Journal Logo


Terfenadine Block of Sodium Current in Canine Atrial Myocytes

Lu, Yanjie*; Wang, Zhiguo*†

Author Information
Journal of Cardiovascular Pharmacology: March 1999 - Volume 33 - Issue 3 - p 507-513
  • Free


Terfenadine, a second-generation and nonsedating histamine-1 receptor blocker, is one of the most widely prescribed medications for relief of symptoms due to allergy and upper respiratory infections. However, terfenadine therapy has been associated with side effects on cardiac electrical activities. The most noticeable cardiac effect of terfenadine is the development of arrhythmias, notable prolonged QT interval, torsades de pointes, and ventricular fibrillation leading to sudden death (1,2). On the other hand, beneficial effects of terfenadine include relieving the ischemia-reperfusion injury by inhibiting the reperfusion arrhythmias in the isolated rat heart (3). Part of the action of terfenadine is attributable to its ability to alter the cardiac action-potential characteristics. The ionic mechanisms by which terfenadine causes cardiotoxicity have been actively studied, and a focus has been placed on K+ channels. Blockade by terfenadine of multiple cardiac potassium currents, particularly the fast component of delayed rectifier K+ channels (IKr) in different species has been well documented, and the K+ channel-blocking property has been thought to account for lengthening of the action-potential duration (APD) and the clinically observed QT prolongation induced by terfenadine (4-11). Recently two separate studies revealed potent blockade of cardiac Ca2+ channels by terfenadine (12,13), which could also contribute to its cardiac side effects in patients.

APD and thereby QT interval is determined by a delicate balance between the inward and the outward currents. In addition to K+ channels such as HERG (IKr), KvLQT, and minK (IKs) (14-17), the role of Na+ channel (INa) in familial long-QT syndrome also was indicated (18,19), and studies clearly demonstrated the importance of INa in determining the likelihood of arrhythmias (20). Moreover, suppression of maximal upstroke velocity and excitability by terfenadine was reported in canine Purkinje fibers at a concentration of 1 μM, which did not suffice to produce APD lengthening (21). Similar results also were observed in guinea pig ventricular muscles (22). These data strongly suggest an ability of terfenadine to block INa. However, controversy exists in the literature. One experiment performed with guinea pig papillary muscle preparations found no changes in the AP amplitude and the maximal upstroke velocity but prolongation of APD (23). A patch-clamp study presented data showing that terfenadine unselectively blocked cationic currents, including Na+ currents in addition to K+ and Ca2+ currents, but the INa blocking effects were manifested only at a depolarized holding potential (HP; −40 mV), and no significant blockade was seen at a membrane potential of −90 mV (12). This again contradicts the depression of upstroke velocity observed in canine Purkinje fibers that presumably had a resting membrane potential of around −90 mV (21).

In this study, we sought to clarify whether terfenadine indeed blocks INa. The experiments were therefore designed to assess in detail the effects of terfenadine on INa in single canine atrial myocytes by using whole-cell patch-clamp techniques.


Myocyte isolation

The isolation procedures for dog atrial myocytes were same as previously described (24). In brief, adult mongrel dogs of either sex (18-26 kg) were anesthetized with pentobarbital sodium (30 mg/kg i.v.), and their hearts quickly removed and immersed in Tyrode's solution (see Solutions). All solutions were equilibrated with 100% O2. The right coronary artery was cannulated, and the right atrium was dissected free and perfused with Ca2+-containing Tyrode's solution at 37°C for ∼10 min until the effluent was clear of blood. Any leaking arterial branches were ligated with silk thread to assure adequate perfusion. The tissue was then perfused at 12 ml/min with Ca2+-free Tyrode's solution for 20 min, followed by perfusion with the same solution containing collagenase (100 U/ml CLS II collagenase; Worthington Biochemical, Freehold, NJ, U.S.A.) and 0.1% bovine serum albumin (Sigma Chemicals, St. Louis, MO, U.S.A.). Tissue samples of ∼2-3 mm in diameter were removed every 5 min beginning 40 min after the onset of exposure to collagenase. Samples were minced into small chunks (1.5 mm3), and cells were obtained by trituration with a Pasteur pipette. Dispersed cells were kept at 4°C in a high-K+ storage solution (see Solutions) before being used.


