You are here:
Computational Approach using Mouse Embryonic Stem Cells to Define a Mechanistic Applicability Domain for Prenatal Developmental Toxicity
Citation:
Hunter, S., R. Judson, Mitch Rosen, N. Baker, AND T. Knudsen. Computational Approach using Mouse Embryonic Stem Cells to Define a Mechanistic Applicability Domain for Prenatal Developmental Toxicity. Society of Toxicology, New Orleans, LA, March 13 - 17, 2016.
Impact/Purpose:
Identification of mechanisms responsible for adverse developmental effects is the first step in creating predictive toxicity models. Identification of putative mechanisms was performed by co-analyzing three datasets for the effects of ToxCast phase Ia and II chemicals: 1.In vitro cellular and molecular effects used ToxCast GeneScore; 2.In vivo phenotypic effects used ToxRefDB; 3.our data of effects on mouse Embryonic Stem Cell (mESC) differentiation and cytotoxicity at gastrulation (D4) or cardiomyocyte (D9) stages. Putative mechanisms of in vivo phenotypes are as diverse as the organ systems affected by chemical exposure. mESC respond to many but not all of the genes associated with abnormal development establishing a biological applicability domain for their use in predictive toxicity models. This Abstract does represent EPA policy.
Description:
Identification of mechanisms responsible for adverse developmental effects is the first step in creating predictive toxicity models. Identification of putative mechanisms was performed by co-analyzing three datasets for the effects of ToxCast phase Ia and II chemicals: 1.In vitro cellular and molecular effects used ToxCast GeneScore; 2.In vivo phenotypic effects used ToxRefDB; 3.our data of effects on mouse Embryonic Stem Cell (mESC) differentiation and cytotoxicity at gastrulation (D4) or cardiomyocyte (D9) stages. In vitro ToxCast-Gene associations for in vivo and mESC endpoints were determined. When the gene-activity was overrepresented (>2-fold) in the chemicals that produced an endpoint, that gene was designated as a putative mechanism. There are 19 in vivo developmental phenotypes that range from 22-107 genes. >60% of genes associated with jaw hypoplasia, ureter, orofacial clefts, appendicular skeletal defect genes were also correlated with other phenotypes. <26% of genes renal, eye, splanchnic viscera, body wall, heart, major vessel defect genes correlated with other phenotypes. 75 genes were common to 8 phenotypic endpoints including FGFR1, FLT4, GSK3B, IGF1R, KDR, RXRA. Genes for D9_Cardiomyocyte differentiation showed the best overlap with in vivo phenotype genes for a single mESC endpoint. The combination of D4_Cytotox and D9_Cardiomyocyte correlated to 75-89% of genes for 13/19 phenotypes. 31 genes correlate to heart defects in vivo; 27 (87%) affected mESC. 13 GPCR-pathway genes, 2 CYPs, RXRA, H2AFX, TP53 are common between these endpoints. Putative mechanisms of in vivo phenotypes are as diverse as the organ systems affected by chemical exposure. mESC respond to many but not all of the genes associated with abnormal development establishing a biological applicability domain for their use in predictive toxicity models. This Abstract does represent EPA policy.