During gastrulation, epiblast cells are pluripotent and their fate is thought to be constrained principally by their position. Cell fate is progressively restricted by localised signalling cues from areas including the primitive streak. However, it is unknown whether this restriction accompanies, at the individual cell level, a reduction in potency. Investigation of these early transition events in vitro is possible via the use of epiblast stem cells (EpiSCs), self-renewing pluripotent cell lines equivalent to the postimplantation epiblast. Strikingly, mouse EpiSCs express gastrulation stage regional markers in self-renewing conditions. Here, we examined the differentiation potential of cells expressing such lineage markers. We show that undifferentiated EpiSC cultures contain a major subfraction of cells with reversible early primitive streak characteristics, which is mutually exclusive to a neural-like fraction. Using in vitro differentiation assays and embryo grafting we demonstrate that primitive streak-like EpiSCs are biased towards mesoderm and endoderm fates while retaining pluripotency. The acquisition of primitive streak characteristics by self-renewing EpiSCs is mediated by endogenous Wnt signalling. Elevation of Wnt activity promotes restriction towards primitive streak-associated lineages with mesendodermal and neuromesodermal characteristics. Collectively, our data suggest that EpiSC pluripotency encompasses a range of reversible lineage-biased states reflecting the birth of pioneer lineage precursors from a pool of uncommitted EpiSCs similar to the earliest cell fate restriction events taking place in the gastrula stage epiblast.

No abstract

Transcription factor (TF) networks are thought to regulate embryonic stem cell (ESC) pluripotency. However, TF expression dynamics and regulatory mechanisms are poorly understood. We use reporter mouse ESC lines allowing non-invasive quantification of Nanog or Oct4 protein levels and continuous long-term single-cell tracking and quantification over many generations to reveal diverse TF protein expression dynamics. For cells with low Nanog expression, we identified two distinct colony types: one re-expressed Nanog in a mosaic pattern, and the other did not re-express Nanog over many generations. Although both expressed pluripotency markers, they exhibited differences in their TF protein correlation networks and differentiation propensities. Sister cell analysis revealed that differences in Nanog levels are not necessarily accompanied by differences in the expression of other pluripotency factors. Thus, regulatory interactions of pluripotency TFs are less stringently implemented in individual self-renewing ESCs than assumed at present.

Continuous analysis of single cells, over several cell divisions and for up to weeks at a time, is crucial to deciphering rare, dynamic and heterogeneous cell responses, which would otherwise be missed by population or single-cell snapshot analysis. Although the field of long-term single-cell imaging, tracking and analysis is constantly advancing, several technical challenges continue to hinder wider implementation of this important approach. This is a particular problem for mammalian cells, where in vitro observation usually remains the only possible option for uninterrupted long-term, single-cell observation. Efforts must focus not only on identifying and maintaining culture conditions that support normal cellular behavior while allowing high-resolution imaging over time, but also on developing computational methods that enable semiautomatic analysis of the data. Solutions in microscopy hard- and software, computer vision and specialized theoretical methods for analysis of dynamic single-cell data will enable important discoveries in biology and beyond.