Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

A prion-like domain in ELF3 functions as a thermosensor in Arabidopsis


Temperature controls plant growth and development, and climate change has already altered the phenology of wild plants and crops1. However, the mechanisms by which plants sense temperature are not well understood. The evening complex is a major signalling hub and a core component of the plant circadian clock2,3. The evening complex acts as a temperature-responsive transcriptional repressor, providing rhythmicity and temperature responsiveness to growth through unknown mechanisms2,4,5,6. The evening complex consists of EARLY FLOWERING 3 (ELF3)4,7, a large scaffold protein and key component of temperature sensing; ELF4, a small α-helical protein; and LUX ARRYTHMO (LUX), a DNA-binding protein required to recruit the evening complex to transcriptional targets. ELF3 contains a polyglutamine (polyQ) repeat8,9,10, embedded within a predicted prion domain (PrD). Here we find that the length of the polyQ repeat correlates with thermal responsiveness. We show that ELF3 proteins in plants from hotter climates, with no detectable PrD, are active at high temperatures, and lack thermal responsiveness. The temperature sensitivity of ELF3 is also modulated by the levels of ELF4, indicating that ELF4 can stabilize the function of ELF3. In both Arabidopsis and a heterologous system, ELF3 fused with green fluorescent protein forms speckles within minutes in response to higher temperatures, in a PrD-dependent manner. A purified fragment encompassing the ELF3 PrD reversibly forms liquid droplets in response to increasing temperatures in vitro, indicating that these properties reflect a direct biophysical response conferred by the PrD. The ability of temperature to rapidly shift ELF3 between active and inactive states via phase transition represents a previously unknown thermosensory mechanism.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Fig. 1: The polyQ repeat of ELF3 is embedded within a predicted prion domain that is essential for thermal responsiveness and modulated by ELF4.
Fig. 2: High temperature induces the formation of ELF3 speckles in vivo.
Fig. 3: The PrD of ELF3 undergoes a reversible phase transition in response to temperature.

Data availability

Sequencing data for gene-expression analysis (RNA-seq) and protein–DNA interactions (ChIP–seq) have been deposited in the publicly available Gene Expression Omnibus (GEO) under accession code GSE137264 ( The raw data used in this study are available at

Code availability

The code to produce the figures (Fig. 1d–f and Extended Data Figs. 5, 79) from the processed files is available at To enable easier browsing, a static site is hosted at The inhouse pipeline for mapping is available at


  1. 1.

    Scheffers, B. R. et al. The broad footprint of climate change from genes to biomes to people. Science 354, aaf7671 (2016).

    PubMed  Article  Google Scholar 

  2. 2.

    Nusinow, D. A. et al. The ELF4–ELF3–LUX complex links the circadian clock to diurnal control of hypocotyl growth. Nature 475, 398–402 (2011).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  3. 3.

    Ezer, D. et al. The evening complex coordinates environmental and endogenous signals in Arabidopsis. Nat. Plants 3, 17087 (2017).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  4. 4.

    Box, M. S. et al. ELF3 controls thermoresponsive growth in Arabidopsis. Curr. Biol. 25, 194–199 (2015).

    CAS  PubMed  Article  Google Scholar 

  5. 5.

    Mizuno, T. et al. Ambient temperature signal feeds into the circadian clock transcriptional circuitry through the EC night-time repressor in Arabidopsis thaliana. Plant Cell Physiol. 55, 958–976 (2014).

    CAS  PubMed  Article  Google Scholar 

  6. 6.

    Raschke, A. et al. Natural variants of ELF3 affect thermomorphogenesis by transcriptionally modulating PIF4-dependent auxin response genes. BMC Plant Biol. 15, 197 (2015).

    PubMed  PubMed Central  Article  Google Scholar 

  7. 7.

    Nieto, C., López-Salmerón, V., Davière, J. M. & Prat, S. ELF3-PIF4 interaction regulates plant growth independently of the Evening Complex. Curr. Biol. 25, 187–193 (2015).

    CAS  PubMed  Article  Google Scholar 

  8. 8.

    Undurraga, S. F. et al. Background-dependent effects of polyglutamine variation in the Arabidopsis thaliana gene ELF3. Proc. Natl Acad. Sci. USA 109, 19363–19367 (2012).

