (APD₅₀) was increased by H₂O₂ from 204 ± 15 to 280 ± 18 ms; ranolazine decreased the APD₅₀ to 235 ± 13 ms and attenuated by 58 ± 8% the prolongation of APD caused by H₂O₂ (n = 8; p < 0.001; Fig. 2C). In another set of experiments, H₂O₂ increased APD₅₀ from 177 ± 4 to 232 ± 9 ms, and TTX decreased APD₅₀ to 198 ± 6 ms in the continued presence of H₂O₂ (n = 5; p < 0.001; Fig. 2D). When EADs were induced by H₂O₂, addition of either 10 μM ranolazine (n = 3) or 2 μM TTX (n = 4) resulted in suppression of the EADs (Figs. 3 and 4, respectively).

Pretreatment of myocytes with ranolazine significantly blunted an increase of APD caused by H₂O₂ and prevented induction by H₂O₂ of EADs. In these experiments, myocytes were treated with either saline (control) or 10 μM ranolazine 3 min before application of H₂O₂ and throughout the exposure to 200 μMH₂O₂. Exposure of myocytes to H₂O₂ led to an increase of APD₅₀ (Fig. 5). APD prolongation usually became apparent after 5 min of exposure to H₂O₂ and increased progressively thereafter (Fig. 5). After a 10-min exposure to H₂O₂, APD₅₀ was increased by 61 ± 7%, and EADs were induced in 7 of 10 cells (action potentials with EADs were not included in APD₅₀ calculation). In contrast, APD₅₀ was increased by only 11 ± 6% (p < 0.01; Fig. 5) in ranolazine-treated myocytes, and EADs were not observed in any of the eight myocytes studied.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 4.

Inhibition by 2 μM TTX of 200 μM H₂O₂-induced EADs of a guinea pig ventricular myocyte. The myocyte was treated with no drug (control) (a), H₂O₂ (b), H₂O₂ plus TTX (c), and H₂O₂ alone (d) (to wash out TTX). Each panel shows three superimposed, consecutive action potentials. Arrows indicate EADs.

Ranolazine Reversed H₂O₂-Induced Contractile Dysfunction of Guinea Pig Ventricular Myocytes. H₂O₂ (200 μM) initially increased the amplitude of myocyte contractile shortening and induced aftercontractions that delayed contractile relaxation (Fig. 6A, b). Prolonged treatment of myocytes with H₂O₂ eventually caused spontaneous contractions and decreases of contractile amplitude and rate and extent of relaxation (Fig. 6A, d). Ranolazine (10 μM) did not significantly change contractile amplitude in the presence of H₂O₂ (Fig. 6B), but it suppressed aftercontractions (Fig. 6A, c). H₂O₂ increased myocyte contractile amplitude from 4.7 ± 0.8 to 6.3 ± 1.1 μm and decreased the rate of contractile relaxation from 12.2 ± 1.0 to 7.3 ± 1.3 μm/s (Fig. 6B). Ranolazine accelerated the relaxation rate from 7.3 ± 1.3 to 15.0 ± 1.9 μm/s (n = 5; p < 0.05) in the presence of H₂O₂ (Fig. 6B).