The standard Tyrode's solution for cell isolation contained the following (in mM): 136.0 NaCl, 5.4 KCl, 1.0 MgCl2, 0.33 NaH2PO4, 2.0 CaCl2, 10 dextrose, and 10 N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid (HEPES; pH adjusted to 7.4 with NaOH). The high-K+ storage solution had the following composition (in mM): 20 KCl, 10 KH2PO4, 25 dextrose, 40 mannitol, 70 potassium glutamate, 10 β-hydroxybutyric acid, 20 taurine, 10 ethylene glycol-bis (β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid (EGTA), and 0.1% albumin (pH adjusted to 7.4 with KOH). The extracellular solution used to record INa in dog cells contained (in mM): 132.5 CsCl, 5.0 NaCl, 1.0 MgCl2, 1.0 CaCl2, 11 dextrose, and 20 HEPES. The internal solution contained (in mM) 135.0 CsF, 5.0 NaCl, 5.0 HEPES, 10.0 EGTA, and 5.0 Mg2-ATP. The pH of internal and external solutions was adjusted to 7.2 and 7.35 with CsOH. CdCl2 (100 μM) was added to the perfusate to block calcium current (ICa). All chemicals were obtained from Sigma Chemical.

Patch-clamp recording

Patch-clamp techniques used in this study were described in detail elsewhere (25,26). Only quiescent rod-shaped cells lacking membrane deformities and showing clear cross-striations were studied. A small aliquot of the solution containing the isolated cells was placed in a 1-ml chamber mounted on the stage of an inverted microscope. Five minutes was allowed for cell adhesion to the bottom of the chamber, and then the cells were superfused at 3 ml/min with the standard extracellular solution. The bath temperature was kept constant at 17°C to study INa by a Peltier-effect device.

The whole-cell currents were recorded in the voltage-clamp mode. Borosilicate glass electrodes (1.0 mm, OD) were used, with tip resistance of 1.0-1.5 MΩ when filled with internal solution. Currents were recorded with an Axopatch 200B amplifier (Axon Instruments, Foster City, CA, U.S.A.). Command pulses were generated by a 12-bit digital-to-analog converter controlled by pCLAMP software (Axon Instruments). Recordings were low-pass filtered at 5 KHz for INa. Recordings were simultaneously digitized with digiData (1200 series interface, Axon Instruments) and stored on the hard disk of an IBM-compatible computer.

Junction potentials were zeroed before formation of the membrane-pipette seal in Tyrode's solution. Mean seal resistance averaged 21.7 ± 4.0 GΩ (n = 18). Several minutes after seal formation, the membrane was ruptured by gentle suction to establish the whole-cell configuration for voltage clamping. The capacitance and series resistance (Rs) was electrically compensated to minimize the duration of the capacitive surge on the current recording and the voltage decrease across the clamped cell membrane. Rs along the clamp circuit was estimated by dividing the time constant obtained by fitting the decay of the capacitive transient by the calculated membrane capacitance (the time integral of the capacitive response to a 5-mV hyperpolarizing step from a HP of −60 mV divided by the voltage decrease). Before Rs compensation, the decay of the capacitive surge was a single exponential function of time with a time constant of 422 ± 31 μs (cell capacitance, 82 ± 5 pF; n = 20 cells). Precompensation Rs values averaged 5.2 ± 0.5 MΩ. After compensation, the time constant was reduced to 121 ± 5 μs (cell capacitance, 67 ± 3 pF), and Rs was reduced to 1.3 ± 0.2 MΩ. Currents recorded during this study rarely exceeded 2.0 nA. The mean maximum voltage decrease across the Rs was thus in the range of 3 mV. Cells with changing leak current (indicated by >10 pA changes in holding current at −140 mV) were rejected. Initial recordings of INa were not made until 30 min after membrane rupture to allow complete equilibration of pipette solutions with cytosol so that the time-dependent alterations of sodium channels (including the activation and inactivation parameters) could fall into steady states (27). Under our experimental conditions, 20 min was adequate for this purpose, similar to other reports (27,28).

