"Conventional" antihistamines slow cardiac repolarization in isolated perfused (Langendorff) feline hearts

Hydroxyzine-induced Torsade de Pointes in a Patient with Complete Atrioventricular Block

An 82-year-old woman was admitted to our hospital because of dyspnea and bradycardia during exertion. Electrocardiography revealed complete atrioventricular block. During pacemaker implantation, a small dose (12.5 mg) of hydroxyzine was injected for sedation, and torsade de pointes (Tdp) occurred. The QT interval was prolonged after administration of hydroxyzine, and Tdp was observed after the R on T phenomenon occurred, indicating that hydroxyzine was capable of prolonging the QT interval and causing Tdp. Therefore, we must be cautious when administering hydroxyzine for sedation during surgery, especially in patients with bradycardia.

Torsadogenic potential of a novel remyelinating drug clemastine for multiple sclerosis assessed in the rabbit proarrhythmia model

We assessed the torsadogenic effects of a novel remyelinating drug clemastine for multiple sclerosis using an in vivo proarrhythmia model of acute atrioventricular block rabbit, since the drug has been demonstrated to suppress the human ether-á-go-go related gene (hERG) K⁺ channels. Bradycardia was induced by atrioventricular node ablation in isoflurane-anesthetized New Zealand White rabbits (n = 5), and the ventricle was electrically driven at 60 beats/min throughout the experiment, except when extrasystoles appeared. Intravenous administration of clinically relevant dose of 0.03 mg/kg of clemastine and 10-times higher dose of 0.3 mg/kg hardly affected the QT interval or duration of the monophasic action potential (MAP) of the ventricle. Additional administration of clemastine at 3 mg/kg significantly increased the QT interval, MAP duration and the short-term variability of repolarization. Meanwhile, the premature ventricular contractions with R on T phenomenon were observed in 3 out of 5 animals, and torsades de pointes arrhythmias were detected in 1 out of 5 animals. These results suggest that the torsadogenic potential of clemastine is obviously observed in the acute atrioventricular block rabbit, which will not appear within the prescribed dose for multiple sclerosis.

The conventional antihistamine drug cyproheptadine lacks QT-interval-prolonging action in halothane-anesthetized guinea pigs: comparison with hydroxyzine

Antihistamines are known to belong to the chemical class that may induce long QT syndrome. Among them, cyproheptadine has been shown to exert multifaceted actions on the ventricular repolarization phase; namely, shortening of the action potential duration at supra-therapeutic concentrations of 2 - 8 μM and prolongation of the QT interval at ≥ 10 μM. Since information is limited regarding the in vivo electrophysiological effects of cyproheptadine, we assessed it using the halothane-anesthetized guinea-pig model, which was compared with effects of another antihistamine drug, hydroxyzine. Sub-therapeutic to therapeutic doses of hydroxyzine at 1 and 10 mg/kg, i.v. prolonged the QT interval and duration of monophasic action potential, whereas therapeutic to supra-therapeutic doses of cyproheptadine at 0.1 and 1 mg/kg, i.v. hardly affected the indices of ventricular repolarization. These results suggest that cyproheptadine may be categorized into antihistamines with little effect on the ventricular repolarization.

Effects of the histamine H(1) receptor antagonist hydroxyzine on hERG K(+) channels and cardiac action potential duration

AIM

To investigate the effects of hydroxyzine on human ether-a-go-go-related gene (hERG) channels to determine the electrolphysiological basis for its proarrhythmic effects.

METHODS

hERG channels were expressed in Xenopus oocytes and HEK293 cells, and the effects of hydroxyzine on the channels were examined using two-microelectrode voltage-clamp and patch-clamp techniques, respectively. The effects of hydroxyzine on action potential duration were examined in guinea pig ventricular myocytes using current clamp.

