Receptor kinases of the Catharanthus roseus RLK1-like (CrRLK1L) family have emerged as important regulators of plant reproduction, growth and responses to the environment1. Endogenous RAPID ALKALINIZATION FACTOR (RALF) peptides2 have previously been proposed as ligands for several members of the CrRLK1L family1. However, the mechanistic basis of this perception is unknown. Here we report that RALF23 induces a complex between the CrRLK1L FERONIA (FER) and LORELEI (LRE)-LIKE GLYCOSYLPHOSPHATIDYLINOSITOL (GPI)-ANCHORED PROTEIN 1 (LLG1) to regulate immune signalling. Structural and biochemical data indicate that LLG1 (which is genetically important for RALF23 responses) and the related LLG2 directly bind RALF23 to nucleate the assembly of RALF23–LLG1–FER and RALF23–LLG2–FER heterocomplexes, respectively. A conserved N-terminal region of RALF23 is sufficient for the biochemical recognition of RALF23 by LLG1, LLG2 or LLG3, and binding assays suggest that other RALF peptides that share this conserved N-terminal region may be perceived by LLG proteins in a similar manner. Structural data also show that RALF23 recognition is governed by the conformationally flexible C-terminal sides of LLG1, LLG2 and LLG3. Our work reveals a mechanism of peptide perception in plants by GPI-anchored proteins that act together with a phylogenetically unrelated receptor kinase. This provides a molecular framework for understanding how diverse RALF peptides may regulate multiple processes, through perception by distinct heterocomplexes of CrRLK1L receptor kinases and GPI-anchored proteins of the LRE and LLG family.
Subscribe to Journal
Get full journal access for 1 year
only $3.90 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
The atomic coordinates and structure factors have been deposited in the RCSB Protein Data Bank (PDB). The PDB codes of ANX1ECD, FERECD, ANX2ECD, LLG1 and RALF23–LLG2–FERECD structures are 6A5A, 6A5B, 6A5C, 6A5D and 6A5E, respectively. For gel source images, see Supplementary Figs. 1–3. All other data or materials can be obtained from the corresponding author upon request.
Franck, C. M., Westermann, J. & Boisson-Dernier, A. Plant Malectin-Like Receptor Kinases: From Cell Wall Integrity to Immunity and Beyond. Annu. Rev. Plant Biol. 69, 301-328 (2018).
Pearce, G., Moura, D. S., Stratmann, J. & Ryan, C. A. Jr. RALF, a 5-kDa ubiquitous polypeptide in plants, arrests root growth and development. Proc. Natl Acad. Sci. USA 98, 12843-12847 (2001).
Escobar-Restrepo, J. M. et al. The FERONIA receptor-like kinase mediates male-female interactions during pollen tube reception. Science 317, 656-660 (2007).
Haruta, M., Sabat, G., Stecker, K., Minkoff, B. B. & Sussman, M. R. A peptide hormone and its receptor protein kinase regulate plant cell expansion. Science 343, 408-411 (2014).
Stegmann, M. et al. The receptor kinase FER is a RALF-regulated scaffold controlling plant immune signaling. Science 355, 287-289 (2017).
Zhao, C. et al. Leucine-rich repeat extensin proteins regulate plant salt tolerance in Arabidopsis. Proc. Natl Acad. Sci. USA 115, 13123-13128 (2018).
Feng, W. et al. The FERONIA Receptor Kinase Maintains Cell-Wall Integrity during Salt Stress through Ca2+ Signaling. Curr. Biol. 28, 666-675 e665 (2018).
Ge, Z. et al. Arabidopsis pollen tube integrity and sperm release are regulated by RALF-mediated signaling. Science 358, 1596-1600 (2017).
Capron, A. et al. Maternal control of male-gamete delivery in Arabidopsis involves a putative GPI-anchored protein encoded by the LORELEI gene. Plant Cell 20, 3038-3049 (2008).
Liu, X. et al. The Role of LORELEI in Pollen Tube Reception at the Interface of the Synergid Cell and Pollen Tube Requires the Modified Eight-Cysteine Motif and the Receptor-Like Kinase FERONIA. Plant Cell 28, 1035-1052 (2016).
Li, C. et al. Glycosylphosphatidylinositol-anchored proteins as chaperones and co-receptors for FERONIA receptor kinase signaling in Arabidopsis. elife 4, e06587 (2015).
