Zhengfeng Zhou

Properties of HERG Channels Stably Expressed in HEK 293 Cells Studied at Physiological Temperature

We have established stably transfected HEK 293 cell lines expressing high levels of functional human ether-a go-go-related gene (HERG) channels. We used these cells to study biochemical characteristics of HERG protein, and to study electrophysiological and pharmacological properties of HERG channel current at 35 degrees C. HERG-transfected cells expressed an mRNA band at 4.0 kb. Western blot analysis showed two protein bands (155 and 135 kDa) slightly larger than the predicted molecular mass (127 kDa). Treatment with N-glycosidase F converted both bands to smaller molecular mass, suggesting that both are glycosylated, but at different levels. HERG current activated at voltages positive to -50 mV, maximum current was reached with depolarizing steps to -10 mV, and the current amplitude declined at more positive voltages, similar to HERG channel current expressed in other heterologous systems. Current density at 35 degrees C, compared with 23 degrees C, was increased by more than twofold to a maximum of 53.4 +/- 6.5 pA/pF. Activation, inactivation, recovery from inactivation, and deactivation kinetics were rapid at 35 degrees C, and more closely resemble values reported for the rapidly activating delayed rectifier K+ current (I(Kr)) at physiological temperatures. HERG channels were highly selective for K+. When we used an action potential clamp technique, HERG current activation began shortly after the upstroke of the action potential waveform. HERG current increased during repolarization to reach a maximum amplitude during phases 2 and 3 of the cardiac action potential. HERG contributed current throughout the return of the membrane to the resting potential, and deactivation of HERG current could participate in phase 4 depolarization. HERG current was blocked by low concentrations of E-4031 (IC50 7.7 nM), a value close to that reported for I(Kr) in native cardiac myocytes. Our data support the postulate that HERG encodes a major constituent of I(Kr) and suggest that at physiological temperatures HERG contributes current throughout most of the action potential and into the postrepolarization period.

HERG Channel Dysfunction in Human Long QT Syndrome INTRACELLULAR TRANSPORT AND FUNCTIONAL DEFECTS

Mutations in HERG are associated with human chromosome 7-linked congenital long QT (LQT-2) syndrome. We used electrophysiological, biochemical, and immunohistochemical methods to study the molecular mechanisms of HERG channel dysfunction caused by LQT-2 mutations. Wild type HERG and LQT-2 mutations were studied by stable and transient expression in HEK 293 cells. We found that some mutations (Y611H and V822M) caused defects in biosynthetic processing of HERG channels with the protein retained in the endoplasmic reticulum. Other mutations (I593R and G628S) were processed similarly to wild type HERG protein, but these mutations did not produce functional channels. In contrast, the T474I mutation expressed HERG current but with altered gating properties. These findings suggest that the loss of HERG channel function in LQT-2 mutations is caused by multiple mechanisms including abnormal channel processing, the generation of nonfunctional channels, and altered channel gating.

Mechanism of Block and Identification of the Verapamil Binding Domain to HERG Potassium Channels

Calcium channel antagonists have diverse effects on cardiac electrophysiology. We studied the effects of verapamil, diltiazem, and nifedipine on HERG K+ channels that encode IKr in native heart cells. In our experiments, verapamil caused high-affinity block of HERG current (IC50=143.0 nmol/L), a value close to those reported for verapamil block of L-type Ca2+ channels, whereas diltiazem weakly blocked HERG current (IC50=17.3 micromol/L), and nifedipine did not block HERG current. Verapamil block of HERG channels was use and frequency dependent, and verapamil unbound from HERG channels at voltages near the normal cardiac cell resting potential or with drug washout. Block of HERG current by verapamil was reduced by lowering pHO, which decreases the proportion of drug in the membrane-permeable neutral form. N-methyl-verapamil, a membrane-impermeable, permanently charged verapamil analogue, blocked HERG channels only when applied intracellularly. Verapamil antagonized dofetilide block of HERG channels, which suggests that they may share a common binding site. The C-type inactivation-deficient mutations, Ser620Thr and Ser631Ala, reduced verapamil block, which is consistent with a role for C-type inactivation in high-affinity drug block, although the Ser620Thr mutation decreased verapamil block 20-fold more than the Ser631Ala mutation. Our findings suggest that verapamil enters the cell membrane in the neutral form to act at a site within the pore accessible from the intracellular side of the cell membrane, possibly involving the serine at position 620. Thus, verapamil shares high-affinity HERG channel blocking properties with other class III antiarrhythmic drugs, and this may contribute to its antiarrhythmic mechanism.