RESULTS

Hydroxyzine (0.2 and 2 μmol/L) significantly increased the action potential duration at 90% repolarization (APD(90)) in both concentration- and time-dependent manners. Hydroxyzine (0.03-3 μmol/L) blocked both the steady-state and tail hERG currents. The block was voltage-dependent, and the values of IC(50) for blocking the steady-state and tail currents at +20 mV was 0.18±0.02 μmol/L and 0.16±0.01 μmol/L, respectively, in HEK293 cells. Hydroxyzine (5 μmol/L) affected both the activated and the inactivated states of the channels, but not the closed state. The S6 domain mutation Y652A attenuated the blocking of hERG current by ~6-fold.

CONCLUSION

The results suggest that hydroxyzine could block hERG channels and prolong APD. The tyrosine at position 652 in the channel may be responsible for the proarrhythmic effects of hydroxyzine.

H(1) antihistamine drug promethazine directly blocks hERG K(+) channel

Promethazine is a phenothiazine derivative with antihistaminic (H(1)), sedative, antiemetic, anticholinergic, and antimotion sickness properties that can induce QT prolongation, which may lead to torsades de pointes. Since block of cardiac human ether-a-go-go-related gene (hERG) channels is one of the leading causes of acquired long QT syndrome, we investigated the acute effects of promethazine on hERG channels to determine the electrophysiological basis for its proarrhythmic potential. Promethazine increased the action potential duration at 90% of repolarization (APD(90)) in a concentration-dependent manner, with an IC(50) of 0.73microM when action potentials were elicited under current clamp in guinea pig ventricular myocytes. We examined the effects of promethazine on the hERG channels expressed in Xenopus oocytes and HEK293 cells using two-microelectrode voltage-clamp and patch-clamp techniques. Promethazine induced a concentration-dependent decrease of the current amplitude at the end of the voltage steps and hERG tail currents. The IC(50) of promethazine dependent hERG block in Xenopus oocytes decreased progressively relative to the degree of depolarization. The IC(50) for the promethazine-induced block of the hERG currents in HEK293 cells at 36 degrees C was 1.46microM at +20mV. Promethazine affected the channels in the activated and inactivated states but not in the closed states. The S6 domain mutations, Y652A and F656A partially attenuated (Y652A) or abolished (F656A) the hERG current block. These results suggest that promethazine is a blocker of the hERG channels, providing a molecular mechanism for the arrhythmogenic side effects during the clinical administration of promethazine.

Hydroxyzine, a first generation H(1)-receptor antagonist, inhibits human ether-a-go-go-related gene (HERG) current and causes syncope in a patient with the HERG mutation

QT prolongation, a risk factor for arrhythmias, can result from genetic variants in one (or more) of the genes governing cardiac repolarization as well as intake of drugs known to affect a cardiac K(+) channel encoded by human ether-a-go-go-related gene (HERG). In this paper, we will report a case of drug-induced long QT syndrome associated with an H(1)-receptor antagonist, hydroxyzine, in which a mutation was identified in the HERG gene. After taking 75 mg of hydroxyzine for several days, a 34-year-old female began to experience repetitive syncope. The deleterious effect of hydroxyzine was suspected because QTc interval shortened from 630 to 464 ms after cessation of the drug. Later on, the patient was found to harbor an A614V-HERG mutation. By using the patch-clamp technique in the heterologous expression system, we examined the functional outcome of the A614V mutation and confirmed a dominant-negative effect on HERG expression. Hydroxyzine concentration-dependently inhibited both wild-type (WT) and WT/A614V-HERG K(+) currents. Half-maximum block concentrations of WT and WT/A614V-HERG K(+) currents were 0.62 and 0.52 microM, respectively. Thus, accidental combination of genetic mutation and intake of hydroxyzine appeared to have led to a severe phenotype, probably, syncope due to torsade de pointes.

