Cells "feel" their environment and constantly adapt to it. In response to an external signal, they have the ability to "polarize" themselves, i.e. to reorganize their internal skeleton (cytoskeleton) and reposition their organelles along an axis defined by the signal. Is it possible to design an artificial cell capable of adopting this behavior? Researchers are trying to answer this question both to better understand the self-organization of living organisms and to explore the feasibility of materials that react to external stimuli (intelligent materials).
Polarity is a primitive biological feature that allows unicellular organisms to move towards a nutrient source or to escape from a predator. In multicellular organisms, it allows cells to direct their secretion, absorption or signal transmission activities according to the position and shape of neighboring cells. Although the molecular components vary from one organism to another, the mechanisms involved in polarization appear to be evolutionarily conserved. All of them are based on the reorganization of the cytoskeleton.
In animal cells, the highly dynamic actin cytoskeleton reacts first by locally adjusting its organization in relation to the signal.
The microtubule network then adapts to the multiple local structures of the actin network. The radial organization of microtubules around an "organizing center", i.e. the centrosome, allows the network to integrate information at the scale of the whole cell and to develop a global response. Although the mechanism of information integration is still unknown, it involves a repositioning of the centrosome towards the signal, in response to a reorganization of forces in the microtubule network.
Where and how are forces exerted on the microtubules? How are they integrated at the centrosome? And how does the actin network influence these forces?
To investigate these questions, Irig researchers used purified proteins in cell-sized microwells to reconstruct in vitro the interaction of an "aster" of microtubules with actin networks of various architectures.
In the absence of actin filament, aster positioning is very sensitive to variations in microtubule length. Actin networks limit the sensitivity of centrosome positioning to microtubule length and reinforce their centering (or decentering), depending on the isotropy (or anisotropy) of their geometry.
These results are an important step towards reconstituting polarity in an artificial cell. They were the subject of a cover article in "The EMBO Journal".