Functional expression of L- and T-type Ca2+ channels in murine HL-1 cells

In the search for a readily available source of native cardiac cells, we investigated the molecular and pharmacological properties of the immortalized cardiac atrial myocyte cell line, HL-1 cells. This work focused on the expression pattern of voltage-gated Ca2+ channels (VGCC). Reverse transcription-polymerase chain reaction analysis revealed that HL-1 cells have mRNA for several types of Ca2+ channels including the L-types, alpha1C and alpha1D, as well as T-types, alpha1H and alpha1G, but are lacking N-type, alpha1B and the T-type, alpha1I. Western blot analysis demonstrated significant alpha1C protein subunit expression, with less alpha1D subunit apparent, while alpha1A, alpha1B and alpha1E subunit expression was undetectable. Immunocytochemical staining showed that the alpha1C protein subunit is expressed predominantly on the cell surface, whereas the alpha1D protein is expressed mostly intracellularly. Whole-cell patch-clamp measurements demonstrated the presence of low (ICa,T) and high (ICa,L) voltage-activated Ca2+ currents, with preferential sensitivity to mibefradil and nimodipine, respectively. Addition of increasing external Ca2+ concentrations, [Ca2+]o, resulted in Ca2+ influx measured by fluorometric imaging with an EC50 of 0.8 mM [Ca2+]o. At a fixed [Ca2+]o of 0.125 mM, Ca2+ influx was also triggered by increasing the extracellular K+ concentration, [K+]o, with an EC50 of 3.7 mM [K+]o. As increasing [K+]o depolarizes the cell, this latter result is consistent with Ca2+ influx through a voltage-dependent mechanism. L-type (nimodipine and verapamil) and T-type (mibefradil and pimozide) Ca2+ channel blockers inhibited Ca2+ influx with IC50s of 1, 2, 0.4 and 0.2 microM, respectively. Antagonists of N-type (omega-conotoxins GVIA) and P/Q-type (MVIIC or omega-agatoxin IVA) did not inhibit Ca2+ influx, consistent with the lack of expression of N-, P-, or Q-type channels observed in the molecular studies. Taken together, these findings indicate that HL-1 cells express L- and T-subtypes of VGCC and are a unique in vitro model system for the study of native, mammalian cardiac Ca2+ channels.

Left ventricular hypertrophy decreases slowly but not rapidly activating delayed rectifier potassium currents of epicardial and endocardial myocytes in rabbits

BACKGROUND

Delayed rectifier K(+) currents are critical to action potential (AP) repolarization. The present study examines the effects of left ventricular hypertrophy (LVH) on delayed rectifier K(+) currents and their contribution to AP repolarization in both epicardial (Epi) and endocardial (Endo) myocytes.

METHODS AND RESULTS

VH was induced in rabbits by a 1-kidney removal, 1-kidney vascular clamping method. Slowly (I(Ks)) and rapidly (I(Kr)) activating delayed rectifier K(+) currents were recorded by the whole-cell patch-clamp technique, and APs were recorded by the microelectrode technique. In normal rabbit left ventricular myocytes, I(Ks) densities were larger in Epi than in Endo (1.1+/-0.1 versus 0.43+/-0.07 pA/pF), whereas I(Kr) density was similar between Epi and Endo (0.31+/-0.05 versus 0.36+/-0.07 pA/pF) at 20 mV. LVH reduced I(Ks) density to a similar extent (approximately 40%) in both Epi and Endo but had no significant effect on I(Kr) in either Epi or Endo. Consequently, I(Kr) was expected to contribute more to AP repolarization in LVH than in control. This was confirmed by specific I(Kr) block with dofetilide, which prolonged AP significantly more in LVH than in control (31+/-3% versus 18+/-2% in Epi; 53+/-6% versus 32+/-4% in Endo at 2 Hz). In contrast, L-768,673 (a specific I(Ks) blocker) prolonged AP less in LVH than in control. The very small I(Ks) density in Endo with LVH is consistent with the greater incidence of early afterdepolarizations induced in this region by dofetilide.

CONCLUSIONS

LVH induces a decrease in I(Ks) density and increases the propensity to develop early afterdepolarizations, especially in Endo.

Synthesis and class III type antiarrhythmic activity of 4-aroyl (and aryl)-l-aralkylpiperazines

The synthesis and in vitro Class III antiarrhythmic activity of several 4-aroyl (and aryl)-1-aralkylpiperazine and piperidine derivatives are described. Among several potent compounds identified in the series, RWJ-28810 (3), with its EC20 of 3 nM, ranks as one of the most potent (in vitro) compounds reported.

Modification of cardiac Na(+) current by RWJ 24517 and its enantiomers in guinea pig ventricular myocytes

We examined the effects of the cardiotonic agent RWJ 24517 (Carsatrin, racemate) and its (S)- and (R)-enantiomers on action potential duration, Na(+) current (I(Na)), and delayed rectifier K(+) current (I(K)) of guinea pig ventricular myocytes. RWJ 24517 (0. 1 and 1 microM) prolongation of action potential duration could not be accounted for by suppression of either the rapid (I(Kr)) or slow (I(Ks),) component of I(K), although RWJ 24517 did reduce I(Kr) at concentrations of 1 microM. A more dramatic effect of RWJ 24517 (0.1-1 microM) and the (S)-enantiomer of RWJ 24517 (0.1-3 microM) was an increase in peak I(Na) and slowing of the rate of I(Na) decay, eliciting a large steady-state current. Neither RWJ 24517 nor the (S)-enantiomer affected the fast time constant for I(Na) decay, but both significantly increased the slow time constant, in addition to increasing the proportion of I(Na) decaying at the slow rate. Both agents elicited a use-dependent decrease of peak I(Na) (3-10 microM), which probably resulted from a slowing of both fast and slow rates of recovery from inactivation. In contrast, the (R)-enantiomer of RWJ 24517 did not induce a steady-state component I(Na) or increase peak I(Na) up to 10 microM, but it decreased peak I(Na) at 30 microM. The (R)-enantiomer displayed little use-dependent reduction of I(Na) during trains of repetitive pulses and had no effect on rates of inactivation or recovery from inactivation. These actions of the racemate and the (S)-stereoisomer to slow inactivation and to prolong both Na(+) influx and action potential duration may contribute to the positive inotropic actions of these agents because the resulting accumulation of intracellular Na(+) would increase intracellular Ca(2+) via Na(+)/Ca(2+) exchange.