Shen, Q., Bourdais, G., Pan, H., Robatzek, S. & Tang, D. Arabidopsis glycosylphosphatidylinositol-anchored protein LLG1 associates with and modulates FLS2 to regulate innate immunity. Proc. Natl Acad. Sci. USA 114, 5749-5754 (2017).
Du, S., Qu, L. J. & Xiao, J. Crystal structures of the extracellular domains of the CrRLK1L receptor-like kinases ANXUR1 and ANXUR2. Protein Sci. 27, 886-892 (2018).
Moussu, S., Augustin, S., Roman, A. O., Broyart, C. & Santiago, J. Crystal structures of two tandem malectin-like receptor kinases involved in plant reproduction. Acta Crystallogr. D 74, 671-680 (2018).
Schallus, T. et al. Malectin: a novel carbohydrate-binding protein of the endoplasmic reticulum and a candidate player in the early steps of protein N-glycosylation. Mol. Biol. Cell 19, 3404-3414 (2008).
Pearce, G., Yamaguchi, Y., Munske, G. & Ryan, C. A. Structure-activity studies of RALF, Rapid Alkalinization Factor, reveal an essential—YISY—motif. Peptides 31, 1973-1977 (2010).
Liu, P., Haruta, M., Minkoff, B. B. & Sussman, M. R. Probing a Plant Plasma Membrane Receptor Kinase's Three-Dimensional Structure Using Mass Spectrometry-Based Protein Footprinting. Biochemistry 57, 5159-5168 (2018).
Chinchilla, D., Bauer, Z., Regenass, M., Boller, T. & Felix, G. The Arabidopsis receptor kinase FLS2 binds flg22 and determines the specificity of flagellin perception. Plant Cell 18, 465-476 (2006).
Sun, Y. et al. Structural basis for flg22-induced activation of the Arabidopsis FLS2-BAK1 immune complex. Science 342, 624-628 (2013).
Mecchia, M. A. et al. RALF4/19 peptides interact with LRX proteins to control pollen tube growth in Arabidopsis. Science 358, 1600-1603 (2017).
Fabrice, T. N. et al. LRX Proteins Play a Crucial Role in Pollen Grain and Pollen Tube Cell Wall Development. Plant Physiol. 176, 1981-1992 (2018).
Sede, A. R. et al. Arabidopsis pollen extensins LRX are required for cell wall integrity during pollen tube growth. FEBS Lett. 592, 233-243 (2018).
Stegmann, M. & Zipfel, C. Complex regulation of plant sex by peptides. Science 358, 1544-1545 (2017).
Wang, Q. S. et al. Upgrade of macromolecular crystallography beamline BL17U1 at SSRF. Nucl. Sci. Tech. 29, 68 (2018).
Otwinowski, Z. & Minor, W. Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 276, 307-326 (1997).
McCoy, A. J. et al. Phaser crystallographic software. J. Appl. Crystallogr. 40, 658-674 (2007).
Emsley, P. & Cowtan, K. Coot: model-building tools for molecular graphics. Acta Crystallogr. D 60, 2126-2132 (2004).
Adams, P. D. et al. PHENIX: Building new software for automated crystallographic structure determination. Acta Crystallogr. D 58, 1948-1954 (2002).
Deslauriers, S. D. & Larsen, P. B. FERONIA is a key modulator of brassinosteroid and ethylene responsiveness in Arabidopsis hypocotyls. Mol. Plant 3, 626-640 (2010).
Duan, Q., Kita, D., Li, C., Cheung, A. Y. & Wu, H. M. FERONIA receptor-like kinase regulates RHO GTPase signaling of root hair development. Proc. Natl Acad. Sci. USA 107, 17821-17826 (2010).
Nakagawa, T. et al. Improved Gateway binary vectors: high-performance vectors for creation of fusion constructs in transgenic analysis of plants. Biosci. Biotechnol. Biochem. 71, 2095-2100 (2007).
