Characterisation of the human liver in vitro metabolic pattern of artemisinin and auto-induction in the rat by use of nonlinear mixed effects modelling.

AIMS: The aims of the study were to characterise the metabolic pattern of artemisinin in human and rat liver microsomes and to assess the magnitude of auto-induction in the rat. METHODS: (14)C-artemisinin was incubated with human liver microsomes and with liver microsomes from rats pretreated with oral artemisinin or placebo. The metabolic fate of (14)C-artemisinin in microsomes from human B-lymphoblastoid cell lines transformed with CYP2A6, CYP2B6 and CYP3A4 was also investigated. The human liver microsome data and the rat liver microsomes data were analysed by nonlinear mixed effects modelling and naïve pooling using NONMEM, respectively. RESULTS: Four metabolites were radiometrically detected in experiments with rat liver microsomes. The model that best described the data involved three primary metabolites of which one metabolite was further metabolised to a secondary metabolite. The formation of the four metabolites was induced 2.8, 7.2, 4.8 and 2.5-fold, respectively, in liver microsomes from rats pre-treated with artemisinin. Three metabolites were formed in human liver microsomes; having the same retention times as three of the metabolites formed in the rat. The final model consisted of two primary metabolites and a secondary metabolite with CYP2B6 and CYP2A6 influencing the formation rates of the major and minor primary metabolites, respectively. CONCLUSIONS: CYP2B6 and CYP2A6 activities described variability in the formation of the major and minor primary metabolites, respectively, in human liver microsomes. All artemisinin metabolic pathways in rat liver microsomes were induced in artemisinin pretreated animals. We suggest modelling as a method for the discrimination and detection of more complex metabolic patterns from in vitro metabolism rate data.