Toxicological evaluation of acyl glucuronides utilizing half-lives, peptide adducts, and immunostimulation assays.

Chemical reactivity of acyl glucuronides (AGs) is believed to be involved in the toxicity of carboxylic acid-containing drugs. Both direct and immune-mediated toxicity have been suggested as possible mechanisms of toxicity; however, it remains unclear. In the present study, we performed assays of half-lives, peptide adducts, and immunostimulation to evaluate the potential risk of AGs of 21 drugs and analyzed the relationship to the toxic category. AGs of all withdrawn drugs tested in this study showed short half-lives and peptide adducts formation, but so did those of several safe drugs. In contrast, only AGs of withdrawn and warning drugs induced interleukin-8 (IL-8) in human peripheral blood mononuclear cells (hPBMCs). Using a DNA microarray assay, we found that zomepirac AG induced the mRNAs of 5 genes, including IL-8 in hPBMCs. In addition, withdrawn and warning drugs were distinguished from safe drugs by an integrated score of relative mRNA expression levels of 5 genes. The immunostimulation assay showed higher sensitivity, specificity, and accuracy compared with other methods. In preclinical drug development, the evaluation of the reactivity of AGs using half-lives and peptide adducts assays followed by the evaluation of immunostimulation by highly reactive AGs using hPBMCs can contribute to improved drug safety.

Impact of intestinal glucuronidation on the pharmacokinetics of raloxifene.

Raloxifene is extensively glucuronidated in humans, effectively reducing its oral bioavailability (2%). It was also reported to be glucuronidated in preclinical animals, but its effects on the oral bioavailability have not been fully elucidated. In the present study, raloxifene and its glucuronides in the portal and systemic blood were monitored in Gunn rats deficient in UDP-glucuronosyltransferase (UGT) 1A, Eisai hyperbilirubinemic rats (EHBRs), which hereditarily lack multidrug resistance-associated protein (MRP) 2, and wild-type rats after oral administration. The in vitro-in vivo correlation (IVIVC) of four UGT substrates (raloxifene, biochanin A, gemfibrozil, and mycophenolic acid) in rats was also evaluated. In Gunn rats, the product of fraction absorbed and intestinal availability and hepatic availability of raloxifene were 0.63 and 0.43, respectively; these values were twice those observed in wild-type Wistar rats, indicating that raloxifene was glucuronidated in both the liver and intestine. The ratio of glucuronides to unchanged drug in systemic blood was substantially higher in EHBRs (129-fold) than in the wild-type Sprague-Dawley rats (10-fold), suggesting the excretion of raloxifene glucuronides caused by MRP2. The IVIVC of the other UGT substrates in rats displayed a good relationship, but the oral clearance values of raloxifene and biochanin A, which were extensively glucuronidated by rat intestinal microsomes, were higher than the predicted clearances using rat liver microsomes, suggesting that intestinal metabolism may be a great contributor to the first-pass effect. Therefore, evaluation of intestinal and hepatic glucuronidation for new chemical entities is important to improve their pharmacokinetic profiles.