Blockade of HERG K+ channel by an antihistamine drug brompheniramine requires the channel binding within the S6 residue Y652 and F656

A number of clinically used drugs block delayed rectifier K+ channels and prolong the duration of cardiac action potentials associated with long QT syndrome. This study investigated the molecular mechanisms of voltage-dependent inhibition of human ether-a-go-go-related gene (HERG) delayed rectifier K+ channels expressed in HEK-293 cells by brompheniramine, an antihistamine. Brompheniramine inhibited HERG current in a concentration-dependent manner with the half-maximal inhibitory concentration (IC50) value of 1.7 microm at 0 mV. A block of HERG current by brompheniramine was enhanced by progressive membrane depolarization and showed significantly negative shift in voltage-dependence of channel activation. Inhibition of HERG current by brompheniramine showed time-dependence. The S6 residue HERG mutant Y652A and F656C largely reduced the blocking potency of HERG current. These results indicate that brompheniramine mainly inhibited the HERG potassium channel through the residue Y652 and F656 and these residues may be an obligatory determinant in inhibition of HERG current for brompheniramine.

Cardiotoxic effects of antihistamines: from basics to clinics (...and back)

Drug-induced arrhythmias, particularly those caused by a prolonged QT interval, have become a critical safety issue for compound selection during development by pharmaceutical companies and for health care regulators. The last two decades have witnessed enormous progress in the definition of the clinical conditions that facilitate the occurrence of such serious adverse effects, of its molecular basis, and in the preclinical strategies aimed at early identification of the cardiotoxic liability of compounds undergoing investigation or already used in the clinic. Moreover, despite the fact that acquired factors play an obvious role in drug-induced arrhythmias, it has become evident that the disease is often manifested upon the interaction of strong environmental stressors with specific genetic determinants of the affected individuals; in that sense, few examples can illustrate the existing interaction between acquired and genetic factors in disease manifestation better than drug-induced arrhythmogenesis. Progress in this field has been mainly driven by a strong interaction among various disciplines, including medicinal chemistry, pharmacology, electrophysiology, molecular genetics, and clinical cardiology; such an interdisciplinary approach has often generated unexpected discoveries of great clinical value, allowing clinicians to drive drug selection toward compounds of proven efficacy and safety. Historically, studies on antihistamines have paved the way for much of our current understanding of the mechanisms and problems associated with QT prolongation and drug-induced arrhythmogenesis; therefore, in this perspective, we will attempt to summarize how basic research studies have helped the interpretation of clinically relevant phenomena (from basics to clinics...) and how this information has prompted new emphasis in preclinical studies aimed at predicting the cardiotoxic potential of compounds (...and back). The current availability of several strategies provided with great predictive potential, together with an increased awareness of physicians, pharmaceutical industries, and health care regulators to this potentially serious cardiovascular side effect, has significantly decreased the risk associated with drug-induced arrhythmias caused by drugs newly introduced into the market; nevertheless, given the large number of cases of QT prolongation still occurring during treatment with a wide variety of congeners, it seems appropriate to review the issue of the cardiotoxic actions of antihistamines, as a better comprehension of the underlying mechanisms and risk factors is likely to contribute to the improvement of the risk/benefit ratio for pharmacological treatment in several therapeutic areas.

Evaluation of the guinea pig monophasic action potential (MAP) assay in predicting drug-induced delay of ventricular repolarisation using 12 clinically documented drugs

INTRODUCTION

While the dog in vivo model is commonly employed in the later phase of discovery for assessing drug-induced QT prolongation, an early screening assay is valuable when selecting compounds for further development and when compound availability usually is low. One such screening assay is the anaesthetised guinea pig monophasic action potential (MAP) model. The aim of the present study was to evaluate the ability of this model to detect proarrhythmic properties by testing a set of reference compounds with known clinical profile. Moreover, these results were compared to data previously obtained using in vivo canine QT assays (QT PRODACT study).

METHODS

Anaesthetised and ventilated male guinea pigs were vagotomised and pretreated with propranolol. After thoracotomy, a pacing electrode was clipped to the left atrial appendage and a suction MAP electrode positioned on the left ventricular epicardium. The drug or corresponding vehicle was injected intravenously in cumulative doses and MAP duration at 90% repolarisation (MAPD90) was recorded during cardiac pacing.

