Cyclosporin A (CsA), a powerful immunosuppressive agent, is orally administered to prevent rejection in new organ transplants and to treat a variety of autoimmune diseases.1) The intestinal absorption of CsA following oral administration is poor (typically 20-50%) and varies greatly among patients, 2) owing to its relatively high molecular weight (1202.6), poor solubility in aqueous fluids, 3) site-specific absorption in the small intestine, 4) and intestinal firstpass metabolism by the Cyp450 system. 5) Furthermore, Pglycoprotein (P-gp), an ATP-dependent efflux transporter, has been proposed as a factor in the poor absorption of CsA [6][7][8] via various experiments, including the effect of single nucleotide polymorphism of P-gp 9) and P-gp related herb-CsA drug interaction. 10) However, little is known to date concerning the quantitative contribution of P-gp to the absorption of CsA. Only a limited number of experiments, such as an in vitro transport study employing the Caco-2 cell system, 11) a speculative in vivo experiment in mice 12) and human clinical trials, 13) have been reported. Thus, the quantitative contribution of P-gp to the absorption of CsA was investigated in the present study. For this purpose, the absorption of CsA in P-gp knockout (Mdr1a/1b knockout) mice and wildtype mice was compared, and the concentration dependency of the transport of CsA across Caco-2 cell monolayers was investigated.
MATERIALS AND METHODS
Chemicals and Dosage Forms3 H-CsA (Amersham, Okaville, ON, U.S.A.), CsA (Galena Co., Czecho) and Neoral ® (a CsA microemulsion by Norvatis Pharma AG, Basle, Switzerland) were used. 3 H-CsA was diluted with the Neoral ® solution in saline to give a specific activity of 10 mCi/ml. All other chemicals were of reagent grade and were used without further purification.Animal Experiments P-gp knockout (Mdr1a/1b knockout) mice (Tarconic, U.S.A.) and wild-type control, FVB mice, weighing 28Ϯ5 g (9-18 weeks of age), were used in all experiments. For the pharmacokinetic study, CsA (as Neoral ® solution, 1 mg/kg) was intravenously administered to mice via the tail vein or orally.8,12) This dose was comparable to an usual dose of CsA in human (i.e., 175 mg/body). In order to avoid possible chronopharmacokinetic changes, 14) the administration was always performed at around 10:30 a.m. Blood samples (10 ml) were collected from the tail artery at designated time intervals, solubilized using a Soluene350 ® : propanol mixture (1 : 1, v/v) and decolorized by the addition of 30% H 2 O 2 . Brain was removed 24 h after the each administration, and solubilized in a similar manner. Total radioactivity was determined by liquid scintillation counting (LS 6000SE; Beckman Coulter, Fullerton, CA, U.S.A.), and used for the calculation of blood and brain concentration of CsA.The area under the blood CsA concentration from time 0 to 24 h after administration (AUC 0-24 ) was determined by the standard log-linear trapezoidal rule. The apparent absolute oral bioavailability (F, %) was calculated by comparing the AUC 0-...