Ranolazine Attenuated H₂O₂-Induced Intracellular Na⁺and Ca²⁺Overload of Rabbit Ventricular Myocytes. In this series of experiments, the effects of 200 μM H₂O₂ on [Na⁺]_i, [Ca²⁺]_i, and contractions of rabbit ventricular myocytes were determined in the absence or presence of 10 μM ranolazine. H₂O₂ caused time-dependent increases of [Na⁺]_i and [Ca²⁺]_i (Fig. 7) that led to a state of hypercontracture within 14.8 ± 1.0 min (data not shown). In the presence of ranolazine, time to H₂O₂-induced hypercontracture was significantly increased to 19.3 ± 1.1 min (n = 14; p < 0.05). Consistent with this finding, ranolazine blunted the time-dependent increases of [Na⁺]_i and [Ca²⁺]_i during exposure to H₂O₂. There was a trend of reduction by ranolazine of baseline [Na⁺]_i, from 5.0 ± 0.8 to 2.4 ± 1.0 mM, and baseline diastolic [Ca²⁺]_i, from 179 ± 44 to 99 ± 26 nM, respectively. However, these changes were not statistically significant. After a 12-min incubation of myocytes with H₂O₂, [Na⁺]_i was increased to 16.7 ± 2.8 mM, whereas in the presence of ranolazine, [Na⁺]_i was significantly reduced to 8.0 ± 2.2 mM (p < 0.05; Fig. 7, A and B). Likewise, H₂O₂-induced increase of [Ca²⁺]_i was significantly attenuated by ranolazine. At the end of a 12-min incubation with H₂O₂, diastolic [Ca²⁺]_i was 569 ± 106 and 338 ± 61 nM, respectively, in the absence and presence of ranolazine (p < 0.05; Fig. 7, C and D).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 5.

Pretreatment of guinea pig ventricular myocytes with 10 μM ranolazine prevented 200 μM H₂O₂-induced prolongation of action potential duration. H₂O₂ was applied in the absence (closed circle) or presence (open circle) of ranolazine. The action potential duration at APD₅₀ was normalized as percentage of control (before application of H₂O₂) and was plotted against H₂O₂ exposure time. Each point represents data collected from 3 to 10 cells. *, p < 0.01 versus ranolazine-treated group.

Discussion

The major finding of this study is that blocking late I_Na with either ranolazine or TTX attenuates H₂O₂-induced arrhythmic activity and contractile dysfunction in cardiac myocytes. In addition, ranolazine was found to significantly reduce the rise in [Na⁺]_i and [Ca²⁺]_i caused by H₂O₂. The results of the study suggest that inhibition of late I_Na may attenuate reactive oxygen species-induced cardiac dysfunction.

The hypothesis tested in the present study can be summarized as follows (Fig. 8): ROS, such as H₂O₂, increase late I_Na and thereby cause 1) prolongation of APD and induction of EADs; and 2) a rise in [Na⁺]_i, which, because intracellular Na⁺ is exchanged for extracellular Ca²⁺ via NCX, causes cellular Ca²⁺ overload. Ca²⁺ overload of cardiomyocytes is associated with electrical instability (i.e., arrhythmias) and contractile dysfunction (i.e., impaired relaxation). Hence, by reducing the increase in late I_Na, the deleterious effects of H₂O₂ on cardiomyocyte function (electrical and contractile) can be attenuated. In support of the hypothesis (Fig. 8), results of other studies have shown that H₂O₂ can cause increases of late I_Na (Ward and Giles, 1997; Ma et al., 2005), intracellular Na⁺ and Ca²⁺ (Wagner et al., 2003), and ventricular diastolic tension and pressure (Hara et al., 1993; Zeitz et al., 2002).

To verify the hypothesis just described (Fig. 8), we measured late I_Na, action potentials, cell contractions, and intracellular concentrations of Na⁺ and Ca²⁺. We found that ranolazine and TTX effectively reduced H₂O₂-induced late I_Na (Fig. 1), action potential prolongation (Fig. 2), EADs (Figs. 3 and 4), and cell contractile dysfunction (Fig. 6). Pretreatment of myocytes with ranolazine significantly delayed or prevented action potential prolongation (Fig. 5), cell contracture, and increases of [Na⁺]_i and [Ca²⁺]_i caused by subsequent exposure of cells to H₂O₂ (Fig. 7). Thus, the hypothesis presented in Fig. 8 is fully supported by the present results. In addition, ranolazine has been reported to reduce H₂O₂-induced contractile dysfunction of rat isolated hearts (Matsumura et al., 1998). The results of the present study using ventricular myocytes thus provide an ionic mechanism (i.e., inhibition of late I_Na) to explain the protective action of ranolazine against ROS-induced increases of left ventricular end-diastolic and coronary perfusion pressures (Matsumura et al., 1998).

