How plants sense temperature is a fundamental question in plant biology. In this project we are investigating the molecular and atomic level determinants of temperature sensing mediated by the protein EARLY FLOWERING 3 (ELF3). ELF3 undergoes reversible liquid-liquid phase separation (LLPS) as a function of temperature. We are determining the structural basis of LLPS, how temperature effects ELF3 phase separation and introducing targeted mutations to alter LLPS in vitro and in vivo.
The initial objectives of the project were the design, production and purification of different constructs of ELF3, in particular the prion-like domain (PrD) region, which is essential for liquid-liquid phase separation (LLPS). The objectives for the first 18 mo. of the project focus on the in vitro characterization of the ELF3 protein and PrD region with respect to the conditions necessary for LLPS. We successfully produced many constructs of ELF3 from Arabidopsis thaliana and other species (Brachypodium distachyon, Solanum tuberosum) with and without GFP tags. These proteins were characterized in vitro using light and fluorescence microscopy and their dynamics investigated using fluorescence recovery after photobleaching (FRAP). Their behavior with respect to concentration, temperature, pH and ionic strength were determined and phase diagrams (dispersed, condensed and precipitate phases) generated as a function of these different variables. Reversible LLPS was shown for AtELF3 PrD and a series of AtELF3 PrD mutants. The characteristics of LLPS as a function of temperature were investigated using turbidity assays and small angle X-ray scattering experiments. Initial SAXS models have been generated using data from the dispersed and condensed phases as a function of temperature. These structural studies will be published within the next year. Optimisation of samples for high resolution NMR studies in ongoing. ELF3 mutants with altered LLPS as a function of temperature have been identified and will be tested in planta in the second part of the project. These experiments will correlate in vitro LLPS with in vivo function of ELF3 and temperature sensitive phenotypes.
The project uses different in vitro and in vivo techniques to investigate thermosensing mechanisms at the molecular and organism level in the model plant Arabidopsis thaliana. The primary in vitro techniques used include structural biology techniques, namely small angle X-ray scattering and nuclear magnetic resonance spectroscopy. These techniques are performed in solution, allowing the investigation of the dynamic properties of the protein under study, EARLY FLOWERING 3 (ELF3). We have demonstrated by SAXS that ELF3 undergoes structural changes as a function of temperature during the formation of the liquid droplet phase. The protein molecules adopt specific conformations in the condensed phase. We are currently preparing samples for detailed atomic resolution studies by NMR to better understand the intra and intermolecular interactions during phase separation. To further characterise the protein in vitro, we have generated fluorescent protein tagged constructs of the ELF3 PrD region. As in the untagged version, ELF3 PrD acts as a driver of LLPS, forming a condensed phase as a function of different variables including pH, protein concentration, ionic strength of the solution and temperature. Using confocal microscopy and fluorescence recovery after photobleaching (FRAP), we are able to investigate the dynamic nature of the phase separated state. ELF3 PrD tagged with green fluorescent protein undergoes LLPS in a temperature dependent manner and the liquid droplets formed are of a liquid-like nature, exhibiting recovery after partial and full photobleaching, indicating dynamics exchange within the droplet and with the surrounding solution.Our collaborators have further shown that this behaviour is largely recapitulated in vivo, suggesting that temperature dependent LLPS is indeed a mechanism for direct thermosensing in Aranidopsis thaliana and related species that possess a PrD in the ELF3 protein.
The TEMPSENS project focuses on the molecular mechanisms of plant thermosensing, in particular the role of the protein ELF3 as a direct thermosensor. We have determined that ELF3 PrD undergoes liquid-liquid phase separation (LLPS) as a function of different variables including pH, protein concentration, ionic strength and increasing temperatures. The protein in solution at low temperatures is monodispersed, which we have determined from initial DLS and SAXS studies. As the temperature is increased from 4°C to 35°C, ELF3 PrD forms liquid droplets. These droplets exhibit some internal ordering as observed by the appearance of a peak in the SAXS scattering curve.
The dynamics of the droplet phase have been probed using fluorescence recovery after photobleaching (FRAP). The droplets exhibit highly mobile behavior upon formation and rapidly age, with a low mobility phase dominating after 5-10 minutes. This behavior is reproducible over numerous experiments and under different buffer conditions (pH, ionic strength).
To further examine the determinants of LLPS driven by the PrD of ELF3, we have begun to systematically vary the polyQ regions present the protein. Lengthening the polyQ’s and deletion of these amino acids alters the LLPS of the PrD based on turbidity assays. We are able to alter the temperature at which LLPS occurs in vitro by manipulating the length and distribution of polyQ’s in the PrD of the protein.
Based on these in vitro studies, we are designing constructs for transformation in the elf3 mutant background. We will test our PrD polyQ mutants in vivo in order to correlate in vitro and in vivo function. GFP tagged ELF3 will allow us to determine the localization of the protein in vivo and whether the protein undergoes LLPS under physiological conditions. Plant phenotypes (hypocotyl length and flowering time with respect to different growth conditions, i.e. 22°C vs. 27°C of mutants and wild type will be assessed.
A major focus in the coming years will be to correlate our in vitro studies with in planta function. We have characterised the ELF3 PrD region and demonstrated reversible and temperature sensitive phase separation. This is an attractive mechanism for in vivo temperature sensing, however this has not been proven. In order to answer whether or not ELF3 LLPS is a primary thermosensing mechanism in Arabidopsis, we will focus on the generation of highly targeted ELF3 mutations that exhibit differential LLPS formation in vitro as a function of temperature. These variants will be characterized in a heterologous system such as yeast and then stable transgenic lines will be generated in the elf3 loss-of-function mutant background. GFP-tagged ELF3 mutants will be examined for localization and LLPS formation as a function of temperature. This will be correlated with phenotypes of plants observed at different growth temperatures. Temperature sensitive growth including hypocotyl elongation and flowering time will be determined for different ELF3 variants. This series of experiments will robustly correlate LLPS with temperature sensing and response.
We have two major peer reviewed publications related to this project:
Silva, C. S.; Nayak, A.; Lai, X.; Hutin, S.; Hugouvieux, V.; Jung, J.-H.; López-Vidriero, I.; Franco-Zorrilla, J. M.; Panigrahi, K. C. S.; Nanao, M. H.; Wigge, P. A.; Zubieta, C. Molecular Mechanisms of Evening Complex Activity in Arabidopsis. Proc. Natl. Acad. Sci. U. S. A. 2020, 117 (12), 6901–6909.
Jung, J.-H.; Barbosa, A. D.; Hutin, S.; Kumita, J. R.; Gao, M.; Derwort, D.; Silva, C. S.; Lai, X.; Pierre, E.; Geng, F.; Kim, S.-B.; Baek, S.; Zubieta, C.; Jaeger, K. E.; Wigge, P. A. A Prion-like Domain in ELF3 Functions as a Thermosensor in Arabidopsis. Nature 2020, 585 (7824), 256–260.
In addition to peer reviewed publications, we have performed various outreach activities to disseminate our results at the CEA, Grenoble (CEA «fait marquant«) and to the general public including contributions on this subject to «The Conversation,« interviews for «Silence, ça pousse,« and local newsletters.
Day length and temperature are primary environmental factors that affect plant growth, development and reproduction. Due to global warming, plant species must adapt their lifecycle to increasing ambient temperatures. The effects of climate change have already had a profound impact on the phenology of wild and cultivated plant species. Thus, understanding how plants sense and respond to temperature are fundamental challenges in the field of plant biology. Different mechanisms have been proposed as contributing to this, including changes in the structure and function of protein molecules, allowing them to act as direct thermosensors. A few putative protein thermosensors have been identified in Arabidopsis, including EARLY FLOWERING 3 (ELF3), a promiscuous scaffold protein that interacts with many different partners including phytochrome B (phyB), PHYTOCHROME INTERACTING FACTOR4 (PIF4) and the circadian Evening Complex proteins, LUX ARRYTHMO (LUX) and EARLY FLOWERING 4 (ELF4). All of these proteins are implicated in thermomorphogenesis, a suite of adaptive changes that include hypocotyl elongation and an early transition to flowering. These and other data, including Quantitative Trait Loci mapping and forward and reverse genetics, suggest that ELF3 acts as a direct thermosensor and coordinator of thermoresponsive growth. How ELF3 perceives and transduces temperature signals is not known. The proposed project will directly address the molecular mechanisms of ELF3 function and thermosensory activity in Arabidopsis.
ELF3 is an intrinsically disordered protein (IDP) with a prion like domain (PrLD) that forms liquid droplets in vivo and in vitro. Intrinsically disordered proteins and PrLDs are able to undergo phase transitions to highly dense liquid droplet phases. In vivo, liquid droplets act as membraneless organelles and are able to concentrate cargo molecules and alter their activity, allowing compartmentalization of biological processes. Liquid droplets have been implicated in diverse processes including transcription, superenhancer formation, RNA processing and metabolism, stress response and the formation of inactive storage sites for proteins and RNA. Although their physical basis remains very poorly understood, particularly in plants, an important characteristic of liquid droplets is their sensitivity to environmental cues. Our preliminary results demonstrate that the ELF3 PrLD is sufficient to form liquid droplets and that droplet formation is temperature-sensitive. We hypothesize that the ELF3 PrLD is required for the thermosensory activity of the protein in Arabidopsis and that thermosensing requires liquid droplet formation.
The project will address the underlying molecular mechanisms of ELF3 liquid droplet formation in vitro, determine how this affects activity, including interactions with different protein partners, and examine divergent ELF3 proteins from species that exhibit altered thermosensitive flowering responses. The project leverages an integrated in vitro, structural and in vivo approach to address the fundamental question of the biological role of ELF3 liquid droplet formation in plant thermosensing and thermoresponse. In vitro biophysical, fluorescence microscopy and solution state structural techniques (small angle x-ray scattering and nuclear magnetic resonance) will provide the molecular basis for droplet formation, interaction with cargo proteins and determine the effects of temperature on droplet stability and dynamics. In vivo experiments in protoplasts and Arabidopsis will correlate in vitro droplet behavior with in vivo function at the cellular and organismal level.
The proposed project will provide a fundamental understanding of ELF3 liquid droplet formation and determine the role of liquid droplets in thermosensing in Arabidopsis. These studies will have broad applications to other species due to the ubiquitous nature of liquid droplets and their potential role as environmental sensors.
Madame Chloe Zubieta (CEA)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
IBS INSTITUT DE BIOLOGIE STRUCTURALE
Help of the ANR 456,840 euros
Beginning and duration of the scientific project: September 2019 - 48 Months