A novel benzodiazepine that activates cardiac slow delayed rectifier K+ currents

The slowly activating delayed rectifier K+ current, IKs, is an important modulator of cardiac action potential repolarization. Here, we describe a novel benzodiazepine, [L-364,373 [(3-R)-1, 3-dihydro-5-(2-fluorophenyl)-3-(1H-indol-3-ylmethyl)-1-methyl-2H- 1,4-benzodiazepin-2-one] (R-L3), that activates IKs and shortens action potentials in guinea pig cardiac myocytes. These effects were additive to isoproterenol, indicating that channel activation by R-L3 was independent of beta-adrenergic receptor stimulation. The increase of IKs by R-L3 was stereospecific; the S-enantiomer, S-L3, blocked IKs at all concentrations examined. The increase in IKs by R-L3 was greatest at voltages near the threshold for normal channel activation, caused by a shift in the voltage dependence of IKs activation. R-L3 slowed the rate of IKs deactivation and shifted the half-point of the isochronal (7.5 sec) activation curve for IKs by -16 mV at 0.1 microM and -24 mV at 1 microM. R-L3 had similar effects on cloned KvLQT1 channels expressed in Xenopus laevis oocytes but did not affect channels formed by coassembly of KvLQT1 and hminK subunits. These findings indicate that the association of minK with KvLQT1 interferes with the binding of R-L3 or prevents its action once bound to KvLQT1 subunits.

Blockade of HERG channels by the class III antiarrhythmic azimilide: mode of action

1. The class III antiarrhythmic azimilide has previously been shown to inhibit I(Ks) and I(Kr) in guinea-pig cardiac myocytes and I(Ks) (minK) channels expressed in Xenopus oocytes. Because HERG channels underly the conductance I(Kr), in human heart, the effects of azimilide on HERG channels expressed in Xenopus oocytes were the focus of the present study. 2. In contrast to other well characterized HERG channel blockers, azimilide blockade was reverse use-dependent, i.e., the relative block and apparent affinity of azimilide decreased with an increase in channel activation frequency. Azimilide blocked HERG channels at 0.1 and 1 Hz with IC50s of 1.4 microM and 5.2 microM respectively. 3. In an envelope of tail test, HERG channel blockade increased with increasing channel activation, indicating binding of azimilide to open channels. 4. Azimilide blockade of HERG channels expressed in Xenopus oocytes and I(Kr) in mouse AT-1 cells was decreased under conditions of high [K+]e, whereas block of slowly activating I(Ks) channels was not affected by changes in [K+]e. 5. In summary, azimilide is a blocker of cardiac delayed rectifier channels, I(Ks) and HERG. Because of the distinct effects of stimulation frequency and [K+]e on azimilide block of I(Kr) and I(Ks) channels, we conclude that the relative contribution of block of each of these cardiac delayed rectifier channels depends on heart frequency. [K+]e and regulatory status of the respective channels.

IK of rabbit ventricle is composed of two currents: evidence for IKs

The delayed rectifier K+ current (IK) in rabbit heart has long been thought to consist of only a single, rapidly activating, dofetilide-sensitive current, IKr. However, we find that IK of rabbit ventricular myocytes actually consists of both rapid and slow components, IKr and IKs, respectively, that can be isolated pharmacologically. Thus, after complete blockade of IKr with dofetilide, the remaining current, IKs, is homogeneous as judged by an envelope of tails test. IKs activates and deactivates slowly, continues to activate during sustained depolarizations, has a half-activation potential of 7.0 +/- 0.8 mV and slope factor of 11.0 +/- 0.7 mV, reverses at -77.2 +/- 1.3 mV (extracellular K+ concentration = 4 mM), is increased by removing extracellular K+, and is enhanced by isoproterenol and stocked by azimilide. Northern analysis demonstrates that the minK (IsK) gene, which encodes a subunit of the channel that underlies the IKs current, is expressed in rabbit heart. Expression of the rabbit protein in Xenopus oocytes elicits a slowly activating, voltage-dependent current, IsK, similar to those expressed previously from mouse, rat, guinea pig, and human genes. The results demonstrate that IKs is present in rabbit ventricle and therefore contributes to cardiac repolarization in this species.

Mechanism of action potential prolongation by RP 58866 and its active enantiomer, terikalant. Block of the rapidly activating delayed rectifier K+ current, IKr

BACKGROUND

The class III antiarrhythmic agent RP 58866 and its active enantiomer, terikalant, are reported to selectively block the inward rectifier K+ current, IK1. These drugs have demonstrated efficacy in animal models of cardiac arrhythmias, suggesting that block of IK1 may be a useful antiarrhythmic mechanism. The symmetrical action potential (AP)-prolonging and bradycardic effects of these drugs, however, are inconsistent with a sole effect on IK1.

METHODS AND RESULTS

We studied the effects of RP 58866 and terikalant on AP and outward K+ currents in guinea pig ventricular myocytes. RP 58866 and terikalant potently blocked the rapidly activating delayed rectifier K+ current, IKr, with IC50S of 22 and 31 nmol/L, respectively. Block of IK1 was approximately 250-fold less potent; IC50S were 8 and 6 mumol/L, respectively. No significant block of the slowly activating delayed rectifier, IK1, was observed at < or = 10 mumol/L. The phenotypical IKr currents in mouse AT-1 cells and Xenopus oocytes expressing HERG were also blocked 50% by 200 to 250 nmol/L RP 58866 or terikalant, providing further conclusive evidence for potent block of IKr. RP 58866 < or = 1 mumol/L and dofetilide increased AP duration symmetrically, consistent with selective block of IKr. Only higher concentrations (> or = 10 mumol/L) of RP 58866 slowed the rate of AP repolarization and decreased resting membrane potential, consistent with an additional but substantially less potent block of IK1.

CONCLUSIONS

These data demonstrate that RP 58866 and terikalant are potent blockers of IKr and prompt a reinterpretation of previous studies that assumed specific block of IK1 by these drugs.