Data analysis

Group data are expressed as mean ± SEM. Statistical comparisons were performed with a paired t test (for comparisons between group means when only two groups were compared) and analysis of variance (ANOVA) with Scheffé Contrasts was used for multigroup comparisons. A two-tailed p < 0.05 was taken to indicate statistical significance. A nonlinear curve-fitting technique (Marquardt's procedure) was performed with the use of the CLAMPfit routine in pCLAMP or Sigmaplot software (Jandel Scientific, San Rafael, CA, U.S.A.).


Concentration dependence

To ensure adequate voltage control, experiments were conducted with a low external Na+ concentration (5 mM) and at a low temperature (17°C). Under such conditions, there was a negative shift of the steady-state voltage-dependent inactivation. Thus a negative HP of −140 mV was applied to the cells throughout the experiments. Recordings were obtained with the voltage protocols shown in the inset of Fig. 1. After baseline recording, various concentrations of terfenadine ranging from 100 nM to 10 μM were sequentially applied to the bath solution. A 15-min perfusion was allowed for each concentration to let the drug effects reach steady-state level. Only data obtained from individual cells successfully exposed to all five concentrations of the drug were used for analysis. Terfenadine produced a clear decrease in INa relative to the baseline amplitude at various test potentials ranging from −60 to −25 mV, in a concentration-dependent fashion (Fig. 1). Figure 1a shows representative current traces recorded in one cell under control conditions, in the presence of 1 and 5 μM terfenadine, and after washout of the drug. Statistically significant inhibition of INa was seen at a concentration of as low as 100 nM (p < 0.05; n = 6 cells). The maximal effects were obtained at 10 μM (91.8 ± 2.5% decrease, p < 0.01; n = 6). INa was peaked at −45 mV, regardless of the drug concentrations (Fig. 1c). The concentration dependence of INa blockade induced by terfenadine at −45 mV was quantified by fitting the experimental data to the Hill equation, as illustrated in Fig. 1b. The calculated value of IC50 was 0.92 ± 0.12 μM (n = 6 cells). Terfenadine decreased INa by ∼14.1, 28.6, 38.4, 78.1, and 91.8% at concentrations of 100 nM, 300 nM, 1 μM, 5 μM, and 10 μM, respectively. The value for Hill coefficient was 0.98.

FIG. 1
FIG. 1:
Concentration-dependent block of sodium current (INa) by terfenadine in canine atrial myocytes. a: Representative traces showing terfenadine blockade of INa. Currents were elicited by 40-ms depolarizing pulses from a holding potential of −140 mV to test potentials between −75 and 0 mV with 5-mV increments at a frequency of 0.1 Hz. Current traces, obtained at −45 mV under control conditions, 20 min after superfusion with 1 and 5 μM terfenadine, and 10 min after washout of the drug, are superimposed for better comparison. b: Concentration-response curve for terfenadine inhibition of INa at a test potential of −45 mV (mean ± SEM, n = 6). Curve shown is best-fit estimates to the equation E = 1/(1 + (IC50/[D]n), where E is the effect at a concentration [D], IC50 is the concentration for half-maximal action, and n is the Hill coefficient. c: Current-voltage (I-V) relation (mean ± SEM, n = 6 cells) for INa in canine atrial myocytes. *p < 0.05, **p < 0.01, control versus 5 μM terfenadine (statistical results are shown only for 5 μM for clarity). d: Fractional block of INa determined by the equation: f = (1 − ITerfenadine/IControl), as a function of test potentials (between −60 and −25 mV. No accurate analyses can be obtained at voltages negative to −60 mV and positive to −25 mV because of the small amplitude of the currents at these potentials). Terf, terfenadine.

Current-voltage (I-V) relation without and with varying concentrations of terfenadine is displayed in Fig. 1c, and quantification of the effects is expressed as fractional block (1-ITerfenadine/Icontrol) against test potentials, as shown in Fig. 1d. Voltage dependence of INa blockade was analyzed by F test and significant voltage-dependence was seen only at concentrations higher than 1 μM (p < 0.05).

