Allosteric activation of the co-receptor BAK1 by the EFR receptor kinase initiates immune signaling

eLife assessment

This manuscript reports important in vitro biochemical and in planta experiments to study the receptor activation mechanism of plant membrane receptor kinase complexes through the non-catalytic function of an active protein kinase. Several lines of evidence convincingly show that one such receptor kinase with pseudokinase-like function, the immune receptor EFR achieves an active conformation following phosphorylation by a co-receptor kinase, and then in turn activates the co-receptor kinase allosterically to enable it to phosphorylate down-stream signaling components. This manuscript will be of interest to scientists focusing on cell signalling and allosteric regulation.

https://doi.org/10.7554/eLife.92110.4.sa0

Significance of the findings:

Important: Findings that have theoretical or practical implications beyond a single subfield

Landmark
Fundamental
Important
Valuable
Useful

Strength of evidence:

Convincing: Appropriate and validated methodology in line with current state-of-the-art

Exceptional
Compelling
Convincing
Solid
Incomplete
Inadequate

During the peer-review process the editor and reviewers write an eLife Assessment that summarises the significance of the findings reported in the article (on a scale ranging from landmark to useful) and the strength of the evidence (on a scale ranging from exceptional to inadequate). Learn more about eLife Assessments

Abstract
Introduction
Results
Discussion
Materials and methods
Appendix 1
Data availability
References
Article and author information
Metrics

Abstract

Transmembrane signaling by plant receptor kinases (RKs) has long been thought to involve reciprocal trans-phosphorylation of their intracellular kinase domains. The fact that many of these are pseudokinase domains, however, suggests that additional mechanisms must govern RK signaling activation. Non-catalytic signaling mechanisms of protein kinase domains have been described in metazoans, but information is scarce for plants. Recently, a non-catalytic function was reported for the leucine-rich repeat (LRR)-RK subfamily XIIa member EFR (elongation factor Tu receptor) and phosphorylation-dependent conformational changes were proposed to regulate signaling of RKs with non-RD kinase domains. Here, using EFR as a model, we describe a non-catalytic activation mechanism for LRR-RKs with non-RD kinase domains. EFR is an active kinase, but a kinase-dead variant retains the ability to enhance catalytic activity of its co-receptor kinase BAK1/SERK3 (brassinosteroid insensitive 1-associated kinase 1/somatic embryogenesis receptor kinase 3). Applying hydrogen-deuterium exchange mass spectrometry (HDX-MS) analysis and designing homology-based intragenic suppressor mutations, we provide evidence that the EFR kinase domain must adopt its active conformation in order to activate BAK1 allosterically, likely by supporting αC-helix positioning in BAK1. Our results suggest a conformational toggle model for signaling, in which BAK1 first phosphorylates EFR in the activation loop to stabilize its active conformation, allowing EFR in turn to allosterically activate BAK1.

Introduction

Plants and metazoans respond to extracellular signals through different sets of plasma membrane receptors. Whereas the G-protein coupled receptor family expanded in metazoans, the receptor kinase (RK) and receptor protein (RP) families expanded in plants (Shiu and Bleecker, 2003; Shiu and Bleecker, 2001; Trusov and Botella, 2016). Plant RKs resemble metazoan receptor tyrosine kinases (RTKs) in their organization, containing an extracellular ligand-sensing domain that is coupled to an intracellular kinase domain by a single-pass transmembrane helix (Hohmann et al., 2017; Lemmon and Schlessinger, 2010). In contrast to RTKs, however, the intracellular kinase domain of plant RKs is a serine/threonine kinase domain most closely related to interleukin-1 receptor associated kinases (IRAKs)/Pelle kinases (Shiu and Bleecker, 2003; Shiu and Bleecker, 2001) – although dual-specificity kinase activity has been reported in some cases (Liu et al., 2018; Luo et al., 2020; Macho et al., 2014; Oh et al., 2009; Perraki et al., 2018).

The leucine-rich repeat (LRR)-RKs are the most extensively studied sub-family of Arabidopsis thaliana (hereafter Arabidopsis) RKs. They primarily sense peptide ligands that regulate growth and development, or molecular patterns that are released during damage or microbial infection (Couto and Zipfel, 2016). Ligand-binding LRR-RKs have long ectodomains (>20 LRRs), and recruit short LRR ectodomain-containing co-receptor kinases (coRKs) upon ligand perception (Bender and Zipfel, 2023; Hohmann et al., 2017), most of which belong to the SERK (somatic embryogenesis receptor kinase) family. One of the most well characterized LRR-RKs is BRI1 (brassinosteroid insensitive 1), which recruits the coRKs SERK1 and SERK3/BAK1 (BRI1-associated kinase 1) upon brassinosteroid (BR) perception (Albrecht et al., 2008; He et al., 2000; Hothorn et al., 2011; Nam and Li, 2002; Santiago et al., 2013). After ligand perception, the intracellular kinase domains of BRI1 and BAK1 reciprocally trans-phosphorylate each other in their activation loops (Hohmann et al., 2017), and BAK1 then phosphorylates BRI1 in its juxtamembrane segment and C-tail to propagate BR signaling (Wang et al., 2008; Wang et al., 2005). BRI1 exemplifies a set of BAK1-dependent LRR-RKs that have RD-type intracellular kinase domains and require their catalytic activity to signal (Cao et al., 2013; Kosentka et al., 2017; Taylor et al., 2016). Other LRR-RKs such as EFR (elongation factor Tu receptor) instead have non-RD intracellular kinases, and are thought not to trans-phosphorylate their associated coRK following ligand perception (Schwessinger et al., 2011). Moreover, kinase inactive mutants of EFR (D849N or K851E) retain signaling function, a result that challenges the generality of the reciprocal trans-phosphorylation model (Bender et al., 2021).

EFR perceives the pathogen-associated molecular pattern (PAMP) elongation factor Tu, or its active peptide epitope elf18 (Zipfel et al., 2006). PAMP perception triggers heterodimerization with the coRK BAK1, resulting in phosphorylation of the EFR intracellular kinase domain (Bender et al., 2021; Roux et al., 2011; Schulze et al., 2010; Schwessinger et al., 2011). The signal is subsequently relayed to the cytoplasmic kinases BIK1 (botrytis-induced kinase 1) and PBL1 (PBS1-like 1) (Li et al., 2014; Lu et al., 2010; Ranf et al., 2014). The resulting immune signaling activation elicits a battery of cellular responses, including an apoplastic oxidative burst (hereafter oxidative burst), Ca²⁺-influx, callose deposition, MAPK (mitogen activated protein kinase) activation, and transcriptional reprogramming (DeFalco and Zipfel, 2021).

