Abstract
Nortriptyline, a second-generation tricyclic antidepressant, is an active metabolite of amitriptyline. Amitriptyline induces QT prolongation and torsades de pointes (TdP), which causes sudden death. We studied the cardiovascular safety of nortriptyline, including QT prolongation risk. We examined the effects of nortriptyline on the cardiovascular system in vivo and in vitro in accordance with the ICH-S7B guideline. We tested its effect on QT interval in conscious telemetered dogs. We also performed in vitro electrophysiological studies on hERG tail currents using stably transfected human embryonic kidney 293 (HEK293) cells. Action potential parameters were studied in isolated rabbit purkinje fibers. Nortriptyline dose-dependently blocked hERG current, with a tail IC50 value of 2.20 ± 0.09 μM (n = 4). In the APD assay, total amplitude, Vmax, and resting membrane potential were not significantly changed by 1 μM nortriptyline, but nortriptyline at 0.3 and 1 μM shortened APD50 and APD90. Nortriptyline did not affect QTcV at 2 or 6 mg/kg, but slightly increased QTcV at 20 mg/kg. In conclusion, it is unlikely that nortriptyline affects the ventricular repolarization process at therapeutic dosages.
Introduction
Tricyclic antidepressants (TCAs) are a class of antidepressant drugs first used in the 1950s and named after the drugs' molecular structure, which contains three rings of atoms. TCAs are commonly used for psychiatric and other disorders. Many psychoactive drugs bind to cardiac ion channels and modulate the cardiac action potential (AP), which can produce cardiovascular side effects. 1 Clinical observations suggest that TCA-induced ventricular tachyarrhythmia is like ‘quinidine toxicity,2,3 namely facilitation of the reentrant excitation as a result of decreased conduction velocity and temporal dispersion of action potential duration (APD).4,5
Nortriptyline (Figure 1) is a second-generation tricyclic antidepressant marketed as the hydrochloride salt. It is FDA-approved for the treatment of depressive disorders. It is also used off-label for the treatment of panic disorder, irritable bowel syndrome, prevention of migraine headaches, and chronic pain or neuralgia modification. 6 Nortriptyline is a separately marketed active metabolite that is also metabolized in the liver from amitriptyline. 7 Amitriptyline and nortriptyline are distributed into the lungs, heart, brain, and liver. Amitriptyline induces QT prolongation and torsades de pointes (TdP), which causes sudden death, 8 and increases action potential duration to induce early after depolarization in canine purkinje fibers. 9 Tricyclic antidepressants increased the corrected QT (QTc) interval significantly by 6.9 milliseconds (ms) between consecutive ECGs in comparison with consecutive ECGs of participants not using TCAs, in particular amitriptyline (8.5 ms) and nortriptyline (35.3 ms). 10 Although nortriptyline is an active metabolite of amitriptyline, few studies have been performed on its cardiovascular side effects.
Structure of nortriptyline.
The potential proarrhythmic effects of pharmaceuticals that increase QT interval prolongation have resulted in the adoption of the International Conference on Harmonization (ICH) S7B Guideline 11 to ‘describe a nonclinical testing strategy for assessing the potential of test substance to delay ventricular repolarization, which could be used to estimate risk for delayed ventricular repolarization and QT interval prolongation in humans.’
The purpose of this study was to profile the cardiovascular safety of nortriptyline, including QT prolongation risk. We examined the effects of nortriptyline on the cardiovascular system according to the ICH-S7B guideline. We tested its effects on blood pressure, heart rate, and ECG in conscious telemetered dogs. In addition, we performed in vitro electrophysiological studies on hERG tail currents recorded from stably transfected human embryonic kidney 293 (HEK293) cells and in action potential parameters in isolated rabbit purkinje fibers.
