Quantitative relationships between .alpha.-adrenergic activity and binding affinity of .alpha.-adrenoceptor agonists and antagonists

Quantitative relationships between in vitro affinity for alpha 1- and alpha 2-adrenoceptors (specific binding sites in rat brain membranes of [3H]prazosin and [3H]clonidine, respectively) and in vitro and in vivo alpha 1/alpha 2-adrenoceptor agonist/antagonist activities were derived for a series of 11 alpha-adrenergic antagonists and 35 agonists of dissimilar chemical structure. For the antagonists, the alpha 1/alpha 2-binding selectivity ratio most significantly correlated with the functional alpha 1/alpha 2-blocking selectivity ratios assessed in vitro (rabbit isolated pulmonary artery: antagonism of alpha 1-adrenoceptor-induced vasoconstriction and alpha 2-adrenoceptor-evoked facilitation of transmitter release) and in vivo (antagonism of alpha 1- and alpha 2-adrenoceptor-mediated vasoconstriction in pithed normotensive rats). These results show that the in vitro alpha 1- and alpha 2-adrenoceptor binding affinities of the antagonists provide adequate information concerning their functional alpha 1- and alpha 2-adrenoceptor blocking potencies against agonists. For the agonists, the central, alpha 2-adrenoceptor-elicited, hypotensive activity was not correlated with alpha 1-adrenoceptor binding affinity but was most significantly described in terms of affinity for alpha 2-adrenoceptors and a parabolic dependence on log P' (octanol/buffer; pH 7.4; 37 degrees C). The relevance of log P' in the regression is explained by the difference in accessibility to the membrane-bound alpha 2-adrenoceptors in the radioligand displacement experiments and the central medullary (hypotensive) alpha 2-adrenoceptors in the intact animal. In contrast, the affinity parameters for alpha 1- and alpha 2-adrenoceptors were found to be poor descriptors of the hypertensive potency of the agonists in which alpha 1- and alpha 2-adrenoceptors are known to play a role. The correlations in which the individual binding parameters and the combination of both variables were included reached only a moderate significance level.