Transient diastolic dilatation of the isolated canine left ventricle predictably elicits arrhythmias. To test the hypothesis that such arrhythmias may be mediated by sarcolemmal stretch-activated channels, we attempted to inhibit stretch-induced arrhythmias with gadolinium (Gd3+), a potent stretch-activated channel blocker. In experiments with six isolated canine hearts, left ventricular volume was increased for 50 msec during early diastole and then returned to initial volume by a computerized servopump. The stretch volume was adjusted to yield a probability of eliciting a stretch-induced arrhythmia of 95 +/- 2% before treatment with Gd3+. When Gd3+ (1-10 microM) was administered, dose-dependent suppression of stretch-induced arrhythmias was observed. The probability of a stretch-induced arrhythmia was reduced to 13 +/- 10% (p less than 0.05) with 10 microM Gd3+. Washout of Gd3+ completely reversed this effect. Since Gd3+ is known to be a calcium channel antagonist, we compared the effect of Gd3+ on stretch-induced arrhythmias with that of verapamil and nifedipine. These calcium channel blockers produced no demonstrable inhibition of stretch-induced arrhythmias when administered at concentrations (1 microM) that substantially depressed left ventricular pressure development. Thus, our results indirectly implicate stretch-activated channels in the genesis of stretch-induced arrhythmias and provide preliminary evidence for a potential new mode of antiarrhythmic drug action--blockade of stretch-activated channels.

Transient diastolic stretch of the left ventricle predictably elicits arrhythmias. To investigate the mechanism of such stretch-induced arrhythmias, monophasic action potentials were recorded from six blood-perfused isolated canine left ventricles with an epicardial contact electrode. Stretch-induced arrhythmias were elicited using a computerized servo-pump system that increased left ventricular volume for 250 ms during early diastole. Depolarizations that coincided with the onset of stretch were observed that always preceded the stretch-induced arrhythmia. As stretch volume (delta V) increased from 10 to 30 ml, the amplitude of the stretch-induced depolarization increased progressively and the probability of eliciting an arrhythmia rose from 30 to 94%. To exclude motion artifact, additional recordings were made after the heart was depolarized by increasing the perfusate K+ concentration to 154 mM (K arrest). After K arrest, the stretch-induced depolarizations were reduced by 95% or more (P less than 0.05) at all stretch volumes. Thus the change in monophasic action potential signal during transient diastolic stretch reflects actual depolarization of the myocardium with negligible motion artifact. When the stretch-activated channel blocker, Gd3+ (10 microM), was administered, which produces potent inhibition of stretch-induced arrhythmias in our model, the stretch-induced depolarizations were substantially reduced in magnitude. Our results show that as diastolic stretch increases, stretch-induced depolarizations become larger and reach threshold potential more often; consequently, the probability of eliciting a stretch-induced arrhythmia increases. This mechanism of arrhythmogenesis may be particularly important in patients with regionally or globally dilated left ventricles.

Gadolinium-sensitive stretch-activated channels have been implicated in the process of mechanotransduction signaling of ventricular myocardium. Such channels nonspecifically transport Na+ and Ca2+ in the inward direction. We tested the hypothesis that Na+ and Ca2+ influx are important in the genesis of stretch-induced arrhythmias (SIAs) in an isolated, blood-perfused canine ventricle. To elicit SIAs, left ventricular volume was transiently increased in early diastole using a computerized servo-pump system. Monophasic action potential recordings revealed stretch-induced depolarizations (SIDs) that preceded the arrhythmias. In five ventricles, raising the perfusate Ca2+ concentration from 1 to 3 mM increased ventricular sensitivity to SIAs, manifested by a decrease in the volume change required to precipitate an arrhythmia 50% of the time (delta V50) from 19.5 +/- 2.7 to 15.2 +/- 1.9 ml (P < 0.05). When the perfusate Na+ concentration was decreased from 150 to 90 mM in seven ventricles, delta V50 greatly increased (31.1 +/- 14.4 vs. 17.7 +/- 5.3 ml, P < 0.05), and SID amplitude decreased by 47% (P = 0.002). The suppression of SIAs with low extracellular Na+ is unlikely to be mediated by voltage-gated Na+ channels because lidocaine (5 mg/dl) did not alter SID amplitude. Thus the transsarcolemmal Na+ gradient (and probably that of Ca2+) modulates the amplitude of SIDs, which, in turn, initiate SIAs. These data provide initial evidence that Na+ and Ca2+ help mediate the mechanotransduction processes that underly the genesis of SIAs.

Ventricular arrhythmias can be initiated by a mechanism of transient diastolic dilation. To test the hypothesis that Ca2+ release from sarcoplasmic reticulum (SR) is important in initiation of such stretch-induced arrhythmias (SIAs), we studied effects of ryanodine in an isolated canine heart model. Arrhythmias were induced by a computerized ventricular volume servo-pump system that transiently increased left ventricular volume by precise amounts (delta V) during diastole. The probability of eliciting an SIA (PSIA) was compared at the minimum delta V that resulted in PSIA of > or = 90% under baseline conditions. Block of SR Ca2+ release with 10(-5) M ryanodine in 11 ventricles produced mild inhibition of SIAs, reducing PSIA by 19.4% (P = 0.039). Because ryanodine produces leakage of SR Ca2+ at low concentration and block of SR Ca2+ release at high concentration, ryanodine concentration was varied from 10(-9) to 10(-5) M in six ventricles. Ryanodine had minimal effect on PSIA over this concentration range. In six ventricles with elevated intracellular Ca2+ produced by pretreatment with 0.1-0.3 microM strophanthidin, 10(-5) M ryanodine did not significantly reduce PSIA. Probability of inducing ventricular pairs or nonsustained ventricular tachycardia was greater in strophanthidin-treated ventricles than in controls, but induction of these repetitive ventricular beats in the strophanthidin group was virtually abolished by addition of 10(-5) M ryanodine.(ABSTRACT TRUNCATED AT 250 WORDS)