Transcriptional initiation and mechanically driven self-propagation of a tissue contractile wave during axis elongation

Anais Bailles, Claudio Collinet, Jean-Marc Philippe, Pierre-François Lenne, Edwin Munro, Thomas Lecuit

Preprint posted on September 29, 2018

A mechanical relay mechanism drives a wave of polarized constriction during Drosophila endoderm morphogenesis.

Selected by Sundar Naganathan


Embryonic tissues undergo intricate reorganizations and deformations that ultimately give rise to a specific shape and form. In many contexts, tissue-scale reorganization is driven by spatiotemporally regulated actomyosin activation and dynamics. For example, myosin activation through Rho1 in the presumptive mesoderm cells of a Drosophila embryo drives apical constriction, leading to tissue bending and furrow formation. A similar process, i.e. Rho1 and Myosin II dependent apical constriction, drives invagination of the posterior endoderm in Drosophila embryos. Interestingly, following this initial invagination in the posterior, the furrow continues invaginating in a wave-like fashion and moves in a polarized fashion towards the anterior. What are the mechanisms by which this polarized furrow movement occurs? The authors uncover a mechanical coupling pathway, where cell shape changes triggered by the initial invagination leads to a wave of myosin activation and associated constriction.

Key findings

The authors performed time-lapse imaging of the invaginating posterior endoderm and quantified the dynamics of apical constriction. In the initial phase of constriction, myosin as well as its upstream regulators, Rho1 and Rok, were found to be uniformly activated across a group of cells in the posterior called the primordium region by regulated transcription of fog, a GPCR ligand. In the second phase, a wave of Rho1-MyosinII activation and associated constriction propagated towards the anterior in a domain that comprised about 8 rows of cells, called the propagation zone.

To what extent is fog transcription involved in myosin activation in the propagation zone? Through live analysis of fog transcripts and by injection of alpha-aminitin, a potent inhibitor of RNA polymerase II, the authors demonstrate that myosin wave propagation does not require gene transcription in the propagation zone. Moreover, diffusion of fog away from the primordium towards the propagation zone also does not play a role, as overexpression of fog in the primordium did not change the wave dynamics in the propagation region. Thus, patterned Fog signaling does not determine the dynamics of Myosin II wave propagation.

Given that invagination involves extensive cell shape changes, can the wave propagate through mechanical feedback with the neighboring cells? By performing a global inhibition of Rok in the embryo, the authors show that anteriorward movement of constriction is mechanically driven. No change in Rho1 activation in the primordium was observed, however, Rok global inhibition severely perturbed Rho1 wave propagation indicating that this requires MyoII activity. The authors then perturbed the mechanical environment of the dorsal epithelium by stitching dorsal epithelial cells that are 30 cell rows away from the primordium to the vitelline membrane or by using dorsalizing mutants. In both cases, cells recruited higher levels of myosin, however, wave propagation significantly slowed down suggesting higher resistance to the invagination. Taken together, wave propagation is dependent on mechanical changes to the tissue influenced by the invaginating epithelium. The authors finally quantified 3D cell deformations during wave propagation and show the existence of a mechanical relay mechanism where neighboring rows of cells sequentially deform that ultimately leads to a wave of myosin activation and constriction.

Why I chose this preprint?

An increasingly number of articles are discussing the importance of mechanical coupling across cells in a tissue and across tissues in an embryo that ultimately drives morphogenesis. The highlighted article is yet another example of mechanical coupling that clearly demonstrates how local activation of myosin through controlled transcription factor expression, leads to a wave of mechanical changes in cells tens of micrometers away from the activated region. This mechanical coupling is important for the resultant self-organization that shapes the tissue into a specific form.

Exciting times are ahead for developmental biophysics, where such quantitative analyses will enable intricate theoretical models to be developed that can explain morphogenesis across length and time scales.

 Open questions

  1. Different tissues undergo diverse transformations and end up in quite different shapes and forms during embryogenesis. Interestingly, the basic constituents, actin, myosin and upstream activators such as Rho remain the same across developmental contexts. How does tissue shape diversity emerge from the same building blocks? How different are the mechanical properties of the surrounding tissues in different contexts? Could a change in the properties of the immediate environment explain the different dynamics observed across contexts?
  2. The authors propose that sequential 3D cell deformations lead to a wave of myosin activation and constriction. It is not clear what exactly the authors mean by deformation. Do the authors observe 3D cell compression or just a change in cell shape with no change in volume? Also, how does cell deformation activate myosin?
  3. Why does the tissue undergo sequential constriction? What is the advantage of sequential constriction when compared to global constriction?
  4. Does attachment to the vitelline membrane provide some traction to the constricting tissue? Can the material properties of the vitelline membrane be specifically targeted and modified?
  5. Can the authors predict what happens if myosin is activated prematurely in the propagation zone building some prestress in the tissue before activation of myosin in the primordium?


  1. Martin AC and Goldstein B, Apical constriction: themes and variations on a cellular mechanism driving morphogenesis, Development, 2014.

Tags: actomyosin, contractility, fly, morphogenesis, self-organization, tension

Posted on: 2nd November 2018 , updated on: 8th November 2018

Read preprint (1 votes)

  • Author's response

    Claudio Collinet shared

    We are delighted of seeing our work highlighted here as a preLights. We share Sundar’s view on the importance of mechanics in relying information across cells and tissues during development. We believe that our results reveal that beyond genetic control remarkable spatial patterns and dynamics also emerge through mechanical control and self-organization.

    Regulation of actomyosin contractions have been shown to play key roles in generating specific patterns of forces involved in morphogenesis in many organisms, however, clearly the final tissue dynamics also critically depends on how forces are resisted by surrounding cells and tissues. Recent work highlighted that cells and tissues actively regulate their mechanical properties thereby impacting on tissue dynamics in different contexts.

    In our case we observe a sequential cycle of 3D cell deformations that is associated with the wave of MyoII activation and cell invagination. We have not carefully quantified cell volume but we believe that this does not change during the process. Our data indicate that cells are subjected to compressive forces exerted by the moving invaginating furrow and are deformed just like water balloons under compression. The whole process requires constant MyoII activity to keep going, however, it is not yet clear how these events (compressions and deformations) are directly linked to MyoII activation in cells of the propagation zone. One possibility is that adhesion/contact with the overlaying vitelline membrane might be implicated in this. Interestingly, the contact/adhesion with the vitelline membrane might as well provide traction to the constricting tissue such to induce and sustain the polarized movement of the invaginating furrow. In this scenario sequential cell deformation, in our case a spreading of the apico-lateral cortex onto the overlying vitelline membrane, followed by constriction and detachment would effectively allow the tissue to “walk” on substrate and produce the observed polarized movement. This would not be possible if all cells constricted at the same time.

    Finally, it is hard to predict what would happen if MyoII was activated prematurely in the propagation zone before the primordium. One possibility would be observing a reverse wave (e.g. from anterior to posterior), however experimentally testing this is not easy. Approaches aiming at locally controlling MyoII activation (e.g. MyoII activation with optogenetic tools) may not be able to reproduce the complex cycle of cell deformation preceding MyoII activation in the propagation zone.

    Have your say

    Your email address will not be published. Required fields are marked *

    This site uses Akismet to reduce spam. Learn how your comment data is processed.

    Sign up to customise the site to your preferences and to receive alerts

    Register here