Mechanical heterogeneity along single cell-cell junctions is driven by lateral clustering of cadherins during vertebrate axis elongation

Background

Convergence-extension (CE) is an evolutionarily conserved process that elongates the body axis during embryonic development. At the cellular scale, this process occurs through a combination of mediolateral intercalation and directed cell migration¹. This polarized behaviour is set up by establishment of polarity at the molecular scale through the Wnt-PCP pathway, first discovered in Drosophila embryos². In addition to these descriptions at cellular and molecular scales, a coarse-grained approach has also been undertaken to understand large-scale tissue movements, which has enabled understanding the hydrodynamics of tissue flows³ and macroscopic tissue stiffening⁴. Despite these diverse approaches and advances, it remains unknown how the polarized molecular patterns enable the observed cellular behaviours during CE. Huebner RJ et al. tackle this problem and bridge the missing link by performing high-resolution imaging of cell junctions and cadherin clustering during CE.

Key results

The authors analysed CE in Xenopus embryos, which can be predominantly considered in terms of four-cell neighbour exchanges, where mediolaterally (ML)-aligned cell-cell junctions shorten with concomitant elongation of perpendicularly-aligned junctions in anteroposterior (AP) direction. The authors first observed that junction shortening in the ML direction was asymmetric, with one vertex (termed ‘active’) shortening significantly more than the other (termed ‘passive’) as previously observed in Drosophila embryos⁵. Comparisons with a mechanical model revealed that this asymmetric shortening is essential for CE and that the active vertex displayed a fluid-like motion, whereas the passive vertex exhibited a glass-like dynamics. Furthermore, the increased local stiffness along an active vertex, as required by the model for CE, was consistent with in vivo observations where active vertices exhibited directed motion in the ML direction with significantly lesser transverse fluctuations. Taken together, mechanical heterogeneity at sub-cellular scales at the level of individual vertices is critical for CE in embryos.

The authors next addressed the molecular mechanisms that underlie this heterogeneous mechanical signature by imaging Cdh3, a cell-cell adhesion protein. Cdh3 is known to undergo both cis– (i.e within the same cell) and trans-clustering (i.e across neighbouring cells). Strikingly, the Cdh3 cluster size was larger along active vertices compared to passive vertices and this asymmetry was not observed in non-shortening junctions. When cis-clustering was specifically targeted through point mutations in the hydrophobic pocket of Cdh3, cluster formation was affected. Importantly, under these mutant conditions, axis elongation was specifically perturbed. The authors then showed that this effect on axis elongation is due to the elimination of mechanical heterogeneity along shortening junctions, thus providing the first functional significance for Cdh3 cis-clustering in vivo. Finally, the PCP pathway was shown to act upstream of Cdh3 cis-clustering, thus providing a mechanistic link between polarity establishment and the observed heterogeneous junctional dynamics.

Why I chose this preprint

1. This is the first demonstration of asymmetric junctional shortening in vertebrate embryos. Moreover, the authors perform in-depth quantifications and compare their observations with a mechanical model to describe mechanistically the importance of mechanical heterogeneities along shortening junctions.
2. While cadherin adhesion proteins have long been known to undergo cis-clustering, this preprint provides the first functional significance of such widely observed clustering events.
3. The preprint discusses interesting observations of fluid-like and glass-like dynamics occurring concurrently in shortening junctions at sub-cellular scales. The transition between these contrasting motions have so far only been observed and described at macroscopic tissue-scales.

Open questions

1. Given a 4-cell configuration, what determines whether the left junction or the right junction exhibits active shortening? On a larger scale, how are these active and passive junctions spatially distributed in a tissue?
2. What percentage of cells in a converging tissue exhibit this heterogeneous active-passive shortening?
3. It is not clear if the quantification of Cdh3 cluster size reported in Fig. 4 exclusively represents cis-clustering or it represents a combination of cis- and trans-clustering? If it does represent cis-clustering, how were the contributions from trans-clustering removed from this analysis? Does cis-clustering also simultaneously result in an increase in trans-interactions?
4. In the model, junctions failed to shorten if the viscoelastic parameter was equal and small for both vertices. How did previous models in the CE field achieve CE even without explicitly considering heterogeneous shortening?
5. Can you explain in simple terms how despite an increased local stiffness along an active vertex compared to a passive vertex, the active vertex still exhibits more fluid-like motion?
6. Any hypothesis on how does the PCP pathway enables asymmetric junctional cis-clustering?

References

1. Keller R. et al., Mechanisms of convergence and extension by cell intercalation, Phil. Trans. Royal Society London. Series B: Biological Sciences, 2000
2. Vinson C. R. and Adler P. N., Directional non-cell autonomy and the transmission of polarity information by the frizzled gene of Drosophila, Nature, 1987
3. Popović M. et al., Active dynamics of tissue shear flow, New J. Phys., 2017
4. Mongera A. et al., A fluid-to-solid jamming transition underlies vertebrate body axis elongation, Nature, 2018
5. Vanderleest T. E. et al., Vertex sliding drives intercalation by radial coupling of adhesion and actomyosin networks during Drosophila germband extension, eLife, 2018

Tags: adherens junctions, frog, morphogenesis, tissue flows

doi: https://doi.org/10.1242/prelights.26039

Read preprint (No Ratings Yet)