Effects of cocaine and its major metabolites on the HERG-encoded potassium channel.

Cocaine abuse has been reported to result in QT prolongation in humans; however, the mechanisms underlying this effect are still poorly understood. In this study we compared the direct effects of cocaine and its major metabolites in human embryonic kidney 293 cells stably transfected with human ether-a-go-go-related gene (HERG). Cocaine blocked HERG-encoded potassium channels with an IC50 of 4.4 +/- 1.1 microM (22 degrees C). Cocaethylene (a metabolite formed in the presence of ethanol) had a significantly lower IC50 of 1.2 +/- 1.1 microM (P < 0.0001), and cocaine's primary pyrolysis metabolite methylecgonidine blocked HERG with a higher IC50 of 171.7 +/- 1.2 microM. In contrast, 1 mM ecgonine methylester or benzoylecgonine produced only a minimal block (21 +/- 4 and 15 +/- 8%, respectively). Blockade of HERG by cocaine, cocaethylene, and methylecgonidine increased significantly over the voltage range where HERG activates, but became constant at voltages where HERG activation was maximal, indicating that all three drugs block open channels, but by a mechanism that is not highly sensitive to voltage per se. Cocaine and cocaethylene also significantly slowed the time course of deactivation at -60 mV, an effect consistent with open channel block. We conclude that cocaethylene is slightly more potent than cocaine as a blocker of HERG, whereas methylecgonidine has much lower potency, and both benzoylecgonine and ecgonine methyl ester are essentially inactive at clinically relevant concentrations.

Quinidine interactions with human atrial potassium channels: developmental aspects.

Clinical studies have suggested that quinidine is less effective when used for the treatment of atrial arrhythmias in pediatric patients compared with its clinical effectiveness in the adult patient population. Age-related changes in the cardiac actions of quinidine on action potential duration and interaction with potassium channels in several mammalian species also have been reported. We investigated the effects of postnatal development on quinidine's interaction with major repolarizing currents (Ito, IKur, Ins, and IK1) in human atrial myocytes, using the whole-cell configuration of the voltage-clamp technique. Our results indicate that there are age-related changes in both the IC50 for quinidine blockade of Ito, as well as the mechanism of quinidine unblocking. In contrast, quinidine was found to inhibit both adult and pediatric IK1 and IKur in an age-independent manner, whereas the nonselective cation current (Ins), which contributes to the sustained outward current (Isus), was insensitive to quinidine. The results from this study help to clarify the electrophysiological mechanism by which quinidine elicits its antiarrhythmic effect in the pediatric and adult human population.

L-type calcium current in pediatric and adult human atrial myocytes: evidence for developmental changes in channel inactivation.

Animal studies have documented the presence of marked, species-dependent, developmental changes in the properties of the L-type calcium current in cardiac myocytes. In an effort to understand the postnatal changes which occur in the calcium current in human heart, we characterized the calcium current in atrial myocytes isolated from 17 pediatric and older children (ages 3 d to 14 y) and 12 adult (ages 43-79 y) human hearts using the whole-cell patch clamp technique. In contrast to animal models, we found no evidence for age-related changes in calcium current density, steady-state inactivation, or kinetics of recovery from inactivation, suggesting that, in human atrium, calcium channels are in many aspects functionally mature at the time of birth. However, statistically significant differences were found in the kinetics of calcium current inactivation, with calcium current measured in cells isolated from pediatric human atria inactivating approximately 2-fold faster than cells isolated from adult hearts. These results suggest a possible role for age-related changes in calcium current inactivation in the shortened action potential duration observed in pediatric compared with adult human atrium.

Analysis of the ionic basis for cocaine's biphasic effect on action potential duration in guinea-pig ventricular myocytes.

