Skip Navigation

Cardiovascular Research 1999 44(3):568-578; doi:10.1016/S0008-6363(99)00258-8
© 1999 by European Society of Cardiology
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Ulens, C.
Right arrow Articles by Tytgat, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ulens, C.
Right arrow Articles by Tytgat, J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Copyright © 1999, European Society of Cardiology

Norpropoxyphene-induced cardiotoxicity is associated with changes in ion-selectivity and gating of HERG currents

Chris Ulens, Paul Daenens and Jan Tytgat*

Laboratory of Toxicology, Faculty of Pharmaceutical Sciences, University of Leuven, Van Evenstraat 4, B-3000 Leuven, Belgium

* Corresponding author. Tel.: +32-16-32-3403; fax: +32-16-32-3405 jan.tytgat{at}farm.kuleuven.ac.be

Received 30 June 1999; accepted 10 August 1999


   Abstract
Top
 Abstract
1 Introduction
 2 Materials and methods
 3 Results
 4 Discussion
 References
 
Objective: Norpropoxyphene (NP) is a major metabolite of propoxyphene (P), a relatively weak µ-opioid receptor agonist. Toxic blood concentrations ranging from 3 to 180 µmol/l have been reported and the accumulation of NP in cardiac tissue leads to naloxone-insensitive cardiotoxicity. Since several lines of evidence suggest that not only block of INa but also IK block may contribute to the non-opioid cardiotoxic effects of P and NP, we investigated the effects of P and NP on HERG channels. HERG presumably encodes IKr, the rapidly-activating delayed rectifier K+ current, which is known to have an important role in initiating repolarization of action potentials in cardiac myocytes. Methods: Using the 2-microelectrode voltage clamp technique we investigated the interaction of P and NP with HERG channels, expressed in Xenopus oocytes. Results: Our experiments show that low drug concentrations (5 µmol/l) facilitate HERG currents, while higher drug concentrations block HERG currents (IC50-values of approx. 40 µmol/l) and dramatically shift the reversal potential to a more positive value because of a 30-fold increased Na+-permeability. P and NP also alter gating of HERG channels by slowing down channel activation and accelerating channel deactivation kinetics. The mutant S631C nullifies the effect of P and NP on the channel's K+-selectivity. Conclusion: P and NP show a complex and unique drug-channel interaction, which includes altering ion-selectivity and gating. Site-directed mutagenesis suggests that an interaction with S631 contributes to the drug-induced disruption of K+-selectivity. No specific role of the minK subunit in the HERG block mechanism could be determined.

KEYWORDS Gene expression; Ion channel, K-channel; Long QT syndrome; Gene expression; QT-dispersion


   1 Introduction
Top
 Abstract
 1 Introduction
2 Materials and methods
 3 Results
 4 Discussion
 References
 
Propoxyphene is a synthetic opioid analgesic that is frequently implicated in fatal drug overdose. Propoxyphene (P) is a relatively weak µ-opioid receptor agonist with an analgesic potency 20 to 40 times less than that of (–)methadone [1]. Toxic blood concentrations ranging from 3 to 180 µmol/l have been reported [2] and it is known that norpropoxyphene (NP), a major metabolite of P, accumulates in cardiac tissue and may cause cardiac toxicity [3]. While symptoms of central nervous system depression, i.e., coma, convulsions and respiratory depression can be reversed by opiate antagonists, cardiotoxic effects remain unaffected by naloxone administration and therefore, may be unrelated to opioid receptor occupancy by P and NP. For instance, various naloxone-insensitive cardiovascular effects associated with P poisoning have been particularly well documented and include bradycardia, asystole, bundle branch block, widening of QRS complex, reduced myocardial contractility and hypotension [4]. In several cases the mechanism underlying the non-opioid action of opioid agonists has been shown to involve a direct interaction between the drug and voltage-dependent Na+ channels. Whitcomb et al. [5] demonstrated that P and NP block INa in a manner similar to local anesthetics, exhibiting both use dependency and slow recovery from block. Whitcomb et al. further demonstrated that the empirically observed reversal of QRS widening by administration of lidocaine, a local anesthetic, was caused by an increase of recovery from INa block by P. It was concluded that lidocaine is clinically effective at reversing some non-opioid receptor mediated effects (QRS widening and bradycardia) and that the cardiotoxicity of P may be attributed, in part, to block of the Na+ channel pore.

In addition, several lines of evidence suggest that not only block of INa but also IK block may contribute to the non-opioid cardiotoxic effects of P and NP. Horrigan and Gilly [6] demonstrated that methadone and P block K+ channels in squid neurons and GH3 cells and that this effect, at least in case of P poisoning, contributes to the cardiotoxic effects. Pugsley et al. [7,8] showed that the antiarrhythmic actions of several {kappa}-opioid receptor agonists, including U50488H, are associated with block of voltage-activated Na+ and K+ currents in isolated rat cardiac myocytes.