    ADS  CAS  PubMed  Article  Google Scholar 

  9. 9.

    Tajima, T., Oda, A., Nakagawa, M., Kamada, H. & Mizoguchi, T. Natural variation of polyglutamine repeats of a circadian clock gene ELF3 in Arabidopsis. Plant Biotechnol. 24, 237–240 (2007).

    CAS  Google Scholar 

  10. 10.

    Jiménez-Gómez, J. M., Wallace, A. D. & Maloof, J. N. Network analysis identifies ELF3 as a QTL for the shade avoidance response in Arabidopsis. PLoS Genet. 6, e1001100 (2010).

    PubMed  PubMed Central  Article  Google Scholar 

  11. 11.

    Press, M. O. & Queitsch, C. Variability in a short tandem repeat mediates complex epistatic interactions in Arabidopsis thaliana. Genetics 205, 455–464 (2017).

    CAS  PubMed  Article  Google Scholar 

  12. 12.

    Lancaster, A. K., Nutter-Upham, A., Lindquist, S. & King, O. D. PLAAC: a web and command-line application to identify proteins with prion-like amino acid composition. Bioinformatics 30, 2501–2502 (2014).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  13. 13.

    Herrero, E. et al. EARLY FLOWERING4 recruitment of EARLY FLOWERING3 in the nucleus sustains the Arabidopsis circadian clock. Plant Cell 24, 428–443 (2012).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  14. 14.

    Doyle, M. R. et al. The ELF4 gene controls circadian rhythms and flowering time in Arabidopsis thaliana. Nature 419, 74–77 (2002).

    ADS  CAS  PubMed  Article  Google Scholar 

  15. 15.

    Silva, C. S. et al. Molecular mechanisms of Evening Complex activity in Arabidopsis. Proc. Natl Acad. Sci. USA 117, 6901–6909 (2020).

    CAS  PubMed  Article  Google Scholar 

  16. 16.

    Wallace, E. W. J. et al. Reversible, specific, active aggregates of endogenous proteins assemble upon heat stress. Cell 162, 1286–1298 (2015).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  17. 17.

    Franzmann, T. M. et al. Phase separation of a yeast prion protein promotes cellular fitness. Science 359, eaao5654 (2018).

    PubMed  Article  Google Scholar 

  18. 18.

    Dignon, G. L., Zheng, W., Kim, Y. C. & Mittal, J. Temperature-controlled liquid–liquid phase separation of disordered proteins. ACS Cent. Sci. 5, 821–830 (2019).

    CAS  PubMed  PubMed Central  Google Scholar 

  19. 19.

    Si, K. Prions: what are they good for? Annu. Rev. Cell Dev. Biol. 31, 149–169 (2015).

    CAS  PubMed  Article  Google Scholar 

  20. 20.

    Jaeger, K. E., Pullen, N., Lamzin, S., Morris, R. J. & Wigge, P. A. Interlocking feedback loops govern the dynamic behavior of the floral transition in Arabidopsis. Plant Cell 25, 820–833 (2013).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  21. 21.

    Barbosa, A. D. et al. Lipid partitioning at the nuclear envelope controls membrane biogenesis. Mol. Biol. Cell 26, 3641–3657 (2015).

    MathSciNet  CAS  PubMed  PubMed Central  Article  Google Scholar 

  22. 22.

    Guilligay, D. et al. The structural basis for cap binding by influenza virus polymerase subunit PB2. Nat. Struct. Mol. Biol. 15, 500–506 (2008).

    CAS  PubMed  Article  Google Scholar 

  23. 23.

    Tarendeau, F. et al. Structure and nuclear import function of the C-terminal domain of influenza virus polymerase PB2 subunit. Nat. Struct. Mol. Biol. 14, 229–233 (2007).

    CAS  PubMed  Article  Google Scholar 

  24. 24.

    Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Nat. Methods 9, 676–682 (2012).

    CAS  PubMed  Article  Google Scholar 

  25. 25.

    Koulouras, G. et al. EasyFRAP-web: a web-based tool for the analysis of fluorescence recovery after photobleaching data. Nucleic Acids Res. 46 (W1), W467–W472 (2018).