Use-dependent effects of the class III antiarrhythmic agent NE-10064 (azimilide) on cardiac repolarization: block of delayed rectifier potassium and L-type calcium currents

We studied the effects of NE-10064 (azimilide), a new antiarrhythmic agent reported to be a selective blocker of the slowly activating component of the delayed rectifier, IKs. In ferret papillary muscles, NE-10064 increased effective refractory period (ERP) and decreased isometric twitch tension in a concentration-dependent manner (0.3-30 microM). Increases in ERP showed reverse use-dependence, and were greater at 1 than at 3 Hz. In contrast, changes in tension were use dependent, with larger decreases observed at 3 than at 1 Hz. In guinea pig ventricular myocytes, NE-10064 (0.3-3 microM) significantly prolonged action potential duration (APD) at 1 Hz. At 3 Hz, NE-10064 (0.3-1 microM) increased APD only slightly, and at 10 microM decreased APD and the plateau potential. NE-10064 potently blocked the rapidly activating component of the delayed rectifier, IKr (IC50 0.4 microM), and inhibited IKs (IC50 3 microM) with nearly 10-fold less potency. NE-10064 (10 microM) did not block the inward rectifier potassium current (IKl). NE-10064 (10 microM) blocked the L-type calcium current (ICa) in a use-dependent manner; block was greater at 3 than at 1 Hz. We conclude that (a) NE-10064's block of potassium currents is relatively selective for IKr over IKs, (b) NE-10064 inhibits ICa in a use-dependent fashion, and (c) NE-10064's effects on ERP and tension in papillary muscle as well as APD and action potential plateau level in myocytes may be explained by its potassium and calcium channel blocking properties.

Comparison of binding to rapidly activating delayed rectifier K+ channel, IKr, and effects on myocardial refractoriness for class III antiarrhythmic agents

Saturation binding studies in guinea pig ventricular myocytes with 3H-dofetilide, a radioligand for the cardiac rapidly activating delayed rectifier K+ IKr channel, indicated specific high-affinity binding with a Kd of 83 nM and a Bmax of 0.18 pmol/mg cellular protein (1.36 x 10(6) sites/cell). Using displacement of high-affinity 3H-dofetilide binding as a measure of interaction with the IKr channel, potencies (Ki values) for binding to the IKr channel in guinea pig myocytes for six class III antiarrhythmic agents were characterized and compared to indices of functional electrophysiologic activity in isolated guinea pig papillary muscles [EC25 values, concentration required to increase effective refractory period (ERP) 25% above baseline]. Dofetilide, E-4031, sematilide, and d-sotalol, which have been characterized previously as selective IKr blockers, displayed good agreement between Ki values for displacement of 3H-dofetilide binding (47 +/- 7 nM, 38 +/- 8 nM, 12 +/- 5 microM, and approximately 100 microM, respectively) and EC25 values for increasing ERP in papillary muscles (45.0 nM, 76.9 nM, 20.2 microM and 63.5 microM, respectively). Ibutilide and RP58866, which have been reported to act via mechanisms other than IKr block, had Ki values for displacement of 3H-dofetilide binding (16 +/- 7 nM and 17 +/- 2 nM, respectively) that were approximately 10-fold lower than EC25 values for increasing ERP in papillary muscles (185.8 nM and 223.5 nM, respectively). The potent displacement of high-affinity 3H-dofetilide binding by ibutilide and RP58866 strongly suggest a role for interaction with IKr in their actions.(ABSTRACT TRUNCATED AT 250 WORDS)

Cardiac electrophysiological actions of the histamine H1-receptor antagonists astemizole and terfenadine compared with chlorpheniramine and pyrilamine

We compared the cardiac electrophysiological actions of two types of H1-receptor antagonists--the piperidines, astemizole and terfenadine, and the nonpiperidines, chlorpheniramine and pyrilamine-in vitro in guinea pig ventricular myocytes and in vivo in chloralose-anesthetized dogs. Astemizole and terfenadine significantly increased action potential duration of guinea pig myocytes. This concentration-dependent prolongation of action potential duration was reverse frequency dependent and led to development of early afterdepolarizations, which occurred more frequently at higher concentrations and slower pacing frequencies. Astemizole and terfenadine potently blocked the rapidly activating component of the delayed rectifier, IKr, with IC50 values of 1.5 and 50 nmol/L, respectively. At 10 mumol/L, terfenadine but not astemizole blocked the slowly activating component of the delayed rectifier, IKs (58.4 +/- 3.1%), and the inward rectifier, IK1 (20.5 +/- 3.4%). Chlorpheniramine and pyrilamine blocked IKr relatively weakly (IC50 = 1.6 and 1.1 mumol/L, respectively) and IKs and IK1 less than 20% at 10 mumol/L. Astemizole and terfenadine (1.0 to 3.0 mg/kg IV) significantly prolonged the QTc interval and ventricular effective refractory period in vivo. Chlorpheniramine and pyrilamine (< or = 3.0 mg/kg) did not significantly affect these parameters. Block of repolarizing K+ currents, particularly IK1, by astemizole and terfenadine produces reverse rate-dependent prolongation of action potential duration and development of early afterdepolarizations, delays ventricular repolarization, and may underlie the development of torsade de pointes ventricular arrhythmias observed with the use and abuse of these agents.

Cardiac electrophysiologic and antiarrhythmic actions of two long-acting spirobenzopyran piperidine class III agents, L-702,958 and L-706,000 [MK-499]

