Population model analysis of chiral inversion and degradation of bupropion enantiomers, and application to enantiomer specific fraction unbound determination in rat plasma and brain.

Pharmacologic effects elicited by drugs most directly relate to their unbound concentrations. Measurement of binding in blood, plasma and target tissues are used to estimate these concentrations by determining the fraction of total concentration in a biological matrix that is not bound. In the case of attempting to estimate R- and S-bupropion concentrations in plasma and brain following racemic bupropion administration, reversible chiral inversion and irreversible degradation of the enantiomers were hypothesized to confound attempts at unbound fraction estimation. To address this possibility, a kinetic modeling approach was used to quantify inversion and degradation specific processes for each enantiomer from separate incubations of each enantiomer in the two matrices, and in pH 7.4 buffer, which is also used in binding experiments based on equilibrium dialysis. Modeling analyses indicated that chiral inversion kinetics were two to four-fold faster in plasma and brain than degradation, with only inversion observed in buffer. Inversion rate was faster for S-bupropion in the three media; whereas, degradation rates were similar for the two enantiomers in plasma and brain, with overall degradation in plasma approximately 2-fold higher than in brain homogenate. Incorporation of degradation and chiral inversion kinetic terms into a model to predict enantiomer-specific binding in plasma and brain revealed that, despite existence of these two processes, empirically derived estimates of fraction unbound were similar to model-derived values, leading to a firm conclusion that observed extent of plasma and brain binding are accurate largely because binding kinetics are faster than parallel degradation and chiral inversion processes.