Abstract

T-wave alternans, characterized by a beat-to-beat change in T-wave morphology, amplitude, and/or polarity on the ECG, often heralds the development of lethal ventricular arrhythmias in patients with left ventricular hypertrophy (LVH). The aim of our study was to examine the ionic basis for a beat-to-beat change in ventricular repolarization in the setting of LVH. Transmembrane action potentials (APs) from epicardium and endocardium were recorded simultaneously, together with transmural ECG and contraction force, in arterially perfused rabbit left ventricular wedge preparation. APs and Ca2+-activated chloride current (ICl,Ca) were recorded from left ventricular myocytes isolated from normal rabbits and those with renovascular LVH using the standard microelectrode and whole cell patch-clamping techniques, respectively. In the LVH rabbits, a significant beat-to-beat change in endocardial AP duration (APD) created beat-to-beat alteration in transmural voltage gradient that manifested as T-wave alternans on the ECG. Interestingly, contraction force alternated in an opposite phase (“out of phase”) with APD. In the single myocytes of LVH rabbits, a significant beat-to-beat change in APD was also observed in both left ventricular endocardial and epicardial myocytes at various pacing rates. APD alternans was suppressed by adding 1 μM ryanodine, 100 μM 4,4′-diisothiocyanostilbene-2,2′-disulfonic acid (DIDS), and 100 μM 4-acetamido-4′-isothiocyanostilbene-2,2′-disulfonic acid (SITS). The density of the Ca2+-activated chloride currents (ICl,Ca) in left ventricular myocytes was significantly greater in the LVH rabbits than in the normal group. Our data indicate that abnormal intracellular Ca2+ fluctuation may exert a strong feedback on the membrane ICl,Ca, leading to a beat-to-beat change in the net repolarizing current that manifests as T-wave alternans on the ECG.

  • T-wave alternans

left ventricular (LV) hypertrophy (LVH) is associated with a high risk of sudden cardiac arrhythmic death (13, 26, 39). One important clinical observation in patients with LVH and failure is the appearance of T-wave alternans, an ECG phenomenon characterized by a significant beat-to-beat change in T-wave morphology, amplitude, and/or polarity. T-wave alternans has been found to precede the onset of malignant ventricular arrhythmias and may serve as an important prognostic indicator of polymorphic ventricular tachycardia in patients with LVH and heart failure (3, 9, 18, 25).

Clearly, T-wave alternans is the consequence of an alternation of ventricular action potential (AP) duration (APD) at the cellular level (29, 42, 45). Multiple studies have suggested that APD alternation is related to intracellular Ca2+ overload and fluctuation, in which the sarcoplasmic reticulum (SR) function plays an important role (11, 21, 36). Direct evidence has shown that an alternation of the intracellular Ca2+ transit can modulate electrical activation and induce APD alternans (20, 28). Recent studies have shown that Ca2+/calmodulin-dependent protein kinase II, an enzyme that phosphorylates several Ca2+ transport proteins, can initiate intracellular Ca2+ alternation that induces APD alternans (5, 22). All of this suggests that intracellular Ca2+ oscillation plays a critical role in the genesis of APD alternans.

Several hypotheses have been proposed to explain the mechanism of Ca2+ oscillations and its coupling to APD alternans, including a beat-to-beat change in the refractoriness of SR Ca2+ release channel, L-type Ca2+ channel, and Na+/Ca2+ exchange current (INa/Ca) (33, 35, 36). However, the precise ionic mechanism for T-wave alternans or APD alternations is still poorly understood. The prominent feature of myocytes from a failing heart is the alteration of intracellular Ca2+ handling (2). We hypothesize that abnormal intracellular Ca2+ oscillation in a failing heart may cause a strong augmentation of Ca2+-activated membrane currents, such as K+ current (IK), INa/Ca, and Ca2+-activated chloride currents (ICl,Ca). A recent study from our laboratory has shown that LVH is associated with a decrease in slow delayed rectifier K+ current (IKs) (44), which excludes the possibility that Ca2+-activated IK is responsible for APD alternans. LVH-induced remodeling of INa/Ca in myocytes of various animals has been studied in detail (8, 30, 37); however, little is known about ICl,Ca channel remodeling and its role in the genesis of T-wave alternans in LVH and heart failure. The present study was designed to evaluate LVH-induced changes in ICl,Ca and to delineate the underlying ionic mechanisms responsible for APD alternans.