The effects of cocaine on the duration of the cardiac action potential were investigated in isolated guinea-pig ventricular myocytes at 37 degrees C. Following a 10-min exposure of cells to 3 microM cocaine, APD90 increased significantly by +22 +/- 5% (n = 6). In contrast, following a ten minute exposure to 30 or 100 microM cocaine, APD90 was reduced by -24 +/- 6% (n = 5) and -53 +/- 2% (n = 8), respectively. The ionic basis for cocaine's effects on the APD was investigated using the whole cell voltage-clamp technique at 37 degrees C. Cocaine produced a concentration-dependent reduction in the amplitude of IK tail currents with an estimated IC50 of 4 microM. The kinetics and voltage dependence of the cocaine-sensitive current indicate that cocaine selectively blocks a current identical to the E-4031 sensitive current IKr. No significant reduction of the slow component of IK (IKs) was observed during exposure to 30 or 100 microM cocaine. High (30 and 100 microM) concentrations of cocaine also produced a significant reduction of both the L-type calcium current and the TTX-sensitive plateau current. Pre-treatment of cells with 10 microM TTX also converted the APD-shortening effect of 30 microM cocaine to one of APD-prolonging. This implies that cocaine block of a TTX-sensitive window current contributes to the APD-shortening effects produced by high concentrations of cocaine. We conclude that: (1) cocaine produces a biphasic concentration-dependent effect on repolarization in guinea-pig ventricular myocytes; and (2) this biphasic effect on repolarization results from differences in the sensitivity of inward and outward currents to the blocking effects of cocaine.

Description of a nonselective cation current in human atrium.

Ion currents were examined in isolated human atrial myocytes by using the whole-cell patch-clamp technique. When currents were recorded with a K(+)-containing pipette solution, depolarizing voltage pulses elicited a rapidly activating outward current that decayed to an apparent steady state. Exposure of cells to 10 mmol/L 4-aminopyridine markedly reduced current amplitude; however, a rapidly activating current that was approximately 30% of the steady state current amplitude remained. When pipette K+ was replaced with Cs+, a similar rapidly activating current that reversed polarity at approximately 0 mV was recorded. This current was seen in 100% of the cells tested from 17 different hearts (n = 142), and its amplitude was approximately 40% of the amplitude of the steady state current recorded in the presence of pipette K+. The current amplitude was not significantly different in cells isolated from adult (6.31 +/- 1.35 pA/pF, n = 8) and pediatric (5.54 +/- 1.04 pA/pF, n = 9) hearts. Studies designed to determine the charge-carrying species indicated that changes in bath Cl- concentration had no effect on either the amplitude or the reversal potential of this current, whereas removal of pipette Cs+ and bath Na+ dramatically reduced this current. In addition, this current was not modulated by either isoproterenol (1 mumol/L, 22 degrees C) or cell swelling. This study provides the first description of a nonselective cation current in human atrial myocytes, which may play an important role in repolarization in human atria.

The pH dependence of cocaine interaction with cardiac sodium channels.

Previous in vitro and in vivo studies have provided evidence implicating cocaine block of cardiac sodium channels as a putative mechanism for cocaine-induced arrhythmias and sudden death. Cocaine also has been shown to cause seizures which can result in respiratory and/or metabolic acidosis. In this study we investigated how changes in both internal pH (pHi) and external pH (pHo) over the range of 6.6 to 9.2 modify the sodium channel blocking properties of cocaine in isolated guinea pig ventricular myocytes by using the whole-cell variant of the patch clamp technique. Use-dependent block produced by a train of 1-sec pulses to -20 mV was not affected by changes in pHi, but both the amplitude and time constant for approaching steady-state block were significantly affected by changes in pHo. Characterization of the time course of cocaine binding during a depolarizing pulse indicated that the kinetics of drug interaction with inactivated channels were independent of pHi, but were significantly affected by changes in pHo. The rate of recovery from channel block at a holding potential of -140 mV also was independent of pHi, but strongly dependent on pHo, with the unblocking time constant decreasing exponentially as pHo was increased. The results of this study indicate that cocaine's effect on cardiac sodium channels can be modulated significantly by changes in pHo, and provide further support for previously poorly tested assumptions of the modulated receptor hypothesis.

Comparison of Ito in young and adult human atrial myocytes: evidence for developmental changes.