The cardiac electrophysiologic and antiarrhythmic actions of two Class III ketone- and alcohol-containing spirobenzopyran piperidine analogs, L-702,958 and L-706,000 [MK-499], respectively, were assessed in vitro and in vivo. L-702,958 and L-706,000 [MK-499] selectively blocked the rapidly activating component of the delayed rectifier K+ current in guinea pig isolated ventricular myocytes (IC50 values, 14.6 and 43.9 nM, respectively), and prolonged effective refractory period in ferret isolated papillary muscles (EC25 values, 10.5 and 53.8 nM, respectively). In anesthetized dogs, L-702,958 and L-706,000 [MK-499] increased ventricular refractory periods (ED20 values, 3.3 and 9.2 micrograms/kg i.v., respectively) and concomitantly increased ECG QT interval and left ventricular+dP/dt. Cumulative i.v. administrations of up to 100 micrograms/kg of L-702,958 and 300 micrograms/kg L-706,000 [MK-499] in anesthetized dogs increased atrial and ventricular refractoriness and prolonged the ECG QT interval, but did not alter atrial, atrioventricular nodal, His-Purkinje or ventricular conduction indices. In anesthetized dogs studied chronically (9.2 +/- 1.1 days) after anterior myocardial infarction, the cumulative i.v. administrations of 100 micrograms/kg of L-702,958 and 300 of micrograms/kg L-706,000 [MK-499] suppressed the induction of ventricular tachyarrhythmia by programmed ventricular stimulation (suppression rates: 8 of 10, 80% and 9 of 11, 82%, respectively) and reduced the incidence of lethal ventricular arrhythmias (incidence of lethal ischemic arrhythmias: 4 of 10, 40% and 1 of 11 9%, respectively, compared to 34 of 40, 85%, in vehicle controls. L-702,958 and L-706,000 [MK-499] (cumulative 100 and 300 micrograms/kg i.v., respectively) did not facilitate the induction of arrhythmias by programmed ventricular stimulation in postinfarction dogs. After equivalently effective p.o. doses in conscious dogs, L-702,958 (10 micrograms/kg) and L-706,000 [MK-499] (30 micrograms/kg) increased ECG QT interval with long durations of action of approximately 9 and 14 hr, respectively. L-706,000 [MK-499] elicited a more consistent and sustained prolongation of the QT interval than L-702,958. These findings show that both L-702,958 and L-706,000 [MK-499] are potentially useful agents for the prevention of malignant ventricular arrhythmias in the setting of myocardial ischemic injury.

K+ currents expressed from the guinea pig cardiac IsK protein are enhanced by activators of protein kinase C

We have isolated cardiac cDNA and genomic clones encoding the guinea pig IsK protein. The deduced amino acid sequence is approximately 78% identical to the rat, mouse, and human variants of this channel, and the structure of the gene encoding the protein is also similar to that in other species. For example, the gene is present only once in the haploid genome, the protein-coding sequence is present on a single uninterrupted exon, an intron exists in the 5' untranslated domain, and multiple alternative polyadenylation sites are used in processing the transcript. Expression of the guinea pig protein in Xenopus oocytes results in a slowly activating, voltage-dependent K+ current, IsK, similar to those expressed previously from the rat, mouse, and human genes. However, in sharp contrast to the rat and mouse currents, activation of protein kinase C with phorbol esters increases the amplitude of the guinea pig IsK current, analogous to its effects on the endogenous IKs current in guinea pig cardiac myocytes. Mutagenesis of the guinea pig cDNA to alter four cytoplasmic amino acid residues alters the phenotype of the current response to protein kinase C from enhancement to inhibition, mimicking that of rat and mouse IsK currents. This mutation is consistent with reports that phosphorylation of Ser-102 by protein kinase C decreases the current amplitude. These data explain previously reported differences in the regulatory properties between recombinant rat or mouse IsK channels and native guinea pig IKs channels and provide further evidence that the IsK protein forms the channels that underlie the IKs current in the heart.

Species variants of the IsK protein: differences in kinetics, voltage dependence, and La3+ block of the currents expressed in Xenopus oocytes

We have compared the slowly activating K+ currents (IsK) resulting from the expression of the human, mouse, or rat IsK proteins in Xenopus oocytes, utilizing natural, species-dependent sequence variations to initiate structure-function studies of this channel. Differences were found between the human and rodent currents in their voltage dependence, kinetics, and sensitivity to external La3+. The current/voltage relationships of the human and rat IsK currents differed significantly, with greater depolarizations required for activation of the human channel. The first 30 s of activation during depolarizations to potentials between -10 and +40 mV was best described by a triexponential function for each of the three species variants. The activation rates were, however, significantly faster for the human current than for either of the rodent forms. Similarly, deactivation kinetics were best described as a biexponential decay for each of the species variants but the human currents deactivated more rapidly than the rodent currents. The human and the rodent forms of IsK were also differentially affected by external La3+. Low concentrations (10, 50 microM) rapidly and reversibly reduced the magnitude of the mouse and rat currents during a test depolarization and increased the deactivation rates of the tail currents. In contrast, the magnitude and deactivation rates of the human IsK currents were unaffected by 50 microM La3+.

RWJ 26629, a new potassium channel opener and vascular smooth muscle relaxant: a potential antihypertensive and antianginal agent

The effects of trans-5,6-dihydro-6-hydroxy-5,5-dimethyl-2-nitro-7-(2-oxopiperidin -1-yl)-7H- thieno[3,2-b]pyran (RWJ 26629) were compared with those of the standard potassium channel opener cromakalim and several standard calcium channel blockers. RWJ 26629 lowered the mean arterial blood pressure in conscious spontaneously hypertensive (ED30 = 10 micrograms/kg p.o. or 8 micrograms/kg i.v.) and renal hypertensive (15 micrograms/kg p.o.) rats, conscious renal hypertensive (ED20 = 4 micrograms/kg p.o.) and normotensive (ED20 = 5 micrograms/kg p.o. or 2 micrograms/kg i.v.) dogs and anesthetized rhesus monkeys (ED20 = 6 micrograms/kg i.v.). RWJ 26629 was more potent than cromakalim and had a maximal activity greater than the calcium channel blockers. At antihypertensive doses, RWJ 26629 had no significant effect on cardiac force, cardiac output, stroke volume or stroke work in dogs and had little or no effect on renal, carotid or femoral blood flow or vascular resistance. RWJ 26629 was also selective for antihypertensive activity in rats compared with its ability to inhibit intestinal motility. However, RWJ 26629 did relax contracted pulmonary smooth muscle in vivo at antihypertensive doses. All compounds tested caused reflex tachycardia in conscious dogs, although this effect was lowest for RWJ 26629. Most importantly, RWJ 26629 potently and selectively increased coronary blood flow with a potency and duration of action greater than that of cromakalim or nifedipine without affecting contractile force. In vitro, RWJ 26629 selectively relaxed precontracted coronary arteries compared with its effect on femoral arteries.(ABSTRACT TRUNCATED AT 250 WORDS)

In vivo cardiac electrophysiologic effects of RWJ 29009, a new potassium-channel activator, in comparison to cromakalim and nicardipine

