Author for correspondence: Bao-Feng Yang, Department of Pharmacology, Harbin Medical University, Baojian Road 157, Harbin, 150086, P. R. China (fax +86 451 86675769, e-mail firstname.lastname@example.org).
Abstract: Arsenic trioxide (As2O3) has been found to be effective for relapsed or refractory acute promyelocytic leukaemia, but its clinical use is burdened by QT prolongation, Torsade de pointes tachycardias, and sudden cardiac death. The aim of the present study was to elucidate the ionic mechanisms of As2O3-induced abnormalities of cardiac electrophysiology and the therapeutic action of choline on As2O3-caused QT prolongation in guinea pig. Intravenous administration of As2O3 prolonged the QT interval in a dose- and time-dependent manner in guinea pig hearts, and the QT prolongation could be modulated by choline. By using whole-cell patch clamp technique and confocal laser scanning microscopy, we found that As2O3 significantly lengthened action potential duration measured at 50 and 90% of repolarization, enhanced L-type calcium currents (ICa-L), inhibited delayed rectifier potassium currents (IK), and increased intracellular calcium concentration ([Ca2+]i) in guinea pig ventricular myocytes. Choline corrected As2O3-mediated alterations of action potential duration, ICa-L and [Ca2+]i, but had no effect on the IK inhibition. As2O3 markedly disturbed the normal equilibrium of transmembrane currents (increasing ICa-L and suppressing IK) in guinea pig cardiomyocyte, and induced prolongation of action potential duration, further degenerated into QT prolongation. Choline normalized QT interval abnormality and corrected lengthened action potential duration by inhibiting the elevated ICa-L and [Ca2+]i in ventricular myocytes during As2O3 application.
Arsenic trioxide (As2O3) has been believed to be poisonous which may result in worldspread organ damage when combined with neighbouring thiol groups in many proteins. However, in the mid 1990s Zhang et al. (1996) reported that the effective use of As2O3 resulted in complete revovery in as many as 90% of acute promyelocytic leukaemia patient. Since then, As2O3 has been used widely for the treatment of relapsed or refractory acute promyelocytic leukaemia. Treatment with As2O3 was often accompanied by QT interval prolongation on the electrocardiogram (ECG) and, in some instances, by Torsade de pointes tachycardias that can degenerate into ventricular fibrillation and cause sudden cardiac death (Ohnishi et al. 2000; Unnikrishnan et al. 2001; Westervelt et al. 2001). This kind of cardiac toxicity due to therapeutic use of As2O3 has received more and more attention. Recently, clinically relevant concentrations of As2O3 has been shown to prolong the action potential duration in isolated guinea pig papillary muscle (Chiang et al. 2002). In another study, the molecular mechanisms leading to As2O3-induced abnormalities of cardiac electrophysiology were analyzed, and the results indicate that reduced trafficking of hERG channels to the cell surface may be one of the reasons for QT prolongation and Torsade de pointes in patients treated with As2O3 (Ficker et al. 2004). However, the exact mechanism of As2O3 in causing QT prolongation has not been characterized, so in this study, the potential mechanism of QT interval abnormality by As2O3 were explored.
It is well known that the cardiac membrane repolarization is controlled by the delicate balance between the inward and outward currents, particularly the plateau phase of the action potential. Therefore, K+ currents and Ca2+ currents (Li et al. 1999; Yang et al. 2002; Dong et al. 2004a) are crucial in determining the rate of membrane repolarization and thereby the action potential duration. Alterations of cardiac K+ channel function and density may have profound pathophysiological consequences in a variety of myocardial diseases, including myocardial ischaemia, heart failure, and life-threatening arrhythmias (Wang et al. 1999). Activation of an outward K+ current would be expected to accelerate repolarization (or shorten action potential duration), to hyperpolarize membrane potential, and to decrease Ca2+ entry into the cells indirectly. In our laboratory, the existence of a M3 receptor in cardiac tissues of canine, rats and guinea pigs has been identified. Accumulation of electrophysiological evidence suggest that stimulation of cardiac M3 receptor is coupled to a novel type of delayed rectifier K+ currents. We named this K+ current IKM3 (M3 receptor mediated K+ current) (Wang et al. 1999; Yang et al. 2005). Meanwhile, we have confirmed that the cardiac M3 receptor and IKM3 can be activated by choline (Shi et al. 1999a & b). In the light of the ability of choline to activate the M3 receptor and to induce a K+ current, it is quite conceivable that choline could affect the cardiac electrical activity. So, based on these rationales, we designed experiments to investigate whether choline can correct the QT prolongation mediated by As2O3 and what the possible mechanism might be.