Initial studies in vitro identified phosphorylation sites in EFR that result either from auto-phosphorylation or from trans-phosphorylation by BAK1 (Wang et al., 2014). More recently, in vivo phosphorylation sites on EFR were identified by immunoprecipitating EFR-GFP from elf18-treated seedlings (Bender et al., 2021). One was a serine (S888) in the activation loop (A-loop), at which phosphorylation was consistently observed in vitro and in vivo (Bender et al., 2021; Wang et al., 2014). Despite the non-RD nature of EFR, and its ability to signal independently of catalytic activity, ligand-inducible phosphorylation in the A-loop (S887/S888) surprisingly proved indispensable for signaling (Bender et al., 2021). Furthermore, a functionally important tyrosine in EFR (Y836) – conserved in subdomain VIa of many eukaryotic protein kinases (Lai et al., 2016; Luo et al., 2020; Perraki et al., 2018) – was found to be phosphorylated after ligand treatment in vivo (Macho et al., 2014). A phospho-ablative Y836F mutation in EFR blocks immune signaling and resistance against the phytopathogenic bacterium Pseudomonas syringae (Macho et al., 2014). Thus, although EFR function does not require its catalytic activity, both A-loop phosphorylation and phosphorylation of the VIa subdomain tyrosine (VIa-Tyr) appear to be crucial. The mechanistic importance of these phosphorylation events remains poorly understood, but we hypothesized that they switch the EFR kinase domain into an active-like conformation that allosterically activates BAK1 in the EFR-BAK1 complex through a mechanism similar to that described for pseudokinases (Mace and Murphy, 2021; Sheetz and Lemmon, 2022) – ultimately promoting BAK1’s activity toward its substrate BIK1.

Here, we tested the hypothesis that EFR is an allosteric regulator of BAK1 using a range of different approaches. Our in vitro studies revealed that forcing dimerization of the EFR and BAK1 intracellular domains allosterically enhances BAK1 activity. Using a homology-guided approach, we designed mutations to stabilize the active-like conformation of the EFR kinase domain and found that they restore functionality of EFR variants that cannot be phosphorylated in the A-loop (EFR^SSAA) or at Y836 (EFR^Y836F). We also used hydrogen-deuterium exchange mass spectrometry (HDX-MS) to analyze conformational dynamics, revealing that the Y836F mutation hampers the ability of the EFR kinase domain to adopt an active-like conformation. Collectively, our findings argue that the active conformation of the EFR kinase domain is required for allosteric activation of the BAK1 kinase domain. Finally, we present evidence suggesting that EFR activates BAK1 allosterically by supporting αC-helix positioning in BAK1.

Results

The EFR intracellular domain allosterically activates BAK1 in vitro

To test the hypothesis that EFR enhances BAK1 catalytic activity allosterically, we used rapamycin (Rap)-induced dimerization (RiD) to induce a complex between the isolated EFR and BAK1 intracellular domains (Banaszynski et al., 2005; Kim et al., 2021). The RiD system was previously applied in planta, maintaining membrane-association by N-terminal myristoylation (Kim et al., 2021). For our in vitro experiments, the myristoylation sites were excluded to facilitate the purification of recombinant protein. We tested two LRR-RKs: EFR and BRI1, fusing GFP C-terminally and FKBP (FK506 binding protein) N-terminally to their intracellular domains. BAK1 instead was N-terminally tagged with FRB (FKBP-Rap-binding). We first confirmed that adding Rap induced formation of FKBP-EFR/FRB-BAK1 dimers in size exclusion chromatography experiments (Figure 1A). We then assessed the effect of inducing RK/coRK kinase complex formation on their ability to phosphorylate BIK1^D202N. BIK1 was chosen as it is a reported substrate of both, EFR/BAK1 and BRI1/BAK1 complexes (Lin et al., 2013). Rap addition increased BIK1^D202N phosphorylation when the BRI1 or EFR kinase domains were dimerized with BAK1 (Figure 1B and C). Kinase-dead variants with the catalytic residue (HRD-aspartate) replaced by asparagine (EFR^D849N and BRI1^D1009N), had distinct effects. BRI1^D1009N failed to enhance BIK1 phosphorylation substantially, whereas EFR^D849N retained some ability to do so (Figure 1B and C). The same trend was observed for phosphorylation of the BAK1 kinase domain itself, indicating that EFR also enhances BAK1 autophosphorylation activity (Figure 1B and C).

Figure 1 with 1 supplement see all

Download asset Open asset

EFR facilitates BIK1 trans-phosphorylation by BAK1 non-catalytically.

The kinase domains of coRK BAK1 and the RKs BRI1 and EFR were tagged with RiD domains and purified from *E. coli* λPP cells. (A) Recombinantly expressed RiD-tagged kinase domains were mixed together at equimolar ratios (2 µM), with or without addition of 10 µM Rap as well as 10 mM MgCl₂ and 1 mM AMP-PNP. (B) RK and coRK were mixed at an equimolar ratio at 50 nM, kinase-dead BIK1^D202N substrate was added at 500 nM. Reactions were carried out at RT for 10 min with 0.5 μCi [γ-³²P]ATP, 100 μM ATP and 2.5 mM each of MgCl₂ and MnCl₂. Addition of 1 μM Rap enhanced transphosphorylation of BIK1 by EFR and BRI1. Kinase-dead BRI1 failed to enhance BIK1 transphosphorylation, but kinase-dead EFR retained some ability to do so. A similar trend was observed for (auto)phosphorylation of BAK1 itself. (C) Quantification of band intensities over three independent experiments of which a representative is shown in B.

Figure 1—source data 1 Raw data for gel images in Figure 1A.: https://cdn.elifesciences.org/articles/92110/elife-92110-fig1-data1-v1.zip
Download elife-92110-fig1-data1-v1.zip
Figure 1—source data 2 Raw data for autoradiography in Figure 1B.: https://cdn.elifesciences.org/articles/92110/elife-92110-fig1-data2-v1.zip
Download elife-92110-fig1-data2-v1.zip

The increased BIK1 trans-phosphorylation observed in these in vitro RiD experiments could arise either from direct enhancement of BAK1 activity by the RK kinase domain or from more efficient BIK1^D202N recruitment to the dimerized RK/coRK complex. We found that BIK1^D202N did not co-elute with the EFR-BAK1 complex in size exclusion studies, indicating that stable EFR/BAK1/BIK1 trimers do not form in vitro (Figure 1A). Our findings therefore support the hypothesis that EFR increases BIK1 phosphorylation by allosterically activating the BAK1 kinase domain. Although EFR’s catalytic activity is dispensable for this effect – and for immune signaling – EFR must be phosphorylated at S887/S888 in the A-loop and Y836 to signal in vivo (Bender et al., 2021; Macho et al., 2014). The phospho-ablative EFR A-loop mutant EFR^SSAA fails to activate the RK/coRK complex, as ligand-induced phosphorylation of BAK1 S612 – a mark for active BAK1-containing receptor complexes (Perraki et al., 2018) – is obstructed despite retaining the ability to associate with BAK1 in a ligand-dependent manner (Bender et al., 2021). To test whether the requirement for Y836 phosphorylation is similar, we immunoprecipitated EFR-GFP and EFR^Y836F-GFP from mock- or elf18-treated seedlings and probed co-immunoprecipitated BAK1 for S612 phosphorylation. EFR^Y836F also obstructed the induction of BAK1 S612 phosphorylation (Figure 1—figure supplement 1), indicating that EFR^Y836F and EFR^SSAA impair receptor complex activation.

