Neural constructs for predicting neural toxicity
2015
Morgridge Institute for Research, Madison, USA
Human pluripotent stem cell-based in vitro models that mirror human physiology have the potential to cost-effectively assess the developmental neurotoxicity of chemicals. Here, human embryonic stem (ES)-derived neural progenitor cells, endothelial cells, mesenchymal stem cells and microglia/macrophage progenitors were combined on synthetic hydrogels and cultured in a serum-free medium to model cellular interactions in the developing brain. The progenitor cells self-assembled into 3D neuronal constructs with distinct populations of neurons and glia, interconnected vascular networks and branching microglia. The replicate constructs were reproducible by RNA sequencing (RNA-Seq) and expressed genes for neurogenesis, vascular development and microglia. Using machine learning, a predictive model was built from these RNA-Seq for the neuronal constructs exposed to a training set of 60 toxic and non-toxic chemicals and then predicted in a blind trial with a set of 10 additional compounds. The model correctly classified 9 of the 10 additional chemicals. This combined strategy demonstrates the value of cell-based assays for predictive toxicology and should be useful for assessing the safety of both drugs and chemicals.
Human pluripotent stem cell-derived neural constructs for predicting neural toxicity
James A. Thomson
Added on: 11-09-2021
[1] https://www.pnas.org/content/112/40/12516
