Abstract We compared the relative potencies of sinorphan and retorphan, the S- and R-enantiomers of acetorphan a potent inhibitor of enkephalinase (EC 220.127.116.11), to inhibit membrane metalloendopeptidase in vivo and to protect exogenous and endogenous ANF after oral administration. In mice, sinorphan was 2–3 fold as potent as retorphan in inhibiting the specific in vivo binding of [ 3H]acetorphan to kidney enkephalinase. The same potency ration was found for the enhancement of trichloroacetic acid-precipitated radioactivity in kidneys of mice that had received 125I-ANF, which is used as a test for the protection of the hormone against inactivation in vivo. In nine healthy human volunteers who had received a low oral dosage of sinorphan or retorphan in a double-blind, placebo-controlled, randomized trial, sinorphan was also 2–3 fold more potent than retorphan in inhibiting plasma enkephalinase activity. These effects were accompanied by a related rise in plasma ANF immunoreactivity, which also reflected the difference in the effectiveness of the two compounds. Sinorphan was also more potent than retorphan in enhancing urinary cyclic GMP excretion and sodium excretion in five of these subjects. These data indicate that, in humans as in rodents, enkephalinase plays a crucial role in the inactivation of ANF, its partial inhibition in vivo being accompanied by a significant protection of the exogenous or endogenous hormone as well as by typical ANF-like responses. Thus orally administered sinorphan appears to be a promising compound for therapeutic use in cardiovascular and renal diseases in which ANF has been postulated to exert beneficial effects.