Voltage dependence

We demonstrated significant voltage-dependent block of INa by terfenadine (Fig. 1d). To investigate whether the observed reduction of INa by terfenadine was the consequence of altered activation parameters or decreased availability of channels for opening at a given voltage or both, conductance for the voltage dependence of both activation and inactivation of INa was evaluated with protocols shown in the inset and the legend of Fig. 2. Exposure of cells to terfenadine resulted in concentration-dependent negative shift of conductance curve for activation (Fig. 2a). The effects were reversible after 30-min washout when terfenadine concentrations were <1 μM. Higher concentrations frequently led to incomplete recovery of INa properties. Data for washout are not shown in the figure for the sake of clarity. Boltzmann fit yielded V1/2 values of −51.4 ± 3.4 mV for control, −56.0 ± 3.6 for 0.1 μM (p > 0.05, compared with control; n = 6), −57.3 ± 3.9 for 0.3 μM (p < 0.05), −59.8 ± 3.1 for 1.0 μM (p < 0.01), and −60.0 ± 2.9 for 5.0 μM (p < 0.01) terfenadine. On washout from 0.3 μM terfenadine, the midpoint voltage for activation shifted back toward control value (52.6 ± 5.5 mV, n = 4). No significant differences in the slope factor were seen with terfenadine relative to the control value. These results may account for the slight shift of the I-V curve in the hyperpolarizing direction in the presence of terfenadine shown in Fig. 1c.

FIG. 2
FIG. 2:
Voltage-dependent block of sodium current (INa) by terfenadine in canine atrial myocytes. a: Conductance curves for activation of INa under control conditions and in the presence of varying concentrations of terfenadine. The conductance curves are constructed by plotting the conductance G as a function of test potentials. G = 1/(V − Vrev), where Vrev is the apparent reversal potential determined from each current-voltage (I-V) curve. Symbols are mean of experimental data, and lines represent Boltzmann fit: G/Gmax = 1/{1 + exp[(V1/2 − V)/k]}, where Gmax represents the maximal conductance at −45 mV, V1/2 is a half-maximal activation voltage, and k is a slope factor. b: Steady-state voltage dependence of INa inactivation under control conditions (solid circles), in the presence of 1 μM terfenadine (solid squares), and 30 min after drug washout (open circles with dashed lines). Inactivation was assessed with a two-pulse protocol, with a 2-s prepulse to voltages between −140 and −50 mV followed by a 40-ms test pulse to −45 mV at 0.1 Hz. The curves shown represent best fits by Boltzmann distribution equations: I/Imax = 1/{1 + exp[(V1/2 − V)/k]}, where I is INa amplitude at a prepulse potential V, V1/2 is the voltage for half-maximal inactivation, and k is a slope factor.

Figure 2b displays mean data of inactivation, along with best-fit Boltzmann distribution curves, under control conditions, in the presence of 5 μM terfenadine and 30 min after washout of the drug (0.3 μM). Terfenadine caused an ∼10-mV shift of the inactivation curve in the negative direction. Mean values for the half-inactivation voltage V1/2 and the slope factor k under control conditions were −98.6 ± 1.2 and −3.9 ± 0.2 mV, respectively. In the presence of 1 μM terfenadine, the corresponding values were changed to −108.9 ± 0.4 mV (p < 0.05 compared with control; n = 5 cells) and −4.6 ± 0.2 mV (p > 0.05), respectively. The negative shift of the inactivation curve seen with the drug was nearly completely converted back to the control values on washout (V1/2 = 108.2 ± 0.7 and k = −4.3 ± 0.4; n = 4). The data indicated that terfenadine blocks INa more at less-negative HPs relative to hyperpolarized potentials.

