Delayed protection against ventricular arrhythmias by monophosphoryl lipid-A in a canine model of ischaemia and reperfusion

Bacterial endotoxin reduces the severity of ventricular arrhythmias which occur when a coronary artery is occluded several hours later. We have now examined in anaesthetised dogs the effects on ischaemia and reperfusion-induced arrhythmias, of a non-toxic derivative component of the endotoxin molecule of the lipid-A (monophosphoryl lipid-A). This was given intravenously, in doses of 10 and 100 microg kg(-1), 24 h prior to coronary artery occlusion. Arrhythmia severity was markedly reduced by monophosphoryl lipid-A. During ischaemia, ventricular premature beats were reduced from 315+/-84 in the vehicle controls to 89+/-60 (with the lower dose of monophosphoryl lipid-A) and 53+/-23 (P<0.05) with the higher dose. The incidence of ventricular tachycardia was reduced from 75% to 25% (P<0.05) and 31% (P<0.05), and the number of episodes of ventricular tachycardia from 13.4+/-4.9 per dog to 1.1+/-1.1 (P<0.05) and 1. 2+/-0.9 (P<0.05) after doses of 10 and 100 microg kg(-1), respectively. The incidence of ventricular fibrillation during occlusion and reperfusion in the control group was 96% (15/16), i.e., only 6% (1/16) dogs survived the combined ischaemia-reperfusion insult. Monophosphoryl lipid-A (100 microg kg(-1)) significantly reduced the incidence of occlusion-induced ventricular fibrillation (from 50% to 7%; P<0.05), and increased survival following reperfusion to 54% (P<0.05). Monophosphoryl lipid-A also significantly reduced ischaemia severity as assessed from ST-segment elevation recorded from epicardial electrodes as well as the degree of inhomogeneity of electrical activation within the ischaemia area. There were no haemodynamic differences prior to coronary occlusion between vehicle controls and monophosphoryl lipid-A-treated dogs. These results demonstrate that monophosphoryl lipid-A reduces arrhythmia severity 24 h after administration. Although the precise mechanisms are still unclear, there is some evidence that nitric oxide and prostanoids (most likely prostacyclin) may be involved because the dual inhibition of nitric oxide synthase and cyclooxygenase enzymes by administration of aminoguanidine and meclofenamate abolished the marked antiarrhythmic protection resulted from monophosphoryl lipid-A treatment 24 h previously.

Effects of pituitary adenylate cyclase-activating polypeptide on the cyclooxygenase pathway of rat platelets and on platelet aggregation

Several data suggest that pituitary adenylate cyclase-activating polypeptide (PACAP) is involved in the regulation of local circulation. One possible role of PACAP in the regulation of circulation is that, it may modify the cyclooxygenase pathway of the arachidonate cascade in platelets. Our study was designed to study the effect of PACAP on the cyclooxygenase pathway of rat platelets and on platelet aggregation. PACAP (10(-7) and 10(-6) M) significantly inhibited the cyclooxygenase pathway of platelets, mostly the thromboxane synthesis. Pretreatment with a PACAP receptor antagonist, PACAP(6-38), or with an inhibitor of protein kinase A, H-89, shows that the effects of PACAP on the cyclooxygenase pathway were diminished. In the aggregation studies, PACAP inhibited both the arachidonic acid-induced and the thrombin-induced platelet aggregation. It can be concluded that PACAP inhibits the cyclooxygenase pathway of rat platelets via a specific PACAP receptor-activated, cAMP-dependent pathway, and these effects of PACAP are involved in the inhibition of platelet aggregation.

Repeated cardiac pacing extends the time during which canine hearts are protected against ischaemia-induced arrhythmias: role of nitric oxide

