AIM: The present study was designed to define the ki-netic behavior of fluoxetine N-demethylation in human liver microsomes and to identify the isoforms of cy-tochrome P-450 (CYP) involved in this metabolic path-way. METHODS: The kinetics of the formation ofnorfluoxetine was determined in human liver microsomes from six genotyped CYP2C19 extensive metabolizers (EM). The correlation studies between the fluoxetine N-demethylase activity and various CYP enzyme activi-ties were performed. Selective inhibitors or chemical probes of various cytochrome P-450 isoforms were also employed. RESULTS: The kinetics of norfluoxetine formation in all liver microsomes were fitted by a single- enzyme Michaelis-Menten equation (mean Km=32umol/L-7umol/L). Significant correlations were found between N-demethylation of fluoxetine at both 25/anol/L and 100 p.mol/L and 3-hydroxylation of tolbu- tamide at 250 panol/L (rl=0.821, P1=0.001; r2=0.668, P2=0.013), respectively, and S-mephenytoin 4'-hydroxylase activity (r=0.717, P=0.006) at high substrate concentration of 100/zmol/L. S-mephenytoin (SMP)(a CC19 substrate) at high concentration and sulfaphenazole (SUL)(a selective inhibitor of CYP2C9) substantially inhibited norfluoxetine formation. The re-action was minimally inhibited by coincubation with chemical probe, inhibitor of CYP3A4 (tfiacetylolean-domycin, TAO). The inhibition of fluoxetine N-demethylation at high substrate concentration (100ptmoL/L) was greater in PM livers than in EM livers (73% vs4, 5%, P<0.01) when the microsomes were precoincu-bated with SUL plus TAO. CONCLUSION: Cy-tochrome P-450 CYP2C9 is likely to be a major CYP iso-form catalyzing fluoxetine N-demethylation in human liv-er microsomes at a substrate concentration close to the therapeutic level, while polymorphic CYP'2C19 may play a more important role in this metabolic pathway at highsubstrate concentration.