Use dependence

A train of 50-ms depolarizing pulses to −45 mV was delivered from −140 mV at 1 Hz, after a quiescent period of ≥10 min at a HP of −140 mV. Effects of terfenadine were assessed after baseline recording. The magnitude of difference in peak INa elicited by the first pulse before and during exposure to terfenadine was defined as tonic block, and magnitude of decrement in the ratio of peak INa (terfenadine over control) elicited by the last pulse of the train relative to that evoked by the first pulse was defined as use-dependent block. Figure 3 illustrates the data from such experiments. The amplitude of INa elicited by the first pulse was considerably reduced from baseline value of 1.2 ± 0.3 to 0.6 ± 0.2 nA in the presence of the drug (5 μM, p < 0.01; n = 5), indicating a substantial tonic block (50%). A slight decrease in INa amplitude with successive pulses was observed before drug application, and this property was remarkably augmented in the presence of terfenadine. This further inhibition of INa on the top of the tonic block was more clearly demonstrated with the normalized data by dividing the values with terfenadine by those without the drug, as shown in Fig. 3b. The decreased INa approached a steady-state level at the thirtieth pulse, indicative of use-dependent block of the channels. A single exponential process was used to fit the data; the resulting rate constant for the use-dependent blockade reaching maximal level was 8.3 ± 1.6 pulses, and the magnitude of the use-dependent block was 26 ± 4%.

FIG. 3
FIG. 3:
Use-dependent blockade of sodium current (INa) by terfenadine in dog atrial myocytes. a: Pulse-by-pulse changes of INa under control conditions and in the presence of 5 μM terfenadine. Currents were elicited by 45 consecutive 50-ms voltage steps to −45 mV from a holding potential (HP) of −140 at a frequency of 1 Hz. b: Ratio of peak INa with terfenadine over control. The symbols are averaged experimental data from six cells, and the line represents single exponential fit.

Pulse-duration dependence

The use-dependent blockade of INa by terfenadine could be a consequence of open-channel block or inactivated-state block or both. To clarify this issue, use-dependent effects of terfenadine on INa were analyzed with different pulse durations ranging from 1 to 50 ms at a test potential of −45 mV. As shown in Fig. 4, little use dependence was seen with 1-ms pulse, which was long enough to let Na+ channels fully activate and short enough to prevent apparent inactivation of the channels. The magnitude of use-dependent block was markedly accentuated with lengthening pulse duration. A 5-ms pulse, which sufficed to allow nearly complete inactivation of INa, rendered an almost maximal magnitude of use-dependent inhibition. The results were more suggestive of inactivated state-dependent block of INa by terfenadine as opposed to open channel block.

FIG. 4
FIG. 4:
Pulse duration-dependent blockade of sodium current (INa) by terfenadine in dog atrial myocytes. INa was elicited by voltage protocols as described in Fig. 3, but with varying pulse length (1, 2, 5, and 50 ms, respectively). a: Raw current traces recorded at −45 mV under control conditions. Only traces obtained with the pulses one and 30 are shown for clarity. b: Analog data acquired in the presence of terfenadine (5 μM) with 1- and 5-ms pulse duration. c: Averaged data on the magnitude of use-dependent reduction of INa in the presence of terfenadine relative to control recording, expressed as percentage block over control. Note that the magnitude of use-dependent block is significantly increased with lengthening pulse duration. **p < 0.01, ***p < 0.001 control versus terfenadine.

To determine the relative affinity of terfenadine to the rested and the inactivated states of Na+ channels, the apparent dissociation constants were calculated as follows. Assuming a 1:1 isotherm between terfenadine binding and response and taking 50% Na+ channel availability after tonic block with terfenadine at a concentration of 1 μM, the rested-state dissociation constant (Kd) for terfenadine was estimated by the equation f = 1/(1 + Kd/[D]), where f = 1 − (Iterfenadine/Icontrol) represents the fractional tonic (or use-dependent) block and [D] is the drug concentration (1 μM in this case). The calculated Kd for the tonic block was 1.2 ± 0.3 μM. Similarly, by using the same equation and taking the magnitude of use-dependent block (26%) at a HP of −140 mV, we calculated that the Kd value for use-dependent block was 2.9 ± 0.7 μM, ∼2.5 times greater than the value for tonic block.