Right ventricular pacing in lightly anaesthetized dogs (4x5 min periods at a pacing rate of 220 beats/min) protects against the consequences of coronary artery occlusion when this is initiated 24 h after the pacing stimulus. The main purpose of the present experiments was to determine whether repeating the pacing stimulus, at a time when protection from the initial stimulus had faded (48 h), prolonged the protection afforded against ischaemia-induced ventricular arrhythmias and other ischaemic changes (epicardial ST-segment mapping; changes in the degree of electrical inhomogeneity in the ischaemic region). Dogs were paced on two occasions, with a 48 h period between and, at different times (48, 72 and 96 h) after the second pacing stimulus, were re-anaesthetized and subjected to occlusion of the left anterior descending coronary artery. There was a marked reduction in the severity of ischaemia-induced arrhythmias 48 and 72 h after the second pacing stimulus (reduction in occlusion-induced and reperfusion-induced ventricular fibrillation, e.g. at 72 h 0/11 during occlusion and only 3/11 following reperfusion, compared to 7/21 and 10/21 respectively in the controls P<0.05). The protection had disappeared 96 h following the second pacing stimulus. Changes in ST-segment elevation and in the degree of inhomogeneity largely followed these changes in the severity of ventricular arrhythmias. The results suggest the possibility of maintaining protection against life-threatening arrhythmias following coronary occlusion by repeating a preconditioning pacing stimulus. We also demonstrate that this prolonged protection afforded by repeated cardiac pacing is mediated by nitric oxide, since the marked antiarrhythmic effect observed, e.g. 72 h after the second pacing stimulus, was abolished when S-(2-aminoethyl)-isothiourea (AEST), a particularly selective inhibitor of iNOS, had been administered before coronary artery occlusion.

Application of neural networks in structure-activity relationships

Methodology and application of artificial neural networks in structure-activity relationships are reviewed focusing on the most frequently used three-layer feedforward back-propagation procedure. Two applications of neural networks are presented and a comparison of the performance with those of CoMFA and a classical QSAR analysis is also discussed.

Vasoactive substances produced by cultured rat brain endothelial cells

The vasoactive substances synthesized by primary cultures of rat brain endothelial cells were investigated and compared to those from two, immortalized cell lines, RBE4 and GP8. The vasoactivity of endothelium-derived substances was measured on isolated canine coronary artery. Vascular tone was significantly decreased by both primary and GP8, but not by RBE4 cells. Indomethacin pretreatment of primary and GP8 cells turned vasorelaxation into contraction while N(omega)-nitro-L-arginine pretreatment decreased the vasorelaxation induced by primary, but not by GP8 cells. Eicosanoid production was determined after incubation with [14C]arachidonic acid. The predominant vasoactive eicosanoid was prostaglandin E2 in both primary and GP8 cells. RBE4 cells synthetized mainly prostaglandin E2 and thromboxane B2 and significantly less prostaglandin E2 than did either primary or GP8 cells. The capacity of cerebral endothelium to regulate vascular tone by production of dilator and constrictor substances can be preserved under certain circumstances in immortalized cell lines.

Comparison of the efficacy of glibenclamide and glimepiride in reperfusion-induced arrhythmias in rats

The effect of glibenclamide and glimepiride, two orally active antidiabetic sulphonylurea derivatives, was investigated on the development of reperfusion-induced arrhythmias and it was compared to their blood glucose lowering action. Arrhythmias were produced by reperfusion following 6 min coronary artery ligation in anaesthetised rats. Glimepiride pretreatment (0.001-0.01-0.1-5.0 mg/kg i.p., 30 min before coronary occlusion) significantly decreased the incidence of irreversible ventricular fibrillation and increased the survival rate during reperfusion (64%, 61%, 60%, and 67% vs. 27% in controls). Glibenclamide produced similar effect (81% survival) only in a dose of 5 mg/kg, while smaller doses were ineffective. The minimal hypoglycaemic dose and the dose required to inhibit significantly the oral glucose loading-induced hyperglycaemia were similar (1 and 0.1 mg/kg, respectively) after glibenclamide and glimepiride. It is concluded that although the blood glucose lowering potency of glibenclamide and glimepiride is rather similar, glimepiride appears to be more potent than glibenclamide in preventing reperfusion-induced cardiac arrhythmias.

Delayed rectifier potassium current in undiseased human ventricular myocytes

OBJECTIVE

The purpose of the study was to investigate the properties of the delayed rectifier potassium current (IK) in myocytes isolated from undiseased human left ventricles.

METHODS

The whole-cell configuration of the patch-clamp technique was applied in 28 left ventricular myocytes from 13 hearts at 35 degrees C.