Download references


We thank M. Perutz for discussions on polyQ proteins; the microscopy facility MuLife of the Interdisciplinary Research Institute of Grenoble (IRIG)/Department of Structural and Cellular Integrative Biology (DBSCI), funded by CEA Nanobio and Labex GRAL, for equipment access and use; and L. Kurzawa and F. Senger for technical assistance and discussions. This work used the platforms of the Grenoble Instruct-ERIC Center (ISBG; grant UMS 3518; CNRS/CEA/UGA/EMBL), with support from the French Infrastructure for Integrated Structural Biology (FRISBI; grant ANR-10-INBS-05-02) and Labex GRAL (grant ANR-CBH-EUR-GS) within the Grenoble Partnership for Structural Biology (PSB). P.A.W. and C.Z. received funding from Instruct-ERIC (PID 2236). This work was also supported by a grant from the National Research Foundation of Korea (NRF), funded by the Korean government (Ministry of Science and ICT; MSIT) (grant 2019R1C1C1010507 to J.-H.J). This work was further funded by project ANR-19-CE20-0021-01 PRC (to C.Z.). P.A.W. received support from the European Research Council (grant EC FP7 ERC 243140) and the Gatsby Charitable Foundation (grant GAT3273/GLB). P.A.W. receives funding from the Leibniz Foundation.

Author information




J.-H.J., A.D.B., S.H., J.R.K., C.Z., K.E.J. and P.A.W. conceived the study and wrote the manuscript. J.-H.J. and M.G. generated transgenic plants and mutants and analysed their phenotypes. A.D.B. performed in vivo imaging experiments in yeasts and plants. S.H., J.R.K., C.S.S., X.L., E.P. and C.Z. performed and analysed in vitro phase-separation assays. K.E.J. performed ChIP–seq and RNA-seq experiments. D.D. and F.G. analysed sequencing data. S.-B.K. and S.B. performed yeast two-hybrid assays and analysed gene-expression levels in transgenic plants. C.Z., K.E.J. and P.A.W. supervised experimental work.

Corresponding author

Correspondence to Philip A. Wigge.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Peer review information Nature thanks Salomé Prat and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data figures and tables

Extended Data Fig. 1 The length of the polyQ repeat within the ELF3 PrD influences temperature responsiveness.

a, Hypocotyl lengths for transgenic plants with altered ELF3 polyQ tracts grown at different temperatures. At 17 °C, ELF3 is required to prevent hypocotyl elongation, but different polyQ lengths do not perturb ELF3 function. At 27 °C, the responsiveness of ELF3 to temperature increases with polyQ length. Q0, Q7 and Q17 refer to the length of the polyQ tract; #15 (for example) denotes a particular transgenic line. Each box is bounded by the lower and upper quartiles; the central bar represents the median; the whiskers indicate minimum and maximum values. b, Alignment of ELF3 amino-acid sequences from three different plant species. The region indicated by an arrow was used to create a chimeric version of Arabidopsis ELF3, with the ELF3 PrD replaced by the corresponding sequence of BdELF3 or StELF3. Conserved amino acids are in white in red-filled rectangles. Similar residues are in red surrounded by blue lines. Dots denote ten-amino-acid spacings.

Extended Data Fig. 2 ELF3 expression in the transgenic lines used here.

All transgenic plants were generated by expressing ELF3 under the control of its native promoter in elf3-1 mutant backgrounds. The elf3-1 phenotypes were perfectly rescued in all ELF3 transgenic lines used here. ELF3 transcript levels were determined by RT–qPCR. Gene-expression values were normalized to EIF4A expression. Data shown as means ± s.d. (n = 3). a, ELF3pro::ELF3 elf3-1 transgenic plants without any tag sequences, used for hypocotyl-elongation and RNA-seq experiments. b, ELF3pro::ELF3-FLAG elf3-1 lines, used for flowering-time measurements and ChIP–seq experiments. c, ELF3pro::ELF3-GFP elf3-1 lines, used to observe ELF3-induced speckle formation in planta.