Novel Mechanism Associated With an Inherited Cardiac Arrhythmia Defective Protein Trafficking by the Mutant HERG (G601S) Potassium Channel

BACKGROUND The congenital long-QT syndrome (LQTS) is an inherited disorder characterized by a prolonged cardiac action potential and a QT interval that leads to arrhythmia. Mutations in the human ether-a-go-go-related gene (HERG), which encodes the rapidly activating component of the delayed rectifier current (IKr), cause chromosome 7-linked LQTS (LQT2). Studies of mutant HERG channels in heterologous systems indicate that the mechanisms mediating LQT2 are varied and include mutant subunits that form channels with altered kinetic properties or nonfunctional mutant subunits. We recently reported a novel missense mutation of HERG (G601S) in an LQTS family that we have characterized in the present work. METHODS AND RESULTS To elucidate the electrophysiological properties of the G601S mutant channels, we expressed these channels in mammalian cells and Xenopus oocytes. The G601S mutant produced less current than wild-type channels but exhibited no change in kinetic properties or dominant-negative suppression when coexpressed with wild-type subunits. To examine the cellular trafficking of mutant HERG channel subunits, enhanced green fluorescent protein tagging and Western blot analyses were performed. These showed deficient protein trafficking of the G601S mutant to the plasma membrane. CONCLUSIONS Our results from both the Xenopus oocyte and HEK293 cell expression systems and green fluorescent protein tagging and Western blot analyses support the conclusion that the G601S mutant is a hypomorphic mutation, resulting in a reduced current amplitude. Thus, it represents a novel mechanism underlying LQT2.

Blockage of the HERG human cardiac K+ channel by the gastrointestinal prokinetic agent cisapride

Cisapride, a gastrointestinal prokinetic agent, is known to cause long Q-T syndrome and ventricular arrhythmias. The cellular mechanism is not known. The human ether-á-go-go-related gene ( HERG), which encodes the rapidly activating delayed rectifier K+current and is important in cardiac repolarization, may serve as a target for the action of cisapride. We tested the hypothesis that cisapride blocks HERG. The whole cell patch-clamp recording technique was used to study HERG channels stably expressed heterologously in HEK293 cells. Under voltage-clamp conditions, cisapride block of HERG is dose dependent with a half-maximal inhibitory concentration of 6.5 nM at 22°C ( n = 25 cells). Currents rapidly recovered with drug washout. The onset of block by cisapride required channel activation indicative of open or inactivated state blockage. Block of HERG with cisapride after channel activation was voltage dependent. At -20 mV, 10 nM cisapride reduced HERG tail-current amplitude by 5%, whereas, at +20 mV, the tail-current amplitude was reduced by 45% ( n = 4 cells). At -20 and +20 mV, 100 nM cisapride reduced tail-current amplitude by 66 and 90%, respectively. We conclude that cisapride is a potent blocker of HERG channels expressed in HEK293 cells. This effect may account for the clinical occurrence of Q-T prolongation and ventricular arrhythmias observed with cisapride.Cisapride, a gastrointestinal prokinetic agent, is known to cause long Q-T syndrome and ventricular arrhythmias. The cellular mechanism is not known. The human ether-á-go-go-related gene (HERG), which encodes the rapidly activating delayed rectifier K+ current and is important in cardiac repolarization, may serve as a target for the action of cisapride. We tested the hypothesis that cisapride blocks HERG. The whole cell patch-clamp recording technique was used to study HERG channels stably expressed heterologously in HEK293 cells. Under voltage-clamp conditions, cisapride block of HERG is dose dependent with a half-maximal inhibitory concentration of 6.5 nM at 22 degrees C (n = 25 cells). Currents rapidly recovered with drug washout. The onset of block by cisapride required channel activation indicative of open or inactivated state blockage. Block of HERG with cisapride after channel activation was voltage dependent. At -20 mV, 10 nM cisapride reduced HERG tail-current amplitude by 5%, whereas, at + 20 mV, the tail-current amplitude was reduced by 45% (n = 4 cells). At -20 and + 20 mV, 100 nM cisapride reduced tail-current amplitude by 66 and 90%, respectively. We conclude that cisapride is a potent blocker of HERG channels expressed in HEK293 cells. This effect may account for the clinical occurrence of Q-T prolongation and ventricular arrhythmias observed with cisapride.

Block of HERG potassium channels by the antihistamine astemizole and its metabolites desmethylastemizole and norastemizole.