RESULTS

An E-4031 sensitive tail current identified the rapid component of IK (IKr) in the myocytes, but there was no evidence for an E-4031 insensitive slow component of IK (IKs). When nifedipine (5 microM) was used to block the inward calcium current (ICa), IKr activation was fast (tau = 31.0 +/- 7.4 ms, at +30 mV, n = 5) and deactivation kinetics were biexponential and relatively slow (tau 1 = 600.0 +/- 53.9 ms and tau 2 = 6792.2 +/- 875.7 ms, at -40 mV, n = 7). Application of CdCl2 (250 microM) to block ICa altered the voltage dependence of the IKr considerably, slowing its activation (tau = 657.1 +/- 109.1 ms, at +30 mV, n = 5) and accelerating its deactivation (tau = 104.0 +/- 18.5 ms, at -40 mV, n = 8).

CONCLUSIONS

In undiseased human ventricle at 35 degrees C IKr exists having fast activation and slow deactivation kinetics; however, there was no evidence found for an expressed IKs. IKr probably plays an important role in the frequency dependent modulation of repolarization in undiseased human ventricle, and is a target for many Class III antiarrhythmic drugs.

Comparison of the cellular electrophysiological characteristics of canine left ventricular epicardium, M cells, endocardium and Purkinje fibres

Electrophysiological differences among M cells, epicardium, endocardium and Purkinje fibres of the canine ventricle were studied over a wide range of stimulation cycle lengths, and the pharmacological response of these cell types to the sodium channel blocker tetrodotoxin, calcium channel blocker nifedipine and ATP-sensitive potassium channel activator pinacidil was compared. The experiments were carried out by applying standard intracellular microelectrode technique in isolated dog left ventricular preparations. The results confirmed the existence of M cells in the canine ventricle, in addition, the distribution of the rate of rise of the action potential upstroke and action potential amplitude values reflecting probably the inhomogeneity of the fast sodium current in these cells was revealed. It was also demonstrated that M cells differ from Purkinje fibres in some aspects which were not expected from previous investigations: (1) The early portion of the action potential duration restitution curve in M cells is more similar to that of endocardial and epicardial cells than to Purkinje fibres. (2) The plateau phase of the action potentials in Purkinje fibres developed at a more negative potential range than that in the other cell types studied. (3) The pharmacological response to tetrodotoxin and pinacidil in M cells resembles to that in the endocardial and epicardial cells more than in the Purkinje fibres. Our results provide further evidence in support of the existence of M cells but also indicate that there are important electrophysiological as well as pharmacological differences between M cells and Purkinje fibres.

Differential efficacy of vasodilators in hypercholesterolaemic rabbits

The effects of hypercholesterolaemia on the endothelium-dependent and -independent vascular reactivity of the superior mesenteric artery has been examined in anaesthetized rabbits in-vivo. Rabbits were fed with either standard or cholesterol-enriched diet for 24 weeks. Plasma lipids and changes in the endothelin content of plasma and vascular tissue were measured in the superior mesenteric artery and in the thoracic aorta. The functional severity of atherosclerosis was determined by examining vascular responses in the isolated thoracic aorta. The blood flow in the superior mesenteric artery was measured by transit-time flowmetry and drugs were injected through an intra-abdominal aortic catheter. Acetylcholine (5, 10, 20 microg kg(-1)) elicited dose-dependent, mesenteric vasodilation in normocholesterolaemic rabbits. In hypercholesterolaemic animals the response to acetylcholine was completely abolished and even became a vasoconstriction. Endothelin levels in plasma and in the vascular tissue were significantly elevated in hypercholesterolaemic animals compared with controls. Cromakalim at a dose of 3 microg kg(-1), elicited similar mesenteric vasodilation in hypercholesterolaemic and normocholesterolaemic animals. These experiments show that the endothelium-dependent responses of the superior mesenteric artery to acetylcholine are functionally impaired by prolonged hypercholesterolaemia, that this altered vascular reactivity is associated with the elevation of endothelin levels in the circulation and in vascular tissues, and that in hypercholesterolaemia the mesenteric vasodilator effect of the K+-channel opener cromakalim is entirely preserved, suggesting that severe hypercholesterolaemia does not depress the function of ATP-sensitive potassium channels in mesenteric vascular smooth muscle.