RWJ 29009 is a new potassium channel activator with prominent coronary and peripheral vasodilating actions. Because of the potential direct cardiac electrophysiologic actions of increased potassium conductance in myocardium, we evaluated the effects of RWJ 29009 on cardiac conduction and refractoriness in comparison to its vasodilator activity in anesthetized, open-chest dogs. We assessed effects during both intrinsic sinus rhythm and during constant atrial pacing. RWJ 29009 markedly increased coronary blood flow and decreased mean arterial blood pressure (MAP) dose dependently (0.3-10 micrograms/kg intravenously, i.v.). RWJ 29009 had no effect on PR interval but decreased AV-nodal conduction time (AH) and Wenkebach cycle length slightly. RWJ 29009 decreased QT interval, left ventricular (LV) monophasic action potential duration (APD), and ventricular and atrial refractory period. These effects were consistent with shortening of cardiac repolarization. RWJ 29009 had no effect on QRS or His-Purkinje conduction time. Cromakalim had a qualitatively similar profile but was much less potent (3-300 micrograms/kg i.v.). In addition, the effects of cromakalim on repolarization parameters were somewhat less marked than those of RWJ 29009. Nicardipine also markedly increased coronary blood flow and decreased arterial pressure (10-300 micrograms/kg i.v.). Unlike the potassium channel activators, nicardipine (100-300 micrograms/kg), did not affect cardiac repolarization, but increased PR and AH interval, and Wenkebach cycle length (WENK) and reduced heart rate (HR) consistent with calcium channel blockade. These results indicate that RWJ 29009, like cromakalim, increases coronary blood flow at low doses without substantial electrophysiologic effects. Electrophysiologic effects observed at higher doses indicated a shortening of repolarization, expectedly produced by potassium channel activation in cardiac tissue.

Synthesis and SAR of 6-substituted purine derivatives as novel selective positive inotropes

A series of purine derivatives was prepared and examined for selective inotropic activity in vitro and in vivo. Thioether-linked derivatives were superior to their oxygen and nitrogen isosteres. Substitution of electron-withdrawing groups on the benzhydryl moiety of these agents increased potency. The best compound of the study, 17 (carsatrin), was examined further and demonstrated selective oral activity as a positive inotrope. These compounds are presumed to act by affecting the kinetics of the cardiac sodium channel by analogy to the prototypic agent DPI 201106 (1). Their high selectivity for increasing contractile force and dP/dt without affecting blood pressure or heart rate is consistent with this mechanism. Carsatrin (17) was selected as a potential development candidate.

Positive inotropic and hemodynamic properties of flosequinan, a new vasodilator, and a sulfone metabolite

Flosequinan, a new orally active vasodilator, and its sulfone metabolite were evaluated for inotropic activity in isolated ferret papillary muscles and pentobarbital anesthetized open-chest dogs. In vitro, flosequinan and its sulfone derivative increased tension development in a concentration-dependent manner (1-50 microM) in electrically stimulated papillary muscles pretreated with the beta-adrenergic blocking agent atenolol (2 microM). Peak increases in tension of 75 +/- 17%, and 111 +/- 46% with potencies (EC50) of 15 and 10 microM were observed for flosequinan and its metabolite, respectively. In vivo, flosequinan increased left ventricular dP/dtmax (74 +/- 12%) and right ventricular contractile force (CF) (104 +/- 10%) after administration of 1.875 mg/kg, i.v. Inotropic activity was dose-dependent and remained elevated for at least 60 min postinfusion. Flosequinan also increased heart rate (HR) (14 +/- 2%) and reduced mean arterial pressure (-9 +/- 3%). The i.v. potency of flosequinan (ED50 = 0.45 mg/kg) and its metabolite (ED50 = 0.38 mg/kg) were similar to that of the inotropic vasodilator amrinone (ED50 = 0.38 mg/kg). Inotropic activity was not significantly altered by pretreatment with propranolol (0.5 mg/kg) and atropine (1.0 mg/kg), further supporting the in vitro data indicating that flosequinan can directly stimulate myocardial contractility independent of beta-adrenergic receptor activation. Additional hemodynamic studies were conducted in an acute heart failure model produced by an overdose of propranolol. Flosequinan (2 mg/kg, i.v.) increased cardiac output (CO) (50 +/- 9%) and stroke volume (SV) (29 +/- 8%) while reducing total peripheral vascular resistance (TPR) (-36 +/- 4%), right atrial pressure (-62 +/- 5%), and left ventricular end-diastolic pressure (LVEDP) (-41 +/- 2%).(ABSTRACT TRUNCATED AT 250 WORDS)

Basis for tetrodotoxin and lidocaine effects on action potentials in dog ventricular myocytes

We studied the effects of tetrodotoxin (TTX) and lidocaine on transmembrane action potentials and ionic currents in dog isolated ventricular myocytes. TTX (0.1-1 x 10(-5) M) and lidocaine (0.5-2 x 10(-5) M) decreased action potential duration, but only TTX decreased the maximum rate of depolarization (Vmax). Both TTX (1-2 x 10(-5) M) and lidocaine (2-5 x 10(-5) M) blocked a slowly inactivating toward current in the plateau voltage range. The voltage- and time-dependent characteristics of this current are virtually identical to those described in Purkinje fibers for the slowly inactivating inward Na+ current. In addition, TTX abolished the outward shift in net current at plateau potentials caused by lidocaine alone. Lidocaine had no detectable effect on the slow inward Ca2+ current and the inward K+ current rectifier, Ia. Our results indicate that 1) there is a slowly inactivating inward Na+ current in ventricular cells similar in time, voltage, and TTX sensitivity to that described in Purkinje fibers; 2) both TTX and lidocaine shorten ventricular action potentials by reducing this slowly inactivating Na+ current; 3) lidocaine has no additional actions on other ionic currents that contribute to its ability to abbreviate ventricular action potentials; and 4) although both agents shorten the action potential by the same mechanism, only TTX reduces Vmax. This last point suggests that TTX produces tonic block of Na+ current, whereas lidocaine may produce state-dependent Na+ channel block, namely, blockade of Na+ current only after Na+ channels have already been opened (inactivated-state block).

