Science Inventory

Engineering a Computable Epiblast for in silico Modeling of Developmental Toxicity

Citation:

Knudsen, T., K. Barham, Nancy C. Baker, K. Carstens, AND R. Spencer. Engineering a Computable Epiblast for in silico Modeling of Developmental Toxicity. 12th World Congress on Alternatives and Animal Use in the Life Sciences, Niagra, Ontario, CANADA, August 27 - 31, 2023. https://doi.org/10.23645/epacomptox.24291223

Impact/Purpose:

This is an invited presentation to the 12th World Congress on Alternatives and Animal Use in the Life Sciences meeting August 2023 (WC12) symposium on 'The Future of Multi-Scale Modleing and Simulation in Human Disease and Toxicology'.

Description:

The developmental potential of human pluripotent stem cells (hPSCs) in culture closely resembles the ‘epiblast’ during gastrulation. ToxCast provides in vitro bioactivity data on over 1000 chemicals from a hPSC assay that predicts developmental toxicity with ~80% balanced accuracy [Zurlinden et al. 2020]. Computer modeling of the epiblast in 3D would parallel the utilization of ToxCast data to track cellular trajectories during simulated chemical exposure. We engineered a fully computable model of the epiblast using the CompuCell3D that simulates primitive streak formation, epithelial-mesenchymal transition of epiblast cells, and self-organization of mesodermal domains (chordamesoderm, paraxial, lateral plate, posterior/extraembryonic). Determination of progenitor cell fate is dependent upon positional information and temporal colinearity of a HOX clock regulated by a control network of morphogenetic signals (FGF, BMP, NODAL, ATRA, CDX). Executing the model renders a quantitative cell-level computation for mechanistic evaluation of mesodermal subpopulations. Consequences of perturbation were shown, for example, on posterior mesoderm cell mass that gives rise to most hemangiogenic precursors in the yolk sac blood islands. Interfering with the signaling network produces effects mirroring those reported in experimental mouse embryology, with 50% reductions in both FGF4 and BMP4 signaling resulting in 88% and 63% reductions, respectively, in the posterior mesodermal population. This cell agent-based model integrates signaling cascades, ToxCast chemical bioactivity data and known embryology to mechanistically predict altered phenotypes through the resulting mesodermal topography in the animal-free zone. Disclaimer: This abstract does not necessarily reflect USEPA policy.