Abstract

This study was designed to establish a strategy to predict drug interactions involving biliary excretion. The interaction between methotrexate and probenecid was examined as an interaction model since this interaction has already been clinically reported. Coadministration of probenecid reduced the biliary clearance of methotrexate in a dose-dependent manner in rats. This inhibition by probenecid was confirmed in vivo both in the uptake and excretion processes of methotrexate across sinusoidal and canalicular membranes, respectively. That is, both hepatic uptake clearance, assessed in integration plot analysis, and steady-state biliary clearance defined with respect to hepatic unbound methotrexate, were reduced in the presence of probenecid. Probenecid inhibited the active transport of methotrexate both in isolated hepatocytes and canalicular membrane vesicles, confirming the interaction at those two membranes. The degree of inhibition of the uptake and excretion processes found in vivo was comparable with the predicted values using the inhibition constant assessed in isolated hepatocytes and canalicular membranes, respectively. This suggests that the interaction at each membrane transport process can be quantitatively estimated from in vitro data. We have also proposed the method to predict the degree of inhibition of the net excretion from circulating plasma into the bile, the predicted values being also comparable with the inhibition actually found in vivo. The present analysis demonstrates a strategic rationale for predicting drug interactions involving biliary excretion using in vitro systems to avoid any false negative predictions.