Comprehensive Assessment of Endocrine Active Mixtures

点击翻译按钮获取中文摘要

DESCRIPTION (Provided by Applicant) Knowledge of individual chemical toxicity is often inadequate for risk assessment since exposure occurs as mixtures. To assess the effects of mixtures, an understanding of the additive, antagonistic, and synergistic interactions that occur at the molecular and tissue levels is required within the context of the whole organism and its genome. This proposal will use rigorous statistics to minimize the full factorial design to test the hypothesis that mixtures induce additive, synergistic and antagonistic interactions in a component ratio-, tissue-, dose- and time-dependent manner. Gene expression and tissue level assays will be used to assess dose- and time-dependent mixture-interactions. Individual chemicals and their mixtures will be examined for estrogenic activity in the mouse brain, liver, mammary gland, uterus and bone. cDNA clones or expressed sequence tags (ESTs) for the estrogen responsive genes will be used to construct a custom estrogen responsive cDNA/EST array. Dose- and time-dependent gene expression patterns in response to 17-beta-estradiol, genistein, methoxychlor, and atrazine will be determined in mouse liver and uterus. Mixtures with different fixed dose ratios of these chemicals will be examined using a ray design and compared to patterns for individual chemicals to identify interactions. Uterine gene expression patterns will be correlated with increases in uterine weight to identify gene biomarkers with greater predictive value. Hepatic gene expression patterns will also be compared to patterns obtained in mouse Hepa 1c1c7 cells to assess differences between in vitro and in vivo models. Brain, mammary gland, and bone will also be examined to investigate the possibility of tissue-specific estrogenic gene expression patterns. The data will be analyzed using principal components analysis to identify linear combinations of genes that are related by dose and time, cluster analysis to identify genes with common behavior, and general linear mixed model to identify statistically significant changes in gene expression relative to controls. Detection and characterization of interactions will be determined using established statistical comparisons between the estimated dose response curves along the fixed-ratio ray. This proposal aims to develop innovative and credible statistical strategies for rigorously assessing additive, synergistic and antagonistic interactions induced by mixtures containing components that use a common mechanism of action. This comprehensive approach used will also provide insights into mechanisms of chemical-initiated molecular perturbations associated with tissue level effects and the etiology of a physiological response.