METHODS

Experimental animals and LVH model.

Adult New Zealand rabbits (1.4–1.8 kg) underwent unilateral nephrectomy with contralateral renal artery banding to produce LVH using techniques reported previously (32). Rabbits with unilateral renal artery banded in this way uniformly develop LVH within 3 mo. Control rabbits were matched for age in the experiment. Data were collected from 16 LVH rabbits (8 of either sex) and 16 control rabbits (8 of each sex). Animal care and protocols were approved by the Institutional Animal Care and Use Committee.

Heart weight measurement and wall thickness.

Rabbits were haparinized (800 u/kg iv) and then anesthetized with ketamine 40 mg/kg iv. When deep anesthesia was achieved, the heart was excised. All hearts were washed in cold bicarbonate-based Ca2+-free solution, containing (in mmol/l) 125 NaCl, 3.5 KCl, 1.5 KH2PO4, 1 MgCl2, 20 NaHCO3, and 20 glucose, saturated with 95% O2-5% CO2 to clear the chambers of blood. After a quick blotting, the heart was weighed. Hearts from rabbits used in the wedge preparations were used to measure the wall thickness. The LV posterior wall thickness was measured with calipers at the level of the papillary muscles.

Arterially perfused rabbit ventricular wedge preparation.

Surgical preparation of the rabbit LV wedge has been described in detail in previous publications (46). Briefly, rabbits weighing ∼3–3.5 kg were haparinized (800 U/kg iv) and then anesthetized with ketamine (40 mg/kg iv). When deep anesthesia was achieved, the heart was excised. After the ratio of heart weight to body weight was calculated, the heart was placed in a cardioplegic solution consisting of cold (4°C) Tyrode solution containing 14 mM extracellular K+ concentration and transported to a dissection tray. The tissue was cannulated via the left circumflex coronary artery and perfused with cardioplegic solution. Unperfused tissues were carefully removed using a razor blade. The preparation was then placed in a small tissue bath and arterially perfused with Tyrode solution of the following composition: (in mM) 129 NaCl, 4 KCl, 0.9 NaH2PO4, 20 NaHCO3, 1.8 CaCl2, 0.5 MgSO4, and 5.5 glucose and 1 U/l insulin, buffered with 95% O2-5% CO2 (36 ± 0.3°C). The preparation was allowed to equilibrate in the tissue bath for 1 h before electrical recordings.

Recording of the transmural ECG, transmembrane APs, and isometric contractile force.

The ventricular wedges were allowed to equilibrate in the tissue bath until electrically stable, usually 1 h. The preparations were stimulated at basic cycle lengths ranging from 500 to 4,000 ms using bipolar silver electrodes insulated, except at the tips, and applied to the endocardial surface. A transmural ECG signal was recorded using extracellular silver/silver chloride electrodes placed near the epicardial and endocardial surfaces of the preparation plugged into a differential direct current amplifier. Transmembrane APs were simultaneously recorded from the epicardial and endocardial sites using two separate intracellular floating microelectrodes (direct current resistance = 10 to 20 MΩ) filled with 2.7 M KCl and connected to a high-input impedance amplifier. Impalements were obtained from the epicardial, endocardial, and transmural cut surfaces of the preparation at positions approximating the transmural axis of the ECG recording. In some experiments in which T-wave alternans appears, isometric contractile force was measured using a force transducer attached to the wedge preparation.

Myocytes isolation.

Single ventricular myocytes were isolated enzymatically from rabbits (New Zealand, 2.8–3.3 kg) of either sex using a method described previously (7, 44). After enzyme perfusion, a thin layer (<1.5 mm) of tissue was dissected from epicardial and endocardial surface of LV free wall and myocytes were dispersed. The isolated cells were stored at 10°C in Tyrode solution containing 1 mM Ca2+. Only quiescent rod-shaped cells showing clear cross striations were used.

Single myocyte AP recording.

Single ventricular myocyte APs were recorded at 36.0 ± 0.3°C using a standard microelectrode technique. The microelectrode had a resistance of 25 to 40 MΩ when filled with 3 M KCl. Cells were perfused with a bath solution containing (in mM) 137 NaCl, 5 KCl, 1 MgCl2, 2 CaCl2, 10 glucose, and 10 HEPES, pH was adjusted to 7.4 with NaOH. AP was recorded at steady state with various stimulus frequencies. APD was measured at 90% repolarization (APD90).