Antihistamines are effective therapy for histamine-mediated conditions, including seasonal and perennial allergic rhinitis and chronic urticaria. Yet, since 1990, evidence has accumulated demonstrating that high doses of the second-generation drugs terfenadine and astemizole prolonged the cardiac QT interval and produced torsades de pointes, a rare but potentially lethal arrhythmia (1,2). Conversely, beneficial effects of terfenadine against reperfusion arrhythmias also were reported in animal models (3). We demonstrated in this study that terfenadine produced concentration-, voltage-, use-, and pulse duration-dependent block of INa in canine atrial myocytes. Terfenadine is also characterized by pronounced tonic block of INa. The IC50 for INa blockade under our experimental conditions was 0.93 ± 0.12 μM. Our study provided additional information for better understanding of ionic mechanisms underlying terfenadine's ability to affect cardiac electrical activity.

Our data provide an ionic mechanism underlying the suppression of upstroke velocity of action potentials (dV/dt) previously observed in canine Purkinje fibers and guinea pig ventricular muscles (21,22). Lang et al. (21) reported an IC50 of 1.3 μM for terfenadine inhibition of dV/dt in dog cardiac Purkinje fibers, in agreement with the IC50 value of 0.93 μM for INa blockade in our study with direct measurement of INa. They also found that this effect became more pronounced with faster rates of stimulation, which is consistent with our finding that terfenadine block of INa is use dependent. Inhibition of INa by terfenadine has also been shown in guinea pig ventricular myocytes (12). The observed effects occurred only at depolarized potential (e.g., −40 mV). Our observations were made at a strongly hyperpolarized HP of −140 mV, although the degree of blocking was greatly accentuated with less-negative potentials. The discrepancy between their study (12) and ours is probably due to the different animal species (guinea pig vs. dog), tissue types (ventricle vs. atrium), bath temperatures (37°C vs. 17°C), and ionic conditions ([Na+]o = 45.7 vs. 5 mM) used for experiments.

Terfenadine has been shown to block a variety of K+ currents, including IKr (fast component of the delayed rectifier K+ current) and HERG-expressed currents (9,17,27), IKur (ultra-rapid delayed rectifier K+ current found in human atrial myocytes), and Kv1.5-expressed K+ currents (4,5,7,11,26), Itol (the transient outward K+ current; 26), and IKl (the inward rectifier K+ current; 28). Concentrations of terfenadine needed for achieving a significant inhibition of these K+ currents were reported in the micromolar range as for INa in our experiments, except for IKr that was blocked by ∼10-fold lower concentration of terfenadine. Interestingly, Lang et al. (21) found no appreciable changes in action-potential duration with 1 μM terfenadine but marked depression of dV/dt in dog Purkinje fibers. The IC50 for ICa blockade was reported to be as low as 142 nM(13). From all these data, it is clear that terfenadine is a nonselective ion channel blocker.

Our data suggest that terfenadine blocks INa from both rested and inactivated states of the channels. Terfenadine caused a 50% tonic block and 26% use-dependent block in our experiments. Although the tonic block most likely represents terfenadine interaction with the rested state of Na+ channels, the use-dependent block could result from either open-channel block or inactivated channel block or both. In the light of voltage-dependence of terfenadine effects (negative shift of the inactivation curve or more pronounced blockade of INa at less-negative membrane potentials), we tend to interpret the observed use dependence as an indication of terfenadine binding to the inactivated states of Na+ channels. This notion was further indicated by the fact that inhibition of INa by terfenadine was augmented with increasing pulse duration, which resulted in more inactivation of INa. Terfenadine block of Ca2+ current reported by Liu et al. (13) demonstrated a similar use dependence, and an interaction with the inactivated state was also taken to explain the effect. In our study, the estimated dissociation constant for the rested state (1.2 μM) is 2.3-fold lower than that for the inactivated state (2.9 μM), implying that terfenadine preferentially interacts with the rested state to the inactivated state of sodium channels. Because terfenadine is a lipophilic tertiary amine with a pKA value of 10 and exists in the charged form at physiologic pH, it can access its binding sites in Na+ channels via both hydrophobic and hydrophilic pathways to produce rested state (tonic block) and inactivated state (use-dependent block) block.