Endothelium-dependent vasorelaxant and anti-aggregatory effect and mechanism of action of some antifibrinogen RGD (Arg-Gly-Asp-containing) peptides

Vasorelaxation caused by some antifibrinogen RGD (Arg-Gly-Asp-containing) peptides and their basic mechanism of action was studied on rabbit isolated thoracic aortic rings preconstricted with 0.25 microM phenylephrine. GRGDS (Gly-Arg-Gly-Asp-Ser-OH) and RGDV (Arg-Gly-Asp-Val-OH) caused dose-dependent relaxation. RGDS (Arg-Gly-Asp-Ser-OH) had a biphasic effect (a transient relaxation followed by a contraction) while GRGDS-[SE] (Gly-Arg-Gly-Asp-Ser(SO3)-OH) did not change the isometric tone of precontracted aortic preparations. GRGDS and RGDV exerted no relaxing effect on endothelium-denuded blood vessels suggesting that the vascular action of these peptides was entirely dependent on the presence of functionally intact endothelium. L-NG-Nitro-arginine (30 microM) attenuated the relaxation induced by GRGDS and abolished that induced by RGDV. All of the four RGD congeners inhibited ADP-induced aggregation of human platelets. These findings indicate that the relaxant effect of RGDV is mediated exclusively by the nitric oxide pathway, but GRGDS could cause, besides nitric oxide release, the release of another substance which is different from nitric oxide. Because the rank order of the vasorelaxant potencies of RGD peptides differed from that found for their anti-aggregatory activities, a vascular effector mechanism mediated by an RGD-recognizing structure other than the known glycoprotein IIb/IIIa-like RGD-binding site is suggested.

Comparison of the antiarrhythmic and the proarrhythmic effect of almokalant in anaesthetised rabbits

In this study the antiarrhythmic and the proarrhythmic activities of almokalant, a selective class III antiarrhythmic agent, were compared. The antiarrhythmic effect of the drug was tested in pentobarbital-anaesthetised rabbits. Arrhythmia was evoked by occluding and releasing the left circumflex coronary artery. Almokalant in a dose of 250 nmol/kg i.v., significantly decreased the incidence of reperfusion induced ventricular fibrillation (21% vs. 75% in the control group) and increased the proportion of surviving animals during reperfusion (86% vs. 42%). The proarrhythmic effect of almokalant was examined during alpha1-adrenoceptor stimulation in chloralose-anaesthetised rabbits. Almokalant (75 nmol/kg per min) triggered torsade de pointes arrhythmias in 8 animals out of 11. The dose of almokalant (mean+/-S.E.M.) required to produce this effect was 1181+/-519 nmol/kg. It is concluded that, although almokalant is an effective antiarrhythmic agent against ischaemia-reperfusion induced arrhythmias, it has marked proarrhythmic activity during alpha1-adrenoceptor stimulation.

The dopamine receptor agonist Z1046 reduces ischaemia severity in a canine model of coronary artery occlusion

Z1046, (S)-6[[6-[[2-(2-methoxyphenoxy)ethyl]amino]propyl]amino]-5,6,7,8-tetra-h ydro-1,2-naphtalenediol dihydrochloride, is an agonist at both dopamine D1 and D2 receptors. Since stimulation of dopamine D2 receptors inhibits noradrenaline release, and because cardiac noradrenaline release has been implicated in the genesis of early ischaemia-induced, life-threatening ventricular arrhythmias, the effect of Z1046 has been examined for its effects on coronary artery occlusion in chloralose urethane anaesthetised mongrel dogs. Z1046 (10 microg kg(-1) intravenously or 1 microg kg(-1) by local intracoronary injection) decreased heart rate and reduced arterial blood pressure and coronary blood flow, effects prevented by the prior administration of domperidone (40 microg kg(-1) i.v.). The ischaemic changes induced by a 25-min occlusion of the left anterior descending coronary artery (including ST-segment elevation and ventricular ectopic activity) were much less marked in those dogs administered Z1046 and survival from the combined ischaemia reperfusion insult was increased from 7% to 36% (P < 0.05). These effects of Z1046 were partly attenuated by domperidone. We conclude that the anti-ischaemic effects of Z1046 are due to inhibition of cardiac sympathetic responses. Studies using rat isolated perfused mesenteric vascular bed preparations subjected to sympathetic nerve stimulation confirmed that Z1046 inhibits synaptic transmission without modifying vascular responses to noradrenaline.

