Cytochrome P450 (CYP) enzymes are mainly involved in oxidative metabolism of xenobiotics in humans. 20(S)-protopanaxadiol (aPPD), a gastrointestinal bacterial metabolite of ginseng saponin, may be responsible for the antidepressant activity of ginseng. Previously, we have identified three monohydroxylated and five minor dihydroxylated products of aPPD following in vitro incubations with human intestinal microsomes (HIM) and human liver microsomes (HLM). However, the enzymes responsible for oxidative metabolism of aPPD are unknown. The present study aims to determine the contribution of individual CYP enzymes in the formation of monohydroxylated aPPD metabolites in HLM and HIM. Reaction phenotyping was carried out by using specific CYP chemical inhibitors and a panel of human recombinant CYP enzymes. CYP chemical inhibitors used in the study were as follows: α-naphthoflavone (10 µM; CYP1A1), furafylline (20 µM; CYP1A2), 8-methoxypsoralen (1 µM; CYP2A6), orphenadrine (200 µM; CYP2B6), sulfaphenazole (20 µM; CYP2C9), quinidine (10 µM; CYP2D6), diethyldithiocarbamate (50 µM; CYP2E1), ketoconazole (1 µM; CYP3A4/5), troleandomycin (20 µM; CYP3A4), and SKF 525A (500 µM; a broad spectrum CYP inhibitor). CYP inhibitors were co-incubated with 4 µM and 11 µM aPPD in HLM (0.15 mg/ml) and HIM (0.25 mg/ml), respectively, in 67 mM phosphate buffer (pH 7.4) and NADPH-regenerating system for 10 min. A panel of baculovirus-expressed human recombinant enzymes (30 pmol/ml; CYP1A1, CYP1A2, CYP2A6, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP2E1, CYP3A4, CYP3A5, and CYP4A11) was incubated with 11 µM aPPD as described above. The hydroxylated metabolites of aPPD were resolved using a liquid chromatography/mass spectrometry (LC/MS) on a BEH C18 column (1.7 µm, 2.1x100 mm) in positive ionization mode. Monohydroxylated metabolites of aPPD were identified by multiple reaction monitoring of the m/z 477>459 and 459>441 transitions in the LC/MS assay. Ketoconazole potently inhibits metabolite formation (85-95% inhibition) in both HLM and HIM, while diethyldithiocarbamate only shows weak inhibition (20-40%). Interestingly, troleandomycin moderately inhibits (65-70%) metabolite formation in HIM but shows weak inhibition (30-35%) in HLM. Among the human recombinant P450 enzymes examined, recombinant CYP3A4 showed the highest activity in metabolizing aPPD, followed by CYP3A5. In summary, our results show that CYP3A isoforms are the predominant enzymes responsible for aPPD monohydroxylation in HLM and HIM.