It is quite plausible that the INa-blocking effect of terfenadine may contribute to its cardiotoxicity in the following ways. First, terfenadine appears a more potent blocker of INa compared with many class I antiarrhythmic agents, such as quinidine (29), lidocaine (30), penticainide (31), and flecainide (32), under comparable conditions. It is known that sodium channel blockers like class I drugs can be either antiarrhythmic or proarrhythmic and arrhythmogenic because of excessive slowing of conduction, which promotes the initiating and sustaining of reentrant arrhythmias. Clinical studies have shown that sodium channel blockade in cardiac tissue could exert severe adverse effects, which might lead to increase in mortality (33). The strong INa-blocking property of terfenadine may also render it proarrhythmic and arrhythmogenic. Second, terfenadine is characterized with a strong tonic block of INa, a property that can produce cardiotoxicity in patients with normal heart rhythm. The additional block due to terfenadine's use dependence and voltage dependence can, as a result, exaggerate terfenadine's adverse effects under certain pathologic situations. Conversely, the INa-blocking property of terfenadine could also contribute in part to its arrhythmia-preventing action during ischemia reperfusion (3). Yet it should be kept in mind that the concentrations of terfenadine used in our experiments are much higher than those observed clinically.

Acknowledgment: This work was supported in part by the Medical Research Council of Canada, the Heart and Stroke Foundation of Quebec, an Establishment Grant for young investigators from the Fonds de Recherche en Sante de Quebec awarded to Dr. Wang, and the Fonds de la Recherche de l'Institut de Cardiologie de Montreal. Dr. Wang is a research scholar of the Heart and Stroke Foundation of Canada.

We thank Dr. Stanley Nattel for his critical review and constructive suggestions to the manuscript and XiaoFan Yang for excellent technical assistance.