Polarographic detection of nitric oxide released from cardiovascular compounds in aqueous solutions

In order to detect the concentration of nitric oxide, known to be one of the biologically active principles of certain cardiovascular compounds, a highly selective polarographic/amperometric device was used. The nitric oxide-releasing properties of sodium nitroprusside, nitroglycerine, nicorandil, and the molsidomine metabolite, 3-morpholinosydnonimine, were compared in the following cell-free experimental solutions in vitro: in Krebs-Henseleit solution with and without a sulfhydryl donor, L-cysteine, in an acidic, reducing medium, and in Krebs-Henseleit solution with superoxide dismutase enzyme. Sodium nitroprusside released similar concentrations of nitric oxide in Krebs-Henseleit solution and in the acidic, reducing medium. L-Cysteine inhibited the release of nitric oxide at physiological pH. In the presence of nitroglycerine, nitric oxide signals were detected in the acidic, reducing environment and in L-cysteine-rich Krebs-Henseleit solution but not in the absence of the sulfhydryl donor. Amperometric signals could not be detected after adding nicorandil in all the experimental conditions used. 3-Morpholinosydnonimine released nitric oxide only in the presence of the superoxide dismutase enzyme. Our results suggest that the polarographic electrode is able to detect the release of nitric oxide from sodium nitroprusside, nitroglycerine, and 3-morpholinosydnonimine in the absence of biological material. The present observations support the importance of the chemical environment during the detection of nitric oxide from donor compounds in the common in vitro bathing systems.

Effect of disopyramide on potassium currents in rabbit ventricular myocytes

The effects of disopyramide (1-30 microM) on the 4-aminopyridine sensitive transient outward current (I(to)), on the rapid component of the delayed rectifier potassium current (I(Kr)) and on the inward rectifier potassium current (I(kl)) were studied in single rabbit ventricular myocytes at 35 degrees C by applying the whole-cell configuration of the patch clamp technique. Disopyramide significantly decreased the amplitude of I(to) (from 1510 +/- 122 pA at control to 1015 +/- 21 pA after 30 microM disopyramide at +50 mV; n = 5). This effect was not voltage- or use-dependent. Disopyramide (10 microM) influenced neither the recovery from inactivation of I(to) nor the steady-state inactivation curve. The drug dose dependently decreased the time constant of the fast component of the decay of I(to) (tau(f) = 6.41 +/- 0.25 ms, n = 24 for control; and 2.20 +/- 0.38 ms, n = 5 after 30 microM disopyramide at +50 mV). The fractional block caused by 30 microM disopyramide as a function of time was well fitted by a single exponential function with time constant of 1.48 +/- 0.18 ms (n = 5), most likely reflecting the binding kinetics of the drug to the open channel. The offset kinetics of the drug was estimated by using a double-pulse protocol and its time constant was 3.9 +/- 0.5 ms. Disopyramide (30 microM) did not influence significantly the onset of inactivation measured at -20 mV. The estimated EC50 value for the I(to) block by disopyramide was 14.1 microM. Our results are consistent with an open-channel block of I(to) by disopyramide, however, a weak, drug-induced increase of the rate of inactivation and a moderate tonic block cannot be excluded. The amplitude of the outward tail current attributed to I(Kr) was depressed dose dependently by disopyramide (after clamping the cells back to the holding potential from +30 mV, 139.5 +/- 10.9 pA for control, and 30.7 +/- 3.2 pA in the presence of 10 microM disopyramide; n = 11). The estimated EC50 was 1.8 microM. I(to) is thus less sensitive to disopyramide than I(Kr). I(kl) was not influenced significantly by disopyramide, even when applied in the highest tested concentration (30 microM). It is concluded that in rabbit ventricular myocytes disopyramide blocks not only I(Kr), but also I(to), both of which may play an important role in the well established repolarization lengthening and antiarrhythmic effects of the drug.

Suppression of erythromycin-induced early afterdepolarizations and torsade de pointes ventricular tachycardia by mexiletine

Erythromycin is a selective IKr-blocking, action potential duration (APD)-prolonging drug, which may induce early afterdepolarizations (EADs) and torsade de pointes ventricular tachycardia. The successful termination of an erythromycin-induced clinical torsades de pointes by the authors with mexiletine prompted them to investigate in vitro whether erythromycin is able to induce EADs in Purkinje fibers and, if so, whether EADs are suppressible or not by mexiletine. Electrically stimulated canine Purkinje fibers (n = 9) were superfused with erythromycin (200 mg/l) and action potentials were recorded by an intracellular microelectrode technique. Erythromycin induced a pronounced prolongation of APD and the appearance of EADs in all Purkinje preparations (9/9). After the addition of mexiletine (10 mM), a marked shortening of APD and the disappearance of EADs (7/9) were observed. Mexiletine, an inhibitor of the tetrodotoxin-sensitive window Na(+)-current, may prevent IKr-blocking drug-induced torsade de pointes ventricular tachycardia by abolishing APD prolongation and EADs.