Although the amplitude of late I_Na recorded in the absence of drugs is small, late I_Na is found to contribute to the regulation of APD (Kiyosue and Arita, 1989; Maltsev et al., 1998; Sakmann et al., 2000). Blocking late I_Na with TTX caused a 10 to 20% decrease of APD (Kiyosue and Arita, 1989; Maltsev et al., 1998; Sakmann et al., 2000) and suppressed EADs of myocytes isolated from failing hearts (Maltsev et al., 1998). We found that in the absence of TTX or ranolazine, the amplitude of late I_Na can be markedly enhanced by H₂O₂ by severalfold (data not shown). Therefore, an increase by H₂O₂ of late I_Na is expected to cause a significant prolongation of APD. Consistent with previous reports (Beresewicz and Horackova, 1991), we found that the early effect of H₂O₂ on myocyte membrane potential was an increase in the duration of action potentials. This effect of H₂O₂ can be largely explained by an increase of late I_Na, because it was significantly attenuated by either ranolazine or TTX. Thus, blocking late I_Na may be the key to reduce H₂O₂-induced arrhythmic activity and contractile dysfunction.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 6.

Ran (10 μM) attenuates 200 μM H₂O₂-induced contractile dysfunction of guinea pig ventricular myocytes. A, amplitudes of twitch shortening of a single myocyte. Records were obtained in the absence (a) and presence (b-d) of drugs as indicated. H₂O₂ (b) increased the amplitude of twitch shortening and induced aftercontractions (additional contractions during relaxation). Ran (c) suppressed the aftercontractions, but had little effect on the twitch amplitude. During washout of Ran in the continued presence of H₂O₂ (d), the aftercontractions resumed, and eventually, spontaneous contractions with a decreased twitch amplitude occurred. B, summary of experiments similar to those shown in A. Each bar represents data collected from five cells. * and **, p < 0.05 versus control and H₂O₂ alone, respectively; NS, no significant difference versus H₂O₂ alone.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 7.

Ranolazine (10 μM) slows 200 μM H₂O₂-induced increases of [Na⁺]_i and diastolic [Ca²⁺]_i in rabbit ventricular myocytes. A to C, effect of H₂O₂ on [Na⁺]_i. A, original traces of SBFI fluorescence in the absence and presence of ranolazine before (0 min) and after a 12-min exposure to H₂O₂ (in gray). B, time course of changes in [Na⁺]_i. Mean values for calibrated signals are normalized to baseline. C, concentrations of Na⁺ measured at the end of a 12-min exposure to H₂O₂. D to F, effect of H₂O₂ on [Ca²⁺]_i. D, original traces of Ca²⁺ transient in the absence and presence of ranolazine before (0 min) and after a 12-min exposure to H₂O₂ (in gray). Please note that in the absence of ranolazine, at the end of 12-min H₂O₂ treatment the Ca⁺ transients had disappeared, consistent with an early hypercontracture. E, time course of changes in diastolic [Ca²⁺]_i. Mean values for diastolic [Ca²⁺]_i are normalized to baseline. F, diastolic [Ca²⁺]_i determined after a 12-min exposure to H₂O₂.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 8.

An action of H₂O₂ to increase late I_Na is a potential mechanism by which H₂O₂ prolongs APD and causes EADs and increases of intracellular [Na⁺] and [Ca²⁺]. TTX and ranolazine inhibit H₂O₂-induced late I_Na (dashed arrow).

In myocytes pretreated with ranolazine, both [Na⁺]_i and [Ca²⁺]_i were found to be lower, before the addition of H₂O₂, than in untreated myocytes. Whether this observation is due to an inherent variation in the intracellular ion concentrations or to a true effect of ranolazine remains to be investigated. However, in another study of rat isolated perfused hearts in which this issue was investigated, ranolazine at a concentration of 10 μM did not lower baseline systolic or diastolic [Ca²⁺]_i (Fraser et al., 2005). Regardless, a possible explanation for the lower [Na⁺]_i and [Ca²⁺]_i in myocytes pretreated with ranolazine is that this piperazine derivative by reducing basal I_Na, albeit small in normal myocytes, could indeed reduce basal [Na⁺]_i and [Ca²⁺]_i. Evidence in support of this explanation is that ranolazine, like TTX, can cause small shortening of the APD. The net effect of ranolazine on ventricular APD depends on the relative contribution of delayed rectifier K⁺ current and late I_Na to the repolarization (Song et al., 2004).