RESULTS

The 8 drugs known to be proarrhythmic in the clinic all displayed dose-dependent prolongation of MAPD90, while the 4 drugs devoid of dysrhythmia in man had no effect. When comparing doses producing a 10% MAPD90 increase with doses reported to increase QTc by 10% in dogs a strong correlation was seen (R(2) 0.94 and 0.58 for anaesthetised and conscious dogs, respectively).

DISCUSSION

The guinea pig MAP assay identified all clinically positive drugs while negative drugs were without effect on ventricular repolarisation. Furthermore, a good concurrence is shown between the guinea pig and dog models in identifying compounds with proarrhythmic properties. Overall, the study reinforces the anaesthetised guinea pig MAP model as a reliable assay predicting QT liability of new chemical entities and as a highly sensitive early screening model for cardiovascular risk.

Structural and functional basis for the long QT syndrome: relevance to veterinary patients

Long QT syndrome (LQTS) is a condition characterized by prolongation of ventricular repolarization and is manifested clinically by lengthening of the QT interval on the surface ECG. Whereas inherited forms of LQTS associated with mutations in the genes that encode ion channel proteins are identified only in humans, the acquired form of LQTS occurs in humans and companion animal species. Often, acquired LQTS is associated with drug-induced block of the cardiac K+ current designated I(Kr). However, not all drugs that induce potentially fatal ventricular arrhythmias antagonize I(Kr), and not all drugs that block I(Kr), are associated with ventricular arrhythmias. In clinical practice, the extent of QT interval prolongation and risk of ventricular arrhythmia associated with antagonism of I(Kr) are modulated by pharmacokinetic and pharmacodynamic variables. Veterinarians can influence some of the potential risk factors (eg, drug dosage, route of drug administration, presence or absence of concurrent drug therapy, and patient electrolyte status) but not all (eg, patient gender/genetic background). Veterinarians need to be aware of the potential for acquired LQTS during therapy with drugs identified as blockers of HERG channels and I(Kr).

Influence of opioid agonists on cardiac human ether-a-go-go-related gene K(+) currents

We have evaluated the ability of various opioid agonists, including methadone, L-alpha-acetylmethadol (LAAM), fentanyl, meperidine, codeine, morphine, and buprenorphine, to block the cardiac human ether-a-go-go-related gene (HERG) K(+) current (I(HERG)) in human cells stably transfected with the HERG potassium channel gene. Our results show that LAAM, methadone, fentanyl, and buprenorphine were effective inhibitors of I(HERG), with IC(50) values in the 1 to 10 microM range. The other drugs tested were far less potent with respect to I(HERG) inhibition. Compared with the reported maximal plasma concentration (C(max)) after administration of therapeutic doses of these drugs, the ratio of IC(50)/C(max) was highest for codeine and morphine (>455 and >400, respectively), thereby indicating that these drugs have the widest margin of safety (of the compounds tested) with respect to blockade of I(HERG). In contrast, the lowest ratios of IC(50)/C(max) were observed for LAAM and methadone (2.2 and 2.7, respectively). Further investigation showed that methadone block of I(HERG) was rapid, with steady-state inhibition achieved within 1 s when applied at its IC(50) concentration (10 microM) for I(HERG) block. Results from "envelope of tails" tests suggest that the majority of block occurred when the channels were in the open and/or inactivated states, although approximately 10% of the available HERG K(+) channels were apparently blocked in a closed state. Similar results were obtained for LAAM. These results demonstrate that LAAM and methadone can block I(HERG) in transfected cells at clinically relevant concentrations, thereby providing a plausible mechanism for the adverse cardiac effects observed in some patients receiving LAAM or methadone.