In an effort to understand the ionic basis for the developmental changes that have been reported to occur in the configuration of the human atrial action potential, we characterized the transient outward current (Ito) and the inward rectifier current in atrial myocytes isolated from 20 young (ages 1 day-2.5 yr) and 8 adult (11-68 yr) human hearts using the whole cell patch-clamp technique. We found evidence for statistically significant (P < 0.05) age-related changes in the Ito, including 1) the presence of an Ito in only 67% of the cells isolated from young hearts vs. 100% of the cells isolated from adult hearts, 2) an almost twofold increase in the current density of Ito in adult cells vs. young cells, and 3) recovery kinetics that are approximately twofold slower in young myocytes relative to adult myocytes. In contrast, there were no age-related changes found in the current density of the inward rectifier current or the sustained current measured after the decay of Ito. These results suggest important current-dependent changes that occur with age in human atria.

Cocaethylene, a metabolite of cocaine and ethanol, is a potent blocker of cardiac sodium channels.

Cocaethylene is an active metabolite of cocaine believed to play a causative role in the increased incidence of sudden death in individuals who coadminister ethanol with cocaine. However, the direct effects of cocaethylene on the heart have not been well defined. In this study, we defined the effects of cocaethylene on the cardiac Na current (INa) in guinea pig ventricular myocytes at 16 degrees C using the whole-cell patch-clamp method. Cocaethylene (10-50 microM) produced both a significant tonic block and a rate-dependent block of INa at cycle lengths between 2 and 0.2 sec. Cocaethylene produced a significantly greater tonic block than cocaine at a concentration of 50 microM and produced a significantly greater use-dependent block over a 5-fold range of drug concentrations (10-50 microM) and cycle lengths (0.2-1.0 sec). Analysis of channel-blocking characteristics revealed that cocaethylene had a significantly higher affinity for inactivated channels (Kdi = 5.1 +/- 0.6 microM, n = 15) compared with cocaine (Kdi = 7.9 +/- 0.5 microM, n = 10) (P < .01) and that cocaethylene produced a significantly greater hyperpolarizing shift of the steady-state INa inactivation curve (P < .05). Cocaethylene also had a significantly longer time constant for recovery from channel block at -140 mV (12.24 +/- 0.88 sec, n = 16) compared with cocaine (8.33 +/- 0.56 sec, n = 14) (P < .01).(ABSTRACT TRUNCATED AT 250 WORDS)

Electrophysiological effects of high cocaine concentrations on intact canine heart. Evidence for modulation by both heart rate and autonomic nervous system.

BACKGROUND: Previous clinical reports have suggested that cocaine intoxication may produce severe ventricular arrhythmias due to a direct effect on the heart. However, the effects of high plasma levels of cocaine on the electrophysiology of the heart have not been well characterized and remain poorly understood. METHODS AND RESULTS: The purpose of this study was to characterize the electrophysiological effects of high doses of cocaine on the in situ dog heart. In dogs anesthetized with morphine and alpha-chloralose, cocaine (2-11 micrograms/mL) increased both atrial and ventricular refractory periods and produced rate-dependent increases in atrial, atrioventricular, His-Purkinje, and ventricular conduction intervals. The time constant for the onset of cocaine's conduction slowing effect following a reduction in pacing cycle length from 400 to 260 msec was approximately two beats, and the time constant for diastolic recovery from conduction slowing was approximately 200 msec, which are similar to values reported for several class Ib antiarrhythmic drugs. Cocaine produced a rate-dependent increase in QT interval that was greatest at high heart rates yet produced no change in the ST (QT-QRS) interval. This suggests that high plasma levels of cocaine delay repolarization primarily via slowing of conduction. Cocaine's effects on both atrioventricular and intraventricular conduction were significantly larger in autonomically blocked than in autonomically intact animals. CONCLUSIONS: We conclude that high plasma levels of cocaine, similar to those reported in autopsy reports following fatal cocaine overdose in humans, produce significant rate-dependent conduction slowing effects on atrial, atrioventricular, and ventricular conduction in the in situ heart. These rate-dependent effects are intensified following autonomic blockade.

Characterization of the rate-dependent effects of ethmozine on conduction, in vivo, and on the sodium current, in vitro, in the newborn heart.