Membrane current recording.

ICl,Ca was recorded in myocytes isolated from LVH rabbits as well as control rabbits at 36.0 ± 0.3°C using the whole cell patch-clamping technique. Pipette solution for recording ICl,Ca contained (in mM) 110 CsCl, 20 tetraethylammonium chloride, 1.0 MgCl2, 5 HEPES, 0.05 EGTA, and 5 MgATP, pH was adjusted to 7.4 with CsOH. Bath solution contained (in mM) 140 N-methyl-d-glucamine (NMDG) chloride, 1.0 CaCl2, 1.0 MgCl2, 10 HEPES, and 10 glucose, pH was adjusted to 7.4 with NMDG-OH. 4-Aminopyridine (4 mM) was used to block the transient outward K+ channel, and BaCl2 (0.5 mM) was added to inhibit the inward rectifier K+ channel. The electrode had a resistance of 2–4 MΩ before compensation. Series resistance was compensated up to 80%. Liquid junction potential was zeroed in the bath but not compensated under the whole cell condition.

Drugs.

4,4′-Diisothiocyanostilbene-2,2′-disulfonic acid (DIDS), 4-acetamido-4′-isothiocyanostilbene-2,2′-disulfonic acid (SITS), and ryanodine were purchased from Sigma Chemical. DIDS, SITS, and ryanodine were dissolved in DMSO to obtain stock solution of 0.1, 0.1, and 0.01 M, respectively. The final DMSO concentrations were 1/1,000 (vol/vol) and 1/10,000 (vol/vol) in DIDS and ryanodine-diluting procedures, respectively. Dilutions of the stock solution were made immediately before the experiment to obtain the desired concentrations.

Data analysis.

Data are expressed as means ± SE. Student's t-test was used to determine the statistical significance of differences between the control and test conditions. Significance was defined as a value of P < 0.05.

RESULTS

Alteration of T wave and QT interval in a LVH rabbit LV wedge.

After 3 mo of surgery (unilateral nephrectomy with contralateral renal artery banding), rabbits developed significant LVH. LVH manifested as an increased LV wall thickness and heart weight. The wall thickness was 0.49 ± 0.02 cm in the LVH group versus 0.34 ± 0.01 cm in the control group (means ± SE, n = 10 hearts, P < 0.01) at 3 mo. The ratio of heart weight to body weight was 2.9 ± 0.01 g/kg in the LVH group versus 2.2 ± 0.01 g/kg in the control group (means ± SE, n = 10 hearts, P < 0.01) at 3 mo.

As seen in Fig. 1, more significant beat-to-beat changes in endocardial APD than in epicardial APD resulted in a beat-to-beat alteration in transmural voltage gradient that manifested as T-wave alternans on the ECG. T-wave alternans was associated with a marked beat-to-beat change in the QT interval as well as in transmural dispersion of repolarization. Similar phenomena as shown in Fig. 1 have been observed in 6 out of 16 LVH rabbits. None of them was found in normal rabbits (0 out of 16). Interestingly, contraction force alternated in an opposite phase (“out of phase”) with APD.

Fig. 1.

Alteration of action potential (AP) durations (APDs) in the epicardium (Epi) and the endocardium (Endo) and T wave and contractile force in a left ventricular hypertrophy (LVH) rabbit left ventricular wedge preparation paced at a basic cycle length (BCL) of 4,000 ms. Please note that longer APD and QT interval were associated with weaker contraction, and vice versa. EAD, early afterdepolarization.

Beat-to-beat APD alternations in LVH myocytes.

