Human blastocysts are comprised of the first three cell lineages of the embryo: trophectoderm, epiblast and primitive endoderm, all of which are essential for early development and organ formation. However, due to ethical concerns and restricted access to human blastocysts, a comprehensive understanding of early human embryogenesis is still lacking. To bridge this knowledge gap, a reliable model system that recapitulates early stages of human embryogenesis is needed. Here we developed a three-dimensional (3D), two-step induction protocol for generating blastocyst-like structures (EPS-blastoids) from human extended pluripotent stem (EPS) cells. Morphological and single-cell transcriptomic analyses revealed that EPS-blastoids contain key cell lineages and are transcriptionally similar to human blastocysts. Furthermore, EPS-blastoids are similar with human embryos that were cultured for 8 or 10 days in vitro, in terms of embryonic structures, cell lineages and transcriptomic profiles. In conclusion, we developed a scalable system to mimic human blastocyst development, which can potentially facilitate the study of early implantation failure that induced by developmental defects at early stage.

No abstract

Primed and naive human embryonic stem cells (hESCs) do not fully recapitulate the X chromosome status observed in human preimplantation epiblast and fail to initiate random X chromosome inactivation (XCI) upon differentiation. Therefore, an ideal system for studying XCI during early human development is yet to be established. We show that incomplete blocking of autocrine fibroblast growth factor 2 (FGF2) signaling in naive hESCs drives significant heterogeneity in X chromosome and pluripotency status. We derived homozygous XaXa naive hESCs with dual allelic XIST expression and high levels of TFCP2L1, whose transcriptome and X chromosome states are similar to human preimplantation epiblast. Random XCI was initiated upon naive-to-primed conversion of these cells, and both pre-and post-XCI primed hESCs were obtained. We observed random XCI in all cells upon further differentiation of pre-XCI primed hESCs. Together, these findings enable derivation of homogeneous naive hESCs and establish a powerful platform to study human XCI.