We evaluated the effects of ethmozine on resting conduction intervals, myocardial refractory periods and His-Purkinje and intraventricular conduction in vivo in 11 neonatal dogs aged 7 to 16 days. Ethmozine produced significant prolongation in resting sinus cycle length (P < .05), in the atrioventricular nodal (P < .001) and His-Purkinje conduction time (P < .01) intervals and in the QRS duration (P < .01). Ventricular, but not atrial refractory periods were significantly prolonged (P < .05). Rate-dependent changes in His-Purkinje and intraventricular conduction were demonstrated after ethmozine by direct pacing of the bundle of His and right ventricular pacing. The development of steady-state conduction delay at a paced cycle length of 200 msec was characterized by a time constant (tau on) of 23.5 beats. The time constant of diastolic recovery (tau off) from rate-dependent conduction delay, determined during His bundle extrastimulation, was 171 msec. Ethmozine was highly proarrhythmic. A total of 7 arrhythmias were induced in 6 out of 12 neonates after administration of ethmozine. We also characterized ethmozine block of cardiac sodium channels in isolated neonatal canine ventricular myocytes using the whole cell variation of the patch clamp technique. Ethmozine (1.3-40 microM) produced a use-dependent block of cardiac Na channels that was dependent upon both drug concentration and pulse duration. Drug binding to inactivated channel states accounted for the observed use-dependent block. The time constant of recovery from use-dependent block ranged between 10 and 30 sec at 16 degrees C.(ABSTRACT TRUNCATED AT 250 WORDS)

Developmental changes in the effects of lidocaine on sodium channels in rat cardiac myocytes.

Previous studies have suggested that there are developmental changes in the sodium channel blocking properties of class I antiarrhythmic drugs, yet this hypothesis has not been well tested using measurements of sodium current. In this study we defined the effects of lidocaine on the cardiac sodium current in neonatal (1-2-day old) and adult rat ventricular myocytes using the whole-cell variation of the patch-clamp technique (16 degrees C, [Na]i = 15 mM, [Na]o = 25 mM). Lidocaine (30 microM) produced significantly more tonic block at negative holding potentials (e.g., -140 mV) in neonatal myocytes (23.2 +/- 7.0%, mean +/- S.E.M., n = 9) compared to adult (6.5 +/- 1.1%, n = 12) (P less than .05). The percentage of use-dependent block obtained during trains of 10-msec pulses at a cycle length of 200 msec was also significantly greater in neonatal myocytes (22.5 +/- 5.6%, n = 9) compared to adult myocytes (6.9 +/- 2.2%, n = 7) (P less than .02). Analysis of the kinetics of block development at -20 mV indicated that neonatal cells have a lower dissociation constant for lidocaine interaction with inactivated channels (10.1 +/- 1.3 microM) compared to adult cells (16.5 +/- 1.9 microM)(P less than .02). A marked difference was found for the time constant of recovery from channel block, where neonates recovered from block approximately twice as slowly as adults (e.g., at -140 mV tau = 1.54 +/- 0.28 sec, n = 11 in neonates vs. tau = 0.64 +/- 0.07 sec, n = 13 in adults) (P less than .05).(ABSTRACT TRUNCATED AT 250 WORDS)

Characterization of the sodium channel blocking properties of the major metabolites of cocaine in single cardiac myocytes.

Clinical reports indicate that almost 30% of cocaine overdose-related deaths occur 2 to 5 hr after administration, when an appreciable amount of cocaine can be expected to have been converted to metabolites. We investigated the effects of the three major metabolites of cocaine: benzoylecgonine, ecgonine methyl ester and norcocaine on sodium channels in isolated guinea pig myocytes using the whole cell variant of the patch clamp technique at 16 +/- 0.5 degrees C. Benzoylecgonine and ecgonine methyl ester did not produce a significant block of sodium current at a concentration of 100 microM. In contrast, 30 microM norcocaine was found to reduce sodium current in a use-dependent manner qualitatively similar to cocaine. The time course of sodium current block development and recovery were characterized. With 30 microM norcocaine, two phases of block development were defined: a rapid phase (tau = 11.9 +/- 11.6 msec) and a slow phase (tau = 2.2 +/- 0.5 sec). Recovery from drug-induced block at -140 mV was approximately twice as fast for norcocaine (tau = 4.6 +/- 1.7 sec, n = 7) compared to cocaine (tau = 8.4 +/- 0.9 sec, n = 6). Norcocaine was found to have a higher affinity for inactivated cardiac sodium channels (Kdi = 5.7 +/- 0.9 microM) than cocaine (Kdi = 7.8 +/- 1.2 microM) (P less than .01); however, norcocaine produced less use-dependent block due to its faster unbinding kinetics. These data indicate that although norcocaine and cocaine are potent sodium channel blockers, benzoylecgonine and ecgonine methyl ester are ineffective blockers at clinically relevant concentrations (i.e., less than or equal to 100 microM).

