Dynamic 3D cell rearrangements guided by a fibronectin matrix underlie somitogenesis

Somites are transient segments formed in a rostro-caudal progression during vertebrate development. In chick embryos, segmentation of a new pair of somites occurs every 90 minutes and involves a mesenchyme-to-epithelium transition of cells from the presomitic mesoderm. Little is known about the cellular rearrangements involved, and, although it is known that the fibronectin extracellular matrix is required, its actual role remains elusive. Using 3D and 4D imaging of somite formation we discovered that somitogenesis consists of a complex choreography of individual cell movements. Epithelialization starts medially with the formation of a transient epithelium of cuboidal cells, followed by cell elongation and reorganization into a pseudostratified epithelium of spindle-shaped epitheloid cells. Mesenchymal cells are then recruited to this medial epithelium through accretion, a phenomenon that spreads to all sides, except the lateral side of the forming somite, which epithelializes by cell elongation and intercalation. Surprisingly, an important contribution to the somite epithelium also comes from the continuous egression of mesenchymal cells from the core into the epithelium via its apical side. Inhibition of fibronectin matrix assembly first slows down the rate, and then halts somite formation, without affecting pseudopodial activity or cell body movements. Rather, cell elongation, centripetal alignment, N-cadherin polarization and egression are impaired, showing that the fibronectin matrix plays a role in polarizing and guiding the exploratory behavior of somitic cells. To our knowledge, this is the first 4D in vivo recording of a full mesenchyme-to-epithelium transition. This approach brought new insights into this event and highlighted the importance of the extracellular matrix as a guiding cue during morphogenesis.This work was supported by the Portuguese Fundação o para a Ciência e a Tecnologia (FCT) and the POCI 2010 program (III Community SupportFramework)/FEDER through project POCI/BIA-BCM/59201/2004 as well as by the Network of Excellence ‘‘Cells into Organs’’(EU/FP6). GGM was supported througha fellowship from FCT (SFRH/BPD/18907/2004), PR was supported by a PhD grant from FCT (SFRH/BD/37423/2007) and RA was supported with IEFP and FCTgrants. The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript