research

We study biological processes of cells and tissues with a biophysical perspective. Using Drosophila as a model system, in investigate the mechanics of collective cell mechanics, the interplay between growth and mechanics. We also develop new methodologies to image and quantify cell shapes, cell movement, and cell metabolism.

Compaction of Drosophila histoblasts in a crowded epidermis is driven by buckling of their apical junctions

In many proliferating epithelia, cells present a polygonal shape that results from tensile forces of the cytoskeletal cortex and from the packing geometry set by the cell cycle \cite{gibson2006emergence, Farhadifar2007}. In the larval Drosophila epidermis, two cell populations, histoblasts and larval epithelial cells, compete for space as they grow on a limited body surface. They do so in the absence of cell divisions.

we show that histoblasts, which are initially polygonal, undergo a dramatic morphological transition in the course of larval development. Histoblasts change from a tensed network configuration, with straight cell outlines at the level of adherens junctions, to a highly folded morphology.

We propose a model in which crowding of the epidermis imposes a compressive load on the growing junctions which induces their buckling. Buckling effectively compacts histoblasts at their apical plane and may serve to avoid physical harm to these adult epidermis precursors during larval life.

Remodeling of histoblast junctions during larval development. Chronic imaging of adherens junctions at different stages of larval development. We observe a transition in the shape of histoblasts

Collective cell migration in the course of Drosophila thorax formation

In this project we study the massively collective migration of Drosophila wing imaginal discs in the course of metamorphosis. Wing discs (~10,000 cells each) move as highly coherent physical units displaying supra-cellular organization of their cytoskeleton (see Figure). Imaginal discs have the ability to find their counterpart with a great precision (Milner, 1984).

Our goals are:

1. The characterization of the cytoskeletal dynamics that underlie tissue guidance in connection with gene patterning.
2. The investigation of the mechanics of migration, to address how the cytoskeletal activity at the migrating front may lead to the mechanical actuation of this massive cluster of cells.
3. The iterative construction of a model of tissue guidance by a constant back and forth between theory and experiments.

A,B: Schematic representation of two imaginal discs (red) moving towards each other on top of the larval epidermis (green). Inset of A: Close up of cytoskeletal dynamics at the front, where lamellipodia and filopodia guide the migrating tissue. A bias at gene expression domain boundaries may provide a means to orient collective cell migration. C: Micrograph of two imaginal discs before they fuse, stained for non-muscular Myosin- II. Each disc spans over ~300μm, while individual cell diameters are in the range of 2-5μm.

The interplay between growth and mechanics

The Role of the extracellular Matrix in shaping growing tissues in 3D

We investigate the shape distorsions of the Drosophila wing imaginal disc upon chemical alteration of the extracellular matrix. We model the mechanical interactions with a bilayered plate theory in the context of finite elasticity.

Folding an epithelium by local ECM degradation.

(a) We locally express in the Drosophila wing imaginal disc the protease Mmp2 (shaded in red). The local softening of the ECM induces a change of the 3D shape of the tissue. (b) Geometry of the bilayer model. $w_0(x,y)$ is the deviation of the initial configuration from a flat surface; $\zeta(x,y)$ is the displacement from this initial configuration as a result of elastic energy minimization under the action of growth and pre-stress ($N_0$). The schema at bottom represent biological context that can be modeled by the theory: the two layers can either be cell or ECM layers of different height, Young modulus and growth rate; the stiff interface can be a cytoskeletal cortex or a thin ECM.

New methodologies for imaging of epithelial structure

The optical microscope is a central tool in biology, but the amount of light involved is often toxic for the sample under investigation. We developed an intelligent microscopes which automatically computes where it should send light to image structures of interest within the sample. By using as little as 1% of the light used by conventional microscopes, the technique will foster a better understanding of very fragile objects such as embryos and organoids.

An adaptive microscope for the imaging of biological surfaces

(a, left) Drawing of a curved biological tissue. The tissue may be overlaid by a second epithelium to be discarded by the imaging process. (a, right) We first estimate the surface of interest (red mesh) using a small fractional pre-scan (green dots). (b) After surface estimation, we either scan a thin shell (blue) around the surface of interest (left side) or we propagate the scanning process along the cell outlines (right side). The zoom in the inset shows the propagation front in red while the previously sample voxels are in green.(c) Comparing image acquisition with the designed smart-scan and a conventional strategy. The same embryonic tissue has been continuously imaged with the smartscan (top) and a conventional scan (bottom). After one hour of continuous imaging the tissue has suffered significant photodamage of the fluorophores in thte conventionally acquired region, but not in the region avquired with the new strategy.

Adaptive scans allow targeted cell-ablations on curved cell sheets

We present a flexible near infrared (NIR) fs-pulsed laser ablation system in which ablation trajectories proceed in 3D and adapt to the curved surface of cell sheets, which are prominent structures in embryos. Trajectories are computed through an unsupervised search for the surface of interest. We demonstrate that, depending on the exact experimental setup, the surface estimation can rely on a high content 3D imaging with a combined confocal microscope, or alternatively on a rapid Lissajou scan of the sample space with a NIR stand-alone setup.

(a) A trajectory (x(t),y(t),z(t)) is computed so that it lies on the estimated surface Zs(x,y). (b) To rapidly estimate the surface of interest using only a NIR laser, we perform a Lissajous scan, which explores a limited number of voxels within the sample space (c,c') 3D representation of an ablated tissue at two time points following the ablation.

New methodologies to image biosynthetic metabolism

Page updated

Google Sites

Report abuse