Materials and methods
Drug
Nortriptyline (lot no. 097K0752 >99% purity) was purchased from Sigma-Aldrich (St. Louis, Missouri, USA). Distilled water was used as a vehicle and a negative control throughout the study.
hERG assay
HEK293 cells (CRL-1573, ATCC) stably expressing the hERG channel were cultured in minimum essential medium supplemented with 10% fetal bovine serum, 1 mM sodium pyruvate, 0.1 mM non-essential amino-acid solution, 100 units/ml penicillin G, 100 μg/mL streptomycin sulfate, and 400 μg/mL geneticin (G418) in an atmosphere of 95% air and 5% CO2. At 60% to 80% confluence, cells were treated with media containing 0.25% trypsin and 0.02% EDTA for 3 minutes, washed with fresh media, and dispensed to new plastic culture dishes. For electrophysiological recording, cells were seeded on 12-mm diameter glass cover slips, incubated for 5 to 24 hours in 30-mm dishes, and then transferred to the recording chamber. The cells were perfused with the medium (NaCl: 137 mmol/L, KCl: 4 mmol/L, MgCl2: 1 mmol/L, CaCl2: 1.8 mmol/L, glucose: 10 mmol/L, HEPES: 10 mmol/L, pH 7.4) at a rate of 1 mL/min at room temperature. Ionic currents were measured under a whole-cell voltage clamp using the patch clamp technique with Axopatch 1-D amplifier (Molecular Devices Corp., Sunnyvale, California, USA). Recording electrodes (actual resistance range: 2-3 MΩ) were filled with a pipette solution (KCl: 130 mmol/L, MgCl2: 1 mmol/L, EGTA: 5 mmol/L, MgATP: 5 mmol/L, HEPES: 10 mmol/L, pH 7.2). Voltage clamp pulses and data acquisition were controlled by computer software (pClamp 10, Molecular Devices Corp.). Once a stable patch had been achieved, cells were initially clamped at a holding potential of −80 mV. The hERG potassium currents were elicited by voltage pulses to +20 mV for 4 sec, −50 mV to form a hERG tail current for 6 sec, and then returned to the holding potential.
Nortriptyline was applied at 0.1, 0.3, 1, 3, 10, and 30 μM. IKr was determined as the maximum tail current induced by the repolarization pulse to −50 mV. The percentage of the peak value 10 min after treatment compared to that before treatment was calculated in three cells for each concentration.
APD assay
Twenty male New Zealand white rabbits (2.5−3 kg) obtained from ORIENT Bio, Inc. (Seongnam, Korea) were used in this experiment. All animal procedures were conducted in accordance with the ethical guidance for animal care for the facility (International Association for Assessment and Accreditation of Laboratory Animal Care, AAALAC) and the experimental protocol was approved by the Animal Care Committee of the institute. Animal rooms were maintained under the following conditions: temperature, 23 ± 3°C; relative humidity, 55% ± 15%; and a 12-h light period (08:00−20:00). The animals were anesthetized by sodium pentobarbital (50 mg/kg) before removal of the heart. The heart was imme-diately placed in a normal tyrode (NT) solution (NaCl: 143 mmol/L, KCl: 5.4 mmol/L, NaH2PO4: 0.33 mmol/L, MgCl2: 0.5 mmol/L, CaCl2: 1.8 mmol/L, glucose: 16.6 mmol/L, pH 7.3-7.5) saturated with 95% O2 and 5% CO2, and incised to obtain a specimen of the left ventricular purkinje fiber The purkinje fiber was superfused with NT solution at 37°C at a rate of 5 mL/min. Under stereoscopic observation, a stimulating electrode was placed on the surface of the muscle and a glass microelectrode filled with 3 mol/L KCl (tip resistance 10–50 mega-Ohm [MΩ]) was inserted into the purkinje fiber for recording. The muscle preparation was electrically stimulated with square wave (frequency: 1 Hz, pulse amplitude: 2 ms, voltage: 1-3 V) derived from an electrical stimulation apparatus (Accupulsere™ pulse generator A310, and Constant Current Stimulus Isolator A360D, World Precision Instruments Inc., Florida, USA). 12
Nortriptyline was applied at 0.1, 0.3, and 1 μM. Each purkinje fiber preparation was treated once with the superfusate containing one concentration of the treatment for 20 min. The action potential was measured in six individual preparations for each group. Wave patterns of the action potential were recorded using an amplifier (Geneclamp 500B, Molecular Devices Corp.) and Notocord program (NOTOCORD INC., France) before and 20 min after treatment. Effects on resting membrane potential (RMP), total amplitude (TA), maximal upstroke velocity (Vmax), and action potential duration at 50% and 90% repolarization (APD50 and APD90) were analyzed with Notocord.