Extended Data Fig. 3 The effects of the elf3-1 allele in A. thaliana are rescued by StELF3 or BdELF3, which lack a detectable PrD.

a, b, Transgenic A. thaliana plants in the elf3-1 background, expressing different forms of ELF3 either constitutively (from the 35S promoter; ‘OE’) or under the control of the endogenous AtELF3 promoter (ELF3pro), were grown in short-photoperiod conditions at 22 °C until bolting. c, Relative expression of FT at ZT8 was analysed by RT–qPCR. Twelve-day-old seedlings grown at different temperatures under short-photoperiod conditions (SDs) were used to analyse transcript accumulation. Data shown as means ± s.d. (n = 3).

Extended Data Fig. 4 ELF4 is required to stabilize the activity of the evening complex at warmer temperatures.

a, At lower temperatures, ELF4 becomes dispensable for controlling flowering, but with increasing temperature, it assumes a greater role. b, ELF4 overexpression greatly reduces the thermal responsiveness of both hypocotyl elongation and flowering time, and this response depends entirely on ELF3. c, At low temperatures, ELF4 is dispensable, and elf4-2 mutants have similar hypocotyl phenotypes to wild-type plants. As temperature increases, the role of ELF4 becomes increasingly important, as measured by hypocotyl length. Overexpressing ELF3 is not sufficient to change thermal responsiveness and ELF3 overexpression has no effect in the elf4-2 background at 27 °C, indicating that ELF4 plays an important part at higher temperatures. In box plots, each box is bounded by the lower and upper quartiles; the central bar represents the median; the whiskers indicate minimum and maximum values. b, c, Scale bars, 5 mm. d, ELF3 constructs used. Numbers indicate residue positions. The domain structure of the ELF3 protein was determined using SMART protein domain annotation ( ELF3 does not contain any specific domains except for low-complexity regions, which are regions in protein sequences that differ from the composition and complexity of most proteins with normal globular structure. e, Interactions of ELF3 with ELF4 in yeast cells. Cell growth on selective medium was examined. The ELF3 fragment containing a low-complexity region, which does not overlap with the PrD, is responsible for the interaction with ELF4. A soluble form of ELF3 peptide, which is used for in vitro experiments, does not include the region required for the interaction with ELF4.

Extended Data Fig. 5 The binding of ELF3 to target genes depends on temperature, and stabilized forms of ELF3 are less temperature responsive than wild-type ELF3.

Average ELF3 ChIP–seq peak signals are measured as fold-enrichment over input (as calculated by MACS2) across multiple transgenic lines expressing the indicated ELF3 variants.

Extended Data Fig. 6 The expression of ELF3-dependent genes is influenced by temperature and the PrD of ELF3.

a, Effects of temperature. We analysed 325 transcripts that show ELF3-dependent expression in RNA-seq datasets from different genotypes at 22 °C and 27 °C. As expected, ELF3-dependent gene expression is generally suppressed at 22 °C (red), except in the elf3-1 background, where genes are upregulated (green). Lines overexpressing BdELF3 show less activation at 27 °C, consistent with their later-flowering phenotypes. Replacing just the Arabidopsis PrD with the corresponding region from BdELF3 (in ELF3pro::BdELF3 at 27 °C) is sufficient to greatly reduce the upregulation of ELF3-dependent genes at this temperature. Upregulation of ELF3-dependent genes also occurs in an elf3-1 mutant when ELF4 is overexpressed, consistent with ELF3 being necessary for ELF4 action. b, Effects of the PrD. We analysed 325 transcripts that show ELF3-dependent expression in RNA-seq datasets from different polyQ genotypes at 22 °C and 27 °C. Plants expressing ELF3 with a truncated polyQ repeat (ELF3-Q0) show a reduced expression of ELF3-dependent genes at 27 °C, consistent with their shorter-hypocotyl phenotype. c, Heat map showing that ELF3-bound targets that are usually induced by shifting to 27 °C (green) become less temperature responsive in backgrounds in which ELF3 is more stable.

Extended Data Fig. 7 The length of the polyQ repeat within the ELF3 PrD influences temperature-dependent speckle formation in vivo.

a, Arabidopsis seedlings expressed GFP-tagged ELF3 variants with no polyQ repeat (Q0), the wild-type polyQ (Q7), a polyQ with 20 or 30 glutamines (Q20 and Q30, respectively), or the PrD replaced by the corresponding region from B. distachyon ELF3 (BdPrD). Seedlings were grown in short photoperiods for 7 days at 17 °C. Roots were imaged by confocal microscopy before and after incubation at 30 °C for 15 min. Scale bar, 40 μm. b, Quantification of the degree of speckle formation in a. Regions of the roots that correspond to the size of individual cells were selected, and the mean, standard deviation and maximum grey values were measured in ImageJ. We assumed that speckle formation would lead to higher grey values and that a higher frequency of speckles within the analysis region would increase the standard deviation. The lower boundary of each box indicates the 25th percentile; the median is marked by a black line within the box; and the top boundary indicates the 75th percentile. Whiskers above and below each box indicate the largest/smallest value up to 1.5 × IQR (interquartile range) from the hinge, and the red dot indicates the mean. Green dots indicate the value for each root measured. a, b, BdPrD, n = 6; Q0, n = 5; Q7, n = 5; Q20, n = 4; Q30, n = 6; all from two independent experiments. c, Relative FT expression in ELF3pro::ELF3-GFP transgenic plants. Twelve-day-old seedlings grown at different temperatures under short-photoperiod conditions were used to analyse the accumulation of FT transcripts at ZT8 by RT–qPCR. Results shown as means ± s.d. (n = 3). The effect of warm temperatures on the induction of FT was weak in transgenic plants containing ELF3 variants with the BdPrD.

Extended Data Fig. 8 Yeast strains show no growth defect after incubation at temperatures used in the speckle-formation experiments, and express detectable levels of ELF3–GFP.

a, Temperature shifts do not affect yeast viability. Yeast cells were grown overnight at 19 °C and shifted to the indicated temperatures for 30 min (as in the temperature shifts used for speckle inductions; Fig. 2d, e). Serial dilutions were spotted onto YPD plates and incubated at 30 °C for one or two days. b, Yeast cells expressing the indicated ELF3–GFP constructs (Q7, Q35 or BdPrD) or an empty vector were grown overnight in selective medium to exponential phase at 30 °C. Cells (at an optical density (OD)600 of approximately 9) were pelleted, washed with sterile water, and lysed in 100 μl SDS-sample buffer with 0.5-mm-diameter glass beads (BioSpec Products) by two rounds of boiling for 2 min and vortexing for 30 s. Protein extracts were centrifuged at 13,000 r.p.m. for 15 min, and supernatants were analysed by western blot using anti-GFP antibody at 1:1,500 dilution (a gift from A. Peden). Western blot signals were developed using enhanced chemiluminescence (GE Healthcare).

Extended Data Fig. 9 ELF3 PrD peptides show phase-change characteristics in vitro.

a, SDS gel analysis (12% polyacrylamide) of the BdELF3 PrD, ELF3 PrD and ELF3 PrD–GFP. M, molecular-weight marker. Proteins were expressed and purified at least ten times with highly reproducible results. b, Phase diagram for the ELF3 PrD peptide with respect to salt and protein concentration. Examples of each phase are shown on the right. c, Droplet formation is dynamic, with droplets re-entering the soluble phase over time, as measured in two biological samples (mean shown) by changes in A340 after droplet formation is induced by dilution from a high-salt to low-salt buffer (50 mM CAPS, pH 9.7, 1 mM TCEP, 500 mM to 150 mM NaCl).

Extended Data Fig. 10 ELF3 PrD droplets fuse.

a, Fusion of two droplets over time, with intensity profiles of each droplet shown below the images. b, Fusion of ELF3 PrD droplets. Two examples are shown (one in each row). c, Example of photobleaching and recovery over time. Images were taken (left to right) before, after and at time points 30 s and 240 s post-photobleaching. d, FRAP recovery curves for c, showing means (green) ± s.d. (tan). Droplet fusions and FRAP experiments were performed five times with reproducible results.

Supplementary information

Supplementary Table 1

| List of genes bound by ELF3 and genes whose expression is controlled by ELF3.

Reporting Summary

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Jung, JH., Barbosa, A.D., Hutin, S. et al. A prion-like domain in ELF3 functions as a thermosensor in Arabidopsis. Nature 585, 256–260 (2020).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing