Cell mechanics-based modelling and in toto imaging describe a new mode of lumen formation in chicken neural tube

Lumen formation creates epithelial tubes and vesicles that underlie the morphology and function of many major organs, such as ear, brain, and lung. The specific morphogenetic pathways taken may vary between organs and species, but the resultant single lumen remains constant. In chick neurulation, lumen formation is thought to employ two disparate processes that happen sequentially, as the neural tube forms from anterior to posterior. Anteriorly, and early in neurulation, the neural tube closes through primary neurulation, where an already specified neural plate folds into a tube shape that seals. Posteriorly and later in development, the neural tube forms through secondary neurulation where the caudal cell mass forms multiple de novo lumens that then merge to form a central lumen. Second neurulation is also thought to be present in several other species, such as mouse and human, whilst zebrafish takes a different approach of lumen opening that shows differences from amniote primary or secondary. How the anterior and posterior neural tubes integrate to form a complete organ containing a single, continuous lumen is still contested in the field, with alternate theories on the junctional region being proposed. Those against the theory of a connecting region suggest a hard boundary where anterior and posterior lumens later fuse. Others speculate a combined behaviour of neural plate thickening and cellular invagination allowing a graded transition from folding to cavitation. The challenge in solving this process is partly due to the lack of live imaging access to the cell dynamics of the region, and more importantly an insufficient understanding of the cellular interactions creating the lumen, particularly for cavitation-based processes such as secondary neurulation. In this work we developed a computational model using simple cell-cell mechanical interactions as foundations, to simulate lumen formation as an emergent property of cell differentiation (epithelialization) in the context of various tissue geometries such as plates and p. The aim of this model is to recapitulate lumen formation dynamics and test if a cohesive underlying set of cell behaviours are sufficient to tie primary and secondary dynamics together across the contested junctional region. The model strategy is agent(cell)-based collective energy minimisation and the cell behaviours are motivated by known processes involved in neurulation, including epithelialization, cell to cell adhesion, apical supracellular actin tension and cellular cortical tension. Our simple 2D model is capable of reproducing primary and secondary neurulation behaviours, and interestingly, both under some shared parameter regimes. The model also unexpectedly produces a third intermediate behaviour, where the field of cells form an epithelial roof from one end that then extends to generate a single central lumen by gradually driving out inner cells. To compare the modelling results with the real tissue, I optimised a clearing based light-sheet imaging technique and captured cellular resolution in toto datasets of the chick neural tube. These data allow a close examination of tissue geometry and cell organization across the junctional region, enabling a pseudo-time reconstruction of lumen formation. I find no evidence of the invaginating cellular phenotype proposed as a mechanism of junctional neurulation at Hamburger–Hamilton stage 8 (HH8), but instead evidence of primary folding and closure up to HH11. Strikingly, the neural tube tissue subsequently transitions to a distinct lumenogenesis process that forms a dorsal epithelial surface first and a single central lumen that enlarges ventrally from HH12-14. This intermediate behaviour matches our model prediction and confirms a separate de novo cavitation behaviour to the classically described primary and secondary neurulation in chick. Our work thus suggests that apparently distinct modes of lumen formation can have common underlying cellular interactions, that play out differently and transition flexibly under spatial-temporally patterned cell fields. Ongoing perturbation experiments aim at generating transitions between these modes predicted by our model in the chicken embryo. Our work potentially offers a unified mechanical model of lumen formation in a highly diverse process that plays a key role in animal development.

Name dc:type.qualificationname

Doctor of Philosophy (PhD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

University of Cambridge

Year dc:date.issued

2024

Author dc:creator

• Moon, Lauren

Advisor dc:contributor.advisor

• Xiong, Fengzhu

• Developmental biology
• Neurulation

Licence

• https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/0360dfee-e260-465f-ad2e-186119f70f8d/download
• http://purl.org/NET/rdflicense/allrightsreserved

Author Identifier

0000-0003-3075-5968

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/386294