EFR VIa-Tyr mutation affects dynamics of regulatory kinase subdomains

Y836 in EFR corresponds in sequence and structural alignments with Y156 in the canonical kinase PKA (protein kinase A) (Figure 2—figure supplements 1 and 2A), which is conserved in many protein kinases. We therefore speculated that Y836 may function as a pivot for key movements of the regulatory αC-helix, controlling conformational toggling of inactive/active transitions as described for its PKA counterpart (Tsigelny et al., 1999). To test whether an EFR Y836F mutation in EFR interferes with assembly of the active-like EFR kinase conformation we used HDX-MS. Plotting differential HDX (Δ%EX) between unphosphorylated EFR and EFR^Y836F (aa684-1031) identifies one region that is stabilized in EFR^Y836F compared to wild-type and two that are destabilized (Figure 2A and B). The region stabilized in EFR^Y836F contains the N-terminal part of the A-loop, implying that this region makes more extensive contacts with the kinase core than in the unphosphorylated wild-type kinase domain. Alternatively, the decreased deuterium uptake in the EFR^Y836F A-loop could arise from stabilization of a short A-loop α-helix, as seen in the inactive conformation of many kinases (such as in epidermal growth factor receptor). The two regions that become more structurally flexible in EFR^Y836F include the β3-αC loop and the catalytic loop plus C-terminal end of the αE-helix – where Y836 is located (Figure 2A and B). Because these two regions are both important for conformational switching, the HDX-MS results suggest that Y836 is important in regulating kinase domain allosteric transitions. Consistent with this, crystal structures as well as AlphaFold2 models of active kinase conformations (Faezov and Dunbrack, 2023) suggest that the side-chain of this tyrosine forms hydrogen-bonds (H-bond) with the αC-β4 loop backbone to establish an inter-lobe connection (Figure 2—figure supplement 1). Loss of this inter-lobe connection may underlie the observed alterations in EFR^Y836F conformational dynamics in EFR^Y836F, resulting from inaccessibility of the active-like conformation.

Figure 2 with 3 supplements see all

Download asset Open asset

EFR^Y836F and EFR^SSAA impair the active kinase conformation, which is required for signaling function.

(A) (left) HDX-MS results for unphosphorylated EFR and EFR^Y836F protein. The difference in percent H/D exchange in wild type EFR and EFR^Y836F is expressed as the Δ%EX (wild type EFR – EFR^Y836F), with the positive and negative Δ%EX indicating more stabilized and destabilized regions in EFR^Y836F, respectively, compared to wild-type EFR. The Δ%EX values at different labeling time points are shown as colored lines, as indicated in the figure. The horizontal dotted black lines indicate the 98% confidence interval for the Δ%EX data (±7.18%, corresponding to ±0.4 Da difference between wild type and Y836F percent exchange) calculated as described previously (Houde et al., 2011). Regions with Δ%EX values that exceed this confidence limit are indicated as colored bars in the figure, including the β3-αC loop (orange), the catalytic loop plus part of αE (purple), and the A-loop (blue). These regions are colored in the AlphaFold2-derived model of the EFR kinase domain shown at right, in which Y836 is shown as a purple sphere. All data are the average of three independent biological repeats (n=3) with three technical repeat experiments each. A summary of the HDX-MS analysis is presented in Table 3. (B) HDX-MS analysis of representative peptides from regions with significantly different HD exchange. Frames are color-coded according to regions in A. Amino acid range of the peptides in full length EFR are indicated in the top left corner and the sequence below. (**C, D**) Secondary site mutation EFR F761[H/M] partially restores function of EFR^Y836F (C) and EFR^SSAA (D). Full length EFR and its variants were expressed transiently in *N. benthamiana* and their function was tested in an oxidative burst assay. EFR F761H partially restored oxidative bursts of EFR^Y836F and EFR^SSAA. Outliers are in indicated by asterisk in addition to the outlier itself and are included in statistical analysis; Statistical test: Kruskal-Wallis test (p<2.2*10^–16 in C, p=1.163*10^–7 in D), Dunn’s post-hoc test with Benjamin-Hochberg correction (p ≤ 0.05) Groups with like lowercase letter designations are not statistically different.

Kinase activating mutations restore partial function of EFR^Y836F and EFR^SSAA

To build on these results, we next sought to rescue the active-like kinase conformation of EFR^Y836F by introducing activating mutations. Kinase activating mutations are well known in human disease, and cause different malignancies (Foster et al., 2016; Hu et al., 2015). Because the HDX-MS data indicated a destabilized αC-helix, we were specifically interested in activating mutations thought to stabilize the kinase αC-helix in a ‘swung-in’ state, with the goal of making homologous changes in EFR. Such mutations were systematically identified in BRAF/CRAF by searching oncogenes using phenylalanine substitutions (Hu et al., 2015). We exploited these oncogenic BRAF mutations for homology-based design of putative activating EFR mutations at corresponding positions that could function as intragenic suppressors of EFR^Y836F (Table 1 and Figure 2—figure supplement 2A and B). Since EFR already carries a phenylalanine at the position corresponding to L505 in BRAF, we generated EFR^F761[H/M] to resemble mutations known to potently activate BRAF (Hu et al., 2015).

Table 1

Homology-based design of putative intragenic suppressor mutations for EFR.

The list contains the residue number of EFR and the analogous oncogenic mutation in BRAF, as well as a short description of the mode of action of the oncogenic mutation. See Figure 2—figure supplement 1B for structural locations.