Rate-dependent effects of lidocaine on His-Purkinje conduction in the intact neonatal heart--characterization and amplification by N-acetylprocainamide.

According to mathematical models of antiarrhythmic drug-receptor interactions, lidocaine binds preferentially to the sodium channel when the membrane is depolarized (i.e., the "inactivated" channel state). Therefore, the effect of lidocaine on conduction should be greater when the action potential duration is prolonged. To test this prediction in vivo we evaluated the rate-dependent effects of lidocaine on His-Purkinje conduction in the intact newborn canine heart. Lidocaine's effect was assessed alone and then when given in combination with N-acetylprocainamide, a Class III antiarrhythmic agent. Utilizing intracardiac electrical stimulation and electrogram recording techniques, changes in the steady-state His-Purkinje conduction time during atrial pacing at increasingly rapid cycle lengths, changes in the conduction time of pacing trains delivered directly to the His bundle and changes in conduction time during His bundle extra stimulation were measured. After bilateral sectioning of the vagi and the administration of propranolol (1.0 mg/kg i.v.), His-Purkinje conduction was assessed in newborn canines (ages 5-15 days) under the following conditions: 1) control (n = 18); 2) after an i.v. infusion of lidocaine HCl (serum concentration 3.7 +/- 0.8 micrograms/ml) (n = 12); and 3) after the combined administration of lidocaine and N-acetylprocainamide (serum concentration, 26.4 +/- 6.3 micrograms/ml) (n = 12). Rate-dependent changes in His-Purkinje conduction time were observed in the newborn in response to lidocaine at rapid paced cycle lengths. These changes were significantly amplified by the coadministration of N-acetylprocainamide. Furthermore, the time constant of recovery from rate-dependent conduction delay, determined during His bundle extra stimulation, was 45 msec, which is notably shorter than values reported for lidocaine in the adult.(ABSTRACT TRUNCATED AT 250 WORDS)

Evidence for developmental changes in sodium channel inactivation gating and sodium channel block by phenytoin in rat cardiac myocytes.

The voltage-dependent properties of the voltage-activated sodium channel were studied in neonatal (1-2-day-old) and adult rat ventricular cardiac myocytes using the whole-cell variation of the patch-clamp technique (16 degrees C, [Na]i = 15 mM, [Na]o = 25 mM). The voltage dependence of the sodium conductance-membrane potential relation was similar in both neonatal and adult myocytes except for a difference in slope; the adult sodium conductance-membrane potential relation was slightly more steep. Neonatal cells also differed from adult cells by demonstrating a more negative voltage midpoint of their sodium availability curve, a slower rate of recovery from inactivation at hyperpolarized potentials, and a greater extent of slow inactivation development compared with adult cells. Phenytoin (40 microM) reduced the sodium current in a tonic and use-dependent manner in both adult and neonatal myocytes. However, phenytoin (40 microM) produced significantly more tonic block at negative holding potentials (e.g., -140 mV) in neonatal myocytes (22 +/- 5% [mean +/- SEM], n = 14) than in adult myocytes (10 +/- 2%, n = 11) (p less than 0.05). The amplitudes of use-dependent block obtained during trains of 1-second pulses to -20 mV were also significantly greater in neonatal myocytes than in adult myocytes when the diastolic interval was varied over a range of 0.1-1.5 seconds (p less than 0.05). Definition of the time courses of block development at -20 mV indicated that phenytoin had a slightly higher affinity for inactivated sodium channels in neonatal cells. In addition, the time constant of recovery from use-dependent block by phenytoin was found to be significantly longer in neonatal cells than in adult cells at membrane potentials between -160 and -100 mV (p less than 0.001). The marked differences in phenytoin effect on cardiac sodium channels in neonatal versus adult rat cardiac myocytes suggest that there may be significant developmental changes in the sodium channel blocking effects of class I antiarrhythmic drugs in cardiac tissue.