Frequency-dependent Cardiac Electrophysiologic Effects of Tedisamil: Comparison With Quinidine and Sotalol

BACKGROUND: Tedisamil is a potent bradycardiac/antiischemic drug known to lengthen cadiac repolarization by blocking various potassium channels. Recent in vivo experiments revealed that it is an antiarrhythmic agent. It was therefore of interest to compare the cellular electrophysiologic effects of tedisamil with those of quinidine and sotalol in isolated cardiac preparations. METHODS AND RESULTS: The conventional microelectrode technique was applied in isolated dog cardiac Purkinje and ventricular muscle fibers and in rabbit left atrial muscle. Tedisamil (1 µM) and sotalol (30 µM) lengthened, while quinidine (10 µM) shortened action potential duration in dog Purkinje fibers. The phase 1 repolarization was delayed by tedisamil and quinidine and not changed by sotalol. In dog ventricular muscle and in rabbit atrial muscle, all three drugs studied lengthened repolarization. In dog Purkinje fiber, tedisamil and sotalol lengthened action potential duration more at slow than at high stimulation frequency (reverse use-dependence). In dog ventricular muscle fibers, the effect of the drugs was not clearly frequency dependent. In rabbit atrial muscle fibers, the quinidine-evoked repolarization lengthening was most pronounced at intermediate cycle lengths (500-1000 ms). Tedisamil and quinidine but not sotalol depressed the maximal rate of depolarization (V(max)), which depended on the stimulation frequency (use-dependence). The nature of the use-dependent V(max) block differed between quinidine and tedisamil. Quinidine decreased V(max) at a relatively wide range of stimulation frequencies whle tedisamil. Quinidine decreased V(max) at a relatively wide range of stimulation frequencies while tedisamil decreased V(max) largely at high rate of stimulation. Tedisamil and quiinidine prevented or decreased the pinacidil-evoked action potential shortening in dog ventricular muscle, suggesting block of the ATP-dependent potassium channels (I(KATP)), while with sotalol such effect was not observed. CONCLUSIONS: Although tedisamil, quinidine, and sotalol are known to lengthen the QT interval, their cellular electrophysiologic effects substantially differ. Tedisamil lengthens repolarization and prevents pinacidil-evoked action potential duration shortening, suggesting I(K(ATP)) blockade. Its effect on the V(max) is limited mostly to fast heart rate. These electrophysiologic effects of tedisamil resemble those of chronic amiodarone treatment.

Does the opening of ATP-sensitive K+ channels modify ischaemia-induced ventricular arrhythmias in anaesthetised dogs?

These experiments were designed to determine whether there is any change in the severity of ventricular arrhythmias resulting from coronary artery occlusion in anaesthetised mongrel dogs if ATP-sensitive potassium channels were already open at the time of coronary occlusion. To achieve this we locally infused the K(ATP) channel opener levcromakalim, in a total dose of 3 microg/kg, and given by slow infusion over a 30 min period directly into a side branch of the left anterior descending coronary artery. This dose increased blood flow in that main artery by 30% (and by 7% in the adjacent left circumflex artery). The degree of inhomogeneity of electrical activation, measured from the left ventricular wall distal to the occlusion site, was unaffected by levcromakalim administration but there was significant epicardial ST-elevation, perhaps indicating K+ egression from cells. Following coronary artery occlusion there was no marked difference in the severity of arrhythmias between control and levcromakalim-treated dogs, except for an increased number of episodes of ventricular tachycardia due entirely to effects in two of the nine treated dogs. We conclude that opening cardiac K(ATP) channels with levcromakalim, at this one dose level, and administered directly to the left ventricular wall, does not significantly modify arrhythmia severity during ischaemia. These results cannot be extrapolated to studies in which such drugs markedly reduce coronary perfusion pressure.