EFR mutation	Analogous oncogenic mutation	Mode of action	Source
L743F	BRAF L485F	Extended hydrophobic interaction network along the αC-helix	Hu et al., 2015
F761[H/M]	BRAF L505[F/H/M]	Enforcement of hydrophobic interactions in the regulatory spine	Hu et al., 2015
ΔNLLKH	BRAF ΔNVTAP	Shortening of β3-αC-helix loop, pulling ‘in’ the αC-helix	Foster et al., 2016
L873E	BRAF V600E	Upward bending of C-terminal αC-helix end	Hu et al., 2015

The putative activating mutations were introduced into either EFR^WT or EFR^Y836F and transiently expressed in Nicotiana benthamiana to test receptor function. As expected, heterologous expression of EFR^WT but not of EFR^Y836F conferred elf18 sensitivity (Figure 2B, Figure 2—figure supplement 2C). When introduced on their own into EFR^WT, the single mutations L743F, F761[HM], and L873E were all consistent with wild-type EFR function, supporting an elf18-induced oxidative burst, except for the ΔNLLKH deletion (Figure 2—figure supplement 2C). Intriguingly, the F761[H/M] mutations were also able to partially restore the ability of EFR^Y836F to support an elf18-induced oxidative burst (Figure 2B and Figure 2—figure supplement 2C). This observation also extended to EFR^SSAA (Figure 2C), arguing that the F761[H/M] mutations support the ability of EFR to trans-activate BAK1.

EFR A-loop phosphorylation was reported previously to be indispensable for EFR function, and led to the hypothesis that it controls conformational switching, despite EFR being a non-RD kinase (Bender et al., 2021). Our finding that the F761H mutation restores EFR^SSAA function further supports this hypothesis. Nevertheless, how EFR A-loop phosphorylation facilitates conformational switching is unclear. In our AlphaFold2 models of the EFR kinase domain, we noted that two basic residues from the β3-αC loop and the αC-helix extend toward the A-loop and, similar to PKA H87 (Meharena et al., 2016), may coordinate A-loop phosphorylation (Figure 2—figure supplement 3).

EFR F761H restores function of EFR^Y836F and EFR^SSAA in Arabidopsis

We next generated stable Arabidopsis complementation lines expressing pEFR::EFR(variant)-GFP::HSP18t constructs in the null efr-1 background. Because the F761H mutation showed greater ability than F761M to rescue function of both EFR^Y836F and EFR^SSAA (Figure 2B and C), we chose this secondary mutation for the generation of complementation lines. Two independent homozygous complementation lines were isolated, with similar accumulation of EFR-GFP protein in the T3 generation. For EFR^F761H/Y836F, only one homozygous T3 line could be isolated, so a heterozygous T2 line with two T-DNA insertion events was used for physiological experiments. Similar to experiments in N. benthamiana, the oxidative burst was partially restored in stable Arabidopsis complementation lines that express EFR^F761H/Y836F or EFR^F761H/SSAA (Figure 3—figure supplement 1A).

The oxidative burst is only one of multiple cellular responses triggered by elf18. We therefore also tested whether other immune signaling responses are restored in the EFR^F761H/Y836F and EFR^F761H/SSAA complementation lines. Consistent with previous results, EFR^Y836F and EFR^SSAA lines were both less sensitive to treatment with 5 nM elf18 than those with wild-type EFR in seedling growth inhibition (SGI) assays (Bender et al., 2021; Macho et al., 2014; Figure 3—figure supplement 1B). As expected, both independent complementation lines of EFR^F761H/Y836F and EFR^F761H/SSAA exhibited enhanced SGI compared to EFR^Y836F and EFR^SSAA, respectively (Figure 3—figure supplement 1B). We also observed that MAPK activation was abolished or severely impaired in both EFR^Y836F and EFR^SSAA complementation lines (Figure 3—figure supplement 1C) and recovered in EFR^F761H/Y836F and EFR^F761H/SSAA complementation lines (Figure 3—figure supplement 1C).

The recovery of multiple immune responses in EFR^F761H/Y836F and EFR^F761H/SSAA complementation lines suggested that they retain fully functional elf18 signaling and that effective resistance against bacteria can be established. To confirm this, we tested the transgenic lines for resistance against Agrobacterium tumefaciens. Infection by this bacterium is restricted in Arabidopsis by EFR, so loss-of-function mutants like efr-1 are more susceptible (Zipfel et al., 2006). A. tumefaciens carrying a plasmid with an intronic version of the β-glucuronidase (GUS) gene was used to evaluate infection success. GUS activity in plant protein extracts following infection correlates with the ability of the bacteria to transiently transform plant cells. Significant GUS activity was detected in efr-1 (Figure 3A), consistent with successful A. tumefaciens infection, but only little was seen in wild-type EFR complementation lines, reflecting resistance to infection (Bender et al., 2021; Zipfel et al., 2006). The EFR^Y836F and EFR^SSAA complementation lines were both more susceptible to A. tumefaciens transformation than wild-type EFR complementation lines, as indicated by elevated GUS activity (Figure 3A), but this was greatly diminished in the EFR^F761H/Y836F and EFR^F761H/SSAA complementation lines. Hence, these experiments show that the EFR F761H mutation restores full signaling function of EFR^Y836F and EFR^SSAA and thus resistance against A. tumefaciens.

Figure 3 with 1 supplement see all

Download asset Open asset

EFR^F761H/Y836F and EFR^F761H/SSAA recover receptor complex activation.

(A) In infection assays, GUS activity was high in the positive control *efr-1* line. GUS activity level was reduced in the EFR^WT and EFR^F761H complementation lines, but much less so in the EFR^Y836F and EFR^SSAA complementation lines. By contrast, EFR^F761H/Y836F and EFR^F761H/SSAA complementation lines displayed substantially repressed GUS activity. Each experiment was repeated three times with similar results. Outliers are indicated by an additional asterisk and included in statistical analysis. Statistical test: Kruskal-Wallis test (P=5.704*10^–7), Dunn’s post-hoc test with Benjamin-Hochberg correction (P ≤ 0.05) Groups with like lowercase letter designations are not statistically different. (B) In IP kinase assays, ligand-induced interaction of EFR^WT and EFR^F761H with BAK1 increased transphosphorylation of BIK1^D202N, but this was abolished for EFR^Y836F and EFR^SSAA. Both EFR^F761H/Y836F and EFR^F761H/SSAA showed partially restored BIK1^D202N trans-phosphorylation as well as BAK1 S612 phosphorylation (across four replicates for EFR^F761H/SSAA and in two out of four replicates for EFR^F761H/Y836F). Samples were also probed for MAPK phosphorylation for effective ligand treatment. Treatment: 100 nM elf18 for 10 min. (C) Quantification of BIK1^D202N band intensity observed in autoradiographs from the four independent replicates performed. Dotted red line indicates unchanged band intensity in mock vs. elf18 treatment.