Telemetry study
Four male beagle dogs (6−7 month old, 8−9 kg) obtained from Woojung BSC, Inc. (Suwon, Korea) were used in this experiment. All animal procedures were conducted in accordance with the ethical guidance for animal care for the facility and the experimental protocol was approved by the Animal Care Committee of the institute. The experimental animals were housed in individual stainless steel cages (W: 70 cm, D: 110 cm, H: 84 cm) in an air-conditioned room under the following conditions: temperature, 23 ± 3°C; relative humidity, 55% ± 15%; and a 12-hr light period (07:00-19:00). Animals were surgically fitted with telemetry transmitters (TL11M2-D70-PCT, Data Sciences International, St. Paul, Minnesota, USA) in the right abdominal subcutaneous area under pentobarbital anesthesia. A catheter for measuring blood pressure was passed subcutaneously and inserted into the abdominal aorta via the femoral artery. The animals were given cephalosporin subcutaneously after surgery and were allowed to recover for at least 3 weeks before experiments. The ECG was measured with M-X and R-L leads using a Holter electrocardiograph (Data Sciences International).
On the day of dosing, nortriptyline at 2, 6, and 20 mg/kg and a vehicle were orally administered in the Latin square crossover design with four animals. Each dose was given 14 days apart. ECG waveforms, BP waves, and HR of conscious, unrestrained animals were continuously obtained from the day before dosing to the day after dosing. Mean blood pressure (MBP), HR, and ECG were analyzed at before and 1, 2, 3, 7, 8, 24 hours after dosing. Stable tracings were collected for at least 2 min at each time-point. The starting and ending locations of PR, QRS, and QT intervals of each beat were visually specified on the monitor of the ECG analyzer, and the duration measured as the mean of 10 successive beats. The QT interval was corrected for the RR interval by Fridericia's formula (QTcF) and Van de Water’s formula (QTcV). 13
Statistical analysis
All data are expressed as the mean ± SD. The baseline values for each parameter in hERG and action potential studies were compared among test groups using one-way analysis of variance (ANOVA). The nortriptyline groups were compared with vehicle control by parametric or non-parametric Dunnett’s test. Student’s t-test was used to compare each group of telemetry studies. Values of p < 0.05 were considered significant.
Results
hERG assay
The effect of nortriptyline (0, 0.3, 1, 3, 10, 30 μM) on hERG current was studied using voltage clamp of HEK293 cells (Figure 2). These results indicate that nortriptyline dose-dependently blocked hERG currents, with tail IC50 values of 2.20 ± 0.09 μM and steady-state IC50 values of 2.27 ± 0.13 μM (n = 4).
Concentration dependence of nortriptyline-induced inhibition of hERG channels stably expressed in human embryonic kidney 293 (HEK293) cells. (A) Superimposed current traces were obtained by applying a 4-s depolarizing pulse from a holding potential of −80 to +20 mV every 15 s, followed by a repolarizing pulse to −50 mV for 6 s in the absence and presence of 0.3, 1, 3,10, and 30 μM nortriptyline, as indicated. The dotted line indicates zero current. (B) Concentration-dependent inhibition hERG channels by nortriptyline for steady-state current amplitude measured at the end of depolarizing pulse to +20 mV and peak tail current (n = 4). Currents in the presence of nortriptyline were normalized to the control amplitudes and plotted as a function of drug concentration. Solid lines represent fits with the Hill equation: Idrug/Icontrol = 1/ (1+ (IC50/D) n ), where D is the drug concentration, IC50 is the drug concentration for 50% block, and n is the Hill coefficient. Data represent the mean ± SEM.