Materials and Methods
Reagents. As2O3 was donated by Harbin YI-DA Pharmaceutical Limited Company. Choline was purchased from Sigma Chemical Co. Fluo-3-acetoxymethyl (Fluo-3/AM) (Molecular Probes, Eugene, OR, USA) was dissolved in DMSO and stored at −20 ° in the dark. All chemicals were from Sigma Chemical Co.
Electrocardiogram record. Adult guinea pigs (300∼400 g, provided by the Experimental Animal Center of Harbin Medical University, Grade II) were anaesthetized with sodium pentobarbitone (40 mg/kg intraperitoneally). Electrodes were penetrated subcutaneously into four limbs and in correct positions on the chest and connected with electrocardiograph. Guinea pigs were randomly divided into five groups: control group, As2O3 groups (0.4 mg/kg, 0.8 mg/kg and 1.6 mg/kg) and choline+As2O3 group. The ECG was recorded before drug infusion, at 10 min. after, and every 30 min. after infusion for 2 hr. The QT interval was measured by another person who was blinded to the experimental procedures from the initial rise of the QRS complex to the end-point of the T-wave, which was at the cross-point of the tangent line to the descending T-wave and the TP-isoelectric line (between the end of the T-wave and the following P-wave). When the heart rate was rapid, the QT interval was estimated by extrapolating the downslope of the T-wave. The corrected QT interval (QTc) was calculated with the Bazett formula: QTc=QT/(RR)1/2, an accepted method for correcting QT interval for rate in guinea pigs (Hayes et al. 1994).
Cell isolation. Single guinea pig ventricular myocytes were isolated as previously described in detail (Dong et al. 2004b). Briefly, adult guinea pigs of either sex were sacrificed by a blow on the head. Their hearts were quickly removed, washed in cool, oxygenated Tyrode solution, and cannulated on a Langendorff perfusion apparatus and retrogradely perfused via the aorta with normal Tyrode solution (mM: NaCl 126, KCl 5.4, HEPES 10, NaH2PO4 0.33, MgCl2 1.0, CaCl2 1.8 and glucose 10; pH was adjusted to 7.4 with NaOH) until the effluent was clear of blood. Then the heart was perfused with nominally Ca2+-free Tyrode solution at a constant rate of 5 ml/min., followed by perfusion with the same solution containing collagenase II (7.8 mg/50 ml) and bovine serum albumin (7.8 mg/50 ml). The ventricular tissue was minced after it softened and placed in KB medium (mM: glutamic acid 70, taurine 15, KCl 30, KH2PO4 10, HEPES 10, MgCl2 0.5, glucose 10 and EGTA 0.5; pH was adjusted to 7.4 with KOH). Single cell was obtained by gentle pipetting and stored at 4 ° for 1∼2 hr. All solutions were gassed with 100% oxygen and warmed to (37±0.5) °. Only Ca2+-tolerant, quiescent and rod-shaped myocytes with clear crossstriations were selected for electrophysiological recording and intracellular calcium measurement.
Electrophysiological recording. Whole-cell patch-clamp techniques used have been described elsewhere (Li et al. 2001). Ionic currents and action potentials were recorded in the voltage-clamp and current-clamp mode using an Axopatch-200B amplifier (Axon Instruments). Borosilicate glass electrodes (1 mm OD) pulled with a Brown-Flaming puller (model P-87) had tip resistance of 2 to 4 MΩ when filled with the pipette solution (mM: KCl 20, potassium aspartate 110, MgCl2 1.0, HEPES 5, EGTA 10, Na2ATP 5; pH was adjusted to 7.2 with KOH). Command pulses were generated by a 12-bit D/A converter controller by pCLAMP6 software. Recordings were low-pass filtered at 1 kHz. Junction potentials were zeroed before formation of the membrane-pipette seal in Tyrode solution. Five min. after seal formation, the membrane was ruptured by gentle suction to establish the whole-cell configuration. The capacitance and series resistance (Rs) were compensated and the leak currents were subtracted. Experiments were conducted at room temperature (20 °∼22 °). Currents were measured before drugs and 10 min. after drug application to the bath.