1. Monahan BP, Ferguson CL, Killeavy ES, et al. Torsades de pointes occurring in association with terfenadine use. JAMA 1990;264:2788-90.
2. Smith SJ. Cardiovascular toxicity of antihistamines. Otolaryngol Head Neck Surg 1994;111:348-54.
3. Woosley RL, Chen Y, Freiman JP, Gillis RA. Mechanism of the cardiotoxic actions of terfenadine. JAMA 1993;269:1532-6.
4. Rampe D, Wible B, Brown A, Dage R. Effects of terfenadine and its metabolites on a delayed rectifier K+ channel cloned from human heart. Mol Pharmacol 1993;44:1240-5.
5. Crumb J, William J, Wible B, Arnold DJ, Payne JP, Brown AM. Blockade of multiple human cardiac potassium currents by the antihistamine terfenadine: possible mechanism for terfenadine-associated cardiotoxicity. J Pharmacol Exp Ther 1995;47:181-90.
6. Salata JJ, Jurkiewicz NK, Wallace AA, et al. Cardiac electrophysiological actions of the histamine H1-receptor antagonists astemizole and terfenadine compared with chlorpheniramine and pyrilamine. Circ Res 1995;76:110-9.
7. Yang T, Prakash C, Roden DM, Snyders DJ. Mechanism of block of a human cardiac potassium channel by terfenadine racemate and enantiomers. Br J Pharmacol 1995;115:267-74.
8. Berul C, Morad M. Regulation of potassium channels by nonsedating antihistamines. Circulation 1995;91:2220-5.
9. Roy M, Dumaine R, Brown AM. HERG, a primary human ventricular target of the nonsedating antihistamine terfenadine. Circulation 1996;94:817-23.
10. Ducic I, Ko C, Shuba Y, Morad M. Comparative effects of loratidine and terfenadine on cardiac K+ channels. J Cardiovasc Pharmacol 1997;30:42-54.
11. Valenzuela C, Delpon E, Franqueza L, Gay P, Vicente J, Tamargo J. Comparative effects of nonsedating histamine H1 receptor antagonists, ebastine and terfenadine, on human Kv1.5 channels. Eur J Pharmacol 1997;326:257-63.
12. Ming Z, Nordin C. Terfenadine blocks time-dependent Ca2+, Na+, and K+ channels in guinea pig ventricular myocytes. J Cardiovasc Pharmacol 1995;26:761-9.
13. Liu S, Melchert R, Kennedy R. Inhibition of L-type Ca2+ channel in rat ventricular myocytes by terfenadine. Circ Res 1997;81:202-10.
14. Neyroud N, Denjoy I, Donger C, et al. Mutations in the potassium channel KvLQT1 cause Jervell and Lange-Nissen syndrome. Circulation 1997;96:I-57.
15. Sanguinetti M, Curran M, Spector P, Keating M. Spectrum of HERG K+-channel dysfunction in an inherited cardiac arrhythmia. Proc Natl Acad Sci U S A 1996;93:2208-12.
16. Schulze-Bahr E, Wang Q, Haverkamp W, et al. Identification of a new gene for autosomal recessive long QT syndrome (Jervell Lange-Nielsen syndrome). Circulation 1997;96:I-57.
17. Wang Z, Fermini B, Nattel S. Delayed rectifier outward current and repolarization in human atrial myocytes. Circ Res 1993;73:276-85.
18. Wang Q, Shen J, Li Z, et al. Cardiac sodium channel mutations in patients with long QT syndrome: an inherited cardiac arrhythmia. Hum Mol Genet 1995;4:1063-7.
19. Dumaine R, Wang Q, Keating M, et al. Multiple mechanisms of Na+ channel-linked long-QT syndrome. Circ Res 1996;78:916-24.
20. Hondeghem L, Katzung B. Antiarrhythmic agents: the modulated receptor mechanism of action of sodium and calcium channel-blocking drugs. Annu Rev Pharmacol Toxicol 1987;24:387-423.
21. Lang DG, Wang CM, Wenger TL. Terfenadine alters action potentials in isolated canine Purkinje fibres more than acrivastine. J Cardiovasc Pharmacol 1993;22:438-42.
22. Tanaka H, Masumiya H, Kato Y, Shigenobu K. Inhibitory effects of terfenadine on the rising phase of action potentials and sinus rates in isolated guinea pig myocardium. Gen Pharmacol 1996;27:337-40.
23. Ki I, Inui A, Ito T. Effects of histamine H1 receptor antagonists on action potentials in guinea pig isolated papillary muscles. Arch Intern Pharmacodyn Ther 1996;331:59-73.
24. Yue L, Feng J, Gaspo R, Li G, Wang Z, Nattel S. Ionic remodeling associated with enhanced susceptibility to atrial fibrillation in a dog model. Circ Res 1997;81:512-24.
25. Wang Z, Fermini B, Nattel S. Effects of flecainide, quinidine, and 4-aminopyridine on transient outward and ultrarapid delayed rectifier currents in human atrial myocytes. J Pharmacol Exp Ther 1994;272:184-96.
26. Wang Z, Fermini B, Nattel S. Sustained depolarization-induced outward current in human atrial myocytes: evidence for a novel delayed rectifier K+ current similar to Kv1.5 cloned channel currents. Circ Res 1993;73:1061-76.
27. Sanguinetti M, Jurkiewicz N. Two components of cardiac delayed rectifier K+ current. J Gen Physiol 1990;96:195-215.
28. Ducic I, Neville R, Morad M. Modulation of K+ currents by antihistamines in rat ventricular myocytes. Biophys J 1995;68:A113.
29. Snyders DJ, Hondeghem LM. Effects of quinidine on the sodium current of guinea pig ventricular myocytes: evidence for a drug-associated rested state with altered kinetics. Circ Res 1990;66:565-79.
30. Sanchez-Chapula J, Tsuda Y, Josephson IR. Voltage- and use-dependent effects of lidocaine on sodium current in rat single ventricular cells. Circ Res 1983;52:557-65.
31. Carmeliet E. Activation block and trapping of penticainide: a disopyramide analogue, in the Na+ channel of rabbit cardiac Purkinje fibres. Circ Res 1988;63:50-60.
32. Takafumi A, Hondeghem LM. Interactions of flecainide with guinea pig cardiac sodium channels: importance of activation un-blocking to the voltage-dependence of recovery. Circ Res 1990;66:789-804.
33. Cardiac Arrhythmia Suppression Trial (CAST). Preliminary report: effect of encainide and flecainide on mortality in a randomized trial of arrhythmia suppression after myocardial infarction. N Engl J Med 1989;321:406-12.

Terfenadine; Sodium current (INa); Patch-clamp; Canine atrial cells

© 1999 Lippincott Williams & Wilkins, Inc.