Active Cell Nematics – ActCellNem

The wide tracks are prepared by conventional chemical treatment. The surface of the track will be chemically modified to adjust the friction. We will also work on low-rigidity substrates. Oriented microtopographies are generated by abrading glass surfaces in a controlled manner. The forces accompanying the formation of the cell cord will be measured by TFM and myoblast differentiation will be monitored in these well-oriented three-dimensional structures. With respect to chirality, the centrosome position of the cells will be monitored for isolated cells plated on micropatterns and for confined populations. Finally, we will increase our experimental throughput by working with the lens-free microscope which can be placed directly in an incubator. In addition to analyzing the orientation field and the velocity field, a computational effort will be made to achieve quantitative measurements on thick structures. Theoretical tools based on the symmetry properties of active gels will be developed in parallel.

Tracks combining micro and meso structures have been fabricated and characterized. Experiments have been carried out on these substrates showing the appearance of a Fréedericksz transition when the two structures are orthogonal (micro and meso contact guidances competition). A theory which assimilates abrasions to an effective external field was developed to account for these results in coherence with the behaviour of the cells in the absence of abrasion. The situation is made complex by the appearance of topological defects which disappear if the abrasions make a finite angle with the track. The transition from 2D to 3D has been addressed on an isolated defect. This work revealed a heterogeneity of the defects’ population, some of which are immobile, probably due to an accumulation of extra-cellular matrix. It is on these defects that bilayers are formed. From a more instrumental point of view, a new algorithmic method allowing the reconstruction of images of multi-cellular layers from acquisitions obtained by lens-free microscopy has been implemented. This method consists of alternating two algorithmic approaches, inverse problem solving and deep neural networks. First results obtained on simulations show that this new method can reconstruct a quantitative phase image of cell clusters (diameter 110-160µm) up to a thickness of 50µm.

Apart from writing articles and theses, we will work on a periodic substrate by seeding the cells on glass fibres. We will work on a theoretical description of this situation in parallel. We will extend our work on the formation of bilayers from an isolated defect to the coordinated formation of cell cords in the middle of the tracks in the presence or not of abrasions at different angles. TFM experiments will provide us with the involved traction forces. This work will be continued by quantifying the efficiency of differentiation of the cells belonging to these cords, under these various conditions. Collective chirality will be measured on tracks and fibres and compared with individual chirality. Chiral terms will be introduced into theoretical expressions to describe the experimental results. The development of lens-free microscopy will be pursued in order to validate on real samples the results obtained so far on simulations.

10 conferences and posters
2 publications. 3 more are in preparation

Confluent spindle-shaped cells tend to line up with each other and adopt a nematic order (i.e. orientational, but not positional, order). When confined to adherent stripes, some cell types align with the stripe direction and show no remarkable dynamic behavior; in contrast, the director of other cell types spontaneously makes angle with the stripe direction while the cells develop antiparallel displacements near each edge amounting to a shear flow. It should be noted that several in vivo observations report such antiparallel displacements, particularly during tumor development. Our first observations showed a major influence of cell-substrate friction; they also evidenced converging flows leading to three-dimensional cellular cords in the center of the stripe, and an extremely robust multicellular chirality. Here, we propose a project combining experiments and theory and aiming at better understanding these different characteristics. In practice, starting from a monolayer, we will monitor the nematic order and the associated cell displacements in the plane and out-of-plane. Along this line, we propose to compare the mesoscopic contact guidance obtained in the relatively stripes described above (typically a few 100 µm wide) with that resulting from microstructures more conventionally used in this application (typically an array of parallel micro-stripes a few µm wide). We will combine structures at these two very different scales on the same device. We anticipate an orientation transition at a critical width of the mesoscopic stripe when the directions imposed by these two fields are different. In addition, the orientation of the microstructures defines an anisotropic friction in the mesoscopic stripes, which results in modifying the respective weights of shear and converging flows; the formation of three-dimensional cellular cords being promoted by dominant convergent flows. In addition to the study of the fundamental mechanisms at the origin of the formation of a 3D structure from a monolayer, this is an important step in the context of cellular differentiation, in particular for muscle cells, and possible applications in this context will be explored. The chirality measured on these confined cellular assemblies poses fundamental questions in a development biology context. We will try to understand how a (weak) chirality at the scale of an individual cell is amplified at the scale of a cellular assembly. From a practical point of view, all the planned experiments require well-suited observation techniques. Faced with the large number of experiments to be undertaken, our choice naturally turned to lens-free microscopy, whose performances at large fields-of-view have recently been demonstrated, including for high density confluent cell layers. In addition, the technique is compatible with our image correlation analysis softwares without having to reconstruct the images. We will use this technology to conduct a large number of experiments in parallel, for example on many stripes of different widths simultaneously. In parallel, we will develop new algorithms to quantitatively reconstruct 3D structures such as cellular strings. Finally, an important feature of this project is its coupling with theory. An active fluid theory has already shown its potential on our initial experiments. It will be necessary to go beyond the initial model to interpret the experiments presented above and develop new ones.