We thank J. He (Shanghai Synchrotron Radiation Facility (SSRF)) and G. Lin for assistance with X-ray diffraction data collection; C. Zhou (Tsinghua University Branch of China National Center for Protein Sciences Beijing) for assistance with SV-AUC data collection and analysis; I. Hang (Vienna Biocenter Core Facilities), Y. Zhu and S. Liu (Tsinghua University) for help with protein purification; and R. Hückelhoven (Technical University Munich), in whose department part of the work was performed. The surface plasmon resonance equipment was kindly provided by the EQ-BOKU VIBT and the BOKU Core Facility for Biomolecular and Cellular Analysis. This research was funded by Projects of International Cooperation and Exchanges NSFC (31420103906), Chinese Ministry of Science and Technology (2015CB910200) and National Natural Science Foundation of China (31421001) (J.C.), the Gatsby Charitable Foundation (C.Z.), the University of Zürich (C.Z.), the European Research Council under the Grant Agreements 309858 and 773153 (grants PHOSPHinnATE and IMMUNO-PEPTALK to C.Z.), the Deutsche Forschungsgemeinschaft (Fellowship STE 2448/1 to M.S.), the Technical University of Munich (M.S.), the European Molecular Biology Organization (EMBO Long-Term Fellowship 100-2017 to T.A.D.) and the Austrian Academy of Science through the Gregor Mendel Institute (Y.B). K.P. was supported by the Vienna Science and Technology Fund Project (LS17-047).
The authors declare no competing interests.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Peer review information Nature thanks Osamu Nureki, Michael R. Sussman and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Extended data figures and tables
Extended Data Fig. 1 Loss of LLG1 phenocopies fer in flg22-triggered formation of the FLS2–BAK1 complex, and ROS production.
a, Tissue-specific expression patterns of FER (AT3G51550), LLG1 (AT5G56170), LLG2 (AT2G20700) and LLG3 (AT4G28280) were obtained using Genevestigator (www.genevestigator.com) and are based on the ‘AT_mRNASeq_ARABI_GL-1’ dataset. FER is ubiquitously expressed throughout most tissues, including leaves. Similarly, LLG1 shows strong expression in vegetative tissue—unlike LLG2 and LLG3, which are predominantly expressed in reproductive tissue. All assays in this study were performed using seedlings or leaf tissue. LLG1, LLG2 and LLG3 expression in these tissues is highlighted by a black box. b, Loss of LLG1 results in reduced flg22-triggered ROS production. ROS burst was measured in leaf discs of the indicated genotypes after elicitation with 100 nM flg22. Values are represented as total photon counts over 40 min. Biologically independent leaf discs, n = 8 for all genotypes. Values shown as mean ± s.e.m. Letters indicate the result of a post hoc Tukey test. Columns with same letters are statistically indistinguishable (>95% confidence); see ‘Supplementary statistical information’ in Supplementary Information for multiple-comparison P values. The experiments were repeated at least three times with similar results. c, LLG1 contributes to flg22-induced formation of the FLS2–BAK1 complex. Arabidopsis Col-0, fer-2 or llg1-2 seedlings were treated with 100 nM flg22 for 10 min before immunoprecipitating FLS2. Western blots for protein detection were probed with anti-FLS2 and anti-BAK1 antibodies. The experiments were repeated at least three times with similar results. See Supplementary Fig. 1 for gel source data. Source Data
a, b, Quantification of the binding affinities of RALF23 with LLG2 or LLG3, or LLG2 or LLG3 with FERECD by ITC. The assays were performed as described in Fig. 1e. c, d, Analyses of LLG2 +FERECD or LLG3 + FERECD, and RALF23 + LLG2 + FERECD or RALF23 + LLG3 + FERECD proteins by SV-AUC. The assays were as described in Fig. 1d. e, Analysis of the anti-FER antibody. The anti-FER antibody was raised against two peptides derived from the extracellular domain of FER in rabbits (Eurogentec). The specificity of the antibody was tested by probing samples of crude protein extract from Col-0, fer-4 and fer-4 FER–GFP seedlings. A specific band for FER in Col-0 plants, migrating between 100 and 130 kDa was detected, which was absent in fer-4 mutants. Furthermore, a shifted band migrating at a higher molecular weight was detected in fer-4 FER–GFP plants, corresponding to the size increase by the addition of the GFP fusion tag. The experiments were repeated at least three times with similar results. See Supplementary Fig. 1 for gel source data.
a, The 2Fo - Fc electron density map (blue mesh) of the final refined RALF23–LLG2–FERECD model contoured at 1.2σ (using COOT) shown in two orientations. The two RALF23–LLG2–FERECD molecules in an asymmetric unit are shown in yellow. b Crystal packing of RALF23–LLG2–FERECD in the P1 space group shown in two orientations. Colour codes are indicated. c, A stereo view of the structure of RALF23–LLG2–FERECD. d, The 2Fo - Fc electron density map (blue mesh) of the final refined model contoured at 1.2σ (using COOT) shows glycosylation of Asn142, Asn305 and Asn345 of FERECD. NAG, N-acetyl-d-glucosamine. e, The 2Fo - Fc electron density map (blue mesh) contoured at 1.2σ (using COOT) shows glycosylation of Asn52 of LLG2. f, Mapping of the glycosylation sites on the structure of RALF23–LLG2–FERECD. Colour codes and the glycosylated residues are indicated.
