In order to examine the importance of metabolic cycles and in particular of reductions of N-hydroxylated compounds, the reversible metabolism at the amidine, guanidine, and amidinohydrazone nitrogen atoms of various drugs and model compounds was investigated. Many of these N-oxygenated metabolites are very easily reduced back into the starting materials. A comparison of the kinetic data for the N-hydroxylation and reduction suggests that the reduction should predominate in vivo. This could be verified by in vivo studies. Thus, N-hydroxylated amidines (amidoximes) can be used as pro-drugs of amidines. Because of their strong basicity, amidines, guanidines, and amidinohydrazones are protonated under physiological conditions, are very hydrophilic, and are usually not absorbed from the gastrointestinal tract. The N-hydroxylated derivatives of amidines (amidoximes), guanidines (N-hydroxyamidines), and amidinohydrazones (N-hydroxyamidinohydrazones) are less basic because of the introduction of the oxygen atom. They are absorbed from the gastrointestinal tract and then reduced to the active amidines, guanidines, and amidinohydrazones. The pro-drug principle was originally developed in our laboratory for pentamidine and then applied to other amidines such as sibrafiban and melagatran (ximelagatran). The enzymatic basis of N-oxidative processes is very well understood, whereas reductions have been less extensively investigated. We purified an enzyme system from pig and human liver consisting of cytochrome b5, its reductase, and a P450 enzyme, which is involved in the reduction of the N-hydroxylated compounds. Similar activities were found in all species studied so far. Furthermore, comparable reductive reactions could also be demonstrated with microsomal fractions from organs other than liver. In addition, mitochondria are highly capable of performing the reductions of these N-hydroxylated compounds. Thus, several organs and cell organelles are involved in the reduction explaining the extensive reduction of the pro-drugs in vivo underlying the suitability of the concept for drug development.