One of our current main goals is to understand the spatio-temporal sequence of events that is required for intracellular colonization of host cells. For this, we are using the model legumes Medicago truncatula and Lotus japonicus and their corresponding symbionts Sinorhizobium meliloti and Mesorhizobium loti, respectively. These bacteria colonize their host predominantly via root hair infection. Although a large number of signalling components including the host receptors that perceive the microbial signalling molecules, the Nod Factors (NF), and a significant number of essential components of the signalling cascade downstream of NF perception have been identified, our knowledge on factors regulating membrane and cell wall dynamics remains limited.
There has been a long lasting debate about the actual size of membrane nanodomains in vivo. While the original lipid raft model suggests sizes of up to 20nm, most data obtained from visualizing nanodomains in living cells point towards larger structures. Some time ago we reported on a number of nanodomains with a mean width of more than 300nm (for ore details see Ott (2017)). These structures can be resolved by standard Confocal Laser-Scanning Microscopy (CLSM). However, we also demonstrated that even though the majority of investigated membrane domains were laterally immobile, their spatio-temporal appearance may vary upon environmental changes. We are using a number of other techniques to visualize and resolve domain dynamics such as Total Internal Reflection Microscopy (TIRF), 3D Structured Illumination Microscopy (3D-SIM), Fluorescence Lifetime Imaging Microscopy (FLIM) as well as photoactivatable/ -convertible fluorophores.
Legumes have the unique ability to undergo symbiotic associations with bacteria belonging to the Rhizobiaceaefamily. These rhizobacteria secret signaling molecules (Nod Factors) that trigger physiological and morphological plant responses. In the course of this interaction, rhizobia invade the host root leading to the formation of a novel symbiotic plant organ: the root nodule. Rhizobia remain surrounded by a plant-derived plasma membrane from the very first moment of host invasion until age-dependent degradation of the symbiotically active bacteroids. This membrane serves as an essential interface for plant-microbe signal transduction and is certainly one of the key determinants for the success of the association.
Remorins have frequently been shown to be activated during plant-microbe interactions (reviewed in Jarsch and Ott, 2011). However, most of these genes are ubiquitiously expressed in plants suggesting that they may serve housekeeping functions. To approach their roles within plant cells we are generating homozygous k.o. mutants against most of the 16 remorins from Arabidopsis thaliana. These mutants are currently under detailled phenotypical investigation and first double mutants have been created. Furthermore all remorins from A. thaliana have been cloned in our lab and expressed as fluorophore-tagged fusion proteins to study their subcellular localisation. We showed that these proteins label a wide range of different, co-existing nanodomains (Jarsch et al., 2014). We have currently placed our main focus on the analysis of remorins during plant-microbe interactions (Bücherl et al., 2017) but recent findings also direct us to investigate aspects of plant developmental biology in more detail.
A key feature of signal transduction is the low affinity/high specificity kinase-substrate interaction, where both partners can dissociate rapidly and efficiently. Hypothetically, such combination could be achieved in a coupled binding and folding mechanism consisting of disorder-to-order transitions. In these transitions, the originally disordered region is locked in the binding conformation instead of exploring different conformations free in solution. This results in a reduction of entropy in the free energy of binding, which translates into a weaker binding (low affinity) compared to interactions that involve already ordered proteins with equivalent exposed surface areas. The specificity is determined by the extension and complementarity of the interaction surface area, which in the case of intrinsically disordered regions, is large even when the interaction domains themselves may be small.