Extended Data Fig. 4 Structural superposition of apo-FERECD with apo-ANX1ECD and ANX2ECD, malectin–nigerose complex from X. laevis and some details of the structure of the RALF23–LLG2–FERECD complex.
a, Structural superposition of apo-FERECD (cyan), apo-ANX1ECD (blue) and apo-ANX2ECD (green). Structural alignment of apo-ANX1ECD(26–410) or apo-ANX2ECD (27–414) with apo-FERECD(29–423) with root mean square deviation (r.m.s.d.) 0.876 and 0.754 Å, respectively. α-helix 3 and β-sheet 16 are labelled. b, Structural comparison of the malectin domains of FERECD (cyan/light blue) with malectin–nigerose (NGR) (grey and yellow, respectively) complex from X. laevis (PDB code 2K46). The malectin domain B of FER (residues 197–447) is used as the template for alignment with malectin domain A of FER (residues 27–196) with an r.m.s.d. of 4.065 Å, and with malectin (residues 1–190) with an r.m.s.d. of 16.157 Å. Black box and arrow show the LLG2-binding region in malectin domain B of FERECD. The red box marks the NGR-binding pocket of malectin–NGR. c, Saccharide-recognizing residues are not conserved among malectins. Sequence alignment of malectin domain A and domain B from FER, and malectin from X. laevis. Red dots mark residues that form an NGR-perception pocket in malectin. d, Structural superposition of FERECD(29–423) and RALF23–LLG2–FERECD(29–423) with an r.m.s.d. of 0.441 Å. Colour codes are indicated. e, The N-terminal region (residues 4–17, brown) of RALF23 binds to a surface groove of LLG2 (shown in electrostatics). Important residues of RALF23 recognized by LLG2 are indicated. f The 2Fo-Fc electron density map (blue mesh) around the N-terminal region of RALF23 (residues 4–17) in the final refined model contoured at 1.2σ (using COOT). The structural model of the N-terminal region is highlighted within the dashed cyan frame, and the LLG2-interacting residues are labelled in white. g, FERECD (cartoon in cyan) interacts with both LLG2 (surface in purple) and RALF23 (surface in brown). Some secondary structures of FERECD are indicated.
Extended Data Fig. 5 Structure-based sequence alignment of proteins of the CrRLK1L family in terms of their extracellular domains.
Secondary structure elements in FERECD are marked above the alignments. The regions related to RALF23–LLG2–FERECD interaction are boxed with green dashed lines. The amino acids of FER that are important for interactions with RALF23 or LLG2 are marked with blue dots.
Extended Data Fig. 6 RALF23N is sufficient to induce interaction between FERECD and LLG1, LLG2 or LLG3, and RALF23C reinforces the RALF23–LLG1–FERECD, RALF23–LLG2–FERECD and RALF23–LLG3–FERECD complexes in vitro.
a, Top, surface plasmon resonance sensorgram for the LLG2–RALF23N binding analyses with the corresponding association (ka) and dissociation (kd) rate constant, and dissociation constants KD. Multi-cycle kinetics were used to validate the interaction. The analyte was injected in a concentration range of 50–800 nM. The sensorgram presents curves after background subtraction (Fc 2 - 1). Bottom, steady‐state curve-fitting analysis for the LLG2–RALF23N interaction was used for the KD calculation. The assays were performed three times independently with similar results. b–e, Quantification of the binding affinities of different variants of RALF23 with LLG1, LLG2 or LLG3 by ITC. The assays were performed three times independently with similar results, as described in Fig. 1e. Excess RALF23N was pre-incubated with LLG1, LLG2 or LLG3 for quantification of interaction with FERECD. Excess LLG1, LLG2 or LLG3 was incubated with RALF23 for interaction with FERECD.