Because the effects of P and NP on cloned cardiac K+ channels are unknown, we investigated the effects of P and NP on IKr, the rapidly-activating delayed rectifier K+ current, which is known to have an important role in initiating repolarization of action potentials in cardiac myocytes. The gene encoding the IKr current was initially identified by Warmke and Ganetzky [9] as the human ether-a-go-go-related gene (HERG). While the inherited long QT syndrome (LQTS), an autosomal dominant disease, has been linked to mutations in the HERG gene, the acquired form of LQTS can result from therapy with drugs that block HERG channels. This interaction with HERG channels is considered as the main mechanism underlying the potential cardiotoxic effects caused by a variety of drugs, including second-generation H1 antihistamines (astemizole [10] and terfenadine [11]), the neuroleptic drug haloperidol [12] and the antidepressant thioridazine [13]. LQTS is associated with syncope and sudden death caused by abnormal repolarization and the onset of a rare but life-threatening polymorphic ventricular tachycardia known as torsade de pointes. Because a contribution of the minK subunit to HERG channel regulation [14] and pharmacological properties [15] has been reported, we investigated the effects of P and NP not only on HERG but also on HERG+minK currents. In vitro synthesized cRNAs were injected into Xenopus laevis oocytes and the effects of P and NP were compared by means of the two microelectrode voltage clamp technique.

In this study, we demonstrate that P and NP show affinity for the HERG cardiac K+ channels at concentrations similar to those reported in P-induced fatalities. We therefore conclude that a drug–HERG interaction, at least in part, may contribute to the cardiotoxic effects of P and NP.


   2 Materials and methods
Top
 Abstract
 1 Introduction
 2 Materials and methods
3 Results
 4 Discussion
 References
 
2.1 Expression in oocytes
The human minK (hIsK) cDNA clone in its original vector, pBLUESCRIPT (Stratagene, USA), was first subcloned into our custom-made high expression vector, pGEMHE [16], by double restriction digest of both pGEMHE and hIsK/pBLUESCRIPT with KpnI and PstI. Both cDNAs were loaded on an agarose gel, fragments of interest were cut out and genecleaned (QIAGEN, USA) for ligation with T4 DNA ligase (Promega, USA). For in vitro transcription, the ligation product, hIsK/pGEMHE, was linearized with PstI. cDNA encoding the human HERG channel (HERG/pSP64) was linearized with EcoRI. Next the cRNAs were synthesized from the linearized plasmids using the ‘RiboMAX large scale RNA production system’ (Promega, USA) with T7 RNA polymerase (for hIsK/pGEMHE) or SP6 RNA polymerase (HERG/pSP64) in the presence of a cap analogue diguanosine triphosphate (Boehringer, Germany). The HERGS631C mutant in its original vector, pALTER (Promega, USA), was first subcloned into pGEMHE. The HERGS631C clone was isolated by a double restriction digest with EcoRI and XbaI and ligated into the corresponding sites of pGEMHE. The ligation product, HERGS631C/pGEMHE, was linearized with XbaI and capped cRNA was synthesized using large-scale T7 mMESSAGE mMACHINE transcription kit (Ambion, USA). The in vitro synthesis of cRNA encoding RCK1 and isolation of Xenopus laevis oocytes was as previously described [16]. Oocytes were injected with HERG or HERGS631C cRNA at a concentration of 5 ng/50 nl. hIsK was coinjected with HERG at a final concentration of 2.5 ng/50 nl each.

2.2 Electrophysiological recordings and analysis
Whole-cell currents from oocytes were recorded from 1 to 4 days after injection using the two microelectrode voltage clamp technique (GeneClamp 500, Axon Instruments, USA). Resistances of voltage and current electrodes were kept as low as possible (approx. 200 k{Omega}) and were filled with 3 mol/l KCl. Currents were filtered at 200 Hz, using a 4-pole low-pass Bessel filter. Capacitative and leak currents were not subtracted. To eliminate the effect of the voltage drop across the bath grounding electrode, the bath potential was actively controlled. All experiments were performed at room temperature (19–23°C). The oocytes were superfused with ND-96 solution (composition in mmol/l: KCl 2, NaCl 96, MgCl2 1, CaCl2 1.8, HEPES 5, pH 7.5). For incubation of oocytes, this solution was supplemented with 50 mg/l gentamicin sulfate. In ion substitution experiments solutions containing different KCl concentrations (0.25–98 mmol/l) were obtained by substitution of KCl for NaCl. Na+-free conditions were obtained by substitution of Na+ for TRIS. The pCLAMP program was used for data acquisition and data files (Axon Instruments, USA) were directly imported, analysed and visualised with a custom made add-in for Microsoft Excel. The percentage reduced current was calculated using the equation:

Formula
Current percentages were used for the calculation of concentration–response curves, using the Hill equation:

Formula
where I represents the current percentage, Imax the maximal current percentage, IC50 the concentration of the agonist that evokes the half-maximal response, A the concentration of agonist, and nH the Hill coefficient. Averaged data are indicated as mean±s.e.mean. Boltzmann activation curves were constructed using the equation:

Formula
where I represents the current, Imax the maximal current, Vtest the test potential, V1/2 the potential that evokes the half-maximal current and k the slope of the Boltzmann curve. Statistical analysis of differences between groups was carried out with Student's t test and a probability of 0.05 was taken as the level of statistical significance.

2.2.1 Compounds
D-propoxyphene hydrochloride was purchased from Federa (Belgium). D-norpropoxyphene maleate was purchased from Sigma Chemical Co (USA). Both compounds were dissolved in ND-96 solution, stored at 5°C until use and extracellularly applied.


