Back

Rapid and efficient generation of human 8-cell-like cells for embryo modelling

Odabas, A.; Unlu, S.; Ozturk, E.; Karasurmeli, N.; Hu, K.; Leleu, M.; Aztekin, C.; Onder, T. T.

2026-08-19 developmental biology
10.64898/2026.08.18.745473 bioRxiv
Show abstract

8-cell blastomeres of human embryos possess broad lineage potential and undergo major zygotic genome activation (ZGA), yet experimental access to this transient cell state remains limited. Rare 8-cell-like cells (8CLCs) arise spontaneously in naive pluripotent stem cell cultures, but their low abundance has constrained mechanistic and functional studies. Here, we develop a chemically defined strategy for rapid and robust induction of 8CLCs. Through sequential small-molecule screens focused on chromatin regulators, we identify five compounds acting through distinct pathways that generate up to 40% 8CLCs within 48 hours. The resulting cells, which we term rapidly induced 8CLCs (ri8CLCs), recapitulate key molecular features of 8-cell blastomeres, including induction of ZGA-associated genes, cleavage-stage transposable elements, and 8-cell-stage transcriptional signatures in bulk and single-cell transcriptomic analyses. Functionally, ri8CLCs exhibit enhanced developmental competence, acquiring the ability for spontaneous extraembryonic differentiation and assembly into well-cavitated blastoids on an accelerated 72-hour timeline. Notably, ri8CLC induction enables blastoid formation even in the absence of MEK inhibition, TGF-{beta}/Activin/Nodal inhibition and exogenous LIF, revealing a developmental competence consistent with an early embryonic state. Together, these findings establish a rapid, defined, and highly efficient platform for generating human ri8CLCs and provide a tractable model for studying early human embryogenesis.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.