RESUMEN
α1-Acid glycoprotein (AGP) is a primary binding protein for many basic drugs in plasma. The number of drugs that bind to AGP, such as molecular target anticancer drugs, has been continuously increasing. Since the plasma level of AGP fluctuates under various pathological conditions such as inflammation, it is important to evaluate the contribution of AGP to drug pharmacokinetics. Here, we generated conventional AGP-knockout (AGP-KO) mice and used them to evaluate the contribution of AGP. The pharmacokinetics of drugs that bind to two AGP variants (F1*S or A variants) or albumin were evaluated. Imatinib (a F1*S-binding drug) and disopyramide (an A-binding drug) or ibuprofen (an albumin-binding drug) were administered to wild-type (WT) and AGP-KO. The plasma level of imatinib and disopyramide decreased rapidly in AGP-KO as compared to WT. In AGP-KO, AUC and t1/2 were decreased, then CLtot was increased. Compared with disopyramide, imatinib pharmacokinetics showed more marked changes in AGP-KO as compared to WT. The results seemed to be due to the difference in plasma level of each AGP variant (F1*S:A = 2-3:1). No differences were observed in ibuprofen pharmacokinetics between the WT and AGP-KO mice. In vitro experiments using plasma from WT and AGP-KO showed that unbound fractions of imatinib and disopyramide were higher in AGP-KO. These results suggest that the rapid elimination of imatinib and disopyramide in AGP-KO could be due to decreased protein binding to AGP. Taken together, the AGP-KO mouse could be a potential animal model for evaluating the contribution of AGP to the pharmacokinetics of various drugs.