   3 Results
Top
 Abstract
 1 Introduction
 2 Materials and methods
 3 Results
4 Discussion
 References
 
3.1 Heterologous expression of HERG and minK in Xenopus oocytes
To investigate the effect of P and NP on HERG and HERG+minK currents, HERG cRNA was injected alone or coinjected with minK cRNA in Xenopus laevis oocytes and currents were investigated 2 to 4 days postinjection by means of the two microelectrode voltage clamp technique. HERG and HERG+minK currents were evoked by application of 2 s test pulses ranging from –50 to +40 mV in 10 mV increments from a holding potential of –70 mV. Tail currents were clamped at a potential of –70 mV during 2 s (Fig. 1A and 1B). Activation curves were constructed using peak tail amplitudes and fitted to a Boltzmann function with V1/2=–21.07±1.81 mV and k=8.71±0.92 mV (n=4) for HERG currents and V1/2=–20.84±0.38 mV and k=10.11±0.69 mV (n=4) for HERG+minK currents. The time course for activation currents at –10 mV were fitted with a single exponential function and the calculated time constant for activation ({tau}) was 714.18±36.81 ms (n=4) for HERG currents and 764.90±44.99 (n=4) for HERG+minK currents. Currents evoked from a holding potential of –70 mV, pulsed to +20 mV during 2 s and stepped to tail potentials ranging from –120 to –30 mV in 10 mV increments, showed inward rectification as illustrated in Fig. 1C and 1D. The calculated Erev was –89.45±0.09 mV (n=6) for HERG currents and –88.12±0.08 mV (n=6) for HERG+minK currents. The fast ({tau}2) and slow ({tau}1) component of deactivation were resolved with a biexponential time fit of tail currents elicited at –70 mV, after a 2 s test pulse to +40 mV. {tau}2 was 246.01±21.12 ms (n=3) for HERG currents and 243.53±47.22 ms (n=3) for HERG+minK currents, while {tau}1 was 968.14±82.61 ms (n=3) for HERG currents and 974.13±120.94 ms (n=3) for HERG+minK currents.


Figure 1
View larger version (16K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Representative currents evoked from oocytes injected with HERG cRNA alone (A and C) or coinjected with minK cRNA (B and D). The membrane potential was held at –70 mV, then stepped to 2 s test pulses varying from –50 to +40 mV in 10 mV increments (A and B). Currents shown in panel C and D were evoked by application of a 2 s test pulse to +20 mV from a holding potential of –70 mV, and stepped to tail potentials ranging from –120 to –30 mV in 10 mV increments. Dashed lines indicate the zero current level.

 
3.2 Drug-induced facilitation and block of HERG currents
In a first series of experiments 2 s test pulses ranging from –50 to +40 mV were applied in 10 mV increments from a holding potential of –70 mV and the effect of 5 µmol/l P and NP was investigated. A steady-state facilitation of HERG and HERG+minK currents was observed on currents evoked during test and tail pulses, with an effect most pronounced on the peak of the tail current (Fig. 2A). Facilitation of HERG and HERG+minK currents in the presence of P amounted 133.24±5.89% (n=4) and 136.68±2.42% (n=4), respectively as calculated on the peak of the tail current which was clamped at –70 mV after a 2 s test pulse to +40 mV. 100% was taken as the control current level. Similarly, the calculated facilitation for NP was 132.140±7.43% (n=4) on HERG currents and 140.19±5.58% (n=4) on HERG+minK currents. Furthermore, a cross-over phenomenon of the currents during the tail pulse was observed, caused by a combination of the agonistic effect and an increase of the deactivation rate, as will be reported in the next section.


Figure 2
View larger version (15K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Representative experiments illustrating the effects of increasing concentrations of NP on HERG+minK currents. The membrane potential was held at –70 mV, then stepped to 2 s test pulses to +40 mV. Panel A: facilitating effect and cross-over phenomenon caused by application of 5 µmol/l NP (NP1). The inset illustrates the effect on the time to the peak of the tail current. Current traces have been scaled for comparison. Panel B: further increase of deactivation rate and block by application of 50 and 125 µmol/l NP (NP2 and NP3). C indicates the control current. Dashed lines indicate the zero current level. Peak tail currents, obtained as in Fig. 1A and 1B, were fitted to a standard Boltzmann equation for control conditions (HERG bullet and HERG+minK {circ}), after application of 100 µM P (Panel C: HERG {blacktriangleup} and HERG+minK {bigtriangleup}) and 125 µM NP (Panel D: HERG {blacksquare} and HERG+minK {square}). V1/2=–21.07±1.81 mV and k=8.71±0.92 mV for HERG currents and V1/2=–20.84±0.38 mV and k=10.11±0.69 mV for HERG+minK currents. V1/2 was shifted to more depolarized potentials by P3 (V1/2=–18.80±1.14 mV for HERG currents and V1/2=–15.36±0.56 mV for HERG+minK currrents) and NP3 (V1/2=–14.80±1.98 mV for HERG currents and V1/2=–10.51±1.62 mV for HERG+minK currrents).

 
In a second series of experiments the effects of higher concentrations P (50 and 100 µmol/l) and NP (50 and 125 µmol/l) were investigated under the same conditions. After application of 50 µmol/l of P or NP, a small level of block was observed on activating currents, with no significant effect on the peak of the tail amplitude. More importantly, the deactivation rate was further increased as illustrated in Fig. 2B. 100 µmol/l P and 125 µmol/l NP caused a steady-state block of HERG currents (36.21±9.53%, n=4 for P and 33.77±4.18%, n=4 for NP) and HERG+minK currents (42.81±4.31%, n=4 for P and 37.39±10.68%, n=4 for NP) as measured on the peak of the tail current. This effect was partially reversible upon 5 minutes washout. Boltzmann activation curves were constructed with normalized peak tail amplitudes obtained after application of 100 µmol/l P and 125 µmol/l NP (Fig. 2C and D). V1/2 was shifted to more depolarized potentials by P (V1/2=–18.80±1.14 mV, n=3 for HERG currents and V1/2=–15.36±0.56 mV, n=3 for HERG+minK currrents) and NP (V1/2=–14.80±1.98 mV, n=4 for HERG currents and V1/2=–10.51±1.62 mV, n=4 for HERG+minK currrents).

3.3 Effect of P and NP on channel gating characteristics
The effect of P and NP on HERG and HERG+minK current activation and deactivation kinetics was analyzed by application of 2 s test pulses to –10 mV from a holding potential of –70 mV. Tail currents were clamped at –70 mV. Experiments show that increasing concentrations of P and NP not only decreased current activation rate constants but also increased deactivation rate constants (Fig. 3A, C and D). The time course of HERG+minK current activation was characterized by a time constant ({tau}) of 946.36±20.29 ms (n=3) after application of 125 µmol/l NP and 1039.21±45.99 ms (n=4) after application of 100 µmol/l P. HERG+minK tail currents evoked after application of 125 µmol/l NP (n=3) were fitted with a biexponential function and a {tau}2-value (fast deactivation phase) of 69.40±4.53 ms and a {tau}1-value (slow deactivation phase) of 292.77±54.32 were calculated. {tau}2 and {tau}1-values after application of 100 µmol/l P (n=3) were 103.61±8.25 ms and 391.67±43.71 ms, respectively. Similar results were obtained for HERG currents (data not shown). The effects of other concentrations tested on HERG+minK current activation and deactivation are summarized in Fig. 3C and 3D, respectively. For comparison we analyzed the effect of 125 µmol/l NP on RCK1 (Kv1.1), another voltage-dependent K+ channel with 6 transmembrane regions but characterized by delayed outward rectification (Fig. 3B). Here, deactivation rate constants of RCK1 tail currents decreased by drug application, resulting in a cross-over phenomenon which is different from the one observed on HERG and HERG+minK current after application of 5 µmol/l P and NP. This effect was not further investigated since it was not considered relevant for the cardiotoxicity of these drugs.


Figure 3
View larger version (13K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 The effect of increasing concentrations NP on activation and deactivation kinetics. Panel A: 125 µmol/l NP (NP3) decreases activation rate constants and increases deactivation rate constants. In contrast, the same concentration of NP on RCK1 (Kv1.1), another voltage-dependent K+ channel with 6 transmembrane regions but characterized by delayed outward rectification, caused a decreased deactivation rate (Panel B), resulting in a cross-over phenomenon which is different from the one shown in Fig. 2A. Dashed lines indicate the zero current level. Panels C and D summarize the effect of increasing concentrations NP on activating and deactivating time constants for HERG+minK currents, respectively. Activating currents were fitted to a single exponential function, while the fast ({tau}2, {square}) and slow ({tau}1, {blacksquare}) component of deactivating currents were resolved with a biexponential time fit.

 
3.4 Drug-induced changes of the reversal potential
When P and NP were applied at the highest concentration tested, 1000 µmol/l and 250 µmol/l respectively, the outwardly deactivating tail current, seen in control conditions, was switched to an inwardly deactivating tail current (Fig. 4A), although the tail was clamped –70 mV, a value more positive than Erev (approx. –89 mV in 2 mmol/l [K+]o). To analyze the effect on Erev, tail currents were studied at tail potentials ranging from –120 to –30 mV in 10 mV increments, after a 2 s test pulse to +20 mV. The application of 250 µmol/l NP shifted the Erev of HERG and HERG+minK currents to –60.30±4.54 mV (n=4) and –59.76±4.36 mV (n=4), respectively. The effect of other concentrations on the Erev of HERG and HERG+minK currents is summarized in Fig. 4B. Similar results were obtained for P (Erev=–77.73±3.30 mV, n=4 after application of 100 µmol/l). In contrast, no effect of P and NP on the Erev of RCK1 currents could be observed (Fig. 4B). Uninjected oocytes were analyzed under the same conditions as HERG currents and no inward currents could be observed upon application of 250 µmol/l NP (n=4, data not shown).


Figure 4
View larger version (13K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 Panel A: a 2 s test pulse to +20 mV was applied from a holding potential of –70 mV. The outwardly deactivating HERG+minK tail current was ‘switched’ to an inwardly deactivating tail current after application of 250 µmol/l NP (NP4), although the tail current was clamped at –70 mV, a value more positive than Erev (approx. –89 mV in 2 mmol/l [K+]o. The dashed line indicates the zero current level. Panel B summarizes the effect of increasing concentrations of NP on the Erev of HERG ({blacksquare}) and HERG+minK ({square}) currents, measured in 2 mmol/l [K+]o. Interestingly, 250 µmol/l NP did not significantly affect the Erev of RCK1 currents ({bigtriangleup}). Panel C: the Erev was measured in external solutions containing different concentrations K+ in the absence ({circ}) and in the presence of 250 µmol/l NP (– {bigtriangleup} –). Data points from both conditions were fitted with the Goldman–Hodgkin–Katz current equation. The calculated K+ over Na+ permeability factor was 75.2 in control conditions and 2.5 in presence of 250 µmol/l NP, meaning a 30-fold increase of the Na+-permeability. The Erev was also determined in external Na+-free solutions (Na+ substituted for TRIS) containing different concentrations K+ (bullet). The Erev, measured in TRIS solution, was not significantly changed in the presence of 250 µmol/l NP ({blacktriangleup}). The dashed curve was drawn according to the Nernst equation (58 mV/decade), indicating the relation for a perfectly selective K+ channel.

