Search

SUMMARY3D spheroids have emerged as powerful drug discovery tools given their high-throughput screening (HTS) compatibility. Here, we established a method for generating functional neural spheroids with differentiated human induced pluripotent stem cell (hiPSC)-derived neurons and astrocytes at cell type compositions mimicking specific regions of the human brain. Recordings of intracellular calcium oscillations were used as functional assays, and the utility of this spheroids system was shown through disease modelling, drug testing, and formation of assembloids to model neurocircuitry. We developed disease models for Alzheimer’s and Parkinson’s Disease through incorporating genetically engineered cells into spheroids, and Opioid Use Disorder by chronically treating spheroids with DAMGO. We reversed baseline functional deficits in each disease model with clinically approved treatments and showed that assembloid activity can be chemogenetically manipulated. Here, we lay the groundwork for brain region-specific spheroids as a robust functional assay platform for HTS studies.

Three-dimensional (3D) organotypic models that capture native-like physiological features of tissues are being pursued as clinically predictive assays for therapeutics development. A range of 3D neural organotypic models are being developed to mimic brain morphology, physiology, and pathology of neurological disease. Biofabrication of 3D gel-based cellular systems is emerging as a versatile technology to produce spatially and cell-type tailored, physiologically complex and native-like tissue models. Here we produce 3D fibrin gel-based functional neural co-culture models with human iPSC differentiated dopaminergic or glutamatergic neurons and astrocytes. We further introduce genetically encoded fluorescence biosensors for real time functional measurements of intracellular calcium and levels of dopamine and glutamate neurotransmitters, in a high throughput compatible plate format. We use optogenetic activation and pharmacological perturbations to demonstrate that functional responses in 3D fibrin gel neural models are as relevant to be expected from in vivo data, compared to the equivalent 2D neural models.