Figure 3—source data 1 Raw data for autoradiography and immunoblotting in Figure 3B.: https://cdn.elifesciences.org/articles/92110/elife-92110-fig3-data1-v1.zip
Download elife-92110-fig3-data1-v1.zip

EFR^F761H/Y836F and EFR^F761H/SSAA restore BIK1 trans-phosphorylation

As shown Figure 3 and Figure 1—figure supplement 1, EFR^Y836F and EFR^SSAA are impaired in elf18-triggered immune signaling at the level of receptor complex activation. In both cases BAK1 S612 phosphorylation is reduced, resulting in compromised ability to trans-phosphorylate BIK1. We therefore asked next whether EFR^F761H/Y836F and EFR^F761H/SSAA show restored ability to induce BAK1 autophosphorylation and resulting BIK1 trans-phosphorylation. We performed semi-in vivo IP-kinase assays in which EFR-GFP variants were immunoprecipitated from 2-week-old seedlings after 10 min mock or 100 nM elf18 treatment, and kinase activity of the complex was assessed by monitoring phosphorylation of recombinant His-BIK1^D202N substrate. The EFR^WT-GFP-BAK1 complex showed pronounced BIK1 phosphorylation compared with the unliganded control (Figure 3B and C). In contrast, although BAK1 co-precipitated with both EFR^Y836F-GFP and EFR^SSAA-GFP after elf18 treatment, these complexes failed to increase BIK1^D202N trans-phosphorylation compared with the unliganded controls. They also showed reduced levels of BAK1 phosphorylation at S612 (Figure 3B and C). However, BAK1 S612 phosphorylation (for EFR^F761H/Y836F in two out of four experiments) and BIK1^D202N trans-phosphorylation were partially restored for EFR^F761H/Y836F and EFR^F761H/SSAA in this assay (Figure 3B and C), arguing that the functional rescue observed in vivo reflects BAK1 activation effects.

Toward a mechanistic understanding of BAK1 allosteric activation by EFR

A phospho-ablative mutation of the BAK1 VIa-Tyr (Y403F) – analogous to Y836F in EFR – has also been reported to compromise elf18-induced signaling by interfering with receptor complex activation (Perraki et al., 2018). We recapitulated this finding using only the intracellular domains of EFR and BAK1, fusing them to FKBP and FRB domains with myristoylation sequences and inducing dimerization with Rap in N. benthamiana (Figure 4 and Figure 4—figure supplement 1A). This approach allowed us to investigate BAK1 variants without interference of endogenous NbSERKs. We wondered whether BAK1^Y403F could be rescued by a mutation analogous to EFR^F761H/M to recover accessibility of the active conformation. Indeed, introducing the analogous mutation (I338H) into BAK1 partly restored the oxidative burst for BAK1^I338H/Y403F when dimerized with EFR^WT (Figure 4—figure supplement 1A), suggesting that BAK1^Y403F is perturbed, like EFR^Y836F, in accessing its active conformation. Analogies with allosteric regulation of other kinases and pseudokinases (Hu et al., 2013; Mace and Murphy, 2021; Sheetz and Lemmon, 2022; Zhang et al., 2006) led us hypothesize that EFR might stabilize the active-like conformation of BAK1 intermolecularly – helping to position the BAK1 αC-helix (in the EFR-BAK1 complex) to activate immune signaling. If this is correct, EFR that is partly ‘locked’ in its active-like conformation may be able to rescue (in trans) the function of a signaling-inactive BAK1 mutant with a destabilized αC-helix (as in BAK1^Y403F). We therefore asked whether EFR^F761H could achieve this, since it exhibited the tendency for increased BIK1^D202N trans-phosphorylation in IP-kinase assays (Figure 3B and C) and for an accelerated oxidative burst in N. benthamiana (Figure 4—figure supplement 2A and B). Intriguingly, when EFR^F761H was paired with BAK1^Y403F, the Rap-induced oxidative burst was partially restored (Figure 4), suggesting that EFR^F761H can partially restore BAK1^Y403F function. The fact that no oxidative burst was seen when EFR^F761H was paired with catalytically inactive BAK1^D416N suggests that this restoration does not occur through a direct catalytic mechanism mediated by EFR^F761H (Figure 4—figure supplement 2C), supporting the hypothesis of allosteric regulation.

Figure 4 with 3 supplements see all

Download asset Open asset

EFR^F761H recovers BAK1^Y403F function.

The cytoplasmic domains of BAK1 and EFR variants with fused RiD-tags were transiently expressed in *N. benthamiana* and leaf discs were treated with Rap to induce dimerization. EFR and EFR^F761H induced a similar total oxidative burst when BAK1 was co-expressed. The co-expression of BAK1^Y403F and EFR diminished the oxidative burst, which was restored partially when EFR^F761H was co-expressed. Outliers are indicated by an additional asterisk and included in statistical analysis. Statistical test: Kruskal-Wallis test (p<8.516 *10^–7), Dunn’s post-hoc test with Benjamin-Hochberg correction (p ≤ 0.05) Groups with like letter designations are not statistically different.

Catalytic independence is only observed for Arabidopsis sp. LRR-RK XIIa kinase domains

EFR belongs to LRR-RK subfamily XIIa (Figure 5A), of which two other RKs are functionally described immune RKs in Arabidopsis: FLS2 and XPS1 (XANTHINE/URACIL PERMEASE SENSING 1) (Gómez-Gómez and Boller, 2000; Mott et al., 2016). This subfamily has been implicated more generally in immune signaling, which is supported by induction of immune signaling of in vivo dimerized FLS2, EFR, or XPS1-LIKE 1 (FEXL1) intracellular domains with BAK1 (Kim et al., 2021). However, the ligands are not known for most subfamily-XIIa RKs (Dufayard et al., 2017; Kim et al., 2021). Because EFR does not require its catalytic activity, we wondered whether other LRR-RK XIIa kinase domains similarly function non-catalytically, in a manner similar to pseudokinases. To test this hypothesis, we fused the EFR ectodomain to the transmembrane helix and intracellular domain of different LRR-RK XIIa members to generate elf18 responsive RKs that can dimerize with BAK1/SERKs (Rhodes et al., 2021). These chimeric proteins were expressed transiently in N. benthamiana leaves, and their immune signaling function was tested by elf18 treatment in oxidative burst assays. All chimeras induced an oxidative burst (Figure 5B), except XIIa2 (the closest FLS2-related kinase), which exhibited only a very minor response. To determine whether their signaling function required kinase activity, we next tested variants with mutations in the catalytic site (replacing the putative catalytic base HRD-aspartate with asparagine). EFR^D849N induced a robust oxidative burst as expected as well as FEXL1^D838N and the closely related XIIa5^D839N (AT3G47570). Similarly, XPS1^D856N and XIIa6^D840N (AT3G47090) induced an oxidative burst independently of their catalytic activity, but the total oxidative burst was reduced compared to the catalytically active variants. Protein levels of catalytic site mutants accumulated comparably to their corresponding wild-type versions (Figure 5—figure supplement 2). In contrast, the catalytically inactive FLS2^D997N showed a diminished oxidative burst (Figure 5B), suggesting that it may be mechanistically distinct. Furthermore, kinetic differences for the oxidative burst of Arabidopsis XIIa kinase domains were observed. XPS1^D856N and XIIa6^D840N showed a delayed oxidative burst (Figure 5—figure supplement 1A and B). Also, EFR^D849N exhibited a delay of approximately 5–7 min compared to wild-type EFR. In contrast, oxidative bursts induced by FEXL1^D838N and XIIa5^D839N were only slightly delayed compared to their corresponding catalytically active kinase domains. We corroborated these findings for XIIa5 by performing in vitro kinase assays, observing that BIK1^D202N and BAK1 phosphorylation increased after Rap application as effectively for XIIa5^D839N as for wild-type XIIa5 (Figure 5—figure supplement 1C and D). Taken together, our results imply that catalytic activity for XIIa5 is almost fully dispensable.

