Abstract
The cell membrane is composed of a myriad of different lipids and proteins. It is known to beboth heterogeneous (juxtaposition of domains) and dynamic (diffusion of these constituents).In addition, the plasma membrane is at the interface of the cell and control the perception andintegration of biological signals. The molecular players can be compartmentalised to respondto different signals (Jaillais and Ott, 2020). This is the case for Rho of Plant (ROP) GTPases,which have been described as a family of key players in the response to different signals: Auxin,ABA, response to pathogen response, osmotic signalling and symbiosis (Basu et al., 2008; Linet al, 2012 ; Poraty-Gavra et al., 2013 ; Choi et al., 2014 ; Lin et al., 2015 ; Engelsdorf et al.,2018; Feiguelman et al., 2018; Platre et al., 2019; Smokvarska et al., 2020). Interestingly, thispleiotropic role also exists at the level of a single ROP isoform. ROP6 has been described as aregulator of auxin and osmotic signalling (Platre et al., 2019; Smokvarska et al., 2020). In roots,we found that ROP6 forms nanodomains during these two signalling processes. During osmoticsignalling, ROP6 forms nanodomains in association with RBOHDs and F allowing theaccumulation of reactive oxygen species (ROS). However, in response to auxin, ROP6nanodomains do not associate with RBOHs (Platre et al., 2019; Smokvarska et al., 2020). Thus,a question is emerging: How can the organisation of ROP6 nanodomains induce a specificresponse downstream of the different signals? We therefore set out to determine how ROP6activators can help encode a specific response at the cellular level. Using a reverse geneticscreen, we found that guanine exchange factor 14 (GEF14) is necessary and specific forinducing osmotic signalling in plant cells. Based on genetic and GTPase activation sensorexperiments, we show that GEF14 acts as an activator of ROP6 in planta. Using super resolution microscopy, we showed that ROP6 nanodomains induced by the osmotic signal, butnot those induced by auxin, require GEF14 to form. Finally, we show that GEF14 restructuresinto clusters after an osmotic signal. We extended our model to immune signalling where wesimilarly identified a specific GEF isoform. In a second part, I sought to determine how GEF14was activated by the osmotic signal. Using a comparative biochemical approach, I identified 5phosphorylation sites in the PRONE domain of GEF14. The GEF14 interactome led us to identifycandidate kinases and phosphatasesthat need to be confirmed by reverse genetic approaches.Our results suggest that the activation of a GEF isoform can determine the early stages ofsignalling and act on the specific recruitment of effector proteins. This mechanism makes itpossible to maintain and control early cell signalling in plants