A comparison of in vitro and in vivo EDSTAC test battery results for detecting antiandrogenic activity

The androgen receptor (AR) transactivation, binding, and Hershberger assays are being developed for large-scale screening of chemicals for endocrine activity. The goal of this study was to evaluate the correlation between in vitro and in vivo antiandrogenicity assays using a variety of compounds (p,p'-DDE, flutamide (FLUT), spironolactone, procymidone, RU486, methoxychlor (MXC), benzo(a)pyrene (BAP), and selected metabolites). For the AR transactivation assay, AR(+) LNCaP prostate carcinoma cells were transfected with an inducible luciferase reporter construct (pGudLuc7ARE) and exposed for 24 h to test materials (less than or equal to 10 muM) in the presence and absence of 1 nM of the AR agonist R-1881. Each of these materials, including the hydroxlated metabolites of BAP and MXC, produced significant antiandrogenic activity in vitro as evidenced by their inhibition of the response to R-1881. Similarly, in vitro AR binding experiments using the recombinant ligand-binding domain (LBD) of the human AR and fluorescence polarization (FP) methodology yielded IC50s comparable to that of testosterone for RU486 and 9-OH-BAP. Other parent compounds and metabolites exhibited lesser binding affinity. In vivo antiandrogenic activity was evaluated with the Hershberger assay, wherein castrated male CD rats were dosed by gavage for 10 days with (mg/kg per day): MXC (10, 50, 100, and 200), BAP (1, 10, 50, and 100), RU486 (1, 5, 10, and 25), and FLUT (10) in the presence of 0.4 mg/kg per day (sc) of testosterone propionate (TP). Neither BAP nor MXC produced significant decreases in accessory sex tissue (AST) weights relative to the TP control. However, 200 MXC resulted in a significant decrease in body weight and 100 BAP significantly increased absolute and relative liver weights. RU486 (25) produced significant decreases in ventral prostate, seminal vesicle, and Cowper's gland weights without affecting body weight. FLUT (10) decreased all AST weights measured. The antiandrogenic activities of the remaining materials (p,p'-DDE, spironolactone, and procymidone) have been demonstrated in previous Hershberger assays. These data indicate the importance of including in vivo results in assessing the endocrine activity of test materials and further stress the importance of a weight of evidence approach in assessing endocrine activity of test materials. (C) 2004 Dow Chemical Company. Published by Elsevier Inc. All rights reserved.