 
To elucidate the cause of the shift of Erev to more depolarized potentials we investigated the effect of NP on the Erev in external ND-96 solutions containing different concentrations of KCl (0.25–98 mmol/l, Fig. 4C). In absence of the drug, the Erev varied over the entire range of extracellular K+ concentrations in a manner well described by the Goldman–Hodgkin–Katz current equation [17,18]. We calculated that HERG+minK in control is selectively permeable to K+ over Na+ by a factor of 75.2. When the Erev was determined under the same experimental conditions, but in presence of 250 µmol/l NP, the K+ over Na+ permeability was 2.5, meaning a 30-fold increase of the Na+-permeability. To confirm that the shift of the Erev is caused by an increased Na+-permeability, we determined the Erev in Na+-free solutions (Na+ substituted for TRIS) supplemented with varying concentrations of KCl, in absence and presence of 250 µmol/l NP. Indeed, no effect of NP on the Erev in Na+-free solutions was observed.

3.5 Pharmacology of P and NP on HERGS631C mutant
Currents of oocytes expressing HERGS631C channels were evoked by application of depolarizing test pulses ranging from –70 to +50 mV in 10 mV increments, from a holding potential of –90 mV. Tail currents were clamped at –70 mV. Before each experiment, oocytes were treated with 0.1% H2O2 for 2–5 min. Representative traces of HERGS613C currents evoked under these conditions are illustrated in Fig. 5A. HERGS631C channels with thiol groups in oxidized states are characterized by disrupted C-type inactivation and K+-selectivity, as very recently described by Fan et al. [25]. Unlike WT HERG channels, mutant channels display inwardly activating currents at test potentials more negative than 0 mV and outwardly rectifying currents at test potentials more positive than 0 mV. Tail currents inwardly deactivate at –70 mV. Erev was determined by clamping tail currents at potentials ranging from –90 to +20 mV in 10 mV increments, after a 1 s test pulse to +40 mV. Under control conditions (Fig. 5B), Erev was –1.67±4.77 mV (n=6), consistent with the low K+-selectivity of HERGS631C channels. Upon application of 250 µmol/l NP, no change of Erev could be observed (n=4, Fig. 5C). Similar results were obtained upon application of 250 µmol/l P (n=4, data not shown).


Figure 5
View larger version (12K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 Representative current traces evoked from oocytes injected with HERGS631C cRNA. Before each experiment, oocytes were treated with 0.1% H2O2 for 2–5 min. HERGS631C channels with thiol groups in oxidized states are characterized by disrupted C-type inactivation and K+-selectivity, as recently described by Fan et al. [25]. Panel A: depolarizing test pulses were applied ranging from –70 to +50 mV in 10 mV increments, from a holding potential of –90 mV. Tail currents were clamped at –70 mV. Panel B: Erev was determined by clamping tail currents at potentials ranging from –90 to +20 mV in 10 mV increments, after a 1 s test pulse to +40 mV. Under control conditions (Fig. 5B), Erev was –1.67±4.77 mV (n=6), consistent with the low K+-selectivity of HERGS631C channels. Panel C: in contrast to WT HERG channels, application of 250 µmol/l NP caused no change of Erev (n=4). The dashed line indicates the zero current level. Currents shown in panel A were corrected by off-line linear leak subtraction. Statistical analysis revealed no significant difference of time constants describing the deactivation process in the absence and presence of drug.

 
3.6 Concentration–response curves
Taking into account the multitude of effects described above, the use of peak tail current amplitudes (Fig. 6, open symbols) to construct a concentration–response relationship was not satisfactory. For instance, a Hill curve would have to be constructed using a combination of current percentage values higher than 100% at 5 µmol/l (due to facilitation) and negative percentages at the highest drug concentration (due to inward tail currents). Therefore, concentration–response curves were constructed using tail current amplitudes, measured at 3500 ms at –70 mV (Fig. 6, filled symbols). We calculated that HERG currents are blocked by P and NP with an IC50-value of 44.70±4.19 µmol/l (n=4) and 33.20±4.36 µmol/l, (n=4) respectively. The Hill coefficients were 1.16±0.15 and 1.03±0.06, respectively. The calculated IC50-values for HERG+minK blockade are 40.92±6.54 µmol/l (n=3) and 40.15±11.91 µmol/l (n=4), respectively, with Hill coefficients of 1.03±0.04 and 1.21±0.27, respectively. Statistical analyses revealed no significant difference (P=0.05) between affinity of P and NP for HERG channel block and no role of the minK subunit could be determined.


Figure 6
View larger version (11K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6 Concentration–response relationships for the inhibitory effect of P and NP on HERG (Panel A) and HERG+minK currents (Panel B). The effect on the peak of the tail current is indicated for P ({bigtriangleup}) and NP ({square}), as well as at 3500 ms, for P ({blacktriangleup}) and NP ({blacksquare}), respectively. The inhibited current was expressed as the percentage inhibition/control current and the calculated curves have been drawn according to a standard logistic equation. We calculated that HERG currents are blocked by P and NP with an IC50-value of 44.70±4.19 µmol/l and 33.20±4.36 µmol/l, respectively. The Hill coefficients were 1.16±0.15 and 1.03±0.06, respectively. The calculated IC50-values for HERG+minK blockade are 40.92±6.54 µmol/l and 40.15±11.91 µmol/l, respectively, with Hill coefficients of 1.03±0.04 and 1.21±0.27, respectively. Statistical analyses revealed no significant difference (P=0.05) between affinity of P and NP for HERG channel block and no role of the minK subunit could be determined.

 

   4 Discussion
Top
 Abstract
 1 Introduction
 2 Materials and methods
 3 Results
 4 Discussion
References
 
It has been reported that a variety of drugs, including psychotropic drugs, exert their cardiotoxic effect by HERG channel blockade and cause an excessive prolongation of action potentials, leading to an acquired form of the LQTS, which can degenerate in fatal torsade de pointes. Propoxyphene (P) is often implicated in fatal drug poisoning and several lines of evidence suggest that not only INa but also IK block may contribute to its naloxone-insensitive cardiotoxic effects. Since it is not known which K+ current is affected by P in native cells and since IKr is a principal repolarizing current in cardiac myocytes, we investigated the effects of P and NP on the HERG channel. In addition, we determined whether the minK subunit contributed to pharmacological characteristics. Abbott et al. [19] have shown very recently that mixed complexes of HERG+MiRP1 (KCNE2) resemble distinctive biphasic inhibition by the class III antiarrhytmic E-4031, as observed with native IKr, unlike channels formed only with HERG or HERG+minK (KCNE1). Therefore, a possible role for MiRP1 might exist for the HERG block mechanism by P and NP.

In the present study, we demonstrated that a concentration of 5 µmol/l P and NP causes a steady-state facilitation of HERG and HERG+minK currents. Kinetic analysis revealed that the same concentration decreases channel activation rate constants and increases channel deactivation rate constants, thereby causing a cross-over phenomenon of the tail currents. A similar effect on Ito deactivation kinetics in rat ventricular myocytes by application of micromolar concentrations of naloxone, a non-selective opioid antagonist, was recently reported [20].

Holland and Steinberg [21] previously reported that superfusion with 3 µmol/l P and NP changes conducting characteristics of canine Purkinje fibers in the heart, by decreasing the maximal rate of rise of the action potential upstroke and decreasing the total action potential duration and cellular refractoriness. While the initial effect on the depolarization phase can be explained by block of INa, the shortening of the repolarization phase correlates remarkably well with our finding that IKr is facilitated, since a drug-induced increase of K+ efflux will return the membrane potential faster to its resting potential. Furthermore, preliminary experiments in our laboratory show that P and NP cause a similar agonistic effect on IKs, the slowly-activating delayed rectifier K+ current, which is regulated by coassembly of KvLQT1 and minK. In contrast, the inwardly rectifying K+ channel IRK1 remains unaffected, while the G protein coupled inwardly rectifying K+ channel (GIRK) is blocked by application of 5 µmol/l P [22].

Horrigan and Gilly [23] recently demonstrated that drug concentrations achieved in mammals during P poisoning are sufficient to block K+ channels in squid neurons and GH3 cells with a KD of 32 µmol/l. It was proposed that IK is blocked by diffusion of the non-protonated form of the drug into the cytoplasm and occlusion of the open K+ channel at the internal quaternary ammonium (QA) site by the protonated form of the drug. We calculated that P and NP block HERG currents with an IC50-value of approx. 40 µmol/l, using current amplitudes measured at 3500 ms. No role of the minK subunit could be determined. Kinetic analysis revealed that P and NP influence channel gating by decreasing channel activation rate constants and increasing channel deactivation rate constants. Boltzmann activation curves constructed, using normalized peak tail amplitudes before and after application of 100 µmol/l P and 125 µmol/l NP, revealed that the V1/2 was shifted to more depolarized potentials by P and NP. A previous study on canines showed that a steady-state plasma concentration of 3.7 µmol/l P decreased heart rate, while higher concentrations increased heart rate and QTc-intervals well above the control level [21]. Therefore, it is plausible to suggest that the HERG channel blockade, which we observed at high concentrations (≥50 µmol/l), contributes to the observed tachycardiac effects and ECG abnormalities, such as QT-prolongation.

The fact that P and NP can facilitate HERG and HERG+minK currents at lower concentrations and cause HERG and HERG+minK channel block at higher concentrations, raises questions about the mechanism by which these effects are exerted. We hypothesize that macroscopic channel block is the result of a drug-induced shift of the V1/2 to more depolarized potentials combined with a drug-induced change of the ion-selectivity (see further). We also suggest that the facilitating effect of low concentrations of P and NP can be explained by an increased rate of recovery from fast inactivation, a decreased rate of fast inactivation or a combination of both. This conclusion is supported with our finding that the time to reach the peak tail current amplitude is decreased in the presence of 5 µmol/l P and NP (Fig. 2A, inset). Additionally, experiments revealed that 5 µmol/l NP increased the recovery from fast inactivation, without any significant effect on the rate of inactivation (Fig. 7).