APD prolongation and instabilities could be easily induced in single myocyte of the LVH rabbit. Figure 2 shows AP recordings in a normal and LVH endocardial myocytes at continuous low frequency (a cycle length of 2,000 ms) stimulation under normal perfusion. In addition to APD prolongation, a significant beat-to-beat change of APD could also be observed in the LVH myocytes (Fig. 2B). APD alternations were associated with an increased susceptibility to early afterdepolarization in LVH myocytes (Fig. 2C). At a basic cycle of length of 2,000 ms, the phenomenon as shown in Fig. 2 was observed more frequently in single myocytes of LVH than in those of the normal rabbits. However, significant APD alternans did occur in single myocytes of the normal rabbits at an extremely high-stimulating frequency (cycle lengths = 150 ms, Fig. 4). The incidence of APD alternans observed in LVH myocytes increased significantly with increasing pacing rates. The incidence of APD alternans in 10 LVH rabbits at pacing cycle lengths of 2,000, 1,000, 500, 250, 200, and 150 ms was 40%, 40%, 50%, 100%, 100%, and 100%, respectively.

Fig. 2.

APD alternation and EAD in a single endocardial myocyte of LVH rabbit at a BCL of 2,000 ms. A: steady-state APs recorded in a myocyte from control rabbit. B and C: APs recorded in the same myocytes from LVH rabbit.

Characteristics of APD alternations in LVH epicardial and endocardial myocytes.

To better understand the mechanism underlying APD alternations in LVH myocytes, we compared APD alternations and their characteristics in epicardial and endocardial myocytes of the LVH rabbits at a cycle length of 250 ms (Fig. 3). Figure 3, A and B, shows the steady-state APs recorded in a single endocardial and epicardial myocyte of LV in control rabbit. AP alternations were not observed in all tested myocytes from control rabbits. However, significant beat-to-beat AP alternations could be induced in all tested endocardial and epicardial myocytes of LVH rabbits (Fig. 3, C and D). Interestingly, beat-to-beat AP alternations were more pronounced in endocardial than in epicardial myocytes (Fig. 3, C and D). Figure 3, E and F, shows the beat-to-beat changes in APD90 of three consecutive beats in endocardial and epicardial myocytes at the cycle length of 250 ms. The APD90 of the first beat (N) and the second beat (N + 1) in endocardial myocytes were 171.4 ± 6.6 and 133.6 ± 5.1 ms (n = 10 cells from 10 rabbits, P < 0.01), respectively. The APD90 values of the first beat (N) and the second beat (N + 1) in epicardial myocytes were 160.6 ± 5.1 and 145.6 ± 4.9 ms (n = 10 cells from 10 rabbits, P < 0.05), respectively.

Fig. 3.

Beat-to-beat APD alternans in LVH myocytes. A and B: steady-state AP recorded from endocardial and epicardial myocytes from left ventricle of a control rabbit at BCL of 250 ms. C and D: steady-state AP recorded from endocardial and epicardial myocytes of LVH rabbit. E and F: beat-to-beat change in APD at 90% polarization (APD90) of 3 successive beats in endocardial and epicardial myocytes of control and LVH rabbits. Data are presented as means ± SE; n = 10 cells from 10 rabbits. *P < 0.05, **P < 0.01 vs. first beat (N) or third beat (N + 2).

Figure 4 summarizes the mean of beat-to-beat alternations in APD90 (APD90 difference between 2 consecutive beats) under control and LVH conditions at various cycle lengths. At cycle lengths of 1,000 and 500 ms, there were no significant beat-to-beat alternations in control rabbits. However, extremely fast pacing rates (such as at cycle lengths of 200 and 150 ms) slightly increased the mean alternans of APD90 both in endocardial and epicardial myocytes. In LVH myocytes, however, the mean alternans of APD90 were significantly increased at all tested cycle lengths in both endocardial and epicardial myocytes. It is noteworthy that beat-to-beat alternations of the APD90 was more profound in endocardial myocytes than in epicardial myocytes under LVH conditions.

Fig. 4.

Summaries of mean alternans of beat-to-beat change of APD90 in endocardial (A) and epicardial (B) myocytes of control and LVH rabbits at various cycle lengths. The mean alternans data obtained from the difference of APD90 between 2 successive beats under steady-state condition. Data are presented as means ± SE; n = 10 cells from 10 rabbits. **P < 0.01 vs. control.

