Stores of transmitter, labelled with [3H]choline, were used to study the negative feedback modulation of the release of acetylcholine by presynaptic M2 muscarinic receptors. The release of acetylcholine was measured by radioassaying the electrical stimulation-evoked release of [3H]acetylcholine from slices of cerebral cortex of the rat and from the Auerbach plexus of the guinea pig ileum. Experimental conditions (2 Hz, 240 shocks) were chosen where the negative-feedback modulation by endogenous acetylcholine was not significant, therefore the presynaptic affinity constant for antagonists was not underestimated. The M2 agonist oxotremorine inhibited the release of acetylcholine in a concentration-dependent manner in both preparations. The IC50 values for oxotremorine were 10.8 +/- 4.89 x 10(-6) M on the cortex and 5.89 +/- 3.85 x 10(-8) M on the Auerbach plexus (n = 4). The effect of oxotremorine was blocked by atropine, similarly to methoctramine, which is a cardioselective muscarinic receptor antagonist. The dose-ratio and dissociation constant were calculated by measuring the righward shift that methoctramine and other antagonists produced on the inhibitory dose-effect curve for oxotremorine. The antagonist equilibrium dissociation constants (pKB) of methoctramine were 5.69 +/- 0.27 and 5.51 +/- 0.37 on the cortical and the myenteric plexus preparations, respectively (n = 4). Postsynaptic antimuscarinic affinity (pA2) of methoctramine on the smooth muscle of the guinea pig ileum was found to be 6.68 +/- 0.11 (n = 4). These findings indicate that, although methoctramine is a cardioselective compound, unlike pancuronium, it may not be a useful tool for discriminating between different presynaptic muscarinic receptors.