Ca2+ fluorescence measurements. Fluorescence measurements in cardiomyocytes have been described previously (Ai et al. 2001). Briefly, single cells were attached to the coverslips of chamber with ConA and incubated with Fluo-3/AM 10 μM working solution containing 0.03% Pluronic F-127 at 37 ° for 45 min. Then the cells were washed twice with Tyrode solution to remove the extracellular Fluo-3/AM. All the Ca2+ measurements were performed at room temperature and carried out within 2 hr of loading. The fluorescent change of the Fluo-3/AM loaded cell was detected by confocal laser scanning microscope (Fluoview-FV300, Olympus, Japan) with 488 nm for excitation from an Argon ion laser and 530 nm for emission and inverted microscope with 40× objective.
Statistic analysis. All data were expressed as mean±S.E., and paired and unpaired Student's t-tests were used as appropriate to evaluate the statistical significance of differences between two group means. A two-tailed P<0.05 was considered to be statistical significance.
Effects of As2O3 on the corrected QT interval (QTc) of ECG in guinea pig in vivo. The effects of different doses (0.4 mg/kg, 0.8 mg/kg and 1.6 mg/kg) As2O3 on QTc were assessed in guinea pig hearts. The results indicated that intravenous administration of As2O3 obviously prolonged QTc of ECG dose-dependently and time-dependently. As shown in fig. 1 that after treatment with As2O3 at doses of 0.8 mg/kg and 1.6 mg/kg, QTc gradually grew longer in the 2 hr observation period from 324±7 msec. of control to 368±11 msec. (P<0.01) and 388±11 msec. (P<0.01), respectively.
Effects of choline on QTc prolongation induced by As2O3 in guinea pig hearts in vivo. To explore whether M3 receptor was responsible for mediating As2O3-induced QT interval prolongation in guinea pig hearts, we studied the effects of choline (8 mg/kg) on QTc. As depicted in fig. 2, lengthened QTc caused by As2O3 was corrected by prior administration of choline.
Effects of choline on action potential duration prolongation induced by As2O3 of ventricular cells in guinea pig. In Tyrode solution, current clamp was used to record action potential in single cells of guinea pig ventricular myocytes. A 2 msec. depolarizing stimulatory pulse was given to trigger action potential. Resting potential, overshoot, 50% and 90% repolarization of action potential duration (APD50 and APD90) were used to describe action potential changes. As2O3 in different concentrations (10, 50, 100 μM) was perfused to the cells after the normal action potential was recorded. Results show that As2O3 prolonged the action potential duration markedly. As shown in table 1, compared with control groups, APD50 was prolonged from 273±29 msec. to 456±13 msec. (n=6, P<0.01) and 513±15 msec. (n=6, P<0.01) for 50 and 100 μM As2O3, respectively. APD90 at the same doses of As2O3 groups was also prolonged from 286±31 msec. to 468±14 msec. (n=6, P<0.01) and 524±15 msec. (n=6, P<0.01). Choline 1 mM was perfused to the bath solution before application of 50 μM As2O3. Action potential was recored in current clamp mode. The results show that choline had significant inhibitory action on APD prolongation induced by As2O3, and APD50 and APD90 were nearly recovered to normal value (table 1). Fig. 3 illustrates the representative traces showing the effects of As2O3 on APD and the therapeutic effect of choline.
Table 1. Effects of choline on action potential duration (APD) prolongation induced by As2O3 of ventricular cells in guinea pig (n=6).
Resting potential (mV)
**P<0.01 versus control group; ## P<0.01 versus 50 μM As2O3 group.
Effects of choline on changed IK induced by As2O3. Currents in guinea pig ventricular myocytes were elicited by 2.5 sec. depolarizing pulses to potentials ranging from 0 mV to +70 mV with 10 mV increment followed by a 1 sec. step to −30 mV in cells perfused with the solution containing CdCl2 0.2 mM. Since there was contamination of IK1 when the test potential was less than 0 mV, the depolarizing pulse was set from 0 mV to +70 mV. Holding potential of −50 mV was used to inactivate INa and T-type Ca2+ current, and CdCl2 to block ICa-L. Fig. 4 (A, B and D) shows the effects of As2O3 on IK from the guinea pig ventricular cell in the absence and presence of As2O3, and the results show that As2O3 decreased the density of IK in a concentration-dependent manner. At the testing potential of +70 mV, As2O3 50 and 100 μM decreased IK from 6.7±0.7 pA/pF to 4.9±0.7 pA/pF (n=6, P<0.05) and 4.5±0.7 pA/pF (n=6, P<0.05), respectively. When pretreatment with choline 1 mM before As2O3 50 μM was applied, our results showed that choline did not produce any changes of As2O3–induced decrease of IK in the guinea pig ventricular cell (fig. 4C and D).