The gastroprokinetic and antiemetic drug metoclopramide is a substrate and inhibitor of cytochrome P450 2D6

Metoclopramide is increasingly prescribed for conditions previously treated with cisapride, but its metabolic enzymology and drug interactions are poorly understood. Using human liver microsomes (HLMs) and recombinant human cytochromes P450 (P450), we identified the major route of metoclopramide oxidation and the P450 isoforms involved. We also documented the ability of metoclopramide to inhibit the P450 system, using isoform-specific substrate reaction probes of CYP1A2, 2C19, 2C9, 2D6, 2E1, and 3A4. Metoclopramide was predominantly N-dealkylated to monodeethylmetoclopramide, a metabolite that has not so far been described in humans. Formation rate of this metabolite followed Michaelis-Menten kinetics (K(m), 68 +/- 16 microM; V(max), 183 +/- 57 pmol/min/mg of protein; n = 3 HLMs). Of the isoform-specific inhibitors tested, 1 microM quinidine was a potent inhibitor of metoclopramide (25 microM) monodeethylation [by an average of 58.2%; range, approximately 38% (HL09-14-99) to 78.7% (HL161)] with K(i) values highly variable among the HLMs tested (K(i), mean +/- S.D., 2.7 +/- 2.8 microM; range, 0.15 microM in HL66, 2.4 microM in HL09-14-99, and 5.7 microM in HLD). Except troleandomycin, which inhibited metoclopramide metabolism in only one HLM (by approximately 23% in HL09-14-99), the effect of other inhibitors was minimal. Among the recombinant human P450 isoforms examined, monodeethylmetoclopramide was formed at the highest rate by CYP2D6 (V = 4.5 +/- 0.3 pmol/min/pmol of P450) and to a lesser extent by CYP1A2 (0.97 +/- 0.15 pmol/min/pmol of P450). The K(m) value derived (approximately 53 microM) was close to that from HLMs (68 microM). Metoclopramide is a potent inhibitor of CYP2D6 at therapeutically relevant concentrations (K(i) = 4.7 +/- 1.3 microM), with negligible effect on other isoforms tested. Further inhibition of CYP2D6 was observed when metoclopramide was preincubated with HLMs and NADPH-generating system before the substrate probe was added (maximum rate of inactivation, K(inact) = 0.02 min(-1), and the concentration required to achieve the half-maximal rate of inactivation, K'(i) = 0.96 microM), suggesting mechanism-based inhibition. Metoclopramide elimination is likely to be slowed in poor metabolizers of CYP2D6 or in patients taking inhibitors of this isoform, whereas metoclopramide itself could reduce the clearance of CYP2D6 substrate drugs.

Comparison of tegaserod (HTF 919) and its main human metabolite with cisapride and erythromycin on cardiac repolarization in the isolated rabbit heart

Tegaserod (HTF 919) is a new drug being developed for gastrointestinal motility disorders. Because other gastrointestinal prokinetic agents, such as cisapride and erythromycin, cause slowing of cardiac repolarization and have been implicated in the development of the potentially fatal ventricular arrhythmia, torsades de pointes, a study was initiated to determine whether tegaserod and its main human metabolite adversely influence cardiac repolarization. By using isolated Langendorff-perfused rabbit hearts, we show that QT intervals remain unchanged at concentrations of tegaserod from 0.5 to 10 microM. It was not until the tegaserod concentration was increased to 50 microM (roughly 500-5,000 times more concentrated than those typically found in human plasma after administration of recommended clinical dosages), that a small, but significant increase in the QT interval (12+/-4%; p < 0.05; n = 4) was observed. No significant changes in QT occurred in the presence of the tegaserod metabolite at any of the concentrations tested (0.5-50 microM). In contrast, cisapride caused QT lengthening at concentrations as low as 0.1 microM, with significant QT increases occurring when 5-50 microM cisapride was used (22+/-4% to >70%, respectively; p < 0.01; n = 4). Erythromycin also caused significant lengthening of QT intervals (11+/-2%; p < 0.001; n = 4), although 100 microM concentrations of this drug were required to achieve this effect. These results demonstrate that both cisapride and erythromycin can slow cardiac repolarization at therapeutic doses and that tegaserod's lack of QT prolongation at therapeutic doses suggests that it has the potential to be a safer alternative to cisapride as a gastrointestinal prokinetic agent.