To test whether intracellular Ca2+ overload plays a critical role in the genesis of APD alternations, we added ryanodine (a specific inhibitor for SR Ca2+ uptake) to the perfusate. Interestingly, ryanodine at the concentration of 1 μM completely suppressed the APD alternations induced in endocardial myocytes of LVH rabbits at a cycle length of 250 ms (Fig. 5, A and B). To our surprise, DIDS (100 μM), a specific ICl,Ca blocker, completely suppressed the APD alternations of endocardial myocytes in LVH rabbits (Fig. 5, C and D). Similarly, SITS at 100 μM, another anion channel blocker, also suppressed the APD alternations (Fig. 5, E and F). In our wedge preparation experiment, DIDS at a concentration of 100 μM did not significantly affect the isometric contractile force (5.45 ± 0.48 g in control vs. 5.70 ± 0.49 g at 100 μM, n = 4 hearts, P > 0.05). It is also noteworthy that DIDS caused a small but statistically significant increase in APD (data not shown).

Fig. 5.

Effects of ryanodine (A and B), DIDS (C and D), and SITS (E and F) on APD alternans in the single endocardial myocyte of LVH rabbit. V, voltage. G and H: APD90 of 3 successive beats in the absence and presence of DIDS and SITS. Data are presented as means ± SE; n = 10 cells from 10 rabbits. **P < 0.01 vs. N or N + 2.

Characteristics of ICl,Ca in LVH endocardial and epicardial myocytes.

ICl,Ca is defined as DIDS-sensitive current and could be recorded in the single ventricular myocytes of rabbit. Figure 6 shows typical ICl,Ca values recorded in endocardial and epicardial myocytes in control rabbits. The ICl,Ca trace demonstrated a transient outward direction and rapidly declined to a zero level within 25–50 ms, which is consistent with previous reports in rabbit ventricular myocytes by other authors (40, 47).

Fig. 6.

Ca2+-activated chloride current (ICl,Ca) in endocardial and epicardial myocytes of rabbit left ventricle. Superimposed current traces of a endocardial (A) and epicardial (B) myocytes in the absence and presence of 0.1 mM DIDS are shown; the depolarizing pulses consisted of a 5-ms prestep to 5 mV followed by a 200-ms depolarizing step to +50 mV; the holding potential was −50 mV. DIDS-sensitive ICl,Ca of Endo (C) and Epi (D) obtained by subtraction of the 2 traces in A and B, respectively.

Figure 7 shows superimposed traces of ICl,Ca from endocardial myocytes of control and LVH rabbits. ICl,Ca values were elicited by depolarizing test pulses from −40–60 in 10-mV increments with a holding potential of −50 mV. There were no significant differences of ICl,Ca densities between endocardial and epicardial myocytes in control rabbits (Fig. 7C). Interestingly, the density of the ICl,Ca was significantly greater in myocytes of the LVH rabbits than in normal rabbits at voltages above 10 mV. Also, there was no significant difference in ICl,Ca density between endocardial and epicardial myocytes of LVH rabbits.

Fig. 7.

Density of ICl,Ca in endocardial and epicardial myocytes of rabbit ventricle. A and B: representative current traces of DIDS-sensitive ICl,Ca in endocardial myocytes of control (A) and LVH (B) rabbits; the currents were elicited by test pulses between +20 and +60 mV in 10-mV increments; holding potential was −50 mV. C: current-voltage relationship of ICl,Ca in endocardial and epicardial myocytes of control and LVH rabbits. Data are presented as means ± SE; n = 10 cells from 10 rabbits. *P < 0.05, **P < 0.01 vs. control in endocardial myocytes.

DISCUSSION

The present study shows that a significant beat-to-beat change in endocardial APD created a beat-to-beat alteration in transmural voltage gradient that manifested as T-wave alternans on the ECG in the intact LV wall of LVH rabbits. In the single myocytes study, our results show sustained APD alternations can be induced in LVH myocytes at various pacing cycle lengths, and the endocardial myocytes are more susceptible to alternation than epicardial myocytes. Our data demonstrate that ryanodine, DIDS, and SITS suppressed the APD alternations in LVH myocytes, indicating that intracellular Ca2+ loading and ICl,Ca contribute importantly to the APD alternations in LVH myocytes. Moreover, a significantly larger ICl,Ca density in LVH myocytes suggests that abnormal intracellular Ca2+ fluctuation may influence the membrane ICl,Ca, leading to beat-to-beat APD alternations in LVH rabbits.

T-wave alternans, defined as a beat-to-beat variation of T-wave morphology, amplitude, and/or polarity is commonly observed in patients with LVH and heart failure. It is a well-established predictor of susceptibility to develop polymorphic ventricular tachycardia and sudden cardiac death in this group of patients (9, 18, 23, 33). Many studies have shown that T-wave alternans in most circumstances results from an alternation of ventricular repolarization (29, 31, 34). Using a canine LV wedge preparation, Shimizu and Antzelevitch (36) suggested that T-wave alternans observed at rapid rates under long QT conditions is largely the result of the M-cell APD alternations, leading to the exaggeration of transmural dispersion of repolarization during alternate beats. In the present study, we also founded that beat-to-beat APD alternations were more pronounced in the endocardium than in the epicardium in LVH rabbits. The dispersion of repolarization between epicardial and endocardial cells results in a transmural voltage gradient that leads to the inscription of T wave. A more significant beat-to-beat change in endocardial APD than in epicardium APD results in a beat-to-beat change in transmural voltage gradient, which manifests as T-wave alternans on the ECG.

There is growing evidence that Ca2+ release from SR plays a critical role in the genesis of APD alternans (20, 21, 36). Hirayama et al. (11) reported that ryanodine and caffeine, which prevent the release of Ca2+ from SR, abolished the APD and mechanical alternans in canine heart. They concluded that delayed intracellular Ca2+ cycling plays an important role in the development of APD alternans (11). In the guinea pig model of T-wave alternans, Pruvot et al. (31) demonstrated that the mechanism underlying T-wave alternans is more closely associated with intracellular Ca2+ cycling. In a hypertrophied and failing heart, the characteristic slow decay of the Ca2+ transient and increased diastolic intracellular Ca2+ may predispose the cell to oscillatory release of Ca2+ from the SR (15, 17). This hypothesis is supported by the observation that APD alternans in LVH was eliminated after the blockade of the SR Ca2+ release by ryanodine in the endocardial myocytes of LVH rabbits in our study. Our results support the notion that intracellular Ca2+ oscillation contributes to the APD alternans in LVH. It is noteworthy that APD alternans in LVH myocytes were seen not only at higher stimulating rates but also at lower stimulating rates (Fig. 2), indicating that APD alternans is an important abnormal electrophysiological feature in hypertrophied hearts.

A single myocyte from LVH and a failing heart have been shown to have an altered intracellular Ca2+ handling (2). The impaired intracellular Ca2+ concentration ([Ca2+]i) handling may potentially augment the membrane Ca2+-modulated cell surface ion channels and transporters, such as IKs, INa/Ca, and ICl,Ca, leading to alternations in cardiac repolarization. A number of studies have investigated the changes in INa/Ca in various animal models of LVH and heart failure. However, the results have always been inconsistent and remained controversial (8, 12, 13, 30, 37). A recent study from our laboratory has shown that LVH is associated with a decrease in IKs (44). The functional downregulation of K+ channels is the most consistent phenomenon and contributes to the APD prolongation in hypertrophied and failing myocytes (13, 39). Therefore, changes in IKs and INa/Ca are not sufficient to explain the APD alternans secondary to intracellular Ca2+ fluctuation under the conditions of LVH and heart failure. At present, little is known about ICl,Ca remodeling and its role in the APD alternans in myocytes of the hypertrophied and failing heart. The presence of ICl,Ca has been demonstrated in ventricular myocytes of rabbits (16, 47). It is activated by an increase in the [Ca2+]i associated with Ca2+-induced Ca2+ release from the SR and plays an important role in AP repolarization in rabbit ventricular myocytes. In our experiment, DIDS at a concentration of 100 μM completely suppressed the induction of APD alternans in myocytes of LVH rabbits (Fig. 5), indicating that ICl,Ca may be an important candidate current involved in APD alternans under the conditions of LVH. Some reports have suggested that DIDS can affect SR Ca2+ handling (10, 24, 27). To rule out this possibility, we tested SITS, another anion channel blocker that has shown no effect on intracellular Ca2+ handling (24), on the effect of APD alternans. Similarly, SITS at the concentration of 100 μM also inhibited the APD alternans (Fig. 5, E and F). Furthermore, DIDS at the concentration of 100 μM showed no significant changes in the contraction force in our wedge preparation. It is also noteworthy that there were significantly different effects on endocardial APD between DIDS and ryanodine. For example, DIDS had an APD-prolonging effect compared with ryanodine (Fig. 5). The direct blocking effect of DIDS on ICl,Ca may contribute to the observed APD prolongation.

