Application of a high-density microelectrode array assay using a 3D human iPSC-derived brain microphysiological...

Unraveling the associations between human exposure to environmental chemicals and potential neurotoxicity presents significant challenges. Evaluation of neurotoxicity potential using animal testing is resource-intensive (financial, labor, and animal use) and faces uncertainties regarding biological relevance to human health outcomes. Therefore, there is a need to develop efficient and human-relevant in vitro new approach methodologies (NAMs) to screen and evaluate chemicals for neurotoxicity potential. Recording of neural network activity using microelectrode array (MEA) technology has been identified as a reliable and reproducible method for evaluating neurotoxicity. Much of this research has been performed in 2D rodent-derived cell models. The 'BrainSpheres MEA assay' described in this study offers a promising functional human induced pluripotent stem cell (iPSC)-derived 3D brain model comprising neurons, astrocytes, and oligodendrocytes. We demonstrate consistent spontaneous neuronal firing and network bursting parameters from 7-week-old BrainSpheres using a high-density MEA technology. The performance of this model as a human-relevant NAM was evaluated by conducting a multi-concentration, 13 day exposure study with a set of ten chemicals. Neural activity metrics were assessed and compared to results from a 2D-MEA assay using rodent cells. Loperamide and domoic acid (two assay positive controls) demonstrated similar bioactivity profiles in the BrainSphere MEA assay to the 2D-MEA assay, while acetaminophen (assay negative control) was inactive in both assays. The 2D-MEA model demonstrated more potent bioactivity for 4/7 chemicals that were active in both assays. In the future, reducing replicate variability and testing a larger set of chemicals will likely improve the accuracy and reliability of the assay. These preliminary findings suggest that the BrainSphere assay could be used alongside the rat network formation assay (rNFA) as part of a tiered strategy, where hits in the rNFA are confirmed and further characterized in the BrainSphere model, helping move toward animal-free toxicological testing.