Electrophysiologic Effects of Astemizole Metabolites. Introduction: The selective H1‐receptor antagonist astemizole (Hismanal) causes acquired long QT syndrome. Astemizole blocks the rapidly activating delayed rectifier K+ current IKr and the human ether‐a go‐go‐related gene (HKRG) K+ channels that underlie it. Astemizole also is rapidly metabolized. The principal metabolite is desmethylastemizole, which retains H1‐receptor antagonist properties, has a long elimination time of 9 to 13 days, and its steady‐state serum concentration exceeds that of astemizole by more than 30‐fold. A second metabolite is norastemizole, which appears in serum in low concentrations following astemizole ingestion and has undergone development as a new antihistamine drug. Our objective in the present work was to study the effects of desmethylastemizole, norastemizole, and astemizole on HERG K+ channels.

Long QT syndrome: cellular basis and arrhythmia mechanism in LQT2.

HERG Channel Dysfunction in LQT2. LQT2 is one form of the congenital long QT syndrome, It results from mutations in the human ether‐a‐go‐go‐related gene (HERG), and more than 80 mutations, usually causing single amino acid substitutions in the HERG protein, are known, HFRG encodes the ion channel pore‐forming subunit protein for the rapidly activating delayed rectifier K+ channel (IKr) in the heart. This review summarizes current findings about mutations causing LQT2, the mechanisms by which mutations may cause the clinical phenotype of a reduction in IKr and a prolonged QT interval, and how this may be involved in the generation of ventricular arrhythmias.

Droperidol lengthens cardiac repolarization due to block of the rapid component of the delayed rectifier potassium current

Droperidol Blocks Cardiac IKr. Introduction: Torsades de pointes have been observed during treatment with droperidol, a butyrophenone neuroleptic agent. Our objectives were (1) to characterize the effects of droperidol on cardiac repolarization and (2) to evaluate effects of droperidol on a major time‐dependent outward potassium current involved in cardiac repolarization (IKr).

Both T- and L-type Ca2+ channels can contribute to excitation-contraction coupling in cardiac Purkinje cells.

Although L-type Ca2+ channels have been shown to play a central role in cardiac excitation-contraction (E-C) coupling, little is known about the role of T-type Ca2+ channels in this process. We used the amphotericin B perforated patch method to study the possible role of T-type Ca2+ current in E-C coupling in isolated canine Purkinje myocytes where both Ca2+ currents are large. T-type Ca2+ current was separated from L-type Ca2+ current using protocols employing the different voltage dependencies of the channel types and their different sensitivities to pharmacological blockade. We showed that Ca2+ admitted through either T- or L-type Ca2+ channels is capable of initiating contraction and that the contractions depended on Ca2+-induced Ca2+ release from the sarcoplasmic reticulum (SR). The contractions, however, had different properties. Those initiated by Ca2+ entry through T-type Ca2+ channels had a longer delay to the onset of shortening, slower rates of shortening and relaxation, lower peak shortening, and longer time to peak shortening. These differences were present even when L-type Ca2+ current amplitude, or charge entry, was less than that of T-type Ca2+ current, suggesting that Ca2+ entry through the T-type Ca2+ channel is a less effective signal transduction mechanism to the SR than is Ca2+ entry through the L-type Ca2+ channel. We conclude that under our experimental conditions in cardiac Purkinje cells Ca2+ entry through the T-type Ca2+ channel can activate cell contraction. However, Ca2+ entry through the L-type Ca2+ channel is a more effective signal transduction mechanism. Our findings support the concept that different structural relationships exist between these channel types and the SR Ca2+ release mechanism.

Differences in action potential and early afterdepolarization properties in LQT2 and LQT3 models of long QT syndrome.

Long OT syndrome has many causes from both acquired and congenital disorders. For the congenital disorders, their presentation and disease course are not identical. We studied two pharmacological models of long QT syndrome (LQT) to identify differences in cellular electrophysiological properties that may account for this. LQT2 was simulated by suppression of the rapidly activating delayed rectifier potassium current (IKr) with the drug E‐4031, and LQT3 was simulated by slowing of the sodium current (INa) decay with the toxin ATX II. Single rabbit ventricular cell action potentials were studied using the amphotericin B perforated patch clamp technique. Action potential and early afterdepolarization (EAD) properties were rigorously defined by the frequency power spectra obtained with fast Fourier transforms. The E‐4031 (n=43 myocytes) and ATX II (n=50 myocytes) models produced different effects on action potential and EAD properties. The major differences are that ATX II, compared with E‐4031, caused greater action potential prolongation, more positive plateau voltages, lower amplitude EADs with less negative take‐off potentials, greater time to the EAD peak voltage, and longer duration EADs. Despite causing greater action potential prolongation, the incidence of EAD induction was much less with the ATX II model (28%) than with the E‐4031 model (84%). Thus these two pharmacological models have strikingly different cellular electrophysiological properties. Our findings provide cellular mechanisms that may account for some differences in the clinical presentation of LQT2 and LQT3.