Back

Reconstructing stochastic cell population trajectories reveals regulators and heterogeneity of endothelial flow-migration coupling driving vascular remodelling

Giese, W.; Rosa, A.; Baumann, E.; Oppenheim, O.; Akmeric, E. B.; Andrade, S.; Hollfinger, I.; Alt, S.; Gerhardt, H.

2023-05-30 developmental biology
10.1101/2023.05.30.542799 bioRxiv
Show abstract

Emerging concepts of developmental vascular remodelling recently identified that selectively labelled sprouting tip cells and/or venous endothelial cells (ECs) accumulate in developing arteries, suggesting directional migration of specific ECs drives artery formation. However, a general population analysis and detailed quantitative investigation of migratory mechanisms is so far lacking. Here, we developed a universally applicable quantitative approach and a computational model allowing to track and simulate stochastically labelled EC populations irrespective of labelling density and origin. Dynamic mapping of EC distributions in a bespoke coordinate system revealed how ECs move during the most active remodelling phases in the mouse retina. Simulation and parameter sensitivity analysis illustrated that the population shift from veins to arteries cannot be explained by random walk, but best fits to a tuneable dual force-field between shear-force induced directionality against blood flow, and hypoxia mediated VEGFA-gradients. High migration rates require only weak flow-migration coupling, whereas low migration rates require strong coupling to the flow direction. Functional analysis identified Cdc42 as the critical mediator of overall population movement from veins to arteries, yet with surprising heterogeneity suggesting the existence of distinct cell populations. This new quantitative understanding will enable future tailored intervention and tuning of the remodelling process.

Matching journals

The top 2 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.