Figure 7
View larger version (7K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 7 Effect of 5 µmol/l NP on the recovery from inactivation. HERG channels were fully activated by application of 20 s depolarizing test pulses to –15 mV. A hyperpolarizing step to –100 mV was applied for 25 ms to allow rapid recovery from inactivation. Upon repolarization to –15 mV the outward current inactivated again to the same current level as before the hyperpolarzing step. 5 µmol/l NP increases the recovery from inactivation, without affecting the rate of inactivation. The dashed line indicates the zero current level.

 
When P and NP were applied at the highest concentration tested, tail currents inwardly deactivated at –70 mV, clearly more positive than the control Erev of approx. –90 mV. Ion substitution experiments (Na+ for TRIS) revealed that this effect is caused by an increased Na+-permeability of HERG. Using the Goldman–Hodgkin–Katz current equation [17,18], we calculated a 30-fold increase of Na+-permeability when 250 µmol/l NP was applied. Although such a high drug concentration might not be achieved during P poisoning, our experiments show that lower and toxicologically relevant concentrations of P and NP significantly shift the Erev to more depolarized potentials. Therefore, we suggest that the increased Na+-permeability might also contribute to the cardiotoxic effects. Interestingly, in a Na+-free solution HERG still does not behave as a perfectly K+-selective channel as predicted by the Nernst equation. This is not surprising since it has been reported that EAG, a K+ channel with marked sequence similarities to HERG, is permeable to Ca2+, which was not substituted in the Na+-free solutions [24].

From a structural point of view, a very interesting parallel exists between the drug-induced Na+-permeability of the HERG pore as seen in our study, and the effect of outer mouth mutations on HERG channel function. Indeed, Fan et al. [25] reported very recently that S631K and S631E mutations disrupted K+-selectivity of the HERG pore, a change not seen in T449K or T449E in Shaker. The fact that P and NP in our study also disrupted K+-selectivity of the HERG pore, but not of Kv1.1 (a member of the Shaker-subfamiliy) is in agreement with the structure-function studies by Fan et al. [25] and suggests that an interaction of P and NP with S631 might contribute to the drug-induced alteration of ion-selectivity observed in our study. To test this hypothesis, we investigated the effects of P and NP on HERGS631C channels. Under oxidizing conditions, this mutant is characterized by disrupted C-type inactivation and K+-selectivity. Application of P and NP did not change Erev, indeed suggesting an important drug interaction with S631.

It is difficult to predict, a priori, how a drug with combined actions on INa and IKr might affect the cardiac action potential, resulting in the clinically observed effects. However, we can conclude from our experiments that 3 drug-induced effects on IKr may contribute to P cardiac toxicity: concentration-dependent facilitation and block of HERG currents as well as an increase of the Na+ permeability. To our knowledge, this phenomenon has not been described for any other drug thus far.

Time for primary review 15 days.


   Acknowledgements
 
We wish to thank Michel Ulens (United Solutions) for the development of a Microsoft Excel add-in supporting the Axon file format.

The human Isk clone was kindly provided by Michel Lazdunski and Jacques Barhanin, CNRS, Valbonne, France.

The HERG clone was donated by Mark Keating, HHMI, University of Utah, Salt Lake City, USA.

The HERGS631C mutant was a gift from Gea-Ny Tseng, Columbia University, New York, USA.

We thank Chantal Maertens for subcloning the human minK in pGEMHE and synthesizing cRNA.

Jan Tytgat is a research associate of the F.W.O. Vlaanderen.


   References
Top
 Abstract
 1 Introduction
 2 Materials and methods
 3 Results
 4 Discussion
 References
 