Figure 5 with 2 supplements see all

Download asset Open asset

Related EFR kinases from LRR-RK XIIa in the *Arabidopsis* genus can function independent of their calatytic activity.

(A) Phylogenetic analysis of LRR-RK subfamily XIIa. Selected LRR-RK XIIa kinase domains are labeled and highlighted with purple points. The EFR-like clade contains all *Arabidopsis* XIIa kinases except FLS2 and XIIa2 and also selected XIIa kinases from *Arabidopsis lyrata* and *Brassica rapa*. (**B, C**) The ectodomain of EFR was fused to the transmembrane and intracellular domain of selected LRR-RK XIIa members to create elf18-responsive chimeras for testing the immune signaling function and catalytic dependency of the related kinase domains. The chimeras were transiently expressed in *N. benthamiana* and tested in oxidative burst assays. All *Arabidopsis* LRR-RK XIIa members induced an oxidative burst except XIIa2, the closest FLS2 related kinase in the subfamily. Catalytic dependency of the kinase domains appears to vary from kinase to kinase, with catalytically dead versions of EFR, FEXL1 and XIIa5 inducing a WT-like oxidative burst and XPS1 and XIIa6 displaying a reduced oxidative burst. FLS2 kinase dead exhibited a diminished oxidative burst. Experiments were repeated three times with similar results.

Because most Arabidopsis LRR-RK XIIa kinase domains execute their signaling function at least to some extent independently of their catalytic activity, we went on to test XIIa kinase domains from other species, that we selected from a phylogenetic tree of LRR-RK XIIa kinases built from previous phylogenetic analysis (Figure 5A, Dufayard et al., 2017). The selected catalytically active XIIa kinase domains from Oryza sativa (XA21), Populus trichocarpa (PtXIIa), Solanum lycopersicum (SlXIIa), Glycine max (GmXIIa), Brassica rapa (BrEFR), and Arabidopsis lyrata (AlEFR), induced an oxidative burst (Figure 5C; Figure 5—figure supplement 1A and B). However, catalytic site mutants of all non-Arabidopsis XIIa RKs accumulated similar to their wild-type counterparts and abolished their function in oxidative burst assays (Figure 5, Figure 5—figure supplement 2).

Discussion

Our study provides insights into the non-catalytic activation mechanisms of the EFR-BAK1 complex. Rather than requiring its kinase activity to transactivate BAK1, our results suggest that the EFR kinase domain undergoes a ‘conformational toggle’ to an active-like state that permits allosteric trans-activation of BAK1. EFR function is impaired by perturbation of structural elements that are important for conformational switching – specifically, mutation of the subdomain VIa Y836 or ablation of A-loop phosphorylation sites – but these effects can be reversed by intragenic suppressor mutations that stabilize the active-like conformation, notably the F761H mutation.

Non-catalytic functions of kinases and pseudokinases are human disease-relevant and have emerged as important signaling regulators in metazoans (Eyers and Murphy, 2013; Mace and Murphy, 2021; Schmidt et al., 2021; Sheetz and Lemmon, 2022). By comparison, very little is known about pseudokinases or non-catalytic kinase functions in plants, although pseudokinases are as prevalent in plant as in metazoan kinomes (Kwon et al., 2019). Our data demonstrate that EFR and potentially other Arabidopsis XIIa kinases function as allosteric regulators of BAK1 kinase activity, implying pseudokinase-like functions. Thus, our findings establish precedence for non-catalytic mechanisms, specifically allosteric regulation, in plant RK signaling.

Phosphorylation of EFR is crucial for signaling function and potentially supports the conformational toggle. Nevertheless, our results do not resolve how phosphorylation of the A-loop or VIa-Tyr facilitate conformational toggling for EFR, and structural studies will be required to address this question. However, our HDX-MS analysis of unphosphorylated protein suggests that EFR^Y836F causes the active-like conformation to become inaccessible, potentially due to a lack of an inter-lobe H-bond – consistent with earlier studies describing the corresponding VIa-Tyr as a pivot for αC-helix movements in PKA (Taylor and Kornev, 2011; Tsigelny et al., 1999). Recent analysis of the PKA hinge region demonstrates the importance of its integrity for coordinated conformational changes in the N- and C-lobe (Olivieri et al., 2023; Wu et al., 2023). Whether EFR VIa-Tyr phosphorylation contributes to these coordinated conformational changes is unclear. We attempted to directly address the role of pY836 in EFR conformational dynamics using HDX-MS but were unable to produce sufficient recombinant protein with a pTyr-analog incorporated at the Y836 site. However, phosphorylation of the VIa-Tyr presumably distorts the hinge region due to its proximity to the αC-β4-loop. Alternatively, the VIa-pTyr and C-terminal end of the αE-helix must rotate outwards (Figure 2—figure supplement 1). Rotation of the αE-helix, however, would have direct impact on the positioning of the catalytic loop and RS1. Of note, functional importance of VIa-Tyr phosphorylation was assigned based on correlation between its phosphorylation and functional impairment by phenylalanine substitution in multiple plant RKs (Liu et al., 2018; Luo et al., 2020; Macho et al., 2014; Perraki et al., 2018), making it possible that functional importance of its phosphorylation confounded a structural role. Future computational protein-structure prediction techniques that consider post-translational modifications will aid in elucidating the effect of EFR VIa-Tyr phosphorylation on EFR’s conformation. Nevertheless, mutation of VIa-Tyr may be a useful tool to break pseudokinase/non-catalytic functions by impeding conformational toggling in a way that is more effective than using conventional kinase-dead mutations.

