Triptolide (TP), a main bioactive component of Tripterygium wilfordii Hook F., is a promising agent for treatment of autoimmune diseases. However, a high incidence of dose-limiting hepatotoxicity was observed in the clinic. Sandwich-cultured rat hepatocyte model was used in this study to identify the involvement of P-glycoprotein (P-gp) in TP disposition and to evaluate TP-induced hepatotoxicity after modulation of P-gp by the known inhibitors, ritonavir and tariquidar, and known inducers, phenobarbital, quercetin, and H 2 O 2 . Our data showed that biliary clearance of TP reduced 73.7% and 84.2% upon treatment of ritonavir (25 mM) and tariquidar (5 mM), respectively. In contrast, increases of 346%, 280%, and 273% in biliary clearance of TP were observed with treatment of phenobarbital (1.0 mM), quercetin (20 mM), and H 2 O 2 (0.5 mM), respectively. The TP-induced hepatotoxicity increased by twofold when CYP activity was blocked by 1-aminobenzotriazole, suggesting that CYP and P-gp may both contribute to the detoxification of TP in the SCRH model. In addition, hepatotoxicity and the expression of apoptosis proteins Bax and Bcl-2 were correlated qualitatively with the TP exposure duration and its intracellular concentration, which, in turn, can be modulated by P-gp inhibitors or inducers. Our results for the first time demonstrated that in addition to CYP-mediated metabolism, P-gp also plays an important role in the disposition of TP and TP-induced hepatotoxicity. Thus, the modulation of canalicular P-gp has a potential to cause drug-drug interaction between TP and the coadministered P-gp inhibitors or inducers in the clinic.

No abstract

Naturally occurring furanocoumarin compounds psoralen (PRN) and isopsoralen (IPRN) are bioactive constituents found in herbaceous plants. They are widely used as active ingredients in several Chinese herbal medicines. In this study, the CYP1A2 inhibitory potential of PRN and IPRN was investigated in rats in vitro and in vivo as well as in human liver microsomes. Both compounds exhibited reversible and time-dependent inhibition toward rat microsomal cyp1a2. The IC 50 , k inact , and K I values were 10.4 6 1.4 mM, 0.060 6 0.002 min 21 , and 1.13 6 0.12 mM for PRN, and 7.1 6 0.6 mM, 0.10 6 0.01 min 21 , and 1.95 6 0.31 mM for IPRN, respectively. In human liver microsomal incubations, potent reversible CYP1A2 inhibition was observed for both compounds, with IC 50 values of 0.26 6 0.01 mM and 0.22 6 0.03 mM for PRN and IPRN, respectively. However, time-dependent inhibition was only observed for IPRN, with k inact and K I values of 0.050 6 0.002 min 21 and 0.40 6 0.06 mM, respectively. Coadministration with PRN or IPRN significantly inhibited cyp1a2 activity in rats, with the area under the curve (AUC) of phenacetin increasing more than 5-fold. Simcyp simulation predicted that PRN would cause 1.71-and 2.12-fold increases in the phenacetin AUC in healthy volunteers and smokers, respectively. IPRN, on the other hand, would result in 3.24-and 5.01-fold increases in phenacetin AUCs in healthy volunteers and smokers, respectively. These findings represent the first detailed report comparing the potential drug-drug interactions of PRN and IPRN, and provide useful information for balancing safe and efficacious doses of PRN and IPRN.