A model for early afterdepolarizations: induction with the Ca2+ channel agonist Bay K 8644

Early afterdepolarizations (EADs) are one mechanism proposed to cause certain cardiac arrhythmias. We studied the effect of the Ca2+ channel agonist Bay K 8644 (1 x 10(-8) to 5 x 10(-5) M) on normally polarized sheep and canine cardiac Purkinje fiber short segments. EADs occurred with higher Bay K 8644 concentrations and had an average take-off potential of -34 mV. The initiation of EADs was preceded by lengthening of action potential duration and flattening of the plateau. Induction of EADs with Bay K 8644 was enhanced by low stimulation frequencies, lowering of [K]o, addition of tetraethylammonium chloride, or application of depolarizing constant current pulses during the plateau. EADs were abolished by increasing stimulation frequency, raising [K]o, the addition of tetrodotoxin, lidocaine, ethmozin, verapamil, and nitrendipine, or application of repolarizing constant current pulses. Using current pulses to modify the action potential plateau, a steep inverse relationship was found between the EAD peak voltage and its take-off potential, and EADs could be initiated over only a narrow range of take-off potentials. Thus, interventions that suppressed EADs shortened action potential duration or shifted the plateau away from the voltage range needed to initiate EADs. These observations suggest that mechanisms dependent on both time and voltage underlie EADs, and provide a unifying hypothesis for the induction of the EADs. We propose that induction of EADs requires 1) lengthening of action potential duration within a plateau voltage range where 2) recovery from inactivation and reactivation of an inward current possibly carried through Ca2+ channels can occur.

Effects of quinidine on action potentials and ionic currents in isolated canine ventricular myocytes

We examined the effects of quinidine (5-20 microM) on transmembrane action potentials and ionic currents of isolated canine ventricular myocytes. Collagenase treatment of canine ventricular tissue produced a yield of 40-60% healthy cells. Myocytes had normal resting and action potentials as measured using conventional microelectrodes. Quinidine decreased Vmax, amplitude, overshoot, and the duration of action potentials stimulated by passage of brief current pulses through the recording pipette. Recovery was complete after washout except that action potential duration was prolonged compared with control. A discontinuous single microelectrode voltage ("switch") clamp was used to measure ionic currents. Quinidine irreversibly reduced steady-state outward current as measured with three different voltage clamp protocols. Quinidine reversibly decreased peak calcium current as well as the slowly inactivating and/or steady-state inward currents in the plateau voltage range, presumably both "late" sodium (tetrodotoxin-sensitive) and calcium (tetrodotoxin-insensitive) currents. The effect on calcium current showed both tonic and use-dependent block. Thus, quinidine has a multitude of actions on both inward and outward currents, which combine to produce the net effect of quinidine on action potential configuration.

Effects of 4-aminopyridine on rate-related depression of cardiac action potentials

Canine cardiac Purkinje fibers and atrial trabeculae and rat and cat papillary muscles superfused with a hyperkalemic, hypoxic, and acidotic Tyrode solution were depolarized to membrane potentials (-70 to -60 mV) at which action potential amplitude declined as the coupling intervals of pacing stimuli were prolonged from 500 to 4,500 ms. The rate-related decline of action potential amplitude appeared to be due to time-dependent recovery of the early outward current rather than to a decrease in inward calcium current, since it was prevented by 4-aminopyridine (1.0 mM), but not by isoproterenol (1.0 microM), caffeine (5.0 mM), or CsCl (5-20 mM) and it was accompanied by an exponential increase of developed tension. Experiments using Purkinje fibers mounted in a single sucrose gap chamber demonstrated that the rate-related decline of action potential amplitude was maximal at membrane potentials between -70 and -40 mV and was negligible at less negative or more negative membrane potentials. These results may pertain to the mechanism for deceleration-dependent bundle branch block.

Effects of sympathetic tone on vagally induced phasic changes in heart rate and atrioventricular node conduction in the anesthetized dog

We examined the effects of stellate ganglia stimulation on the phase-dependent chronotropic and dromotropic responses to brief vagal bursts in open-chest anesthetized dogs. Stellate stimulation affected the phasic vagal effects on heart rate by shortening the latent period, shifting the phase at which maximum decrease in heart rate occurred to earlier phases, and reducing the maximum decrease in heart rate. These effects were due primarily to an increase in the basic heart rate. No significant sympathetic-parasympathetic interaction occurred for heart rate, indicating that accentuated antagonism did not occur with brief vagal bursts. Stellate stimulation primarily decreased the amplitude of the phasic vagal effects on atrioventricular nodal conduction, regardless of the underlying heart rate, and a significant sympathetic-parasympathetic interaction was associated with this effect. The peak of the phase-dependent vagal effects on heart rate and atrioventricular nodal conduction were phase-shifted with one another. From these findings, we postulate the small changes in sympathetic tone might shift the predominant phase-dependent vagal effect from one on heart rate to one on atrioventricular nodal conduction. Furthermore, our results suggest that dynamic vagal control of heart rate and atrioventricular node conduction involves both phase-dependent and phase-independent factors. Sympathetic activity appears to affect only the phase-independent factor(s) in the control of heart rate, whereas it affects both phase-dependent and phase-independent factors in the control of atrioventricular node conduction.

Rate-related suppression and facilitation of conduction in isolated canine cardiac Purkinje fibers

Previous studies have shown that antegrade conduction through damaged His Purkinje tissue may be suppressed following rapid ventricular pacing (overdrive suppression of conduction). We studied this phenomenon using isolated Purkinje fibers placed in a three-chamber bath. Superfusates for the left, middle, and right segments of the fiber were altered to produce action potentials that resembled those of normal bundle branch, damaged His bundle, and normal His bundle, respectively. To produce anisotropic conduction, the left segment of the fiber was adjusted to be three to four times longer than the right segment. Pacing the right segment at intermediate rates produced maximal action potential amplitude in the middle segment and 1:1 right-to-left conduction, whereas pacing at faster or slower rates reduced action potential amplitude and produced block. Pacing the left segment at fast or slow rates also reduced action potential amplitude in the middle segment, but conduction was maintained (anisotropy). After rapid or slow left segment pacing, action potential amplitude in the middle segment remained low during subsequent right segment pacing at intermediate rates, and transient block occurred (overdrive or underdrive suppression of conduction). With time, action potential amplitude normalized and conduction resumed. In other more severely depressed preparations, conduction block occurred even at intermediate right segment pacing rates prior to left segment pacing. Under these conditions, pacing the left segment at intermediate rates increased action potential amplitude in the middle segment and temporarily permitted 1:1 conduction at intermediate right segment pacing rates (overdrive facilitation of conduction).(ABSTRACT TRUNCATED AT 250 WORDS)