Modification of Na channel inactivation by alpha-chymotrypsin in single cardiac myocytes.

The effects of alpha-chymotrypsin and trypsin on the macroscopic Na current in isolated guinea pig ventricular myocytes at 16 degrees C were investigated using the whole-cell voltage-clamp technique. Intracellular application of both enzymes reduced the extent of Na current inactivation during 20- to 50-ms depolarizing pulses. Elimination of fast inactivation by alpha-chymotrypsin was accompanied by a slowing of the rate of Na current decay through changes in both the time constants of current decay and the proportions of current undergoing a fast vs slow rate of decay. Treatment that reduced Na current decay to less than or equal to 10% within 20 ms was accompanied by a hyperpolarizing shift of the Na conductance/voltage relationship and an increase in the time-to-peak current that was most prominent for small depolarizations. Evidence for a significant slow inactivation process was obtained following removal of fast inactivation. The effect of trypsin (0.15-0.3 mg/ml) was less specific than alpha-chymotrypsin in that it also reduced Na conductance and increased leak current.

Electrocardiographic evidence for cocaine cardiotoxicity in cat.

Recent case studies suggest that cocaine overdose may produce life-threatening cardiac arrhythmias. We therefore investigated its effects on the electrocardiogram (leads II and V1) and arterial blood pressure in cats anesthetized with pentobarbital. Cocaine was administered by intravenous infusion over a 2-min interval at 1 mg/kg in 10 cats. In 5 out of 10 cats an additional infusion of 3 mg/kg cocaine was also administered after hemodynamic and electrocardiographic parameters had returned to control values (i.e., within 10 min). During and following infusion of 1 mg/kg cocaine, no significant change in heart rate or systolic or diastolic blood pressure were found, however the QRS duration increased by 38% (from 46 +/- 5 to 64 +/- 12 ms) (p less than 0.01). Evidence for bundle branch block and (or) premature ventricular beats was observed in 9 out of 10 cats after 1 mg/kg cocaine. Infusion of a further 3 mg/kg cocaine in five cats significantly lowered diastolic blood pressure (from 98 +/- 18 to 64 +/- 28 mmHg; 1 mmHg = 133.3 Pa) (p less than 0.01), and further prolonged QRS to 79 +/- 14 ms, a 75% increase from the mean control value (p less than 0.01). In addition, 1st and 2nd degree atrioventricular block, ventricular extrasystoles, and ectopic rhythms (AV junctional or idioventricular) were observed in four out of five cats given 3 mg/kg cocaine. Mean plasma concentrations of cocaine were 1.37 +/- 0.39 micrograms/mL (4.28 +/- 1.22 microM) (n = 5) at the end of a 1 mg/kg infusion and 2.93 +/- 0.43 micrograms/mL (9.16 +/- 1.34 microM) after a 3 mg/kg infusion (n = 3).(ABSTRACT TRUNCATED AT 250 WORDS)

Characterization of cocaine-induced block of cardiac sodium channels.

Recent evidence suggests that cocaine can produce marked cardiac arrhythmias and sudden death. A possible mechanism for this effect is slowing of impulse conduction due to block of cardiac Na channels. We therefore investigated its effects on Na channels in isolated guinea pig ventricular myocytes using the whole-cell variant of the patch clamp technique. Cocaine (10-50 microM) was found to reduce Na current in a use-dependent manner. The time course for block development and recovery were characterized. At 30 microM cocaine, two phases of block development were defined: a rapid phase (tau = 5.7 +/- 4.9 ms) and a slower phase (tau = 2.3 +/- 0.7 s). Recovery from block at -140 mV was also defined by two phases: (tau f = 136 +/- 61 ms, tau s = 8.5 +/- 1.7 s) (n = 6). To further clarify the molecular mechanisms of cocaine action on cardiac Na channels, we characterized its effects using the guarded receptor model, obtaining estimated Kd values of 328, 19, and 8 microM for channels predominantly in the rested, activated, and inactivated states. These data indicate that cocaine can block cardiac Na channels in a use-dependent manner and provides a possible cellular explanation for its cardiotoxic effects.

Stereoselective block of cardiac sodium channels by RAC109 in single guinea pig ventricular myocytes.

The effects of the optical stereoisomers of the local anesthetic RAC109 (RAC109-I and RAC109-II) on sodium current in isolated guinea pig ventricular myocytes were investigated by use of the whole-cell variation of the patch-clamp technique. RAC109-I and RAC109-II produced similar levels of tonic block, but RAC109-I produced a significantly larger use-dependent block on repetitive pulsing to potentials positive to -60 mV. Definition of the time courses of block development at -20 mV and recovery at -140 and -160 mV indicated that RAC109-I had a higher affinity for activated and inactivated channels and dissociated more slowly at hyperpolarized potentials compared with RAC109-II. Removal of fast inactivation by alpha-chymotrypsin intensified tonic block but did not reduce use-dependent block by RAC109-I; this finding suggests that channel inactivation is not necessary for use-dependent block. The guarded-receptor model was used to calculate apparent rate constants of drug binding and unbinding. According to the model, RAC109-I and RAC109-II have significantly different unbinding rate constants for channels when they exist predominantly in rested, activated, or inactivated states, as well as significantly different binding rate constants when channels are activated. However, the apparent rates of drug binding to closed (rested and inactivated) channels are not significantly different for the two isomers; this finding indicates that drug binding to closed channels is not markedly stereospecific, in contrast to unbinding. The effects of RAC109 stereoisomers on cardiac sodium channels were also qualitatively similar to those previously reported in nerve; these findings suggest that the binding sites for local anesthetics in both tissue types have a similar structural topography.

Evidence for voltage-dependent block of cardiac sodium channels by tetrodotoxin.

The effects of tetrodotoxin (TTX) on cardiac sodium channels in guinea-pig ventricular muscle were investigated. Membrane potential was controlled using a single sucrose gap voltage clamp method, and the maximum upstroke velocity of the ventricular action potential (Vmax) was used as an indicator of drug-free sodium channels. Reduction of Vmax by TTX was found to be both voltage- and time-dependent, similar to the effects of many local anesthetic drugs, with the exception that TTX concentrations high enough to produce significant use-dependent block (e.g. 2 microM), also produced significant tonic block, even at potentials negative to -85 mV. The mechanism underlying use-dependent block was determined by defining the time course of block development at potentials between -40 and +20 mV, and the time course of recovery at -85 mV. In 2 microM TTX, the time course of block development at +20 mV contained two phases, a fast phase (tau less than 3 ms) having a mean amplitude of 8.1 +/- 3.2% of control Vmax, and a slow phase (tau = 429 +/- 43 ms) having an amplitude of 35 +/- 2% of control Vmax (n = 5). Recovery from use-dependent block at -85 mV occurred with a time constant of 324 +/- 58 ms (n = 5). The effects of TTX could be well-described by a modulated receptor model with an estimated 12 mV drug-induced shift of inactivation, and state-dependent dissociation constants of 10, 4 and 0.3 microM for rested, activated and inactivated channels. These same drug rate constants could also be used to adequately simulate the reported effects of TTX on plateau sodium currents in a variant model with slow inactivation kinetics.

Evidence for two components of sodium channel block by lidocaine in isolated cardiac myocytes.

The effects of lidocaine on sodium current in cardiac myocytes isolated from cat and guinea pig were investigated using the whole-cell variation of the patch-clamp technique. Lidocaine (43-200 microM) reduced sodium current during repetitive depolarizing pulses in a use-dependent manner. To clarify the nature of the use-dependent block, we characterized the time course of block development using a two-pulse protocol. Two distinct phases of block development were found: a rapid phase (tau = 1-6 msec) having a time course concurrent with the time course of channel activation, and a slower phase (tau = 100-900 msec), which developed after channels inactivated. The amplitude of the block during the rapid phase of development was a steep function of transmembrane voltage over the range of -70 to +20 mV. The voltage-dependence was similar to that for sodium channel activation (sodium conductance) but was too steep to be attributed solely to the passive movement of a singly charged molecule under the influence of the transmembrane voltage gradient. These results suggest that use-dependent block of sodium channels in cardiac tissue may result from an interaction of lidocaine with sodium channels in the activated as well as the inactivated channel states. Possible mechanisms underlying the fast component of block are discussed.