Abstract

Prediction of clinical efficacy, toxicity, and drug-drug interactions may be improved by accounting for the intracellular unbound drug concentration (C_unbound) in vitro and in vivo. Furthermore, subcellular drug distribution may aid in predicting efficacy, toxicity, and risk assessment. The present study was designed to quantify the intracellular C_unbound and subcellular localization of drugs in rat sandwich-cultured hepatocytes (SCH) compared with rat isolated perfused liver (IPL) tissue. Probe drugs with distinct mechanisms of hepatocellular uptake and accumulation were selected for investigation. Following drug treatment, SCH and IPL tissues were homogenized and fractionated by differential centrifugation to enrich for subcellular compartments. Binding in crude lysate and cytosol was determined by equilibrium dialysis; the C_unbound and intracellular-to-extracellular C_unbound ratio (Kp_u,u) were used to describe accumulation of unbound drug. Total accumulation (Kp_observed) in whole tissue was well predicted by the SCH model (within 2- to 3-fold) for the selected drugs. Ritonavir (Kp_u,u ∼1) was evenly distributed among cellular compartments, but highly bound, which explained the observed accumulation within liver tissue. Rosuvastatin was recovered primarily in the cytosolic fraction, but did not exhibit extensive binding, resulting in a Kp_u,u >1 in liver tissue and SCH, consistent with efficient hepatic uptake. Despite extensive binding and sequestration of furamidine within liver tissue, a significant portion of cellular accumulation was attributed to unbound drug (Kp_u,u >16), as expected for a charged, hepatically derived metabolite. Data demonstrate the utility of SCH to predict quantitatively total tissue accumulation and elucidate mechanisms of hepatocellular drug accumulation such as active uptake versus binding/sequestration.