"Fade" of hyperpolarizing responses to vagal stimulation at the sinoatrial and atrioventricular nodes of the rabbit heart

Previous studies have suggested that maintained vagal stimulation or acetylcholine infusion results in a fade of responses in the sinoatrial node but not in the atrioventricular node, implying different muscarinic receptor subtypes in the two regions. We investigated this hypothesis in 23 isolated rabbit atrial preparations made quiescent by continuous superfusion with verapamil (1 microgram/ml). Transmembrane potentials were recorded simultaneously from cells in the sinoatrial pacemaker region and from the "N" region of the atrioventricular node. Postganglionic vagal stimulation was achieved by the application of trains of pulses (50-150 microseconds; 10-20 V; 200 Hz). Simultaneous application of long-lasting (1-10 sec) vagal trains produced hyperpolarizations which were nearly identical for both nodal regions. Maximal hyperpolarizations (approximately or equal to 24 mV for sinoatrial node; 26 mV for atrioventricular node) were reached about 500 msec after initiation of the vagal train. Thereafter, hyperpolarizations faded, following a biphasic time course, and thus displaying two different time constants, one fast (tau fast = 580 msec for sinoatrial node; 550 msec for atrioventricular node), and one slow (tau slow = 9.2 sec for both sinoatrial and atrioventricular nodes). Hyperpolarizations during brief (200-msec) but repetitive vagal trains also faded biphasically, but approached a steady state much more rapidly than responses to long-lasting trains. Recovery from hyperpolarization decay occurred rather slowly and was linear. Our results demonstrate that the membrane potential responses to vagal stimulation in the atrioventricular node are indistinguishable from those in the sinoatrial node, and suggest that similar muscarinic receptors are operative in both regions. These phenomena may play an important role in the response of the cardiac conducting system to direct or reflexly mediated vagal input.

The distribution of acetylcholine and choline in guinea pig heart

The regional distributions of acetylcholine (ACh) and choline (Ch) in the guinea pig heart were investigated with a pyrolysis-mass fragmentography technique. Using ACh as a marker for cholinergic neurons, we have described a pattern of parasympathetic innervation in the guinea pig heart. This distribution is very similar to that suggested by studies using several different cholinergic indicators in various species. Atrial areas receive richer parasympathetic innervation than ventricular areas, with the right portions receiving more than the left. The nodal areas were the most abundantly innervated regions examined. Ch content is not a good indicator for cholinergic innervation as the regional distribution of ACh and Ch throughout the guinea pig heart are not strongly associated.

Vagal control of pacemaker periodicity and intranodal conduction in the rabbit sinoatrial node

Effects of brief vagal bursting on pacemaker periodicity and intranodal conduction were studied by recording transmembrane potentials in isolated rabbit sinoatrial preparations. Postganglionic vagal terminals were activated by applying brief (50- to 150-msec) trains of pulses (duration, 100 mu sec; frequency, 200 Hz) to the endocardial surface of the node. Sympathetic effects were prevented by superfusion with Tyrode's solution containing propranolol (1 microgram/ml). Vagal trains applied singly every 10 seconds produced brief hyperpolarizations in "true" pacemaker and transitional cells, and induced phasic changes in periodicity and conduction. These changes were out of phase with each other, and were dependent on the magnitude and duration of the vagal train, as well as on its position within the pacemaker period. When similar trains were applied repetitively at cycle lengths (200-1200 msec) that were independent of the pacemaker period, complex patterns of vagus-sinoatrial node interactions resulted. Hence, depending on the vagal stimulus cycle length, the sinoatrial pacemaker was forced to beat at stable and predictable harmonic (i.e., 1:1, 1:2, 2:1, etc.), subharmonic (3:2, 4:3, etc.), or more complex entrainment ratios. At some of these ratios, arrhythmic sinus patterns coexisted with intranodal conduction disturbances, including first- and second-degree block, Wenckebach phenomena, or combinations thereof. At other entrainment ratios, arrhythmias occurred in the absence of conduction changes. At still other ratios, conduction disturbances developed in the presence of apparently undisturbed sinus rhythm. These results provide insight into the mechanism of the dynamic vagal control of sinoatrial periodicity and conduction, and may have clinical implications as well.

A model of dynamic vagus-sinoatrial node interactions

Computer simulations of dynamic vagus-sinoatrial (SA) node interactions were performed using an empirical model. The phasic effects of single vagal trains on pacemaker cycle length obtained experimentally in isolated preparations were summarized in phase response curves (PRCs). These PRCs were used to stimulate the interactions of the sinoatrial pacemaker with single or with repetitive vagal input. For single stimuli, the triphasic inhibitory curve describing the time course of a brief vagal burst (G. Brown and J. Eccles. J. Physiol. London 82: 211-241, 1934; and J. Jalife and G. K. Moe. Circ. Res. 45: 595-607, 1979) was used to predict the PRC at any given spontaneous pacemaker cycle length. In simulations of repetitive vagal input the model predicted the entrainment of the pacemaker. The patterns of interaction were dependent on the shape and amplitude of the PRC as well as on the relationship between the spontaneous pacemaker period and the vagal cycle length. At certain vagal frequencies, stable entrainment of the pacemaker occurred, and the entrained pacemaker period held harmonic relations to the vagal input (i.e., 1:1, 2:1, and so on). At other frequencies, zones of instability were found in which arrhythmic patterns developed. These predictions of the model matched the experimental results very closely. Under some conditions, during simulations with fixed sinovagus coupling intervals, the model generated patterns of sinus activity similar to those occurring experimentally or in patients with apparent sinoatrial block. The model was also capable of generating patterns similar to those obtained in cases of isorhythmic atrioventricular dissociation. The study of these interactions may have important bearing on the understanding of the dynamic control of heart rate by the parasympathetic nervous system and may be used to explain certain cardiac dysrhythmias.