APD assay
The effects of nortriptyline (0.1, 0.3, 1 μM) on action potential parameters in rabbit purkinje fibers are summarized in Table 1(n = 6), and representative examples before and after application of nortriptyline at 0.1, 0.3, 1 μM are illustrated in Figure 3. Nortriptyline at 0.1 μM did not affect any parameters, and higher doses did not affect total amplitude (TA), Vmax, or resting membrane potential (RMP). However, nortriptyline (0.3 and 1 μM) dose-dependently shortened APD50 and APD90.
Concentration-dependent effects of nortriptyline on action potential duration recorded in the rabbit Purkinje fibers a
Abbreviations: RMP, resting membrane potential; Vmax, maximal upstroke velocity of phase 0; TA, total amplitude.
a Data are expressed as mean ± SEM and compared by ANOVA followed by Dunnett’s test (n = 6).
b p < 0.01; Con, control; APD50 and APD90; APD at 50 and 90% repolarization
Representative illustrations of nortriptyline activity on the action potential of rabbit Purkinje fiber.
Telemetry study
Table 2 shows the time-course changes from baseline of systemic blood pressure (SBP), diastolic blood pressure (DBP), mean blood pressure (MBP), and heart rate (HR) after oral administration of nortriptyline (0, 2, 6, or 20 mg/kg). Nortriptyline at 20 mg/kg slightly increased MBP.
Blood pressure and heart rate in unrestrained conscious dogs implanted with a telemetry transmitter a
a Nortriptyline was orally administered at 2, 6, and 20 mg/kg. Data are mean and standard deviations of four animals.
b p < 0.05; significantly different from the control by student’s t-test.
ECG-telemetry in unrestrained conscious dogs a
Abbreviations: QTcF: Fridericia's formula, QTcV: Van de Water’s formula.
a Nortriptyline was orally administered at 2, 6, and 20 mg/kg. Data are represented as the mean and standard deviations of four animals.
bp < 0.05; significantly different from the control by student’s t-test.
Figure 4 shows changes in the duration of ECG parameters. Low doses (2 and 6 mg/kg) did not show any statistically significant changes in any parameters, but 20 mg/kg caused significant increases in the QT interval and QTcV interval 7 h after dosing (p = 0.0122, p = 0.0158, respectively, by Student’s t test). When the QT interval was corrected for the RR using the Fridericia formula (QTcF), there was a statistically significant difference in the QTcF at 3, 7, and 8 hours.
Effects of nortriptyline (0, 2, 6, 20 mg/kg) on Van de Water’s formula (QTcV) and Fridericia's formula (QTcF) following oral administration in conscious beagle dogs monitored by telemetry. Data are expressed as mean ± SEM (n = 4). Corrected QT interval (QTc) calculated by Van de Water’s formula: QT = QTvdw = QT- (87 × ((60/HR)-1)) and Fridericia’s formula: QTb = QT/RR1/3. *p < 0.05 vs. control.
Discussion
Nortriptyline is an active metabolite of amitriptyline that peaks in plasma at 7 to 8.5 hours. The plasma half-life of nortriptyline ranges from 16 to more than 90 h. The poor dose-response relationship and narrow therapeutic index of nortriptyline make this drug an excellent candidate for improving therapeutic efficacy through Therapeutic Drug Monitoring. Therefore, the American Psychiatric Association recommends the plasma monitoring in the range 50−150 ng/mL (approximately, 0.2−0.5 μM) for nortriptyline. 14 It has been suggested that drugs with IC50s at least 30-fold greater than the highest achievable free plasma concentration will be free of liability for TdP. 15 The IC50 value (2.27 μM) obtained from HEK293 cell is about 4.5-fold higher than the expected highest achievable free plasma concentration (0.5 μM), suggesting that the inhibition of hERG channels by the drug could contribute to its proarrhythmic effects. A recent study demonstrated that desipramine, another member of tricyclic antidepressants, also blocked hERG channel current. 16 Taken together this report and our presented results, the blockade of hERG channels by tricyclic antidepressants seems quite common events.
