The first part of the project was about plant mechano-perception and long-distance signaling in plants. We wanted to investigate whether plants could use hydraulic signals to carry mecanosensing information rapidly throughout the plant vascular network, thereby enabling fast and long-distance communication between distant organs. To this end, we have adopted a biomimetic strategy and designed soft artificial branches perforated with longitudinal liquid-filled channels that mimic the basic features of natural stems and branches. We found that the bending of the branch generates a strong fluid-pressure in the conductive channels, a phenomenon that cannot be explained using classical elastic theories. Further experiments conducted on natural tree branches revealed the same phenomenology. We proposed a novel poroelastic mechanism to rationalize these observations and built a model that gathers all biomimetic and natural branches on the same master curve (Louf et al, PNAS 2017). The universality of this mechanism was confirmed using the model plant of molecular biologists, Arabidopsis thaliana (Llorens et al, in preparation).
The second part of the project concerned plants gravisensing. We first studied the macroscopic gravitropic response of shoots to different level of gravity and inclination using a centrifugal set-up. The response was shown to be sensitive only to inclination (sine law) but not to the gravitational force as previously believed, which lead us to propose a new scenario of graviperception: the position-sensor hypothesis (Chauvet at al Sci. Rep 2016, Pouliquen et al, Phys. Biol 2017). To validate this scenario, we studied the motion of the statoliths (the grains at the origin of gravity detection) in the gravisensing cells. The striking result was that, unlike a classical granular medium, statoliths move and flow like a liquid even at very low angles of inclination. Experiments biomimetic cells revealed that this behavior comes from the strong agitation of the statoliths, whose origin is biological (involving acto-myosin), and not thermal (Bérut et al PNAS 2018, Bizet et al in preparation). We then extended our work to transient stimulation (inclination) and revealed the existence of a memory process in the gravitropic signaling pathway, independent of statolith dynamics. This led us to build a mathematical model that unifies the different laws found in the literature (Chauvet et al J. Exp. Bot 2019).
The third part of the project was about rapid movements and fast actuation in plants (Forterre et al Eur. News 2016). We first unveiled the intrinsic dynamics of closure of the Venus flytrap and showed that its timescale was too short to be explained by a water transport across the traps, as previously proposed. We then probed in-situ the mechanical response of the trap using a micro-indenter and showed that closure is triggering by a rapid softening of the outer layer, providing a generic mechanism for rapid actuation in plants (Ryu et al, in preparation).
Finally, inspired by our work on plant gravitropism, we study the flow of passive Brownian granular piles and found that, unlike a macroscopic granular material, the medium does not stop at a finite pile angle, but slowly creeps until its free surface becomes horizontal—a result we rationalize within a simplified model (Bérut et al Phys Rev Lett 2019). By using the suspension avalanche angle as a proxy for interparticle friction, we also provided the first experimental evidence of the frictional transition scenario proposed to explain the dramatic shear-thickening observed in some dense suspensions (Clavaud et al, PNAS 2017). We confirmed this result by designing an original pressure-imposed rheometer suitable for studying colloidal suspensions (Clavaud et al J Rheol 2020). During this work, we unexpectedly found that viscous dense suspensions exhibit hysteresis of their avalanche angle, while it was previously believed that inertia was required to observe this phenomenon in granular materials (Perrin et al Phys Rev X 2019).