Effects of choline on increased ICa-L by exposure to As2O3. For ICa-L recording, the myocytes were placed in an experimental chamber and continuously superfused with extracellular solution (in mM: NaCl 137, CsCl 5.4, MgCl2 1, CaCl2 1.8, glucose 10, HEPES 10, TTX 0.01, TEA-Cl 20; pH adjusted to 7.4 with CsOH). INa was inactivated at a holding potential of −50 mV and blocked by TTX. ICa-L was activated by 300 msec. depolarizing steps every 10 sec. from -40 mV to +60 mV with a holding potential of -50 mV. The magnitude of ICa-L was measured to be the peak inward current. Under our experiment conditions, the run-down of ICa-L in control was less than 5% throughout the observation period. The changes of current density of ICa-L was measured after exposure to As2O3 and choline. The results indicate that at testing potential of 0 mV, 50 and 100 μM, As2O3 increased significantly ICa-L density in guinea pig ventricular myocytes from −6.5±1.1 pA/pF to −9.7±1.3 pA/pF (n=5, P<0.01) and −10.6±1.8 pA/pF (n=5, P<0.01), respectively. Choline 1 mM inhibited the increased calcium currents induced by 50 μM As2O3 to −7.0±1.6 pA/pF (n=5, P<0.05) (fig. 5).
Effects of pretreatment with choline on the changes of [Ca2+]i modulated by As2O3 in guinea pig cardiomyocytes. In normal Tyrode solution or Ca2+-free solution, 10, 50 and 100 μM As2O3 was used respectively to determine whether As2O3 was able to modify [Ca2+]i at resting levels. It was found that As2O3 did not induce detectable alteration in the basal level [Ca2+]i (P>0.05 compared with control group, data not shown). Next, we examined whether As2O3 could affect KCl-mediated Ca2+ mobilization. In the confocal microscope, [Ca2+]i was elevated gradually after application of KCl 60 mM in the presence of 1.8 mM Ca2+, and the ratio (FI/FI0) of Fluo-3/AM fluorescence was increased 5.1±0.4 times (peak value) (n=6, P<0.01 compared with resting value). After pretreatment with Tyrode solutions containing As2O3 at the concentration ranging from 10∼100 μM for 5 min., the KCl-induced elevations of [Ca2+]i were enhanced markedly. Fifty and 100 μM As2O3 induced an obvious increase in FI/FI0 8.2±0.3 times and 8.7±1.0 times (n=6, P<0.01 compared with KCl group). With co-application of choline (1 mM) and As2O3 (50 μM), KCl-evoked elevations of [Ca2+]i increase was inhibited compared with 50 μM As2O3 group (n=6, P<0.01), and the average data are summarized in fig. 6A.
The most important finding of the present study is the demonstration of choline-modulating QT interval prolongation caused by As2O3. This conclusion is based on the following evidence. We have, first, established QT prolongation model (Chiang et al. 2002) on ECG in guinea pig by exposure to As2O3, and found that As2O3-mediated prolongation of QT interval could be corrected by choline. Second, the possible ionic mechanisms were further detected in guinea pig cardiomyocyte. The results indicate that choline could dramatically modulate prolongation of action potential duration and normalize increasing ICa-L as well as inhibit elevation of [Ca2+]i in response to As2O3.
