A P450 catalyzed N-para-hydroxy metabolite was suggested to be a prerequisite for N-dephenylation occurrence. Although two mechanisms have been proposed to describe this process as a consequence of either a chemical degradation or P450 lead epoxidation of the hydroxy metabolite, direct evidence has not been demonstrated. In this study, we started with a novel technique using a dipeptide, Lys-Phe, to trap the byproduct of N-dephenylation, a quinone-like compound, forming a peptide adduct to facilitate LC/MS characterization. N-dephenylation via chemical degradation was assessed by LC/MS characterization of the resulting (Lys-Phe)(2)-quinone from 4-hydroxyphenyl-2-naphthylamine following interaction with Lys-Phe in pH 7.4 buffer. N-dephenylation mediated by P450 catalysis proposed was investigated in N-para-hydroxy benzodioxane derivative incubated with mouse liver microsomes in the presence of Lys-Phe in 50/50 H(2)(16)O/H(2)(18)O. LC/MS demonstrated that only one of two hydroxy oxygens in the byproduct was exchanged with water and the MS signal intensity of the (16)O labeled peptide adduct was equal to that of (18)O labeled. These observations suggested us that the origin of the oxygen in the byproduct was from water only, not from O(2). Therefore, it appears that N-dephenylation occurs via a stepwise process, namely the substrate is initially metabolized to a N-para-hydroxy metabolite by P450, which was readily oxidized to a quinone imine/iminium chemically or enzymatically, then hydrolyzed resulting in N-dephenylation. However, in our studies, the proposed P450 mechanism involving epoxidation of a N-para-hydroxy metabolite was disproved.