Basal BAK1 activity in vitro is low compared to its robustly enhanced activity after Rap-induced dimerization (Figure 1B and C; Figure 5—figure supplement 1C and D), which might be further reduced in planta by negative regulators (e.g. protein phosphatases; Segonzac et al., 2014). We propose that, by increasing the local concentration, ligand-induced association of EFR and BAK1 allows partially active BAK1 to phosphorylate the adjacent EFR A-loop. This in turn allows wild-type EFR to adopt the active-like conformation that can allosterically fully activate BAK1 to promote substrate phosphorylation. The activating mutation EFR F761H potentially circumvents the requirement of A-loop phosphorylation since it partially restored EFR^SSAA function. Consistently, EFR^F761H showed elevated BIK1 trans-phosphorylation in IP-kinase assays and an accelerated oxidative burst (Figure 3B and C; Figure 4—figure supplement 2A, B), suggesting constitutive assembly of the active-like conformation. However, EFR^F761H still requires catalytic activity of complexed BAK1 (Figure 4—figure supplement 1).

Our data demonstrate that allosteric BAK1 kinase activation plays a key role in EFR-BAK1-mediated immune signaling. The partial recovery of BAK1^Y403F, which we hypothesize is impaired in αC-helix positioning, by EFR^F761H suggests a mechanism for allosteric regulation involving BAK1 αC-helix positioning. Indeed, multiple metazoan kinases are allosterically regulated by αC-helix positioning (Sheetz and Lemmon, 2022). Moreover, BAK1 has a high propensity for a disordered αC-helix (Moffett et al., 2017), suggesting its correct positioning requires additional support. Different orientations of kinase-kinase dimers, in which allosteric regulation occurs at the αC-helix, were previously described (Mace and Murphy, 2021; Sheetz and Lemmon, 2022). Structural analysis will be required to resolve the exact interaction interface for allosteric activation in the EFR-BAK1 kinase dimer and to design interface-disrupting mutations for structure-function analysis but is beyond the scope of the present work.

Allosteric activation of BAK1/SERKs is partially preserved in Arabidopsis LRR-RK XIIa kinases (Figure 5). Whether these kinases function solely non-catalytically, as suggested by XIIa5^D839N (Figure 5, Figure 5—figure supplement 1), or contribute also catalytically to signaling activation is unclear since oxidative bursts induced by all functional catalytic base mutants of XIIa kinases were delayed (Figure 5—figure supplement 1A and B). Catalytic base mutation may affect conformational dynamics, which we show are important for EFR function (Figure 2), confounding our interpretation. Inactive XIIa kinases with intact conformational dynamics may clarify the extent of catalytic contribution and could be obtained by development of selective inhibitors or engineering analog sensitive kinases. Alternatively, catalytic independency could be a recent innovation in Arabidopsis sp. XIIa kinases, that is taxonomically restricted due to a strong negative selection on the kinase domain (Man et al., 2023) but further study is required to test this hypothesis.

Taken together, our results add to growing evidence that non-catalytic functions of kinases are similar to bona fide kinases controlled by conformational switching (Sheetz et al., 2020; Sheetz and Lemmon, 2022) and further set precedence for discovering more non-catalytic mechanisms in plant RK signaling where pseudokinases are particularly prevalent (Kwon et al., 2019).

Limitations of the study

In our in vitro kinase assays using the RiD system, the proteins were freely diffusing as they were produced as soluble entities. Freely diffusing components, however, do not fully resemble the in planta mode of activation where the plasma membrane restricts the free movement. The colocalization of RK/coRK complexes with their substrates may be a supporting component of signaling activation, which we did not capture in our in vitro assays.

It is yet unclear whether EFR phosphorylates BAK1 efficiently on activation-relevant residues in the context of stabilized proximity. This information will ultimately help to deduce whether the catalytic activity of EFR contributes to signaling activation and whether even marginal catalytic activity of EFR^D849N may be sufficient to support signaling activation.

Further, our study relies heavily on mutations. Mutations can have pleiotropic effects as they impact conformational plasticity which supports catalytic activity and non-catalytic functions. Thus, it is difficult to untangle catalytic from non-catalytic functions using solely mutations. Eventually, it would be beneficial to obtain structural information of native proteins in a complex and monitor site-specific and time-resolved phosphorylation on complex components.

We do not have experimental support for EFR^F761H stabilizing or facilitating the transition into an active-like conformation and relieving the negative effect of the Y836F mutation in EFR^F761H/Y836F. It would have been beneficial to obtain HDX-MS data for these mutants as well, but we could not produce sufficient protein to do so. Thus, we rather derived the biochemical and structural effects of the EFR^F761H mutant by homology to mammalian kinases, for example BRAF.

Protein	T after induction [°C]	t of expression [h]	pH of extraction buffer
For in vitro kinase assay
EFR/EFR D849N	30	4	8.0
BAK1	30	4	8.0
BRI1/BRI1 D1009N	18	Over night	8.0
FLS2/FLS2 D997N	18	Over night	8.0
BIK1 D202N	18	Over night	7.5
For HDX-MS
EFR/EFR Y836F	30	3	8.0

Data Set	Wild-type EFR	EFR^Y836F
HDX reaction details	20 mM HEPES, 100 mM NaCl, pD 7.4	20 mM HEPES, 100 mM NaCl, pD 7.4
HDX time course	0, 10, 60, 600, 3600, and 7200 s	0, 10, 60, 600, 3600, and 7200 s
HDX control samples	8 M urea-D4 for fully deuterated standard	8 M urea-D4 for fully deuterated standard
Back-exchange	46%	49%
# of Peptides	143	153
Sequence coverage	100%	100%
Ave peptide length / Redundancy	12.7/5.44	13.3/6.06
Replicates	n=3 biological repeats (3 technical/n)	n=3 biological repeats (3 technical/n)
Repeatability (Ave SD)	0.129 Da (2.28 %)	0.111 Da (2.09 %)
Significant differences in HDX (98% CL)	0.395 Da (7.18 %)

Protein	Extinction coefficient (reduced state)[Abs 0.1% (=1 g/l)]	Molecular weight[kDa]
FKBP-EFR-mEGFP	7.40	80
FKBP-BRI1-mEGFP	8.26	83
FKBP-FLS2-mEGFP	6.56	79
FKBP-XIIa5-mEGFP	7.59	79.5
FRB-BAK1	15.07	55
6xHis-TEV-BIK1^D202N	9.45	47
6xHis-TEV-EFR(684–1031)	6.97	41.5