Nortriptyline in isolated rabbit purkinje fibers dose-dependently shortened APD50 and APD90 at 0.3 and 1 μM, potentially via Na+ channel blockade. Amitriptyline, the precursor of nortriptyline, is a potent dose-dependent blocker of Na+ channels near the therapeutic plasma concentration and it has an unusually high affinity toward the inactivated state of Na+ channels. These characteristics could, in part, explain its analgesic actions. 17 Also, the IC50 (0.27 ± 0.05 μM) for Na+ channel blockade by nortriptyline is in the range of therapeutic plasma concentrations for the treatment of depression. 18 Action potential duration is influenced by Ca2+ channels, Na+ channels, as well as hERG channels. Nortriptyline inhibited hERG channels and reduced APD probably via Na+ channel blockade.
Nortriptyline did not affect SBP, DBP, MBP, or HR according to the telemetry assay, and low doses (2 and 6 mg/kg) did not affect other parameters. The effects of nortriptyline on QT interval were corrected using classical formulas: Bazett (QTb = QT/RR1/2), Fridericia (QTb = QT/RR1/3), and Van de Water’s (QT = QTvdw = QT-(87 × ((60/HR-1)). The QTcV interval was significantly longer at 7 hours after drug administration at 20 mg/kg by Student’s t-test (p = 0.0158). When the QT interval was corrected for the RR using the Fridericia formula (QTcF), QTcF was significantly different at 3, 7, and 8 hours. In Beagle dogs, the van de Water’s formula shows a better correlation, confirming its usefulness in Beagle dogs. 19 QTcV was significantly different at 7 hours, which coincided with the peak plasma level of nortriptyline. Nortriptyline overdose of over ten times the therapeutic dosage may cause QT prolongation.
Shortened APD and prolonged QT interval seem contradictory, although species differences between rabbits and dogs and the complicated structure of the heart muscle may explain these adverse results. Terfenadine is known to prolong QT interval in electrocardiograms and to induce TdP but does not prolong APD in myocardial tissue preparations. 20 It has been known that aprindine 21 and terodiline 22 show similar results. Therefore, hERG channel inhibition and in vivo QT prolongation cannot be evaluated solely based on their action potential-prolonging activity in isolated myocardial tissue preparations.
In conclusion, we have tested the effects of nortriptyline on the cardiovascular system using safety pharmacology methods according to the ICH-S7B guideline for the first time. We measured nortriptyline (2, 6, and 20 mg/kg) effects in unrestrained conscious dogs with a telemetry transmitter for BP and HR measurement, as well as ECGs including the QTcV interval via body surface electrodes. We also measured action potentials in isolated rabbit purkinje fibers at 0.1, 0.3, and 1 μM of nortriptyline, as well as IKr in HEK-293 cells stably transfected with hERG at 0.1, 0.3, 1, 3, 10, and 30 μM of nortriptyline.
Nortriptyline dose-dependently blocked hERG current and shortened APD50 and APD90 at 0.3 and 1 μM via Na+ channel blockade. Further electrophysiological studies on ion channel activity are needed to confirm these effects. Nortriptyline showed no effect in QTcV at 2, 6 mg/kg, but caused slight increases in QTcV at 20 mg/kg. Taken together, it is unlikely that nortriptyline affects the ventricular repolarization process at clinical doses, but an overdose of nortriptyline, for example ten times the therapeutic dosage, may cause QT prolongation. Further clinical studies are needed to clarify the cardiovascular safety of nortriptyline.
Footnotes
This work was supported by the National Institute of Food and Drug Safety Evaluation (09171KFDA666).