There are potential limitations to this study. First, H₂O₂ might elevate [Na⁺]_i via enhancing Na⁺-H⁺ exchange (Sabri et al., 1998) or reducing Na⁺-K⁺-ATPase (Kim and Akera, 1987) activity. Inhibition of Na⁺-K⁺-ATPase may cause a transient prolongation of action potential (Levi, 1991). However, it is unlikely that modulation of Na⁺-K⁺-ATPase plays a significant role in H₂O₂-induced action potential prolongation and EADs, because the effect of H₂O₂ was blocked by TTX. Furthermore, ranolazine at 20 μM had no effect on the Na⁺-H⁺ exchanger in Madin-Darby canine kidney cells (CV Therapeutics, Inc., 2003).

Second, our data interpretation is dependent on the selectivity of TTX and ranolazine for Na⁺ channels. The late (persistent) I_Na is more susceptible than the peak inward I_Na to the inhibitory effect of TTX (Kiyosue and Arita, 1989; Maltsev et al., 1998). In the present study, TTX at a low concentration of 2 μM significantly attenuated H₂O₂-induced action potential prolongation and EADs, whereas it had little effect on the basal action potentials (in the absence of H₂O₂; data not shown). This result indicates that the prolongation of APD caused by H₂O₂ can be mainly attributed to an increase of late I_Na. Consistent with the findings of the present study, TTX has been shown to markedly attenuate the deleterious effects (Ca²⁺ overload, diastolic dysfunction, and so on) caused by ROS and palmitoyl-l-carnitine, known to increase late I_Na (Ver Donck and Borgers, 1991; Hara et al., 1997). Ranolazine has also been reported to be a rather selective (38-fold) inhibitor of late relative to peak I_Na (Undrovinas et al., 2006). In ventricular myocytes of dogs with chronic heart failure, ranolazine inhibited peak and late I_Na with potencies of 244 and 6.5 μM, respectively (Undrovinas et al., 2006). Ranolazine at a concentration of 10 μM has been shown to effectively attenuate anemone toxin II-induced late I_Na, prolongation of APD and formation of EADs (Song et al., 2004). Alternative explanations for the effect of ranolazine to attenuate H₂O₂-induced action potential prolongation and EADs are an inhibition of the L-type Ca²⁺ current [I_Ca(L)] and/or Na⁺-Ca²⁺ exchange current (I_Na-Ca). However, the IC₅₀ values for ranolazine to inhibit I_Ca(L) and I_Na-Ca were reported to be around 300 and 91 μM, respectively (Antzelevitch et al., 2004; Schram et al., 2004). Moreover, we found that H₂O₂ caused a decrease of I_Ca(L) in guinea pig ventricular myocytes (data not shown). Therefore, it is unlikely that putative inhibition by ranolazine of I_Ca(L), I_Na-Ca, and/or peak I_Na can explain its attenuation of the effect of H₂O₂. It is also unlikely that ranolazine reduces the effects of H₂O₂ via radical scavenging or antioxidant actions, because ranolazine does not decrease H₂O₂-induced lipid peroxidation (Matsumura et al., 1998).

In summary, block of late I_Na by TTX or ranolazine attenuates the deleterious effects of H₂O₂ on electrical and contractile functions and cellular Na⁺ and Ca²⁺ homeostasis of cardiac myocytes. The results of the present study suggest that an increase of late I_Na is a major ionic mechanism underlying the cardiac actions of H₂O₂. Reducing late I_Na may be a critical step to attenuate ROS-induced myocardial dysfunction.