Reagent type (species) or resource	Designation	Source or reference	Identifiers	Additional information
Strain, strain background (Arabidopsis thaliana)	Col-0	ther		wild-type ecotype
Strain, strain background (Nicotiana benthamiana)	WT	Lab stock		Wild-type strain
strain, strain background (Escherichia coli)	BL21(DE3) pLPP	Amid Biosciences	Cat#: BLPP-201	Co-expression of lambda phosphatase
Strain, strain background (Escherichia coli)	BL21(DE3)-V2R-pACYC LamP	PMID: 20436473		Co-expression of lambda phosphatase
Strain, strain background (Agrobacterium tumefaciens)	GV3101	Lab stock
Strain, strain background (Arabidopsis thaliana)	efr-1	PMID: 16713565
Genetic reagent (Arabidopsis thaliana)	EFR-GFP	This paper		See Materials and methods
Genetic reagent (Arabidopsis thaliana)	F761H#2	This paper		See Materials and methods
Genetic reagent (Arabidopsis thaliana)	F761H#7	This paper		See Materials and methods
Genetic reagent (Arabidopsis thaliana)	Y836F#1	This paper		See Materials and methods
Genetic reagent (Arabidopsis thaliana)	Y836F#4	This paper		See Materials and methods
Genetic reagent (Arabidopsis thaliana)	F761H/Y836F#1	This paper		See Materials and methods
Genetic reagent (Arabidopsis thaliana)	F761H/Y836F#5	This paper		See Materials and methods
Genetic reagent (Arabidopsis thaliana)	SSAA#2	This paper		See Materials and methods
Genetic reagent (Arabidopsis thaliana)	SSAA#7	This paper		See Materials and methods
Genetic reagent (Arabidopsis thaliana)	F761H/SSAA#2	This paper		See Materials and methods
Genetic reagent (Arabidopsis thaliana)	F761H/SSAA#7	This paper		See Materials and methods
Sequence-based reagent	EFR-f	This paper	PCR primer	atcgGAAGACaaAATGA AGCTGTCCTTTTCACTTG
Sequence-based reagent	EFR Y836F-f	This paper	PCR primer	ggagtTtctgcacgtt cattgtcatgaccctgta
Sequence-based reagent	EFR SSAA-f	This paper	PCR primer	Gtttgctgctgctggt gtcagaggcaccat
Sequence-based reagent	EFR F761H-f	This paper	PCR primer	acgatgtcgtatacccttgt gggtttcacattccgccata
Sequence-based reagent	EFR F761M-f	This paper	PCR primer	tacgatgtcgtatacccttc atggtttcacattccgccata
Sequence-based reagent	EFR V845F-f	This paper	PCR primer	gcttaatatcacagtgagc gaaagggtcatgacaatgaacgtg
Sequence-based reagent	EFR L743F-f	This paper	PCR primer	gctttaggaggttgaaaac tttaaccgcgacgagttta
Sequence-based reagent	EFR L873E-f	This paper	PCR primer	ctcagGAGctctataaata cgatcgagaatcctttcta
Sequence-based reagent	EFR dNLLKH-f	This paper	PCR primer	taaagttttgggagcgacga aaagctttatggcggaatgtg
Sequence-based reagent	EFR D849N-f	This paper	PCR primer	cactgtAatattaagcca agcaacattcttctagacgat
Sequence-based reagent	BAK1-f	This paper	PCR primer	atcgGAAGACaaAATGGA ACGAAGATTAATGATCC
Sequence-based reagent	BAK1 Y403F-f	This paper	PCR primer	cttgcgtttttacatgat cattgcgacccaaaga
Sequence-based reagent	BAK1 I338H-f	This paper	PCR primer	gagatgCAtagtatggcg gttcacagaaacttgct
Sequence-based reagent	BAK1 D416N-f	This paper	PCR primer	catcgaAatgtgaaagctg caaatattttgttggatgaag
Sequence-based reagent	EFRprom-f	This paper	PCR primer	atcgGgtctctTACGCGTCTC aGTAGatctagacgat taagtaattgagca
Sequence-based reagent	BAK1prom-f	This paper	PCR primer	atcgGgtctctTACGCGTCTCa CCTAtgtcgtgaaaagggcac
Sequence-based reagent	XIIa2_Tmicd-f	This paper	PCR primer	atcgGCTCTTCgAAGGTT CTTCTACCGGTTCTGTTAT
Sequence-based reagent	XIIa3_Tmicd-f	This paper	PCR primer	atcgGCTCTTCgAAGG TTGCGATTGGGGTCA
Sequence-based reagent	XIIa4_Tmicd-f	This paper	PCR primer	atcgGCTCTTCgAAGATAA TCACCATTTGTGTCAGTG
Sequence-based reagent	XIIa5_Tmicd-f	This paper	PCR primer	atcgGCTCTTCgAAGGTT GTGATTGGAGTTAGCGT
Sequence-based reagent	XIIa6_Tmicd-f	This paper	PCR primer	atcgGCTCTTCgAAGGTTG CAATTTTAGTAAGCATAGG
Sequence-based reagent	FLS2_Tmicd-f	This paper	PCR primer	atcgGCTCTTCgAAGGTCA TCCTGATTATTCTTGGATCA GCCGCGGCaCTTCTTCTTGTCCTG
Sequence-based reagent	AlXIIa_D850N-f	This paper	PCR primer	atcgGCTCTTCAGTa ATATTAAGCCAAGCAACG

Share this article

Cite this article

EFR facilitates BIK1 trans-phosphorylation by BAK1 non-catalytically.

Figure 1—source data 1

Figure 1—source data 2

EFRY836F and EFRSSAA impair the active kinase conformation, which is required for signaling function.

Homology-based design of putative intragenic suppressor mutations for EFR.

EFRF761H/Y836F and EFRF761H/SSAA recover receptor complex activation.

Figure 3—source data 1

EFRF761H recovers BAK1Y403F function.

Related EFR kinases from LRR-RK XIIa in the Arabidopsis genus can function independent of their calatytic activity.

Protein expression conditions.

Summary table of HDX-MS analysis.

Extinction coefficients and molecular weights retrieved from ProtParam and used for determination of protein concentration.

Author details

Henning Mühlenbeck

Contribution

Competing interests

Yuko Tsutsui

Contribution

Competing interests

Mark A Lemmon

Contribution

Competing interests

Kyle W Bender

Contribution

For correspondence

Competing interests

Cyril Zipfel

Contribution

For correspondence

Competing interests

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

Categories and tags

Research organism

Further reading

EFR^Y836F and EFR^SSAA impair the active kinase conformation, which is required for signaling function.

EFR^F761H/Y836F and EFR^F761H/SSAA recover receptor complex activation.

EFR^F761H recovers BAK1^Y403F function.