Verkerk et al. (41) recorded a significantly larger outward ICl,Ca in ventricular myocytes from the failing rabbit heart; however, the density of ICl,Ca in failing rabbit cells did not differ significantly from the control rabbit myocytes due to that the cell capacitances found threefolds larger in the failing heart cells. Benitah et al. (1) also reported that a significant chloride current component was induced in hypertrophied cardiac myocytes. They suggested that the outward chloride current could prevent the excessive AP prolongation in hypertrophied and failing heart. Kameyama et al. (15) reported that DIDS attenuated the monophasic AP alternans that was induced with an abrupt shortening of the cycle length from 1,000 to 350 ms in the canine beating heart. They suggested that ICl,Ca might contribute to the appearance of the electrical alternans. Interestingly, we found that the density of ICl,Ca was significantly increased in both endocardial and epicardial myocytes of the LVH rabbits compared with those of the control rabbits (Fig. 7). Our data suggest that impaired intracellular Ca2+ handling of LVH myocytes may directly regulate the cell membrane ICl,Ca. Under conditions of LVH and failure, elevated [Ca2+]i may exert a strong feedback effect on ICl,Ca that shortens APD. APD shortening will in turn limit Ca2+ influx through cell membrane and cause less Ca2+ release from SR in the subsequent cardiac cycle (4). A smaller intracellular Ca2+ transient and delayed ICl,Ca recovery will result in a weaker ICl,Ca that prolongs APD. In an isolated cardiac muscle preparation, an opposite relationship between alternation of APD and alternation of the strength of contraction is frequently observed (35, 38, 43). APD alternans secondary to ICl,Ca fluctuation may provide an excellent explanation for those “out of phase” T-wave alternans with a longer AP associated with a smaller contractibility and a shorter APD corresponding to a larger contractibility.

Limitations of the study.

To date, ICl,Ca has been reported to be present in isolated myocytes from various species, such as rat (14) and rabbit (47) ventricle. However, there is no conclusive evidence that ICl,Ca is also present in human ventricular myocytes (6, 19, 41). Therefore, the role of ICl,Ca in the genesis of cardiac arrhythmias associated with T-wave alternans should be interpreted with caution in the failing human heart. Also, our interpretation of the data is based on the assumption that intracellular Ca2+ undergoes a fluctuating change under conditions of LVH since many studies have demonstrated that impaired intracellular Ca2+ handling is in LVH and heart failure. Therefore, further studies are required to elucidate the relationship between APD alternans and intracellular Ca2+ transient. In addition, studies from hypertrophied and failing heart have demonstrated an increase in both INa/Ca mRNA and protein, suggesting that enhanced INa/Ca function may also be involved in the genesis of APD alternans in LVH rabbit myocytes.

The duration and shape of the AP is the result of a delicate balance between the depolarizing and repolarizing currents that are active during the plateau phase. T-wave alternans, a cellular electrophysiological abnormality associated with LVH, is also the result of the summation of changes in the membrane currents. Therefore, multiple ion current abnormalities (remodeling) should contribute to the genesis of T-wave alternans in LVH rabbits. Our current results only suggest an association between ICl,Ca, hypertrophy and alternans but in no way proves a causal relationship with T-wave alternans.

Some studies have suggested that DIDS could affect SR Ca2+ handling (10, 24, 27) and, hence, it is possible that the inhibition of APD alternans by DIDS may be secondary to its effect on intracellular Ca2+ and not completely dependent on the inhibition of ICl,Ca. In our present study, the perfusion of DIDS at 100 μM in the arterially perfused LVH rabbit wedge preparation did not result in any significantly change in the isometric contractile force, indicating that DIDS unlikely influences [Ca2+]i. However, the possibility that DIDS may directly influence intracellular Ca2+ handling cannot be completely ruled out due to our inability to directly measure [Ca2+]i.

GRANTS

This study was supported by an American Heart Association Scientist Developmental Grant 0530160N (to D. Guo), the Albert M. Greenfield Foundation (to P. R. Kowey and G.-X. Yan), the Sharpe-Strumia Research Foundation (to G.-X. Yan), and W. W. Smith Charitable Trust (to D. Guo and P. R. Kowey).

Footnotes

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

REFERENCES

View Abstract