Combinatorial BMP4 and Activin direct choice between alternate routes to endoderm during human gastrulation [RN25025_RNAseq]

Cellular differentiation requires the proper interpretation of multiple signalling cues which vary in concentration. How exactly the combination and history of signals a cell is exposed to influences a fate decision remains poorly understood. In this study we use hESCs as a tractable model system to explore how combinations of cues guide state transitions during gastrulation. Using single-cell transcriptomics and live-cell imaging of engineered hESCs expressing endogenous cell state reporters, we reconstructed developmental lineages and obtained single-cell measurements of fate specification dynamics during gastrulation. We found that definitive endoderm, one of the three germ layers, arises from two distinct developmental trajectories: a direct route from pluripotency, and an indirect route via a mesoderm progenitor state. Furthermore, by modulating the signalling input we found that the relative concentration of Activin and BMP4 controls the choice between alternate trajectories to endoderm. These findings reveal a lineage convergence event during human gastrulation with multiple routes existing to definitive endoderm dictated by the combination of signalling cues presented. This work shows that the combination cues a cell is exposed to not only directs the final fate outcome it assumes, but the developmental route taken. Overall design: The goal of this study was to characterise the transcriptional differences between definitive endoderm cells derived from two distinct developmental trajectories. This was achieved by ultilising a H1 hESCs (human embryonic stem cells) engineered with a FOXC1 Achilles transcriptional reporter. After 72 hours of differentiation in response to Activin 100 ng/mL and BMP4 10 ng/mL hESCs were fixed with 4% PFA, stained for SOX17 and sorted into four disitnct populations: SOX17+ Achilles+, SOX17+ Achilles-, SOX17- Achilles+ and SOX17+ Achilles+ and RNA extracted for sequencing. Experiments were performed in triplicate. Control cells were maintained in mTeSR1.