  1. Robbins E.B. The pharmacologic effects of a new analgesic a-4-dimethylamino-1,2-diphenyl-3-methyl-4-propionyloxybutane. J Am Pharmaceut Assoc (1955) 44:497–500.
  2. Cravey R.H., Shaw R.F., Nakamura G.R. Incidence of propoxyphene poisoning: a report of fatal cases. J Forensic Sci (1974) 19:72–80.[Medline]
  3. Chan G.L., Matzke G.R. Effects of renal insufficiency on the pharmacokinetics and pharmacodynamics of opioid analgesics. Drug Intell Clin Pharm (1987) 21:773–783.[Abstract]
  4. Amsterdam E.A., Rendig S.V., Henderson G.L., Mason D.T. Depression of myocardial contractile function by propoxyphene and norpropoxyphene. J Cardiovasc Pharmacol (1981) 3:129–138.[Web of Science][Medline]
  5. Whitcomb D.C., Gilliam F.R., Starmer C.F., Grant A.O. Marked QRS complex abnormalities and sodium channel blockade by propoxyphene reversed with lidocaine. J Clin Invest (1989) 84:1629–1636.[Web of Science][Medline]
  6. Horrigan F.T., Gilly W.F. Methadone block of K+ current in squid giant fiber lobe neurons. J Gen Physiol (1996) 107:243–260.[Abstract/Free Full Text]
  7. Pugsley M.K., Saint D.A., Walker M.J. An electrophysiological basis for the antiarrhythmic actions of the kappa-opioid receptor agonist U-50,488H. Eur J Pharmacol (1994) 261:303–309.[CrossRef][Web of Science][Medline]
  8. Pugsley M.K., Hayes E.S., Saint D.A., Walker M.J. Do related kappa agonists produce similar effects on cardiac ion channels? Proc West Pharmacol Soc (1995) 38:25–27.[Medline]
  9. Warmke J.W., Ganetzky B. A family of potassium channel genes related to eag in Drosophila and mammals. Proc Natl Acad Sci USA (1994) 91:3438–3442.[Abstract/Free Full Text]
  10. Roy M., Dumaine R., Brown A.M. HERG, a primary human ventricular target of the nonsedating antihistamine terfenadine. Circulation (1996) 94:817–823.[Abstract/Free Full Text]
  11. Suessbrich H., Waldegger S., Lang F., Busch A.E. Blockade of HERG channels expressed in Xenopus oocytes by the histamine receptor antagonists terfenadine and astemizole. FEBS Lett (1996) 385:77–80.[CrossRef][Web of Science][Medline]
  12. Suessbrich H., Schonherr R., Heinemann S.H., et al. The inhibitory effect of the antipsychotic drug haloperidol on HERG potassium channels expressed in Xenopus oocytes. Br J Pharmacol. (1997) 120:968–974.[CrossRef][Web of Science][Medline]
  13. Babe K.S. Jr., Serafin W.E. Goodman and Gilman's The Pharmacological Basis of Therapeutics. Hardman J.G., Limbird L.E., eds. (1996) IX Edn. 581–600.
  14. McDonald T.V., Yu Z., Ming Z., et al. A minK-HERG complex regulates the cardiac potassium current IKr. Nature (1997) 388:289–292.[CrossRef][Web of Science][Medline]
  15. Busch A.E., Busch G.L., Ford E., et al. The role of the IsK protein in the specific pharmacological properties of the IKs channel complex. Br J Pharmacol (1997) 122:187–189.[CrossRef][Web of Science][Medline]
  16. Liman E.R., Tytgat J., Hess P. Subunit stoichiometry of a mammalian K+ channel determined by construction of multimeric cDNAs. Neuron (1992) 9:861–871.[CrossRef][Web of Science][Medline]
  17. Goldman D.E. Potential, impedance, and rectification in membranes. J Gen Physiol (1943) 27:37–60.[Abstract/Free Full Text]
  18. Hodgkin A.L., Katz B. The effects of sodium on the electrical activity of the giant axon of the squid. J Physiol (1949) 108:37–77.[Free Full Text]
  19. Abbott G.W., Sesti F., Splawski I., et al. MiRP1 forms IKr potassium channels with HERG and is associated with cardiac arrhythmia. Cell (1999) 97:175–187.[CrossRef][Web of Science][Medline]
  20. Hung C.F., Wu M.H., Tsai C.H., et al. Electrophysiological mechanisms for the antiarrhythmic activities of naloxone on cardiac tissues. Life Sci (1998) 63:1205–1219.[CrossRef][Web of Science][Medline]
  21. Holland D.R., Steinberg M.I. Electrophysiologic properties of propoxyphene and norpropoxyphene in canine cardiac conducting tissues in vitro and in vivo. Toxicol Appl Pharmacol (1979) 47:123–133.[CrossRef][Web of Science][Medline]
  22. Ulens C., Robberecht W., Daenens P., Tytgat J. Apparent inverse agonism of opioids caused by direct GIRK1/GIRK2 blockade in oocytes co-expressing opioid receptors. Soc Neurosci Abstr (1998) 24:1094.
  23. Horrigan F.T., Gilly W.F. Methadone block of K+ current in squid giant fiber lobe neurons. J Gen Physiol (1996) 107:243–260.[Abstract/Free Full Text]
  24. Bruggemann A., Pardo L.A., Stuhmer W., Pongs O. Ether-a-go-go encodes a voltage-gated channel permeable to K+ and Ca2+ and modulated by cAMP. Nature (1993) 365:445–448.[CrossRef][Medline]
  25. Fan J.S., Jiang M., Dun W., McDonald T.V., Tseng G.N. Effects of outer mouth mutations on hERG channel function: A comparison with similar mutations in the Shaker channel. Biophys J (1999) 76:3128–3140.[Web of Science][Medline]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 JGPHome page
H. Gang and S. Zhang
Na+ Permeation and Block of hERG Potassium Channels
J. Gen. Physiol., June 26, 2006; 128(1): 55 - 71.
[Abstract] [Full Text] [PDF]

Home page
 Cardiovasc ResHome page
J. Tamargo, R. Caballero, R. Gomez, C. Valenzuela, and E. Delpon
Pharmacology of cardiac potassium channels
Cardiovasc Res, April 1, 2004; 62(1): 9 - 33.
[Abstract] [Full Text] [PDF]

Home page
 Cardiovasc ResHome page
P. Hajdu, C. Ulens, G. Panyi, and J. Tytgat
Drug- and mutagenesis-induced changes in the selectivity filter of a cardiac two-pore background K+ channel
Cardiovasc Res, April 1, 2003; 58(1): 46 - 54.
[Abstract] [Full Text] [PDF]

Home page
 JGPHome page
F. M. Mullins, S. Z. Stepanovic, R. R. Desai, A. L. George Jr., and J. R. Balser
Extracellular Sodium Interacts with the HERG Channel at an Outer Pore Site
J. Gen. Physiol., September 30, 2002; 120(4): 517 - 537.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Ulens, C.
Right arrow Articles by Tytgat, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ulens, C.
Right arrow Articles by Tytgat, J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?