ABSTRACT: Presynaptic imidazoline receptors (IRs) which inhibit norepinephrine (NE) release from sympathetic nerve endings have been identified in cardiovascular tissue of man, rabbit, rat, and guinea pig. They do not belong to one of the classical presynaptic inhibitory receptor classes such as α2-adrenoceptors or H3 histamine receptors, and there is also no relation to I1- and I2-imidazoline binding sites. Segments of human right atrial appendages preincubated with [3H]NE were used to determine unknown pharmacologic properties of the presynaptic IRs. In the presence of 1 μM rauwolscine, S23230, the (−)-enantiomer of the racemic oxazoline derivative S22687 (5-(2(methyl-phenoxy-methyl)-1,3-oxazoline-2-yl)amine) exhibited low potency in inhibiting the electrically evoked [3H]NE release (pIC30%= 4.96), whereas the (+)-enantiomer S23229 and the racemate S22687 were ineffective. The IR-mediated inhibitory effect of the imidazoline BDF 6143 (4-chloro-2-(2-imidazolin-2-ylamino)-isoindoline) and the guanidine aganodine on evoked [3H]NE release from sympathetic nerves in human atrial appendages was counteracted by rauwolscine (with very low potency) and by the cannabinoid CB1-receptor antagonist SR141715A (N-[piperdin-1-yl]-5-[4-chlorophenyl]-2,4-dichlorophenyl]-4-methyl-1H-pyrazole-3-carboxamide). The inhibitory effect of moxonidine on evoked [3H]NE release (which is exclusively mediated via activation of α2-autoreceptors) was antagonized with high potency by rauwolscine, but not by SR141716A. The cannabinoid CB1 receptor agonists CP55,940([(−)-Cis-3-[2-hydroxy-4-(1,1-dimethylheptyl)phenyl]-trans-4-(3-hydroxy-propyl)cyclohexane]) and anandamide inhibited evoked [3H]NE release. Inhibition by CP55,940 and anandamide was abolished by 1 μM SR141716A as well as by 30 μM rauwolscine. In radioligand binding experiments on membranes from human atrial appendages (α2-and σ-binding sites were masked), cannabinoid receptor ligands and IR agonists displaced the radiolabeled guanidine derivative [3H]DTG (1,3-di-o-tolyguanidine, an agonist at presynaptic IRs) from its binding sites. Comparison of the potencies of these drugs determined in the competition experiments with [3H]DTG with those in inhibiting NE release via activation of the presynaptic IRs in the same tissue revealed a correlation. The present results suggest, e.g., that the presynaptic IRs may have certain binding domains in common with presynaptic cannabinoid receptors or that both receptors are different proteins which interact with each other in an unknown manner.