ABSTRACT
(S)-Efavirenz (EFV) is a reverse transcriptase inhibitor and an antiviral drug. In addition, (S)-EFV can interact off target with CYP46A1, the major cholesterol hydroxylating enzyme in the mammalian brain, and allosterically activate CYP46A1 at a small dose in mice and humans. Studies with purified CYP46A1 identified two allosteric sites on the enzyme surface, one for (S)-EFV and the second site for L-glutamate (Glu), a neurotransmitter that also activates CYP46A1 either alone or in the presence of (S)-EFV. Previously, we found that racemic (rac)-7-hydroxyefavirenz, (rac)-8-hydroxyefavirenz, (S)-8-hydroxyefavirenz, and (rac)-8,14-dihydroxyefavirenz, compounds with the hydroxylation positions corresponding to the metabolism of (S)-EFV in the liver, activated CYP46A1 in vitro. Yet, these compounds differed from (S)-EFV in how they allosterically interacted with CYP46A1. Herein, we further characterized (rac)-7-hydroxyefavirenz, (rac)-8-hydroxyefavirenz, (S)-8-hydroxyefavirenz, and (rac)-8,14-dihydroxyefavirenz, and, in addition, (R)-EFV, (S)-7-hydroxyefavirenz, (rac)-7,8-dihydroxyefavirenz, (S)-7,8-dihydroxyefavirenz, and (S)-8,14-dihydroxyefavirenz for activation and binding to CYP46A1 in vitro. We found that the spatial configuration of all tested compounds neither affected the CYP46A1 activation nor the sites of binding to CYP46A1. Yet, the hydroxylation position determined whether the hydroxylated metabolite interacted with the allosteric site for (S)-EFV [(R)-EFV, (rac)-7,8-dihydroxyefavirenz, and (S)-7,8-dihydroxyefavirenz], L-Glu [(rac)- and (S)-8,14-dihydroxyefavirenz], or both [(rac)-7-hydroxyefavirenz, (S)-7-hydroxyefavirenz, (rac)-8-hydroxyefavirenz, and (S)-8-hydroxyefavirenz]. This difference in binding to the allosteric sites determined, in turn, how CYP46A1 activity was changed in the coincubations with (S)-EFV and either its metabolite or L-Glu. The results suggest EFV metabolites that could be more potent for CYP46A1 activation in vivo than (S)-EFV. SIGNIFICANCE STATEMENT: This study found that not only efavirenz but also all its hydroxylated metabolites allosterically activate CYP46A1 in vitro. The enzyme activation depended on the hydroxylation position but not the metabolite spatial configuration and involved either one or two allosteric sites-for efavirenz, L-glutamate, or both. The results suggest that the hydroxylated efavirenz metabolites may differ from efavirenz in how they interact with the CYP46A1 allosteric and active sites.