In order to evaluate the possible advantages of potentiating the effects of the endogenous enkephalins, to obtain analgesia without the serious drawbacks of morphine, it was essential to design systemically active compounds which inhibit the two metabolizing enzymes, aminopeptidase N (APN) and neutral endopeptidase 24.11 (NEP). A new concept combining the idea of "prodrug" and "mixed inhibitor" was therefore developed. Given the high efficiency of beta-mercaptoalkylamines as APN inhibitors and of N-(mercaptoacyl) amino acids as NEP inhibitors, compounds associating these molecules through disulfide or thioester bonds, which are known to increase lipophilicity and to favor passage across the blood-brain barrier, have been synthesized. An HPLC study indicated that the disulfide bridge was resistant to serum enzymes but was cleaved by brain membrane homogenates, suggesting that the active inhibitors were released in the central nervous system. The validity of the approach was verified by the efficient antinociceptive responses obtained in the hot plate test in mice after iv administration of disulfide-containing inhibitors (ED50s of from 4 to 26 mg/kg on the jump latency time). The analgesic potencies of the "mixed inhibitor-prodrug" RB 101 [H2NCH(CH2CH2SCH3)CH2SSCH2CH(CH2Ph)CONHCH( CH2Ph)COOCH2Ph] after iv administration were three times greater than those of a similar combined dose of its two constitutive moieties. The separation of the two diastereoisomers constituting RB 101 showed that the analgesia has a stereochemical dependence, the (S,S,S)-isomer being more active than the (S,R,S)-isomer. Furthermore, in the tail flick test in the rat, RB 101 gave 38% analgesia at a dose of 80 mg/kg. Due to its high efficiency and its longer pharmacological effect, RB 101 was selected for a complete study of its analgesic properties.