Extended Data Fig. 7 LLG2 interacts with RALF peptides from subfamily 1 but not with those from subfamily 2.
a, Full-length RALF23 is required for biological activity. Five-day-old Col-0 seedlings were treated with 1 μM RALF23, 1 μM RALF23N or RALF23 residues 18–50 (RALF23C) for 7 days before measuring fresh weight. The relative fresh weight of seedlings compared to the MS control is shown. n = 12 biologically independent seedlings for all genotypes. Values are shown ± s.d. Letters indicate the result of a post hoc Tukey test; columns with same letters are statistically indistinguishable (>95% confidence). See ‘Supplementary statistical information’ in Supplementary Information for multiple-comparison P values. The experiments were repeated at least three times with similar results. b, N-terminal sequence alignment of mature RALF peptides (subfamily 1) from Arabidopsis. Conserved or similar residues are boxed with a red background or shown in red font, respectively. c, Summary of the sequences of the RALF peptides used. d, Quantification of LLG2 interaction with RALF peptides from subfamily 1 by ITC. The calculated stoichiometries (n) and the dissociation constants (Kd) are indicated. The assays were performed three times independently with similar results as described in Fig. 1e. e, N-terminal sequence alignment of RALF peptides (subfamily 2) from Arabidopsis. Conserved or similar residues are boxed in blue or shown in red font, respectively. f, Quantification of RALF11N, RALF12N, RALF13N or RALF21N interaction with LLG2 by ITC. The calculated stoichiometries (n) and the dissociation constants (Kd) are indicated. The assays were performed three times independently with similar results as described in Fig. 1e. Source Data
Extended Data Fig. 8 Structure-based sequence alignment of the Arabidopsis LRE and LLG family and RALF peptides in subfamily 1 induce LLG1–FERECD interaction in vitro.
a, Analyses of indicated proteins by SV-AUC. The assays were performed three times independently with similar results, as described in Fig. 1d. The profiles of SV-AUV for RALF1, RALF4, RALF14, RALF19 and RALF34 with LLG1 and FERECD are shown in cyan. b, Sequence alignment of LLG1, LLG2, LLG3 and LRE from Arabidopsis. Secondary structure elements in apo-LLG1 are marked on the top of the alignments. Four pairs of disulfide bonds are labelled underneath the alignments in green. The signal peptide, conformationally flexible region and GPI-anchor region are coloured as indicated. The conserved amino acids involved in the interaction of LLG1, LLG2 or LLG3 with RALF23 are marked below the alignment with dots, and the non-conserved Arg120 of LRE is marked below the alignment with a black dot. c, Analyses of indicated proteins by SV-AUC. The assays were performed three times independently with similar results as described in Fig. 1d. The profiles of SV-AUV for RALF23 + LRE + FERECD are shown in cyan.
Extended Data Fig. 9 Mutagenesis analysis of RALF23–LLG1–FERECD interaction and RALF23–LLG2–FERECD interaction.
a, Quantification of RALF23 binding to LLG1(G114R) indicated by ITC. b, Mutagenesis analyses of GST–RALF23N interaction with LLG2 by the assays described in Fig. 4c. Left, SDS–PAGE of LLG2 pulled down by wild-type GST–RALF23N or GST–RALF23N mutants, as indicated. Right, SDS–PAGE of wild-type LLG2 or LLG2 mutants pulled down by GST–RALF23N. See Supplementary Fig. 2 for gel source data. c, Quantification of RALF23N + LLG1(N118Y) binding to FERECD, and RALF23N + LLG1 binding to the four FER mutants indicated, by ITC. d, Quantification of binding of RALF23N + LLG2 to the indicated FER mutants. e, Summary of ITC results on RALF23N + LLG2 interaction with wild-type or mutant FERECD. f, Expression analysis of pFER::FER–GFP and the respective point mutants in the fer-4 mutant background. Western blots for protein detection were probed with anti-GFP antibodies. CBB, Coomassie brilliant blue. See Supplementary Fig. 1 for gel source data. g, Structure-guided LLG1 mutants disrupt the RALF23-induced complex with FER in vivo. N. benthamiana leaves co-expressing FER–GFP and SP–mRFP–LLG1 (wild type or mutant variants, as indicated) were treated with buffer with or without 5 μM RALF23 for 10 min before immunoprecipitation. Experiments were repeated at least three times with similar results. See Supplementary Fig. 3 for gel source data.
About this article
Cite this article
Xiao, Y., Stegmann, M., Han, Z. et al. Mechanisms of RALF peptide perception by a heterotypic receptor complex. Nature 572, 270–274 (2019). https://doi.org/10.1038/s41586-019-1409-7
Molecular Plant-Microbe Interactions® (2021)
Molecular Plant (2021)
Glycosylphosphatidylinositol-anchored proteins mediate the interactions between pollen/pollen tube and pistil tissues
Keeping up with the RALFs: how these small peptides control pollen–pistil interactions in Arabidopsis
New Phytologist (2021)