In Vitro Confirmation of Estrogenic and Antiandrogenic Pesticide Bioactivity using Schild Regression Analysis
A broad range of chemicals display estrogenic (ER) or antiandrogenic bioactivity in high throughout (HTS) in vitro assays. Some agencies have developed tiered in vitro–in vivo endocrine screening batteries in which positive in vitro results automatically “trigger” studies with laboratory animals. Since in vitro assays produce a number of false positive and false negative results, triggering testing could result in the unnecessary use of animals and other resources. The false positive rate is particularly high with androgen receptor (AR) antagonists because there are many nonspecific mechanisms that disrupt competitive AR assays such that they falsely appear to be competitive AR ligands in dose-response studies. Herein, we used in vitro Schild regression to interrogate the in vitro ER and anti-AR bioactivity of nine pesticides; seven positive in the HTS ER and/or AR models (G1) and two negative the HTS models (G3) as an example of the utility of this technique. Schild regression discriminates chemicals that act as true competitive receptor ligands from those that disrupt signaling via noncompetitive mechanisms. Schild regression analyses indicated that 50% of four G1 pesticides were ER agonists and 50% were cytotoxic. Furthermore, 60% of five G1 pesticides were not competitive AR antagonists (HTS false positives) and 100% of two G3 pesticides were competitive AR antagonists (HTS false negatives). Herein, we propose a tiered strategy that includes a more in-depth analysis of in vitro bioactivity using Schild regression to determine if HTS or other in vitro bioactivity results from true competitive receptor antagonism or some nonspecific mechanism to avoid animal testing with chemicals that are not competitive receptor ligands.