Estimation of the Contribution of CYP2C8 and CYP3A4 in Repaglinide Metabolism by Human Liver Microsomes Under Various Buffer Conditions.

We have previously reported that the microsomal activities of CYP2C8 and CYP3A4 largely depend on the buffer condition used in in vitro metabolic studies, with different patterns observed between the 2 isozymes. In the present study, therefore, the possibility of buffer condition dependence of the fraction metabolized by CYP2C8 (fm2C8) for repaglinide, a dual substrate of CYP2C8 and CYP3A4, was estimated using human liver microsomes under various buffer conditions. Montelukast and ketoconazole showed a potent and concentration-dependent inhibition of CYP2C8-mediated paclitaxel 6α-hydroxylation and CYP3A4-mediated triazolam α-hydroxylation, respectively, without dependence on the buffer condition. Repaglinide depletion was inhibited by both inhibitors, but the degree of inhibition depended on buffer conditions. Based on these results, the contribution of CYP2C8 in repaglinide metabolism was estimated to be larger than that of CYP3A4 under each buffer condition, and the fm2C8 value of 0.760, estimated in 50 mM phosphate buffer, was the closest to the value (0.801) estimated in our previous modeling analysis based on its concentration increase in a clinical drug interaction study. Researchers should be aware of the possibility of buffer condition affecting the estimated contribution of enzyme(s) in drug metabolism processes involving multiple enzymes.

Effect of buffer conditions on CYP2C8-mediated paclitaxel 6α-hydroxylation and CYP3A4-mediated triazolam α- and 4-hydroxylation by human liver microsomes.

1. Buffer conditions in in vitro metabolism studies using human liver microsomes (HLM) have been reported to affect the metabolic activities of several cytochrome P450 (CYP) isozymes in different ways, although there are no reports about the dependence of CYP2C8 activity on buffer conditions. 2. The present study investigated the effect of buffer components (phosphate or Tris-HCl) and their concentration (10-200 mM) on the CYP2C8 and CYP3A4 activities of HLM, using paclitaxel and triazolam, respectively, as marker substrates. 3. The Km (or S50) and Vmax values for both paclitaxel 6α-hydroxylation and triazolam α- and 4-hydroxylation, estimated by fitting analyses based on the Michaelis-Menten or Hill equation, greatly depended on the buffer components and their concentration. 4. The CLint values in phosphate buffer were 1.2-3.0-fold (paclitaxel) or 3.1-6.4-fold (triazolam) higher than in Tris-HCl buffer at 50-100 mM. These values also depended on the buffer concentration, with a maximum 2.3-fold difference observed between 50 and 100 mM which are both commonly used in drug metabolism studies. 5. These findings suggest the necessity for optimization of the buffer conditions in the quantitative evaluation of metabolic clearances, such as in vitro-in vivo extrapolation and also estimating the contribution of a particular enzyme in drug metabolism.