In vivo acetylcholine turnover in rat heart

The in vivo uptake of choline (Ch) and synthesis of acetylcholine (ACh) in rat heart were studied using a pyrolysis mass fragmentography (PMF) method. Deuterium labeled Ch was pulse injected (i.v.) into anesthetized rats. Labeled and unlabeled Ch and ACh were measured by PMF in hearts at various times following injection. From these data we calculated the specific activities of Ch and ACh, rate constants for ACh and turnover rates of ACh. After an initial equilibration period of approximately 2 min, the specific activities of Ch and ACh decayed in a parallel manner with half-times of 28.2 and 28.8 min respectively. Between 2 and 60 min the calculated ACh turnover rate was 0.144 nmol/g/min. Unlike brain Ch, heart Ch levels are very stable with time following sacrifice. No advantage was found in using microwave irradiation to stabilize heart ACh and Ch content.

Dynamic vagal control of pacemaker activity in the mammalian sinoatrial node

Dynamic heart rate control by parasympathetic nervous input involves feedback mechanisms and reflex bursting of efferent cardiac vagal fibers. Periodic vagal bursting induces phasic changes in sinoatrial cycle length and can entrain the pacemaker to beat at periods that may be identical to those of the vagal burst. We investigated the electrophysiological basis of these phenomena in isolated sinus node preparations (rabbit, cat, and sheep). In the presence of propranolol (3.9 X 10(-6)M), relatively brief (50-150 msec) trains of stimuli, applied onto the endocardial surface of the preparation, activated postganglionic vagal terminals and induced a brief hyperpolarization of sinoatrial pacemaker cells. This vagally mediated hyperpolarization could alter the pacemaker rhythm by an amount that depended on its duration and its position in the cycle, as well as on the duration of the free-running pacemaker period. When the free-running period was sufficiently long and the hyperpolarization was induced sufficiently early in the spontaneous cycle, a "paradoxical" acceleration of the pacemaker rhythm ensued. Phasic changes were plotted on phase-response curves, constructed by scanning systematically the sinoatrial pacemaker period with single or repetitive vagal trains. These phase-response curves enabled us to predict the entrainment characteristics and the levels of synchronization of the pacemaker to the vagal periodicity. The overall data explain the cellular mechanisms involved in the phasic effects of brief vagal discharges on sinoatrial periodicity, and provide conclusive evidence for the prediction that repetitive vagal input is capable of forcing the cardiac pacemaker to beat at rates that can be faster or slower than the intrinsic pacemaker rate. These data should improve our knowledge of the dynamic control of heart rate by neural reflexes and aid in our understanding of rhythm disturbances generated by the interaction of the cardiac pacemaker with vagal activity.

Dantrolene sodium: effects on isolated cardiac tissues

The effects of dantrolene sodium on dog Purkinje fibers, cat atrial and ventricular muscles were studied. Action potential duration was significantly increased and contractility was significantly decreased by dantrolene in all three types of tissue. The plateau phase of Purkinje fiber and occasionally atrial action potential was slightly depressed. Dantrolene sodium had no significant effect on resting membrane potential, action potential amplitude or upstroke velocity of phase 6. The negative inotropic effects were most pronounced in Purkinje fibers, followed by atrial muscle while papillary muscles were least sensitive. Contractile force of Purkinje fibers was decreased by relatively the same amount at all frequencies of stimulation. At faster rates, atrial and ventricular muscle contractility was depressed relatively more than at slower rates. Slow response action potentials in cat papillary muscle were diminished slightly, but this effect was not significant. All drug effects took 10 to 15 min to develop, reached a steady state after 30 to 40 min, and were irreversible upon washout. Increasing the extracellular calcium concentration reversed the dantrolene-induced changes. These findings suggest that effects of dantrolene are mediated in part by a decrease in the intracellular free calcium concentration in cardiac tissue.

Amantadine-induced diastolic depolarization and automaticity in ventricular muscle

We studied the cardiac effects of amantadine, an antiviral and anti-Parkinson drug. Amantadine hydrochloride (100--800 microM) produced significant changes in the electrophysiological properties of isolated ventricular muscle preparations from frog, rabbit, cat, dog, and calf. At relatively low concentrations (100--300 microM), the drug increased action potential duration, decreased action potential amplitude and maximum diastolic potential, and induced phase 4 depolarization. Amantadine also caused subthreshold diastolic depolarizations, apparent upon cessation of stimulation in all preparations studied. The amplitude of the diastolic depolarizations increased as a function of time and/or concentration of drug, eventually reached threshold, and spontaneous activity ensued. In the steady state, amantadine-induced spontaneous activity was rather stable, and the rate was dependent upon the amantadine and external potassium concentrations, as well as the membrane potential. In the absence of stimulation, amantadine-induced spontaneous activity occurred abruptly or could be triggered by a single stimulus, often occurring in a "bursting" fashion that appeared to originate from multiple discrete foci. All actions of amantadine were rapidly reversed upon washout. Propranolol had no effect on the actions of the drug. Amantadine-induced spontaneous activity was unaffected by lidocaine, diminished by TTX, and reduced or abolished by verapamil. The results indicate that amantadine can directly alter the membrane properties of ventricular muscle, possibly due to an effect on potassium conductance. Furthermore, amantadine can be used as a tool to study the ionic basis of ventricular automaticity and to model cellular mechanisms of ventricular rhythm disturbances.

Effects of dantrolene sodium on the electrophysiological properties of canine cardiac Purkinje fibers

Dantrolene sodium at concentrations between 8.8 and 35.1 microM produced rather selective changes in the electrophysiological properties of isolated dog Purkinje fibers. The action potential duration at 90% repolarization and effective refractory period of normally polarized fibers were increased in a dose- and frequency-dependent fashion. The plateau phase of the action potential was significantly depressed and this effect coincided with a marked decrease in the strength of contraction. Dantrolene sodium had no significant effect on resting membrane potential, upstroke velocity of phase O, conduction velocity or pacemaker activity of Purkinje fibers. Dantrolene diminished or abolished "slow response" action potentials produced by superfusion with 18 mM K+ Tyrode's solution containing 10(-7) M isoproterenol. Drug effects took 10 to 15 min to become apparent, reached a steady-state after 45 to 60 min and were not reversible even after 2 hr of washout with drug-free Tyrode's solution. In contrast, increasing the [Ca++]0 produced nearly a complete reversal of the dantrolene-induced changes. These results suggest that dantrolene produces its effects by interfering with the slow inward current. Thus, dantrolene may be similar in action to other slow channel blocking agents, such as verapamil, and may be useful as an antiarrhythmic agent.