Bradykinin and endothelial-cardiac myocyte interactions in ischemic preconditioning

Myocardial ischemia results in the release of a variety of vasoactive substances from coronary vascular endothelial cells and/or from cardiac myocytes. Some of these substances appear to be protective and include nitric oxide and bradykinin. One hypothesis for the pronounced antiarrhythmic effects of preconditioning involves the early generation of bradykinin and, subsequently, nitric oxide. Evidence for early bradykinin release has come from clinical studies involving patients undergoing coronary angioplasty where, in 4 of 5 patients, there was evidence for elevated kinin levels in coronary sinus blood either during balloon inflation (i.e., ischemia) or deflation (reperfusion). The levels reached are sometimes considerable (increases 10-20 fold). The second piece of evidence comes from dogs subjected to a preconditioning stimulus (2 x 5 min periods of ischemia), followed 20 min later by occlusion of the same artery for a 25-min period. This preconditioning procedure markedly reduces ischemia-induced ventricular arrhythmias and, although under resting conditions there was little difference between arterial and coronary sinus bradykinin levels (125 +/- 22 and 157 +/- 41 pg/mL, respectively), there was a marked increase in coronary sinus levels in preconditioned dogs before the prolonged occlusion (637 +/- 293 pg/mL compared with 114 +/- 18 pg/mL in nonpreconditioned dogs); levels at the end of the prolonged occlusion in the preconditioned dogs were also higher (577 +/- 305 pg/mL compared with 162 +/- 34 pg/mL in control dogs). Other evidence for the involvement of bradykinin and nitric oxide comes from studies in which the generation, or effects, of these mediators have been suppressed (e.g., with the bradykinin B2 receptor blocking agent icatibant, with inhibitors of the L-arginine-nitric oxide pathway, and by methylene blue). The conclusion is that early bradykinin release is protective under conditions of ischemia, is presumably enhanced during therapy with angiotensin-converting enzyme (ACE) inhibitors and is suppressed under conditions of endothelial dysfunction.

Modification by bradykinin B2 receptor blockade of protection by pacing against ischaemia-induced arrhythmias

In dogs, rapid cardiac pacing, by way of a pacing electrode in the right ventricle, protects against ventricular arrhythmias when a coronary artery is occluded immediately after cessation of the pacing period. This represents a form of ischaemic preconditioning. The role of bradykinin in mediating the protective effects of rapid cardiac pacing in this model was investigated using a selective antagonist of bradykinin at B2 receptors (icatibant; HOE 140). In the presence of icatibant cardiac pacing (220 beats min(-1)) resulted in more severe ischaemia (as assessed by ST-segment elevation from the pacing electrode at the end of the stimulus) and to a higher incidence of ventricular arrhythmias during the pacing protocol. When the coronary artery was occluded under such conditions the antiarrhythmic protection afforded by cardiac pacing was not seen although other indices of reduced ischaemia severity (epicardial ST-segment mapping; changes in the degree of inhomogeneity of electrical activation within the ischaemic area) were not affected by icatibant treatment. These results suggest that bradykinin is an important trigger mediator involved in the protective effects of cardiac pacing. Whether this is due to the generation of endothelium-derived protective substances (such as nitric oxide and prostacyclin) or whether it results from a direct effect on B2 receptors in cardiac myocytes is unclear.

KATP channel modulators increase survival rate during coronary occlusion-reperfusion in anaesthetized rats

We investigated the effect of ATP-sensitive K+ channel (KATP) openers (pinacidil and cromakalim), and a KATP blocker (glibenclamide) on reperfusion-induced arrhythmias in pentobarbitone-anaesthetized rats. Arrhythmias were induced by reperfusion following a 6 min ligation of the left main coronary artery. Rats were pretreated with pinacidil (0.1 or 0.3 mg/kg), or cromakalim (28 or 56 micrograms/kg), or glibenclamide (5 mg/kg), or vehicle. Pinacidil and cromakalim produced dose-related reductions in blood pressure. Pinacidil (0.1 mg/kg) and cromakalim (56 micrograms/kg) significantly decreased the incidence of reperfusion-induced ventricular fibrillation and increased survival. Glibenclamide did not decrease ventricular fibrillation incidence, yet improved survival by increasing the possibility of recovery from ventricular fibrillation. The present study suggests that both opening and blocking KATP channels may increase survival during coronary occlusion and reperfusion in anaesthetized rats.

