In vitro teratogenicity testing using a 3D, embryo-like gastruloid system

Pharmaceuticals intended for use in patients of childbearing potential need to be tested for teratogenicity before marketing. Several pharmaceutical companies use animal-free in vitro models which allow a more rapid selection of lead compounds and contribute to 3Rs principles ('replace, reduce and refine') by streamlining the selection of promising compounds submitted to further regulatory studies in animals. Currently available in vitro models typically rely on adherent monolayer cultures or disorganized 3D structures, both of which lack the spatiotemporal and morphological context of the developing embryo. A newly developed 3D 'gastruloid' model has the potential to achieve a more reliable prediction of teratogenicity by providing a robust recapitulation of gastrulation-like events alongside morphological coordination at relatively high-throughput. In this first proof-of-concept study, we used both mouse and human gastruloids to examine a panel of seven reference compounds, with associated in vivo data and known teratogenic risk, to quantitatively assess in vitro teratogenicity. We observed several gross morphological effects, including significantly reduced elongation or decreased size of the gastruloids, upon exposure to several of the reference compounds. We also observed aberrant gene expression using fluorescent reporters, including SOX2, BRA, and SOX17, suggestive of multi-lineage differentiation defects and disrupted axial patterning. Finally, we saw that gastruloids recapitulated some of the known in vivo species-specific susceptibilities between their mouse and human counterparts. We therefore suggest that gastruloids represent a powerful tool for teratogenicity assessment by enabling relevant physiological recapitulation of early embryonic development, demonstrating their use as a novel in vitro teratogenic model system.This work was supported by grants from Medical Research Council (MRC) (MR/R017190/1 and MR/V005367/1) to A.M.A. and N.M., a Leverhulme Trust grant (RPG-2018-356) to AMA, as well as a Gates Cambridge Scholarship awarded to V.M.. The mouse part was supported by an ERC Advanced investigator grant (AdG 834580) to A.M.A. supporting P.B.B. Data derived from human iPSC in vitro model based compound assessment has been supported by the BMBF, EFSA, and the DK-EPA (MST-667-00205) and received funding from the European Union's Horizon 2020 research and innovation program under grant agreements No. 681002 (EU-ToxRisk) and No. 825759 (ENDpoiNTs