Project Summary / Abstract Assessing the health effects of exposure to complex mixtures is a priority for NIEHS: “It is imperative to develop methods to assess the health effects associated with complex exposures in order to minimize their impact on the development of disease.” The vast number of potential mixtures includes environmental chemicals, pharmaceuticals, dietary and endogenous compounds. Concentration addition/dose addition (CA) is a predictive method widely used for compounds that act by similar mechanisms and provides a foundation for risk assessment. However, CA cannot make predictions for mixtures that contain full and partial receptor agonists at effect levels above that of the least efficacious component. Since partial agonists are common, we developed Generalized Concentration Addition (GCA) to address this need. GCA has been applied to systems where ligands compete for a single receptor binding site, successfully predicting experimental data for mixtures of AhR ligands and of PPARγ ligands. This project focuses on ligand-receptor systems as they are biologically important, initiate many toxicity pathways, and are amenable to modeling and rapid testing. Our overall hypothesis is that GCA applies to all receptor systems in which ligands reversibly compete for the same receptor binding sites. Based on mechanistic information, we use pharmacologically-based mathematical modeling to estimate the biological effect of mixtures; we test the predictions with empirical data. Here, we propose to test the ability of GCA to predict the biological effects of more complex receptors and mixture scenarios. Specific Aim 1 tests the ability of GCA to predict receptor activation by mixtures of ligands for receptors that homodimerize. The predictions will be tested using reporter cell lines for AR and ERα and a spectrum of ligands (full agonists, partial agonists, competitive antagonists). Applicability of GCA will be further examined using Tox21 data for single chemicals and mixtures. Specific Aim 2 tests the ability of GCA to predict mixture effects for downstream biological endpoints. We hypothesize that GCA predicts a downstream effect if the effect is a function of receptor activation. This will be tested for proximal and distal effects of mixtures of ER ligands (in vitro) and PPARγ ligands (in vitro and in vivo). Specific Aim 3 examines how similar mechanisms must be for GCA to apply. Models for several “similar” mechanisms will be compared with empirical data: 1) mixtures that contain selective receptor modulators for ERα and PPARγ; 2) heterodimer partners that each bind ligands (ERα:ERβ, PPARγ:RXR) and 3) mixtures containing an aromatase inhibitor (altering the amount of natural ligand) plus ERα ligands. This project builds upon the Tox21 recommendations of examining perturbations of toxicity pathways, increased use of in vitro testing and computational models and will generate a powerful approach for improving risk assessment of mixtures.

Project Narrative Toxicity testing is typically performed on individual chemicals, but people are exposed to mixtures of environmental chemicals, pharmaceuticals and other compounds. Assessing the health effects of chemical mixtures is a crucial and difficult challenge. This project combines information on how chemicals act individually with a mathematical model to predict the biological effects of mixtures.