Three-dimensional differentiation of human pluripotent stem cell-derived neural precursor cells using tailored porous polymer scaffolds

This study investigates the utility of a tailored poly(ethylene glycol) diacrylate-crosslinked porous polymeric tissue engineering scaffold, with mechanical properties specifically optimised to be comparable to that of mammalian brain tissue for 3D human neural cell culture. Results obtained here demonstrate the attachment, proliferation and terminal differentiation of both human induced pluripotent stem cell- and embryonic stem cell-derived neural precursor cells (hPSC-NPCs) throughout the interconnected porous network within laminin-coated scaffolds. Phenotypic data and functional analyses are presented demonstrating that this material supports terminal in vitro neural differentiation of hPSC-NPCs to a mixed population of viable neuronal and glial cells for periods of up to 49 days. This is evidenced by the upreguladon of TUB83, MAP2, SYP and GFAP gene expression, as well as the presence of the proteins beta III-TUBULIN, NEUN, MAP2 and GFAP Functional maturity of neural cells following 49 days 3D differentiation culture was tested via measurement of intracellular calcium. These analyses revealed spontaneously active, synchronous and rhythmic calcium flux, as well as response to the neurotransmitter glutamate. This tailored construct has potential application as an improved in vitro human neurogenesis model with utility in platform drug discovery programs. Statement of significance The interconnected porosity of polyHIPE scaffolds exhibits the ability to support three-dimensional neural cell network formation due to limited resistance to cellular migration and re-organisation. The previously developed scaffold material displays mechanical properties similar to that of the mammalian brain. This research also employs the utility of pluripotent stem cell-derived neural cells which are of greater clinical relevance than primary neural cell lines. This scaffold material has future potential in better mimicking three-dimensional neural networks found in the human brain and may result in improved in vitro models for disease modelling and drug screening applications. (C) 2019 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.