In cardiac myocytes, there are a variety of ion channels including sodium channel, calcium channel and potassium channel, and so on. The functional and structural equilibrium of these various channels is fundamental for maintaining the rhythm and rate of heart beats as well as the contractions of the heart. Any abnormality in channel function or current density (up-regulation or down-regulation) may result in changes of currents and action potential duration of ventricular myocytes that can cause arrhythmia (Yang et al. 1999; Benitah et al. 2002; Clancy & Kass 2002). As2O3 has been found to be effective for relapsed acute promyelocytic leukaemia. Clinical trials have demonstrated QT interval prolongation due to As2O3 therapy (Zhou et al. 2003). In our study, As2O3 was shown to prolong QTc in guinea pig heart, which was in agreement with previous reports (Chiang et al. 2002). Although the ionic mechanisms leading to As2O3-induced QT interval abnormality have been reported, the results were different (Drolet et al. 2004; Ficker et al. 2004). So far, the modification of cardiac electrophysiological profile induced by As2O3 remains uncertain. Our results indicate that As2O3 significantly lengthened action potential duration in guinea pig ventricular myocytes, and that the mechanisms were associated with enhancement of ICa-L, inhibition of IK and further increase of [Ca2+]i. IK is the major outward current responsible for repolarization of cardiac action potential in guinea pig cardiomyocytes. Two components have been identified: a rapidly activating component termed IKr and a slowly activating component termed IKs. In the present study, the IK recorded included both IKr and IKs. According to the theory of best targets-ICa, IK and action potential duration being the potential targets during the occurrence of arrhythmia (Yang et al. 1998, 2002 & 2005), we concluded that increasing ICa-L and reducing IK contributed to action potential duration prolongation, then degenerated into QT prolongation by exposure to As2O3.
The study by Ficker et al. (2004) demonstrated long-term effects of As2O3 on the function of ion channel with no short-term effects at the clinically relevant concentration. However, in the present study, we detected that As2O3 exerted immediate effects on cardiac ion channel at higher concentration (5∼10 times clinical concentration). This discrepancy might be due to the differences of doses given and the experimental conditions. As demonstrated with low dose of As2O3 (10 μM), acute effects were not observed. This is consistent with other reports, but acute As2O3-induced changes in the current amplitudes became more pronounced with increasing doses. Based on the results obtained we propose that high concentrations of As2O3 have immediate cardiac effects on ion channel, but not at clinically relevant doses.
Choline is well known to be a necessary substrate of lipid membrane, and a precursor and metabolite of acetylcholine. A growing amount of evidence has indicated that choline is not limited to be a structural component, but that it is also a functional modulator of the cellular membrane (Shi et al. 1999b). The existence of M3 receptors in cardiac tissues has been convincingly confirmed with both functional and molecular evidence in our laboratory (Shi et al. 2003). Further investigation has demonstrated that choline potentially modulated cardiac electrical activity and accelerated membrane repolarization by activating a novel type of delayed rectifier K+ current (IKM3) via stimulation of M3 receptors (Shi et al. 1999b). In the present experiment, we have studied the effects of choline on As2O3-mediated ionic current changes. Choline showed no any changes of decreased IK in the guinea pig ventricular cell, but had significantly inhibitory effect on increased ICa-L and enhanced [Ca2+]i in response to As2O3. From the above results, we speculate that choline plays an important role in As2O3-induced QT prolongation via stimulation of M3 receptor and activation of IKM3, which may affect K+ efflux tremendously and accelerate cardiac repolarization, thus influencing Ca2+ influx on phase 2 plateau indirectly. Podzuweit (1982) has shown that choline can abolish the ventricular tachycardia induced by subepicardial infusion of norepinephrine and Ca2+ in open-chest pigs. We also demonstrated that one mechanism of the antiapoptotic effect of activation of M3 receptor by choline was due to the down-regulation of [Ca2+]i (Liu et al. 2004). We conclude that the modulating mechanisms of choline on APD prolongation by As2O3 may be mediated by a decrease in Ca2+ entry directly.
In summary, this study represents the first indication that As2O3 interferes with the delicate equilibrium between the inward and outward currents (increasing ICa-L and suppressing IK), and induces action potential duration prolongation; choline can normalize QT interval abnormality on ECG in guinea pig hearts and modulate prolonged action potential duration by inhibiting increased ICa-L and enhancing [Ca2+]i in isolated single ventricular myocytes by application of As2O3. Our results suggest that choline may be useful in modulating QT interval prolongation during As2O3 therapy possible through activation of M3 receptor and IKM3, which are the best target for antiarrhythmia but more detailed studies should be done.
This work was supported by the Key Project of National Natural Science Foundation of the People's Republic of China (30430780).