Anion manipulation, a novel antiarrhythmic approach: mechanism of action

We recently reported that modulation of anion homeostasis by substitution of extracellular chloride by nitrate prevents ischaemia- and reperfusion-induced ventricular fibrillation (VF) in rat and rabbit in vitro by an unknown mechanism independent of haemodynamic changes but related to widening of QT interval (Ridley and Curtis 1991). In the present study we have examined three possible explanations for the mechanism: modification of membrane anion permeability, alteration of cyclic nucleotide homeostasis and alteration of intracellular pH. In isolated Langerdorff-perfused rat heart (n = 12/group), substitution of chloride in modified Krebs perfusion solution by anion surrogates (methylsulphate, bromide, nitrate or iodide) inhibited left regional ischaemia- and reperfusion-induced arrhythmias only when the membrane permeability of the surrogate was greater than that of chloride (e.g., nitrate, bromide, iodide); the least permeant anion, methylsulphate, was proarrhythmic during ischaemia. Rank order of arrhythmia susceptibility correlated with the relative permeability of each anion, with near abolition of both ischaemia- and reperfusion-induced VF (P < 0.05) by the most permeant anions (iodide and nitrate). Arrhythmia suppression occurring in the iodide and nitrate groups was accompanied by significant widening of QT interval at 90% repolarization, with effects substantially more marked during ischaemia than before ischaemia. In separate studies using the same model we determined cardiac cyclic (c) AMP and cGMP content and their molar ratios by radioimmunoassay of biopsies before, during and after ischaemia. There was no meaningful relation between cyclic nucleotide content and rank order of arrhythmia susceptibility ruling out changes in the former as a contributory mechanism to the latter. In further studies we measured intracellular pH in the isolated perfused rat heart by phosphorus NMR spectroscopy. Nitrate caused a slight intracellular acidosis which was exacerbated when hearts were made globally ischaemic, indicating that its antiarrhythmic activity was not a consequence of alkalinisation (e.g., via inhibition of chloride-bicarbonate exchange). To test for inherent adverse effects on cardiac contractile function we analysed Starling curves in isolated rat hearts perfused under conditions equivalent to those used for arrhythmia studies. There was no relationship between perfusion anion composition and systolic (developed pressure at constant intraventricular volume, and pressure-volume slope) or diastolic function (end-diastolic pressure at constant intraventricular volume). In conclusion, alteration of membrane permeability is a mechanism which may be sufficient to explain modulation of arrhythmias by manipulation of extracellular anion content, appears to be devoid of deleterious effects on contractile function, and may represent a focus for future antiarrhythmic drug development.