Increased cardiac workload by adrenoceptor agonists for the estimation of potential antiischemic activity in a conscious rabbit model

The antiischemic effect of drugs can be detected at a lower dose range if the cardiac workload is increased. A brief period of frequency-loading (ventricular overpacing = VOP) results in well-defined, reproducible changes in cardiac parameters in the conscious, chronically instrumented rabbit; however, rapid pacing frequently evoked ventricular tachycardia or even fatal ventricular fibrillation. Therefore, cardiac workload has been increased by i.v. administration of adrenoceptor agonists, such as isoproterenol (ISO), phenylephrine (PHE), and their combination, respectively. The doses applied (especially the combination of 2 micrograms/kg ISO and 16 micrograms/kg PHE, giving optimal changes) were sufficient to produce a marked elevation of both the ST segment in the intracavital electrogram and the left ventricular end-diastolic pressure, without evoking cardiac arrhythmias. We compared the effect of this adrenergic "test" stimulus with that of VOP on hemodynamic and electrophysiological parameters of the heart, and furthermore, on the modification of responses to both "test" stimuli by oral administration of the coronary vasodilator: Isosorbide-5-mononitrate (IS-5-N), given in a dose of 40 mg/kg. Both VOP- and ISO+PHE-induced changes were significantly attenuated by IS-5-N, and a temporal coincidence of the maximal effects was found as well. We reached the following conclusion: The combined administration of ISO and PHE not evoking fatal arrhythmias in the dose range applied can replace the more risky VOP as a "test" workload in the estimation of antiischemic action.

Influence of anesthetics on the incidence of reperfusion-induced arrhythmias and sudden death in rats

PURPOSE

We compared the influence of pentobarbital (P, 60 mg/kg), urethane (U, 1.8 g/kg), and a contribution of diazepam with ketamine hydrochloride (D+K, 10 + 100 mg/kg) anesthesia on the incidence of ischemia/ reperfusion-induced arrhythmias in rats. In anesthetized rats, myocardial ischemia was produced by a 6-min ligation of the left main coronary artery, followed by 5 min of reperfusion. The incidence of reperfusion-induced ventricular fibrillation (VF) and ventricular tachycardia (VT) was markedly decreased in the D+K-anesthetized animals compared with the P-anesthetized group (VF, 46 vs. 89%; VT, 64 vs. 94%). The mean blood pressure (MBP) before coronary ligation was significantly lower in U-anesthetized animals (72 +/- 3.5 vs. 102 +/- 4.1 and 108 +/- 5.9 mm Hg in P and D+K, respectively). The MBP recovery was the best in the D+K-anesthetized group. In experimental work, pentobarbitone anesthesia may be recommended for studying arrhythmias, while the combination of diazepam and ketamine may be useful for hemodynamic investigations.

Electrical restitution in rat ventricular muscle

The mechanism of electrical restitution was studied in isolated rat ventricular muscle using drugs that inhibit specific ion currents. The effect of transient changes in cytosolic Ca concentration and Na/Ca exchange in relation to the restitution process was also studied in single ventricular cardiomyocytes. Conventional microelectrode techniques were applied to record action potentials having gradually increasing coupling intervals, each evoked following a train of stimuli with a frequency of 1 Hz. Ion currents were recorded from enzymatically isolated cells using the whole cell patch clamp technique. Ca transients were monitored in myocytes loaded with the fluorescent dye, indo-1. The electrical restitution process in multicellular rat ventricular preparations at 37 degrees C was described as a sum of three exponential components: an early positive component, a subsequent fast negative component and a late negative component, having time constants of 21.9 +/- 1.9, 73.1 +/- 6.0 and 1053 +/- 61 ms, respectively (n = 9). Inhibition of the transient outward K current, the delayed rectifier K current, or the chloride current did not substantially alter these time constants. The early positive and fast negative components were fully abolished by nifedipine or MnCl2. In the presence of caffeine, the fast negative component was absent, while the time constant of the early positive component increased to 39.5 +/- 5.8 ms (n = 5). In single myocytes loaded with indo-1, the Ca transients decayed with a time constant of 151 +/- 12 ms at room temperature (n = 5). These Ca transients were accompanied by inward current tails, identified as a Na/Ca exchange current, having a decay time constant of 140 +/- 4.5 ms. It is concluded that electrical restitution in rat ventricular muscle is relatively little affected by recovery from voltage-dependent inactivation of ion channels, it is rather governed by transient changes in cytosolic Ca concentration possible via Ca-dependent inactivation of the L-type Ca current and activation of the Na/Ca exchange current.