You are here:
Engineering a Computable Epiblast for in silico Gastrulation and Predictive Modeling of Developmental Toxicity with in vitro Data from the ToxCast Stem Cell Assay
Citation:
Barham, K., R. Spencer, AND T. Knudsen. Engineering a Computable Epiblast for in silico Gastrulation and Predictive Modeling of Developmental Toxicity with in vitro Data from the ToxCast Stem Cell Assay. FutureTox-5: New Technologies to Evaluate Organ-Specific Effects of Drugs and Chemicals, Chapel Hill, NC, May 10 - 11, 2022. https://doi.org/10.23645/epacomptox.19940867
Impact/Purpose:
Poster presented to the FutureTox-5: New Technologies to Evaluate Organ-Specific Effects of Drugs and Chemicals Meeting May 2022 (Contemporary Concepts in Toxicology (CCT) series of the Society of Toxicology). The U.S. EPA is working to develop New Approach Methods (NAMs) to study chemical effects on prenatal development as an alternative to animal testing. One area of interest is virtual tissue modeling in which developmental events are recapitulated in silico. A human pluripotent stem cell (hPSC) assay on over 1,000 chemicals has been utilized to generate the ToxCast portfolio, which can predict developmental toxicity with ~80& balanced accuracy. The developmental potential of hPSCs is closely reflected by the mammalian ‘epiblast’ due to its differentiation into the primary germ layers (endoderm, mesoderm, ectoderm) during gastrulation. An in silico epiblast encompassing cellular dynamics during gastrulation provides a platform for predictive toxicology through dynamical modeling and simulation of developmental phenotypes.
Description:
New Approach Methods (NAMs) for assessing chemical effects on prenatal development with less reliance on animal testing is an important challenge for predictive toxicology at the USEPA [https://www.epa.gov/chemical-research/comptox-chemicals-dashboard]. The ToxCast portfolio provides a vast resource of in vitro data for in silico modeling of developmental toxicity and includes data on over 1000 chemicals in a human pluripotent stem cell (hPSC) assay that predicts developmental toxicity with ~80% balanced accuracy [Zurlinden et al. 2020]. The mammalian epiblast is the in vivo embryonic structure that is most closely reflected by the developmental potential of hPSCs in vitro; therefore, a computer model that recapitulates cellular dynamics of the epiblast in silico provides a means to track normal versus adverse developmental trajectories during gastrulation – a critical process during which the embryonic body plan is decoded from the genome. We engineered a fully computable epiblast model in the www.compucell3d.org modeling environment that recapitulates development of the primitive streak and subsequent epithelial-mesenchymal transition from epiblast stem cells. The model self-organizes progenitors destined for different mesodermal domains (chordamesoderm, paraxial, lateral plate, extraembryonic) formed during mouse (E5.5 – E6.5) and human (3rd week) gestation. Cellular dynamics in the in silico epiblast provides a quasi-normal simulation of gastrulation under the control of various morphogenetic signals (e.g., WNT, FGF, BMP, Nodal) and can be tweaked with bioactivity data from the ToxCast portfolio. This agent-based model uniquely integrates chemical bioactivity data with biological knowledge of the major cell signaling networks and developmental processes driving gastrulation for quantitative mechanistic prediction of altered phenotypes. Case studies are underway using chemical effects data from the ToxCast_STM portfolio to render virtual screening of complex interactions between genetic and/or environmental stressors. This abstract does not necessarily reflect Agency policy.
URLs/Downloads:
DOI: Engineering a Computable Epiblast for in silico Gastrulation and Predictive Modeling of Developmental Toxicity with in vitro Data from the ToxCast Stem Cell Assay
FUTURETOXV POSTER_KAITLYN.PDF (PDF, NA pp, 1475.93 KB, about PDF)